shaba/systemd
1
0
Fork 0
mirror of https://github.com/systemd/systemd.git synced 2024-11-07 01:27:11 +03:00
Code
c1227a1840
systemd/meson.build

2738 lines
104 KiB
Meson
Raw Normal View History

Add license headers and SPDX identifiers to meson.build files So far I avoided adding license headers to meson files, but they are pretty big and important and should carry license headers like everything else. I added my own copyright, even though other people modified those files too. But this is mostly symbolic, so I hope that's OK.
2017-11-18 20:32:01 +03:00
# SPDX-License-Identifier: LGPL-2.1+
#
# Copyright 2017 Zbigniew Jędrzejewski-Szmek
#
# systemd is free software; you can redistribute it and/or modify it
# under the terms of the GNU Lesser General Public License as published by
# the Free Software Foundation; either version 2.1 of the License, or
# (at your option) any later version.
#
# systemd is distributed in the hope that it will be useful, but
# WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
# Lesser General Public License for more details.
#
# You should have received a copy of the GNU Lesser General Public License
# along with systemd; If not, see <http://www.gnu.org/licenses/>.
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
project('systemd', 'c',
meson: increase version numbers
2017-12-11 18:10:25 +03:00
version : '236',
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
license : 'LGPLv2+',
default_options: [
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
'c_std=gnu99',
'prefix=/usr',
'sysconfdir=/etc',
'localstatedir=/var',
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
],
meson: bump required meson version to 0.41 (#7385) We need data_dirs argument for i18n.merge_file, added in 0.40.1-11-g82492f5d76.
2017-11-20 10:08:43 +03:00
meson_version : '>= 0.41',
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
)
meson: increase version numbers
2017-12-11 18:10:25 +03:00
libsystemd_version = '0.20.0'
libudev_version = '1.6.8'
meson: move library version defines to the top (#6939)
2017-09-28 20:24:16 +03:00
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
# We need the same data in three different formats, ugh!
# Also, for hysterical reasons, we use different variable
# names, sometimes. Not all variables are included in every
# set. Ugh, ugh, ugh!
conf = configuration_data()
conf.set_quoted('PACKAGE_STRING', meson.project_name() + ' ' + meson.project_version())
conf.set_quoted('PACKAGE_VERSION', meson.project_version())
substs = configuration_data()
substs.set('PACKAGE_URL', 'https://www.freedesktop.org/wiki/Software/systemd')
substs.set('PACKAGE_VERSION', meson.project_version())
m4_defines = []
#####################################################################
meson: install the git hook (#6425) This was done autogen.sh previously and was dropped in 72cdb3e783174dcf9223a49f03e3b0e2ca95ddb8. Let's add it back. The meson configuration step is the only reasonable place. Note that this only works for the most standard git dirs, e.g. the hook will not be installed if git worktree is used or if $GIT_DIR is specified, etc. I think that's OK because most of the time meson will be run at least once in the original cloned dir.
2017-07-24 11:41:45 +03:00
# Try to install the git pre-commit hook
git_hook = run_command(join_paths(meson.source_root(), 'tools/add-git-hook.sh'))
if git_hook.returncode() == 0
message(git_hook.stdout().strip())
endif
#####################################################################
meson: emit a warning if rootprefix is set It's printed at the end because it's the easiest to spot there. When meson is upgraded to 0.43.1, we'll be able to use warning() instead.
2017-11-28 23:46:53 +03:00
split_usr = get_option('split-usr')
conf.set10('HAVE_SPLIT_USR', split_usr)
Revert "meson: drop rootprefix option" This reverts commit ee4bfa21f62dbf4d8f2be27d3c763c3c16743bfd. Also add comment so that this doesn't get deleted again by mistake. Replaces #7461.
2017-11-28 22:00:10 +03:00
rootprefixdir = get_option('rootprefix')
# Unusual rootprefixdir values are used by some distros
# (see https://github.com/systemd/systemd/pull/7461).
meson: emit a warning if rootprefix is set It's printed at the end because it's the easiest to spot there. When meson is upgraded to 0.43.1, we'll be able to use warning() instead.
2017-11-28 23:46:53 +03:00
rootprefix_default = get_option('split-usr') ? '/' : '/usr'
if rootprefixdir == ''
rootprefixdir = rootprefix_default
Revert "meson: drop rootprefix option" This reverts commit ee4bfa21f62dbf4d8f2be27d3c763c3c16743bfd. Also add comment so that this doesn't get deleted again by mistake. Replaces #7461.
2017-11-28 22:00:10 +03:00
endif
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
sysvinit_path = get_option('sysvinit-path')
sysvrcnd_path = get_option('sysvrcnd-path')
Fix logical error in meson.build (#7658) sysvinit_path and sysvrcnd_path have to be set both to activate Sysv compatibility.
2017-12-15 18:05:25 +03:00
have = sysvinit_path != '' and sysvrcnd_path != ''
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_SYSV_COMPAT', have,
description : 'SysV init scripts and rcN.d links are supported')
m4_defines += have ? ['-DHAVE_SYSV_COMPAT'] : []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
# join_paths ignore the preceding arguments if an absolute component is
# encountered, so this should canonicalize various paths when they are
# absolute or relative.
prefixdir = get_option('prefix')
if not prefixdir.startswith('/')
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
error('Prefix is not absolute: "@0@"'.format(prefixdir))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
bindir = join_paths(prefixdir, get_option('bindir'))
libdir = join_paths(prefixdir, get_option('libdir'))
sysconfdir = join_paths(prefixdir, get_option('sysconfdir'))
includedir = join_paths(prefixdir, get_option('includedir'))
datadir = join_paths(prefixdir, get_option('datadir'))
localstatedir = join_paths('/', get_option('localstatedir'))
rootbindir = join_paths(rootprefixdir, 'bin')
rootlibexecdir = join_paths(rootprefixdir, 'lib/systemd')
rootlibdir = get_option('rootlibdir')
if rootlibdir == ''
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
rootlibdir = join_paths(rootprefixdir, libdir.split('/')[-1])
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
# Dirs of external packages
meson: use join_paths consistently With -Dsplit-usr=true, we set rootprefix to /. This leads to //lib/systemd or //lib/udev for various dir variables. Using join_paths() avoids this.
2017-04-13 17:30:56 +03:00
pkgconfigdatadir = join_paths(datadir, 'pkgconfig')
pkgconfiglibdir = join_paths(libdir, 'pkgconfig')
polkitpolicydir = join_paths(datadir, 'polkit-1/actions')
polkitrulesdir = join_paths(datadir, 'polkit-1/rules.d')
polkitpkladir = join_paths(localstatedir, 'lib/polkit-1/localauthority/10-vendor.d')
varlogdir = join_paths(localstatedir, 'log')
xinitrcdir = join_paths(sysconfdir, 'X11/xinit/xinitrc.d')
meson: do not install RPM macros if rpmmacrosdir is set to no (#7430) Fixes #7427.
2017-11-23 16:20:22 +03:00
rpmmacrosdir = get_option('rpmmacrosdir')
if rpmmacrosdir != 'no'
rpmmacrosdir = join_paths(prefixdir, rpmmacrosdir)
endif
modprobe.d: fix directory of modprobe configuration file (#7149) kmod upstream uses /lib/modprobe.d which means we need to use rootprefix instead of prefix for installing the modprobe configuration file as otherwise split-usr systems are broken. See https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=879191
2017-10-21 09:32:50 +03:00
modprobedir = join_paths(rootprefixdir, 'lib/modprobe.d')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
# Our own paths
meson: use join_paths consistently With -Dsplit-usr=true, we set rootprefix to /. This leads to //lib/systemd or //lib/udev for various dir variables. Using join_paths() avoids this.
2017-04-13 17:30:56 +03:00
pkgdatadir = join_paths(datadir, 'systemd')
environmentdir = join_paths(prefixdir, 'lib/environment.d')
pkgsysconfdir = join_paths(sysconfdir, 'systemd')
userunitdir = join_paths(prefixdir, 'lib/systemd/user')
userpresetdir = join_paths(prefixdir, 'lib/systemd/user-preset')
tmpfilesdir = join_paths(prefixdir, 'lib/tmpfiles.d')
sysusersdir = join_paths(prefixdir, 'lib/sysusers.d')
sysctldir = join_paths(prefixdir, 'lib/sysctl.d')
binfmtdir = join_paths(prefixdir, 'lib/binfmt.d')
modulesloaddir = join_paths(prefixdir, 'lib/modules-load.d')
networkdir = join_paths(rootprefixdir, 'lib/systemd/network')
pkgincludedir = join_paths(includedir, 'systemd')
systemgeneratordir = join_paths(rootlibexecdir, 'system-generators')
usergeneratordir = join_paths(prefixdir, 'lib/systemd/user-generators')
systemenvgeneratordir = join_paths(prefixdir, 'lib/systemd/system-environment-generators')
userenvgeneratordir = join_paths(prefixdir, 'lib/systemd/user-environment-generators')
systemshutdowndir = join_paths(rootlibexecdir, 'system-shutdown')
systemsleepdir = join_paths(rootlibexecdir, 'system-sleep')
systemunitdir = join_paths(rootprefixdir, 'lib/systemd/system')
systempresetdir = join_paths(rootprefixdir, 'lib/systemd/system-preset')
udevlibexecdir = join_paths(rootprefixdir, 'lib/udev')
udevhomedir = udevlibexecdir
udevrulesdir = join_paths(udevlibexecdir, 'rules.d')
udevhwdbdir = join_paths(udevlibexecdir, 'hwdb.d')
catalogdir = join_paths(prefixdir, 'lib/systemd/catalog')
kernelinstalldir = join_paths(prefixdir, 'lib/kernel/install.d')
factorydir = join_paths(datadir, 'factory')
docdir = join_paths(datadir, 'doc/systemd')
bootlibdir = join_paths(prefixdir, 'lib/systemd/boot/efi')
testsdir = join_paths(prefixdir, 'lib/systemd/tests')
systemdstatedir = join_paths(localstatedir, 'lib/systemd')
catalogstatedir = join_paths(systemdstatedir, 'catalog')
randomseeddir = join_paths(localstatedir, 'lib/systemd')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
dbuspolicydir = get_option('dbuspolicydir')
if dbuspolicydir == ''
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
dbuspolicydir = join_paths(datadir, 'dbus-1/system.d')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
dbussessionservicedir = get_option('dbussessionservicedir')
if dbussessionservicedir == ''
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
dbussessionservicedir = join_paths(datadir, 'dbus-1/services')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
dbussystemservicedir = get_option('dbussystemservicedir')
if dbussystemservicedir == ''
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
dbussystemservicedir = join_paths(datadir, 'dbus-1/system-services')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
pamlibdir = get_option('pamlibdir')
if pamlibdir == ''
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
pamlibdir = join_paths(rootlibdir, 'security')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
pamconfdir = get_option('pamconfdir')
if pamconfdir == ''
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
pamconfdir = join_paths(sysconfdir, 'pam.d')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
conf.set_quoted('PKGSYSCONFDIR', pkgsysconfdir)
meson: use join_paths consistently With -Dsplit-usr=true, we set rootprefix to /. This leads to //lib/systemd or //lib/udev for various dir variables. Using join_paths() avoids this.
2017-04-13 17:30:56 +03:00
conf.set_quoted('SYSTEM_CONFIG_UNIT_PATH', join_paths(pkgsysconfdir, 'system'))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
conf.set_quoted('SYSTEM_DATA_UNIT_PATH', systemunitdir)
conf.set_quoted('SYSTEM_SYSVINIT_PATH', sysvinit_path)
conf.set_quoted('SYSTEM_SYSVRCND_PATH', sysvrcnd_path)
meson: fix a bunch of substitutions This makes the diff -r output on autotools and meson installations empty, except for binary files and ## comments.
2017-04-13 02:54:33 +03:00
conf.set_quoted('RC_LOCAL_SCRIPT_PATH_START', get_option('rc-local'))
conf.set_quoted('RC_LOCAL_SCRIPT_PATH_STOP', get_option('halt-local'))
meson: use join_paths consistently With -Dsplit-usr=true, we set rootprefix to /. This leads to //lib/systemd or //lib/udev for various dir variables. Using join_paths() avoids this.
2017-04-13 17:30:56 +03:00
conf.set_quoted('USER_CONFIG_UNIT_PATH', join_paths(pkgsysconfdir, 'user'))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
conf.set_quoted('USER_DATA_UNIT_PATH', userunitdir)
conf.set_quoted('CERTIFICATE_ROOT', get_option('certificate-root'))
meson: use join_paths consistently With -Dsplit-usr=true, we set rootprefix to /. This leads to //lib/systemd or //lib/udev for various dir variables. Using join_paths() avoids this.
2017-04-13 17:30:56 +03:00
conf.set_quoted('CATALOG_DATABASE', join_paths(catalogstatedir, 'database'))
conf.set_quoted('SYSTEMD_CGROUP_AGENT_PATH', join_paths(rootlibexecdir, 'systemd-cgroups-agent'))
conf.set_quoted('SYSTEMD_BINARY_PATH', join_paths(rootlibexecdir, 'systemd'))
conf.set_quoted('SYSTEMD_FSCK_PATH', join_paths(rootlibexecdir, 'systemd-fsck'))
Add x-systemd.makefs option for fstab I opted to completely generate a unit for both mount points and swaps. For swaps, it would be possible to use fixed template unit like systemd-mkswap@.service, because there's no information passed except the device name. For mount points, that's not possible because both the device name and file system type need to be passed. Nevertheless, I expect that options will need to passed to both mkfs and mkswap, in which case it'll be necessary to create units of both types anyway.
2017-11-22 01:18:05 +03:00
conf.set_quoted('SYSTEMD_MAKEFS_PATH', join_paths(rootlibexecdir, 'systemd-makefs'))
Add x-systemd.growfs option for fstab
2017-11-29 22:02:11 +03:00
conf.set_quoted('SYSTEMD_GROWFS_PATH', join_paths(rootlibexecdir, 'systemd-growfs'))
meson: use join_paths consistently With -Dsplit-usr=true, we set rootprefix to /. This leads to //lib/systemd or //lib/udev for various dir variables. Using join_paths() avoids this.
2017-04-13 17:30:56 +03:00
conf.set_quoted('SYSTEMD_SHUTDOWN_BINARY_PATH', join_paths(rootlibexecdir, 'systemd-shutdown'))
conf.set_quoted('SYSTEMD_SLEEP_BINARY_PATH', join_paths(rootlibexecdir, 'systemd-sleep'))
conf.set_quoted('SYSTEMCTL_BINARY_PATH', join_paths(rootbindir, 'systemctl'))
conf.set_quoted('SYSTEMD_TTY_ASK_PASSWORD_AGENT_BINARY_PATH', join_paths(rootbindir, 'systemd-tty-ask-password-agent'))
conf.set_quoted('SYSTEMD_STDIO_BRIDGE_BINARY_PATH', join_paths(bindir, 'systemd-stdio-bridge'))
Revert "meson: drop rootprefix option" This reverts commit ee4bfa21f62dbf4d8f2be27d3c763c3c16743bfd. Also add comment so that this doesn't get deleted again by mistake. Replaces #7461.
2017-11-28 22:00:10 +03:00
conf.set_quoted('ROOTPREFIX', rootprefixdir)
meson: add a few forgotten substitutions v2: - a few more
2017-04-11 02:06:45 +03:00
conf.set_quoted('RANDOM_SEED_DIR', randomseeddir)
meson: use join_paths consistently With -Dsplit-usr=true, we set rootprefix to /. This leads to //lib/systemd or //lib/udev for various dir variables. Using join_paths() avoids this.
2017-04-13 17:30:56 +03:00
conf.set_quoted('RANDOM_SEED', join_paths(randomseeddir, 'random-seed'))
conf.set_quoted('SYSTEMD_CRYPTSETUP_PATH', join_paths(rootlibexecdir, 'systemd-cryptsetup'))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
conf.set_quoted('SYSTEM_GENERATOR_PATH', systemgeneratordir)
conf.set_quoted('USER_GENERATOR_PATH', usergeneratordir)
conf.set_quoted('SYSTEM_ENV_GENERATOR_PATH', systemenvgeneratordir)
conf.set_quoted('USER_ENV_GENERATOR_PATH', userenvgeneratordir)
conf.set_quoted('SYSTEM_SHUTDOWN_PATH', systemshutdowndir)
conf.set_quoted('SYSTEM_SLEEP_PATH', systemsleepdir)
meson: use join_paths consistently With -Dsplit-usr=true, we set rootprefix to /. This leads to //lib/systemd or //lib/udev for various dir variables. Using join_paths() avoids this.
2017-04-13 17:30:56 +03:00
conf.set_quoted('SYSTEMD_KBD_MODEL_MAP', join_paths(pkgdatadir, 'kbd-model-map'))
conf.set_quoted('SYSTEMD_LANGUAGE_FALLBACK_MAP', join_paths(pkgdatadir, 'language-fallback-map'))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
conf.set_quoted('UDEVLIBEXECDIR', udevlibexecdir)
meson: use join_paths consistently With -Dsplit-usr=true, we set rootprefix to /. This leads to //lib/systemd or //lib/udev for various dir variables. Using join_paths() avoids this.
2017-04-13 17:30:56 +03:00
conf.set_quoted('POLKIT_AGENT_BINARY_PATH', join_paths(bindir, 'pkttyagent'))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
conf.set_quoted('LIBDIR', libdir)
conf.set_quoted('ROOTLIBDIR', rootlibdir)
conf.set_quoted('ROOTLIBEXECDIR', rootlibexecdir)
conf.set_quoted('BOOTLIBDIR', bootlibdir)
meson: use join_paths consistently With -Dsplit-usr=true, we set rootprefix to /. This leads to //lib/systemd or //lib/udev for various dir variables. Using join_paths() avoids this.
2017-04-13 17:30:56 +03:00
conf.set_quoted('SYSTEMD_PULL_PATH', join_paths(rootlibexecdir, 'systemd-pull'))
conf.set_quoted('SYSTEMD_IMPORT_PATH', join_paths(rootlibexecdir, 'systemd-import'))
conf.set_quoted('SYSTEMD_EXPORT_PATH', join_paths(rootlibexecdir, 'systemd-export'))
conf.set_quoted('VENDOR_KEYRING_PATH', join_paths(rootlibexecdir, 'import-pubring.gpg'))
conf.set_quoted('USER_KEYRING_PATH', join_paths(pkgsysconfdir, 'import-pubring.gpg'))
conf.set_quoted('DOCUMENT_ROOT', join_paths(pkgdatadir, 'gatewayd'))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
conf.set_quoted('ABS_BUILD_DIR', meson.build_root())
conf.set_quoted('ABS_SRC_DIR', meson.source_root())
substs.set('prefix', prefixdir)
meson: add a few forgotten substitutions v2: - a few more
2017-04-11 02:06:45 +03:00
substs.set('exec_prefix', prefixdir)
substs.set('libdir', libdir)
substs.set('rootlibdir', rootlibdir)
substs.set('includedir', includedir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
substs.set('pkgsysconfdir', pkgsysconfdir)
meson: add a few forgotten substitutions v2: - a few more
2017-04-11 02:06:45 +03:00
substs.set('bindir', bindir)
substs.set('rootbindir', rootbindir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
substs.set('rootlibexecdir', rootlibexecdir)
substs.set('systemunitdir', systemunitdir)
substs.set('userunitdir', userunitdir)
substs.set('systempresetdir', systempresetdir)
substs.set('userpresetdir', userpresetdir)
substs.set('udevhwdbdir', udevhwdbdir)
substs.set('udevrulesdir', udevrulesdir)
meson: add a few forgotten substitutions v2: - a few more
2017-04-11 02:06:45 +03:00
substs.set('udevlibexecdir', udevlibexecdir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
substs.set('catalogdir', catalogdir)
substs.set('tmpfilesdir', tmpfilesdir)
substs.set('sysusersdir', sysusersdir)
substs.set('sysctldir', sysctldir)
substs.set('binfmtdir', binfmtdir)
substs.set('modulesloaddir', modulesloaddir)
substs.set('systemgeneratordir', systemgeneratordir)
substs.set('usergeneratordir', usergeneratordir)
substs.set('systemenvgeneratordir', systemenvgeneratordir)
substs.set('userenvgeneratordir', userenvgeneratordir)
substs.set('systemshutdowndir', systemshutdowndir)
substs.set('systemsleepdir', systemsleepdir)
meson: fix a bunch of substitutions This makes the diff -r output on autotools and meson installations empty, except for binary files and ## comments.
2017-04-13 02:54:33 +03:00
substs.set('VARLOGDIR', varlogdir)
substs.set('CERTIFICATEROOT', get_option('certificate-root'))
meson: use join_paths consistently With -Dsplit-usr=true, we set rootprefix to /. This leads to //lib/systemd or //lib/udev for various dir variables. Using join_paths() avoids this.
2017-04-13 17:30:56 +03:00
substs.set('SYSTEMCTL', join_paths(rootbindir, 'systemctl'))
substs.set('RANDOM_SEED', join_paths(randomseeddir, 'random-seed'))
meson: fix a bunch of substitutions This makes the diff -r output on autotools and meson installations empty, except for binary files and ## comments.
2017-04-13 02:54:33 +03:00
substs.set('SYSTEM_SYSVINIT_PATH', sysvinit_path)
substs.set('SYSTEM_SYSVRCND_PATH', sysvrcnd_path)
substs.set('RC_LOCAL_SCRIPT_PATH_START', get_option('rc-local'))
substs.set('RC_LOCAL_SCRIPT_PATH_STOP', get_option('halt-local'))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
#####################################################################
cc = meson.get_compiler('c')
pkgconfig = import('pkgconfig')
meson: fix checking of linker args Previous checks did nothing, because cc.has_argument only does compilation, without any linking. Unfortunately cc.links() cannot be used, because it does not accept any options. Providing the test file as a static source is easiest, even if not every elegant. https://github.com/mesonbuild/meson/issues/1676
2017-04-25 04:03:35 +03:00
check_compilation_sh = find_program('tools/meson-check-compilation.sh')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
tests,meson: add test-bus-vtable, compiled as C and C++ This test is mostly a compilation test that checks that various defines in sd-bus-vtable.h are valid C++. The code is executed, but the results are not checked (apart from sd-bus functions not returning an error). test-bus-objects contains pretty extensive tests for this functionality. The C++ version is only added to meson, since it's simpler there. Because of the .cc extension, meson will compile the executable with c++. This test is necessary to properly check the macros in sd-bus-vtable.h. Just running the headers through g++ is not enough, because the macros are not exercised. Follow-up for #5941.
2017-05-13 20:23:28 +03:00
cxx = find_program('c++', required : false)
if cxx.found()
# Used only for tests
add_languages('cpp')
endif
meson: use -Wextra if available -Wextra was not added to CFLAGS under meson by default, as it is done by the autotools build. C.f. 218f46711115669c26389a5bad79e57aa3c37f66.
2017-07-05 00:59:15 +03:00
foreach arg : ['-Wextra',
build-sys: require all defines under #if to be present This should help to catch any errors with typos and HAVE/ENABLE mismatches.
2017-10-03 13:11:49 +03:00
'-Werror=undef',
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
'-Wlogical-op',
'-Wmissing-include-dirs',
'-Wold-style-definition',
'-Wpointer-arith',
'-Winit-self',
'-Wdeclaration-after-statement',
'-Wfloat-equal',
'-Wsuggest-attribute=noreturn',
'-Werror=missing-prototypes',
'-Werror=implicit-function-declaration',
'-Werror=missing-declarations',
'-Werror=return-type',
'-Werror=incompatible-pointer-types',
'-Werror=format=2',
'-Wstrict-prototypes',
'-Wredundant-decls',
'-Wmissing-noreturn',
meson: "upgrade" -Wimplicit-fallthrough to 5 5 means that only the explicit attribute introduced in previous commit is accepted. We don't want the comments anymore.
2017-11-20 16:23:40 +03:00
'-Wimplicit-fallthrough=5',
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
'-Wshadow',
'-Wendif-labels',
'-Wstrict-aliasing=2',
'-Wwrite-strings',
'-Werror=overflow',
'-Wdate-time',
'-Wnested-externs',
'-ffast-math',
'-fno-common',
'-fdiagnostics-show-option',
'-fno-strict-aliasing',
'-fvisibility=hidden',
'-fstack-protector',
'-fstack-protector-strong',
'-fPIE',
'--param=ssp-buffer-size=4',
]
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
if cc.has_argument(arg)
add_project_arguments(arg, language : 'c')
endif
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endforeach
meson: fix detection of "-Wno-" options
2017-04-27 17:05:41 +03:00
# "negative" arguments: gcc on purpose does not return an error for "-Wno-"
# arguments, just emits a warnings. So test for the "positive" version instead.
foreach arg : ['unused-parameter',
'missing-field-initializers',
'unused-result',
build-sys: make nonnull warning non-fatal (#6742) Seems to be some kind of confusion in gcc. Insteading of playing whack-a-mole and adding work-arounds in code, let's adjust the compilation options instead. Fixes #6119, replaces #6657.
2017-09-04 19:49:12 +03:00
'format-signedness',
'error=nonnull', # work-around for gcc 7.1 turning this on on its own
]
meson: fix detection of "-Wno-" options
2017-04-27 17:05:41 +03:00
if cc.has_argument('-W' + arg)
add_project_arguments('-Wno-' + arg, language : 'c')
endif
endforeach
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
if cc.compiles('
#include <time.h>
#include <inttypes.h>
typedef uint64_t usec_t;
usec_t now(clockid_t clock);
int main(void) {
struct timespec now;
return 0;
}
meson: use name arg so check result is reported
2017-04-25 04:46:40 +03:00
', name : '-Werror=shadow with local shadowing')
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
add_project_arguments('-Werror=shadow', language : 'c')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
if cc.get_id() == 'clang'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
foreach arg : ['-Wno-typedef-redefinition',
'-Wno-gnu-variable-sized-type-not-at-end',
]
meson: use name arg so check result is reported
2017-04-25 04:46:40 +03:00
if cc.has_argument(arg,
name : '@0@ is supported'.format(arg))
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
add_project_arguments(arg, language : 'c')
endif
endforeach
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
meson: fix checking of linker args Previous checks did nothing, because cc.has_argument only does compilation, without any linking. Unfortunately cc.links() cannot be used, because it does not accept any options. Providing the test file as a static source is easiest, even if not every elegant. https://github.com/mesonbuild/meson/issues/1676
2017-04-25 04:03:35 +03:00
link_test_c = files('tools/meson-link-test.c')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
# --as-needed and --no-undefined are provided by meson by default,
# run mesonconf to see what is enabled
foreach arg : ['-Wl,-z,relro',
'-Wl,-z,now',
'-pie',
]
meson: fix checking of linker args Previous checks did nothing, because cc.has_argument only does compilation, without any linking. Unfortunately cc.links() cannot be used, because it does not accept any options. Providing the test file as a static source is easiest, even if not every elegant. https://github.com/mesonbuild/meson/issues/1676
2017-04-25 04:03:35 +03:00
have = run_command(check_compilation_sh,
cc.cmd_array(), '-x', 'c', arg,
'-include', link_test_c).returncode() == 0
message('Linking with @0@ supported: @1@'.format(arg, have ? 'yes' : 'no'))
if have
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
add_project_link_arguments(arg, language : 'c')
endif
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endforeach
meson: use -f{function,data}-sections for optimized builds
2017-04-25 02:28:04 +03:00
if get_option('buildtype') != 'debug'
foreach arg : ['-ffunction-sections',
'-fdata-sections']
meson: use name arg so check result is reported
2017-04-25 04:46:40 +03:00
if cc.has_argument(arg,
name : '@0@ is supported'.format(arg))
meson: use -f{function,data}-sections for optimized builds
2017-04-25 02:28:04 +03:00
add_project_arguments(arg, language : 'c')
endif
endforeach
foreach arg : ['-Wl,--gc-sections']
meson: fix checking of linker args Previous checks did nothing, because cc.has_argument only does compilation, without any linking. Unfortunately cc.links() cannot be used, because it does not accept any options. Providing the test file as a static source is easiest, even if not every elegant. https://github.com/mesonbuild/meson/issues/1676
2017-04-25 04:03:35 +03:00
have = run_command(check_compilation_sh,
cc.cmd_array(), '-x', 'c', arg,
'-include', link_test_c).returncode() == 0
message('Linking with @0@ supported: @1@'.format(arg, have ? 'yes' : 'no'))
if have
meson: use -f{function,data}-sections for optimized builds
2017-04-25 02:28:04 +03:00
add_project_link_arguments(arg, language : 'c')
endif
endforeach
endif
meson: make cpp invocations cross-compilation friendly This implementation assumes that the arguments in compiler.cmd_array() don't contain any spaces. Since we are only interested in compilation on Linux, I think this is a safe assumption. Solution suggested by Nirbheek Chauhan.
2017-04-11 17:25:34 +03:00
cpp = ' '.join(cc.cmd_array()) + ' -E'
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
#####################################################################
# compilation result tests
meson: use booleans for conf.set and drop unecessary conditionals Using conf.set() with a boolean argument does the right thing: either #ifdef or #undef. This means that conf.set can be used unconditionally. Previously I used '1' as the placeholder value, and that needs to be changed to 'true' for consistency (under meson 1 cannot be used in boolean context). All checks need to be adjusted.
2017-04-28 04:13:08 +03:00
conf.set('_GNU_SOURCE', true)
conf.set('__SANE_USERSPACE_TYPES__', true)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
conf.set('SIZEOF_PID_T', cc.sizeof('pid_t', prefix : '#include <sys/types.h>'))
conf.set('SIZEOF_UID_T', cc.sizeof('uid_t', prefix : '#include <sys/types.h>'))
conf.set('SIZEOF_GID_T', cc.sizeof('gid_t', prefix : '#include <sys/types.h>'))
conf.set('SIZEOF_DEV_T', cc.sizeof('dev_t', prefix : '#include <sys/types.h>'))
conf.set('SIZEOF_INO_T', cc.sizeof('ino_t', prefix : '#include <sys/types.h>'))
conf.set('SIZEOF_TIME_T', cc.sizeof('time_t', prefix : '#include <sys/time.h>'))
conf.set('SIZEOF_RLIM_T', cc.sizeof('rlim_t', prefix : '#include <sys/resource.h>'))
decl_headers = '''
#include <uchar.h>
#include <linux/ethtool.h>
networkd: add support to configure IP Rule (#5725) Routing Policy rule manipulates rules in the routing policy database control the route selection algorithm. This work supports to configure Rule ``` [RoutingPolicyRule] TypeOfService=0x08 Table=7 From= 192.168.100.18 ``` ``` ip rule show 0: from all lookup local 0: from 192.168.100.18 tos 0x08 lookup 7 ``` V2 changes: 1. Added logic to handle duplicate rules. 2. If rules are changed or deleted and networkd restarted then those are deleted when networkd restarts next time V3: 1. Add parse_fwmark_fwmask
2017-09-14 22:51:39 +03:00
#include <linux/fib_rules.h>
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
'''
# FIXME: key_serial_t is only defined in keyutils.h, this is bound to fail
foreach decl : ['char16_t',
'char32_t',
'key_serial_t',
'struct ethtool_link_settings',
networkd: add support to configure IP Rule (#5725) Routing Policy rule manipulates rules in the routing policy database control the route selection algorithm. This work supports to configure Rule ``` [RoutingPolicyRule] TypeOfService=0x08 Table=7 From= 192.168.100.18 ``` ``` ip rule show 0: from all lookup local 0: from 192.168.100.18 tos 0x08 lookup 7 ``` V2 changes: 1. Added logic to handle duplicate rules. 2. If rules are changed or deleted and networkd restarted then those are deleted when networkd restarts next time V3: 1. Add parse_fwmark_fwmask
2017-09-14 22:51:39 +03:00
'struct fib_rule_uid_range',
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
]
meson: use booleans for conf.set and drop unecessary conditionals Using conf.set() with a boolean argument does the right thing: either #ifdef or #undef. This means that conf.set can be used unconditionally. Previously I used '1' as the placeholder value, and that needs to be changed to 'true' for consistency (under meson 1 cannot be used in boolean context). All checks need to be adjusted.
2017-04-28 04:13:08 +03:00
# We get -1 if the size cannot be determined
have = cc.sizeof(decl, prefix : decl_headers) > 0
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_' + decl.underscorify().to_upper(), have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endforeach
foreach decl : [['IFLA_INET6_ADDR_GEN_MODE', 'linux/if_link.h'],
['IN6_ADDR_GEN_MODE_STABLE_PRIVACY', 'linux/if_link.h'],
['IFLA_VRF_TABLE', 'linux/if_link.h'],
['IFLA_MACVLAN_FLAGS', 'linux/if_link.h'],
networkd: Add support for ipvlan L3s and flags (#7726) This works supports to configure L3S mode and flags such as bridge, private and vepa
2017-12-23 20:55:03 +03:00
['IFLA_IPVLAN_FLAGS', 'linux/if_link.h'],
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
['IFLA_PHYS_PORT_ID', 'linux/if_link.h'],
['IFLA_BOND_AD_INFO', 'linux/if_link.h'],
['IFLA_VLAN_PROTOCOL', 'linux/if_link.h'],
['IFLA_VXLAN_REMCSUM_NOPARTIAL', 'linux/if_link.h'],
['IFLA_VXLAN_GPE', 'linux/if_link.h'],
meson: add geneve to build system ``` [107/1793] Compiling c object 'src/basic/basic@sta/log.c.o' In file included from ../src/basic/log.c:44:0: ../src/basic/missing.h:762:6: warning: "HAVE_DECL_IFLA_GENEVE_LABEL" is not defined, evaluates to 0 [-Wundef] ^~~~~~~~~~~~~~~~~~~~~~~~~~~ [108/1793] Compiling c object 'src/basic/basic@sta/path-util.c.o' In file included from ../src/basic/path-util.c:40:0: ../src/basic/missing.h:762:6: warning: "HAVE_DECL_IFLA_GENEVE_LABEL" is not defined, evaluates to 0 [-Wundef] ^~~~~~~~~~~~~~~~~~~~~~~~~~~ ```
2017-04-25 18:00:34 +03:00
['IFLA_GENEVE_LABEL', 'linux/if_link.h'],
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
# if_tunnel.h is buggy and cannot be included on its own
['IFLA_VTI_REMOTE', 'linux/if_tunnel.h', '#include <net/if.h>'],
['IFLA_IPTUN_ENCAP_DPORT', 'linux/if_tunnel.h', '#include <net/if.h>'],
['IFLA_GRE_ENCAP_DPORT', 'linux/if_tunnel.h', '#include <net/if.h>'],
['IFLA_BRIDGE_VLAN_INFO', 'linux/if_bridge.h'],
['IFLA_BRPORT_PROXYARP', 'linux/if_link.h'],
['IFLA_BRPORT_LEARNING_SYNC', 'linux/if_link.h'],
['IFLA_BR_VLAN_DEFAULT_PVID', 'linux/if_link.h'],
networkd: Add support for ipvlan L3s and flags (#7726) This works supports to configure L3S mode and flags such as bridge, private and vepa
2017-12-23 20:55:03 +03:00
['IPVLAN_F_PRIVATE', 'linux/if_link.h'],
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
['NDA_IFINDEX', 'linux/neighbour.h'],
['IFA_FLAGS', 'linux/if_addr.h'],
networkd: add support to configure IP Rule (#5725) Routing Policy rule manipulates rules in the routing policy database control the route selection algorithm. This work supports to configure Rule ``` [RoutingPolicyRule] TypeOfService=0x08 Table=7 From= 192.168.100.18 ``` ``` ip rule show 0: from all lookup local 0: from 192.168.100.18 tos 0x08 lookup 7 ``` V2 changes: 1. Added logic to handle duplicate rules. 2. If rules are changed or deleted and networkd restarted then those are deleted when networkd restarts next time V3: 1. Add parse_fwmark_fwmask
2017-09-14 22:51:39 +03:00
['FRA_UID_RANGE', 'linux/fib_rules.h'],
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
['LO_FLAGS_PARTSCAN', 'linux/loop.h'],
networkd: introduce vxcan netdev. (#7150) Similar to the virtual ethernet driver veth, vxcan implements a local CAN traffic tunnel between two virtual CAN network devices. When creating a vxcan, two vxcan devices are created as pair When one end receives the packet it appears on its pair and vice versa. The vxcan can be used for cross namespace communication.
2017-11-22 10:23:22 +03:00
['VXCAN_INFO_PEER', 'linux/can/vxcan.h'],
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
]
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
prefix = decl.length() > 2 ? decl[2] : ''
have = cc.has_header_symbol(decl[1], decl[0], prefix : prefix)
build-sys: change all HAVE_DECL_ macros to HAVE_ This is a legacy of autotools, where one detection routine used a different prefix then the others. $ git grep -e HAVE_DECL_ -l|xargs sed -i s/HAVE_DECL_/HAVE_/g
2017-10-03 11:32:34 +03:00
conf.set10('HAVE_' + decl[0], have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endforeach
foreach ident : ['secure_getenv', '__secure_getenv']
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_' + ident.to_upper(), cc.has_function(ident))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endforeach
foreach ident : [
meson: use "args" for setting _GNU_SOURCE when checking for functions This reworks how we set _GNU_SOURCE when checking for the availability of functions: 1. We set it for most of the functions we look for. After all we set it for our entire built anyway, and it's usually how Linux-specific definitions in glibc are protected these days. Given that we usually have checks for such modern stuff only anyway, let's just blanket enable it. 2. Use "args" instead of "prefix" to set the macro. This is what is suggested in the meson docs, hence let's do it.
2017-12-25 14:01:14 +03:00
['memfd_create', '''#include <sys/mman.h>'''],
meson: look for gettid() definition where getpid() is defined Hopefully, should gettid() show up one day in glibc it'll show up where getpid() is defined too.
2017-12-25 14:35:28 +03:00
['gettid', '''#include <sys/types.h>
#include <unistd.h>'''],
meson: when pivot_root() is added one day, look for it in <unistd.h> We of course don't know in which header glibc will export pivot_root() and if it ever will. But there's a good chance they'll place it where chroot() is located, given the similarity in the operations, hence let's try our luck and look for it at the same place. If we are lucky this means we don't have to patch our code if glibc decides to expose the call one day.
2017-12-25 14:07:40 +03:00
['pivot_root', '''#include <stdlib.h>
#include <unistd.h>'''], # no known header declares pivot_root
meson: use "args" for setting _GNU_SOURCE when checking for functions This reworks how we set _GNU_SOURCE when checking for the availability of functions: 1. We set it for most of the functions we look for. After all we set it for our entire built anyway, and it's usually how Linux-specific definitions in glibc are protected these days. Given that we usually have checks for such modern stuff only anyway, let's just blanket enable it. 2. Use "args" instead of "prefix" to set the macro. This is what is suggested in the meson docs, hence let's do it.
2017-12-25 14:01:14 +03:00
['name_to_handle_at', '''#include <sys/types.h>
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
#include <sys/stat.h>
#include <fcntl.h>'''],
meson: use "args" for setting _GNU_SOURCE when checking for functions This reworks how we set _GNU_SOURCE when checking for the availability of functions: 1. We set it for most of the functions we look for. After all we set it for our entire built anyway, and it's usually how Linux-specific definitions in glibc are protected these days. Given that we usually have checks for such modern stuff only anyway, let's just blanket enable it. 2. Use "args" instead of "prefix" to set the macro. This is what is suggested in the meson docs, hence let's do it.
2017-12-25 14:01:14 +03:00
['setns', '''#include <sched.h>'''],
meson: hopefully renameat2() will show up where renameat() is defined Should glibc add this eventually, let's try to be smart where to look for it.
2017-12-25 14:35:43 +03:00
['renameat2', '''#include <stdio.h>
#include <fcntl.h>'''],
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
['kcmp', '''#include <linux/kcmp.h>'''],
['keyctl', '''#include <sys/types.h>
#include <keyutils.h>'''],
meson: use "args" for setting _GNU_SOURCE when checking for functions This reworks how we set _GNU_SOURCE when checking for the availability of functions: 1. We set it for most of the functions we look for. After all we set it for our entire built anyway, and it's usually how Linux-specific definitions in glibc are protected these days. Given that we usually have checks for such modern stuff only anyway, let's just blanket enable it. 2. Use "args" instead of "prefix" to set the macro. This is what is suggested in the meson docs, hence let's do it.
2017-12-25 14:01:14 +03:00
['copy_file_range', '''#include <sys/syscall.h>
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
#include <unistd.h>'''],
Add abstraction model for BPF programs This object takes a number of bpf_insn members and wraps them together with the in-kernel reference id. Will be needed by the firewall code.
2016-10-18 18:57:10 +03:00
['bpf', '''#include <sys/syscall.h>
#include <unistd.h>'''],
meson: check string.h not strings.h for explicit_bzero I used strings.h because that's what explicit_bzero(3) says. But glibc defines it in string.h. There are no other available implementations atm (musl does not define it, bionic aliases it to memset in openbsd-compat.h, yikes). We also include just string.h from the code. So just look for the function in string.h, and if it ever appears in other places, we can change the detection and includes.
2017-04-19 23:14:16 +03:00
['explicit_bzero' , '''#include <string.h>'''],
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
]
meson: use "args" for setting _GNU_SOURCE when checking for functions This reworks how we set _GNU_SOURCE when checking for the availability of functions: 1. We set it for most of the functions we look for. After all we set it for our entire built anyway, and it's usually how Linux-specific definitions in glibc are protected these days. Given that we usually have checks for such modern stuff only anyway, let's just blanket enable it. 2. Use "args" instead of "prefix" to set the macro. This is what is suggested in the meson docs, hence let's do it.
2017-12-25 14:01:14 +03:00
have = cc.has_function(ident[0], prefix : ident[1], args : '-D_GNU_SOURCE')
build-sys: change all HAVE_DECL_ macros to HAVE_ This is a legacy of autotools, where one detection routine used a different prefix then the others. $ git grep -e HAVE_DECL_ -l|xargs sed -i s/HAVE_DECL_/HAVE_/g
2017-10-03 11:32:34 +03:00
conf.set10('HAVE_' + ident[0].to_upper(), have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endforeach
meson: use "args" for setting _GNU_SOURCE when checking for functions This reworks how we set _GNU_SOURCE when checking for the availability of functions: 1. We set it for most of the functions we look for. After all we set it for our entire built anyway, and it's usually how Linux-specific definitions in glibc are protected these days. Given that we usually have checks for such modern stuff only anyway, let's just blanket enable it. 2. Use "args" instead of "prefix" to set the macro. This is what is suggested in the meson docs, hence let's do it.
2017-12-25 14:01:14 +03:00
if cc.has_function('getrandom', prefix : '''#include <sys/random.h>''', args : '-D_GNU_SOURCE')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('USE_SYS_RANDOM_H', true)
build-sys: change all HAVE_DECL_ macros to HAVE_ This is a legacy of autotools, where one detection routine used a different prefix then the others. $ git grep -e HAVE_DECL_ -l|xargs sed -i s/HAVE_DECL_/HAVE_/g
2017-10-03 11:32:34 +03:00
conf.set10('HAVE_GETRANDOM', true)
meson: detect getrandom in sys/random.h, fall back to linux/random.h This is analogous to commit e84397d95c3da2b9722ff265843ad4a8874643d8.
2017-04-20 04:20:54 +03:00
else
have = cc.has_function('getrandom', prefix : '''#include <linux/random.h>''')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('USE_SYS_RANDOM_H', false)
build-sys: change all HAVE_DECL_ macros to HAVE_ This is a legacy of autotools, where one detection routine used a different prefix then the others. $ git grep -e HAVE_DECL_ -l|xargs sed -i s/HAVE_DECL_/HAVE_/g
2017-10-03 11:32:34 +03:00
conf.set10('HAVE_GETRANDOM', have)
meson: detect getrandom in sys/random.h, fall back to linux/random.h This is analogous to commit e84397d95c3da2b9722ff265843ad4a8874643d8.
2017-04-20 04:20:54 +03:00
endif
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
#####################################################################
sed = find_program('sed')
awk = find_program('awk')
meson: require newest meson and revert m4 patch This reverts 78b68dcb55.
2017-04-25 15:49:58 +03:00
m4 = find_program('m4')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
stat = find_program('stat')
meson: add TAGS target Quite necessary for all the emacs users out there ;) v2: - fix error with etags or git are not found v3: - move the definition of git and etags up so it's available for man/
2017-04-06 21:33:27 +03:00
git = find_program('git', required : false)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
meson: create various symlinks v2: - remove bashisms
2017-04-10 06:55:50 +03:00
meson_make_symlink = meson.source_root() + '/tools/meson-make-symlink.sh'
meson: create dirs and touch /usr This is the equivalent of $(INSTALL_DIRS) and install-touch-usr-hook. I did not bother to create the directories into which we install files, since they will be created anyway. v2: - remove bashism
2017-04-10 06:55:05 +03:00
mkdir_p = 'mkdir -p $DESTDIR/@0@'
test-efi-create-disk.sh: allow running from separate build dir, hook up to meson This allow test-efi-disk.img to be created under meson. The invocation of qemu is not converted yet, in particular because the command-line used in Makefile.am is outdated.
2017-04-16 19:04:46 +03:00
test_efi_create_disk_sh = find_program('test/test-efi-create-disk.sh')
splash_bmp = files('test/splash.bmp')
meson: create dirs and touch /usr This is the equivalent of $(INSTALL_DIRS) and install-touch-usr-hook. I did not bother to create the directories into which we install files, since they will be created anyway. v2: - remove bashism
2017-04-10 06:55:05 +03:00
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
# if -Dxxx-path option is found, use that. Otherwise, check in $PATH,
# /usr/sbin, /sbin, and fall back to the default from middle column.
progs = [['telinit', '/lib/sysvinit/telinit'],
['quotaon', '/usr/sbin/quotaon' ],
['quotacheck', '/usr/sbin/quotacheck' ],
['kill', '/usr/bin/kill' ],
['kmod', '/usr/bin/kmod' ],
['kexec', '/usr/sbin/kexec' ],
['sulogin', '/usr/sbin/sulogin' ],
['mount', '/usr/bin/mount', 'MOUNT_PATH'],
['umount', '/usr/bin/umount', 'UMOUNT_PATH'],
['loadkeys', '/usr/bin/loadkeys', 'KBD_LOADKEYS'],
['setfont', '/usr/bin/setfont', 'KBD_SETFONT'],
]
foreach prog : progs
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
path = get_option(prog[0] + '-path')
if path != ''
message('Using @1@ for @0@'.format(prog[0], path))
else
exe = find_program(prog[0],
'/usr/sbin/' + prog[0],
'/sbin/' + prog[0],
required: false)
path = exe.found() ? exe.path() : prog[1]
endif
name = prog.length() > 2 ? prog[2] : prog[0].to_upper()
conf.set_quoted(name, path)
substs.set(name, path)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endforeach
meson: add ln --relative check
2017-04-19 02:11:54 +03:00
if run_command('ln', '--relative', '--help').returncode() != 0
error('ln does not support --relative')
endif
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
############################################################
gperf = find_program('gperf')
gperf_test_format = '''
#include <string.h>
const char * in_word_set(const char *, @0@);
@1@
'''
gperf_snippet_format = 'echo foo,bar | @0@ -L ANSI-C'
gperf_snippet = run_command('sh', '-c', gperf_snippet_format.format(gperf.path()))
gperf_test = gperf_test_format.format('size_t', gperf_snippet.stdout())
if cc.compiles(gperf_test)
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
gperf_len_type = 'size_t'
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
gperf_test = gperf_test_format.format('unsigned', gperf_snippet.stdout())
if cc.compiles(gperf_test)
gperf_len_type = 'unsigned'
else
error('unable to determine gperf len type')
endif
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
message('gperf len type is @0@'.format(gperf_len_type))
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
conf.set('GPERF_LEN_TYPE', gperf_len_type,
description : 'The type of gperf "len" parameter')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
############################################################
if not cc.has_header('sys/capability.h')
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
error('POSIX caps headers not found')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
foreach header : ['linux/btrfs.h',
'linux/memfd.h',
'linux/vm_sockets.h',
meson: check for sys/auxv.h This check was present in configure.ac, but was never added under meson. The code under HAVE_SYS_AUX_H has been dead ever since :(.
2017-10-03 13:09:40 +03:00
'sys/auxv.h',
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
'valgrind/memcheck.h',
'valgrind/valgrind.h',
]
meson: use booleans for conf.set and drop unecessary conditionals Using conf.set() with a boolean argument does the right thing: either #ifdef or #undef. This means that conf.set can be used unconditionally. Previously I used '1' as the placeholder value, and that needs to be changed to 'true' for consistency (under meson 1 cannot be used in boolean context). All checks need to be adjusted.
2017-04-28 04:13:08 +03:00
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_' + header.underscorify().to_upper(),
cc.has_header(header))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endforeach
############################################################
conf.set_quoted('FALLBACK_HOSTNAME', get_option('fallback-hostname'))
resolved,nss-myhostname: use _gateway for the gateway This changes the symbolic name for the default gateway from "gateway" to "_gateway". A new configuration option -Dcompat-gateway-hostname=true|false is added. If it is set, the old name is also supported, but the new name is used as the canonical name in either case. This is intended as a temporary measure to make the transition easier, and the option should be removed after a few releases, at which point only the new name will be used. The old "gateway" name mostly works OK, but hasn't gained widespread acceptance because of the following (potential) conflicts: - it is completely legal to have a host called "gateway" - there is no guarantee that "gateway" will not be registered as a TLD, even though this currently seems unlikely. (Even then, there would be no conflict except for the case when the top-level domain itself was being resolved. The "gateway" or "_gateway" labels have only special meaning when the whole name consists of a single label, so resolution of any subdomain of the hypothetical gateway. TLD would still work OK. ) Moving to "_gateway" avoids those issues because underscores are not allowed in host names (RFC 1123, §2.1) and avoids potential conflicts with local or global names. v2: - simplify the logic to hardcode "_gateway" and allow -Dcompat-gateway-hostname=true as a temporary measure.
2017-07-11 09:15:08 +03:00
conf.set10('ENABLE_COMPAT_GATEWAY_HOSTNAME', get_option('compat-gateway-hostname'))
gateway_hostnames = ['_gateway'] + (conf.get('ENABLE_COMPAT_GATEWAY_HOSTNAME') == 1 ? ['gateway'] : [])
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
default_hierarchy = get_option('default-hierarchy')
conf.set_quoted('DEFAULT_HIERARCHY_NAME', default_hierarchy,
description : 'default cgroup hierarchy as string')
if default_hierarchy == 'legacy'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
conf.set('DEFAULT_HIERARCHY', 'CGROUP_UNIFIED_NONE')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
elif default_hierarchy == 'hybrid'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
conf.set('DEFAULT_HIERARCHY', 'CGROUP_UNIFIED_SYSTEMD')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
conf.set('DEFAULT_HIERARCHY', 'CGROUP_UNIFIED_ALL')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
time_epoch = get_option('time-epoch')
if time_epoch == ''
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
NEWS = files('NEWS')
time_epoch = run_command(stat, '-c', '%Y', NEWS).stdout()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
time_epoch = time_epoch.to_int()
conf.set('TIME_EPOCH', time_epoch)
system_uid_max = get_option('system-uid-max')
if system_uid_max == ''
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
system_uid_max = run_command(
awk,
'BEGIN { uid=999 } /^\s*SYS_UID_MAX\s+/ { uid=$2 } END { print uid }',
'/etc/login.defs').stdout()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
system_uid_max = system_uid_max.to_int()
conf.set('SYSTEM_UID_MAX', system_uid_max)
substs.set('systemuidmax', system_uid_max)
meson: use name arg so check result is reported
2017-04-25 04:46:40 +03:00
message('maximum system UID is @0@'.format(system_uid_max))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
system_gid_max = get_option('system-gid-max')
if system_gid_max == ''
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
system_gid_max = run_command(
awk,
'BEGIN { gid=999 } /^\s*SYS_GID_MAX\s+/ { gid=$2 } END { print gid }',
'/etc/login.defs').stdout()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
system_gid_max = system_gid_max.to_int()
conf.set('SYSTEM_GID_MAX', system_gid_max)
substs.set('systemgidmax', system_gid_max)
meson: use name arg so check result is reported
2017-04-25 04:46:40 +03:00
message('maximum system GID is @0@'.format(system_gid_max))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
build-sys: make the dynamic UID range, and the container UID range configurable Also, export these ranges in our pkg-config files.
2017-12-02 14:48:31 +03:00
dynamic_uid_min = get_option('dynamic-uid-min').to_int()
dynamic_uid_max = get_option('dynamic-uid-max').to_int()
conf.set('DYNAMIC_UID_MIN', dynamic_uid_min)
conf.set('DYNAMIC_UID_MAX', dynamic_uid_max)
substs.set('dynamicuidmin', dynamic_uid_min)
substs.set('dynamicuidmax', dynamic_uid_max)
container_uid_base_min = get_option('container-uid-base-min').to_int()
container_uid_base_max = get_option('container-uid-base-max').to_int()
conf.set('CONTAINER_UID_BASE_MIN', container_uid_base_min)
conf.set('CONTAINER_UID_BASE_MAX', container_uid_base_max)
substs.set('containeruidbasemin', container_uid_base_min)
substs.set('containeruidbasemax', container_uid_base_max)
meson: print warnings if the "nobody" user/group name is not compatible with the local system At least on Fedora and Debian systems this not obvious to get right, hence warn.
2017-12-05 13:00:24 +03:00
nobody_user = get_option('nobody-user')
nobody_group = get_option('nobody-group')
getent_result = run_command('getent', 'passwd', '65534')
if getent_result.returncode() == 0
name = getent_result.stdout().split(':')[0]
if name != nobody_user
message('WARNING:\n' +
' The local user with the UID 65534 does not match the configured user name "@0@" of the nobody user (its name is @1@).\n'.format(nobody_user, name) +
' Your build will result in an user table setup that is incompatible with the local system.')
endif
endif
id_result = run_command('id', '-u', nobody_user)
if id_result.returncode() == 0
id = id_result.stdout().to_int()
if id != 65534
message('WARNING:\n' +
' The local user with the configured user name "@0@" of the nobody user does not have UID 65534 (it has @1@).\n'.format(nobody_user, id) +
' Your build will result in an user table setup that is incompatible with the local system.')
endif
endif
getent_result = run_command('getent', 'group', '65534')
if getent_result.returncode() == 0
name = getent_result.stdout().split(':')[0]
if name != nobody_group
message('WARNING:\n' +
' The local group with the GID 65534 does not match the configured group name "@0@" of the nobody group (its name is @1@).\n'.format(nobody_group, name) +
' Your build will result in an group table setup that is incompatible with the local system.')
endif
endif
id_result = run_command('id', '-g', nobody_group)
if id_result.returncode() == 0
id = id_result.stdout().to_int()
if id != 65534
message('WARNING:\n' +
' The local group with the configured group name "@0@" of the nobody group does not have UID 65534 (it has @1@).\n'.format(nobody_group, id) +
' Your build will result in an group table setup that is incompatible with the local system.')
endif
endif
meson: warn if nobody-user and nobody-group are set to different name It may work, but is very strange. So, let's warn about that. v2: Debian uses nobody and nogroup. Do not warn such case.
2017-12-07 11:19:11 +03:00
if nobody_user != nobody_group and not (nobody_user == 'nobody' and nobody_group == 'nogroup')
message('WARNING:\n' +
' The configured user name "@0@" and group name "@0@" of the nobody user/group are not equivalent.\n'.format(nobody_user, nobody_group) +
' Please re-check that both "nobody-user" and "nobody-group" options are correctly set.')
endif
meson: print warnings if the "nobody" user/group name is not compatible with the local system At least on Fedora and Debian systems this not obvious to get right, hence warn.
2017-12-05 13:00:24 +03:00
conf.set_quoted('NOBODY_USER_NAME', nobody_user)
conf.set_quoted('NOBODY_GROUP_NAME', nobody_group)
sysusers: use NOBODY_USER_NAME
2017-12-07 09:49:16 +03:00
substs.set('NOBODY_USER_NAME', nobody_user)
substs.set('NOBODY_GROUP_NAME', nobody_group)
build-sys: make the dynamic UID range, and the container UID range configurable Also, export these ranges in our pkg-config files.
2017-12-02 14:48:31 +03:00
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
tty_gid = get_option('tty-gid')
conf.set('TTY_GID', tty_gid)
meson: fix a bunch of substitutions This makes the diff -r output on autotools and meson installations empty, except for binary files and ## comments.
2017-04-13 02:54:33 +03:00
substs.set('TTY_GID', tty_gid)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
sysusers: Provide meson argument to set gid for 'users' group (#7533) To allow better integration with distributions requiring an explicitly set gid for the `users` group, provide the new `-Dusers-gid` option to set to a new numeric value. In the absence of a specified gid, we'll fallback to the default existing behaviour of `-` as the gid value, to automatically assign the next available gid on the system.
2017-12-03 15:28:23 +03:00
# Ensure provided GID argument is numeric, otherwise fallback to default assignment
if get_option('users-gid') != ''
meson: fix indentation
2017-12-05 08:01:39 +03:00
users_gid = get_option('users-gid').to_int()
sysusers: Provide meson argument to set gid for 'users' group (#7533) To allow better integration with distributions requiring an explicitly set gid for the `users` group, provide the new `-Dusers-gid` option to set to a new numeric value. In the absence of a specified gid, we'll fallback to the default existing behaviour of `-` as the gid value, to automatically assign the next available gid on the system.
2017-12-03 15:28:23 +03:00
else
meson: fix indentation
2017-12-05 08:01:39 +03:00
users_gid = '-'
sysusers: Provide meson argument to set gid for 'users' group (#7533) To allow better integration with distributions requiring an explicitly set gid for the `users` group, provide the new `-Dusers-gid` option to set to a new numeric value. In the absence of a specified gid, we'll fallback to the default existing behaviour of `-` as the gid value, to automatically assign the next available gid on the system.
2017-12-03 15:28:23 +03:00
endif
substs.set('USERS_GID', users_gid)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
if get_option('adm-group')
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
m4_defines += ['-DENABLE_ADM_GROUP']
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
if get_option('wheel-group')
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
m4_defines += ['-DENABLE_WHEEL_GROUP']
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
substs.set('DEV_KVM_MODE', get_option('dev-kvm-mode'))
udev-rules: Permission changes for /dev/dri/renderD* - Remove the uaccess tag from /dev/dri/renderD*. - Change the owning group from video to render. - Change default mode to 0666. - Add an option to allow users to set the access mode for these devices at compile time.
2017-10-31 18:46:24 +03:00
substs.set('GROUP_RENDER_MODE', get_option('group-render-mode'))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
meson: fix a bunch of substitutions This makes the diff -r output on autotools and meson installations empty, except for binary files and ## comments.
2017-04-13 02:54:33 +03:00
kill_user_processes = get_option('default-kill-user-processes')
conf.set10('KILL_USER_PROCESSES', kill_user_processes)
substs.set('KILL_USER_PROCESSES', kill_user_processes ? 'yes' : 'no')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
dns_servers = get_option('dns-servers')
conf.set_quoted('DNS_SERVERS', dns_servers)
substs.set('DNS_SERVERS', dns_servers)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
ntp_servers = get_option('ntp-servers')
conf.set_quoted('NTP_SERVERS', ntp_servers)
substs.set('NTP_SERVERS', ntp_servers)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
conf.set_quoted('GETTEXT_PACKAGE', meson.project_name())
meson: add a few forgotten substitutions v2: - a few more
2017-04-11 02:06:45 +03:00
substs.set('SUSHELL', get_option('debug-shell'))
substs.set('DEBUGTTY', get_option('debug-tty'))
meson: add forgotten debug options v2: -rename -Dextra-debug to -Ddebug to match ./configure --debug
2017-04-28 03:51:34 +03:00
debug = get_option('debug')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
enable_debug_hashmap = false
enable_debug_mmap_cache = false
meson: add forgotten debug options v2: -rename -Dextra-debug to -Ddebug to match ./configure --debug
2017-04-28 03:51:34 +03:00
if debug != ''
foreach name : debug.split(',')
if name == 'hashmap'
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
enable_debug_hashmap = true
meson: add forgotten debug options v2: -rename -Dextra-debug to -Ddebug to match ./configure --debug
2017-04-28 03:51:34 +03:00
elif name == 'mmap-cache'
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
enable_debug_mmap_cache = true
meson: add forgotten debug options v2: -rename -Dextra-debug to -Ddebug to match ./configure --debug
2017-04-28 03:51:34 +03:00
else
message('unknown debug option "@0@", ignoring'.format(name))
endif
endforeach
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('ENABLE_DEBUG_HASHMAP', enable_debug_hashmap)
conf.set10('ENABLE_DEBUG_MMAP_CACHE', enable_debug_mmap_cache)
meson: add forgotten debug options v2: -rename -Dextra-debug to -Ddebug to match ./configure --debug
2017-04-28 03:51:34 +03:00
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
#####################################################################
threads = dependency('threads')
librt = cc.find_library('rt')
libm = cc.find_library('m')
libdl = cc.find_library('dl')
libcrypt = cc.find_library('crypt')
meson: also search for libcap directly
2017-04-27 08:30:30 +03:00
libcap = dependency('libcap', required : false)
if not libcap.found()
# Compat with Ubuntu 14.04 which ships libcap w/o .pc file
libcap = cc.find_library('cap')
endif
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
libmount = dependency('mount',
build-sys: require libmount >= 2.30 (#6795) Fixes #4871. The new libmount has two changes relevant for us: - x-* options are propagated to /run/mount/utab and are visible through libmount (fixes #4817). - umount -c now really works (partially solves #6115).
2017-09-15 15:47:57 +03:00
version : '>= 2.30')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_seccomp = get_option('seccomp')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_seccomp != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libseccomp = dependency('libseccomp',
meson: add version check for libseccomp Compilation fails because of the missing arm64 bits with old seccomp versions.
2017-04-27 17:05:18 +03:00
version : '>= 2.3.1',
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
required : want_seccomp == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libseccomp.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libseccomp = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_SECCOMP', have)
m4_defines += have ? ['-DHAVE_SECCOMP'] : []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_selinux = get_option('selinux')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_selinux != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libselinux = dependency('libselinux',
version : '>= 2.1.9',
required : want_selinux == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libselinux.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libselinux = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_SELINUX', have)
m4_defines += have ? ['-DHAVE_SELINUX'] : []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_apparmor = get_option('apparmor')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_apparmor != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libapparmor = dependency('libapparmor',
required : want_apparmor == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libapparmor.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libapparmor = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_APPARMOR', have)
m4_defines += have ? ['-DHAVE_APPARMOR'] : []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
smack_run_label = get_option('smack-run-label')
if smack_run_label != ''
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
conf.set_quoted('SMACK_RUN_LABEL', smack_run_label)
m4_defines += ['-DHAVE_SMACK_RUN_LABEL']
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
meson: properly conditionalize polkit installation ... including pkla installation on Debian. v2: - fix polkit-gobject-1 pkgconfig name
2017-04-13 02:09:26 +03:00
want_polkit = get_option('polkit')
install_polkit = false
install_polkit_pkla = false
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_polkit != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
install_polkit = true
meson: properly conditionalize polkit installation ... including pkla installation on Debian. v2: - fix polkit-gobject-1 pkgconfig name
2017-04-13 02:09:26 +03:00
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libpolkit = dependency('polkit-gobject-1',
required : false)
if libpolkit.found() and libpolkit.version().version_compare('< 0.106')
message('Old polkit detected, will install pkla files')
install_polkit_pkla = true
endif
meson: properly conditionalize polkit installation ... including pkla installation on Debian. v2: - fix polkit-gobject-1 pkgconfig name
2017-04-13 02:09:26 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('ENABLE_POLKIT', install_polkit)
meson: properly conditionalize polkit installation ... including pkla installation on Debian. v2: - fix polkit-gobject-1 pkgconfig name
2017-04-13 02:09:26 +03:00
meson: add option to disable libacl support
2017-04-21 20:53:59 +03:00
want_acl = get_option('acl')
if want_acl != 'false'
libacl = cc.find_library('acl', required : want_acl == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libacl.found()
meson: add option to disable libacl support
2017-04-21 20:53:59 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: add option to disable libacl support
2017-04-21 20:53:59 +03:00
libacl = []
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_ACL', have)
m4_defines += have ? ['-DHAVE_ACL'] : []
meson: add option to disable libacl support
2017-04-21 20:53:59 +03:00
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_audit = get_option('audit')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_audit != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libaudit = dependency('audit', required : want_audit == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libaudit.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libaudit = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_AUDIT', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_blkid = get_option('blkid')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_blkid != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libblkid = dependency('blkid', required : want_blkid == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libblkid.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libblkid = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_BLKID', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_kmod = get_option('kmod')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_kmod != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libkmod = dependency('libkmod',
version : '>= 15',
required : want_kmod == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libkmod.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libkmod = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_KMOD', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_pam = get_option('pam')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_pam != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libpam = cc.find_library('pam', required : want_pam == 'true')
libpam_misc = cc.find_library('pam_misc', required : want_pam == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libpam.found() and libpam_misc.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libpam = []
libpam_misc = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_PAM', have)
m4_defines += have ? ['-DHAVE_PAM'] : []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_microhttpd = get_option('microhttpd')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_microhttpd != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libmicrohttpd = dependency('libmicrohttpd',
version : '>= 0.9.33',
required : want_microhttpd == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libmicrohttpd.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libmicrohttpd = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_MICROHTTPD', have)
m4_defines += have ? ['-DHAVE_MICROHTTPD'] : []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_libcryptsetup = get_option('libcryptsetup')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_libcryptsetup != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libcryptsetup = dependency('libcryptsetup',
version : '>= 1.6.0',
required : want_libcryptsetup == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libcryptsetup.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libcryptsetup = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_LIBCRYPTSETUP', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_libcurl = get_option('libcurl')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_libcurl != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libcurl = dependency('libcurl',
version : '>= 7.32.0',
required : want_libcurl == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libcurl.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libcurl = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_LIBCURL', have)
m4_defines += have ? ['-DHAVE_LIBCURL'] : []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_libidn = get_option('libidn')
resolved: support libidn2 in addition to libidn libidn2 2.0.0 supports IDNA2008, in contrast to libidn which supports IDNA2003. https://bugzilla.redhat.com/show_bug.cgi?id=1449145 From that bug report: Internationalized domain names exist for quite some time (IDNA2003), although the protocols describing them have evolved in an incompatible way (IDNA2008). These incompatibilities will prevent applications written for IDNA2003 to access certain problematic domain names defined with IDNA2008, e.g., faß.de is translated to domain xn--fa-hia.de with IDNA2008, while in IDNA2003 it is translated to fass.de domain. That not only causes incompatibility problems, but may be used as an attack vector to redirect users to different web sites. v2: - keep libidn support - require libidn2 >= 2.0.0 v3: - keep dns_name_apply_idna caller dumb, and keep the #ifdefs inside of the function. - use both ±IDN and ±IDN2 in the version string
2017-05-10 04:56:34 +03:00
want_libidn2 = get_option('libidn2')
if want_libidn == 'true' and want_libidn2 == 'true'
error('libidn and libidn2 cannot be requested simultaneously')
endif
NEWS: say that libidn2 is experimental (#6335) Handling of "_" and some other details requires more thought: https://gitlab.com/libidn/libidn2/issues/30 Let's switch the default back to libidn and add a note in NEWS.
2017-07-12 10:25:59 +03:00
if want_libidn != 'false' and want_libidn2 != 'true'
libidn = dependency('libidn',
required : want_libidn == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libidn.found()
resolved: support libidn2 in addition to libidn libidn2 2.0.0 supports IDNA2008, in contrast to libidn which supports IDNA2003. https://bugzilla.redhat.com/show_bug.cgi?id=1449145 From that bug report: Internationalized domain names exist for quite some time (IDNA2003), although the protocols describing them have evolved in an incompatible way (IDNA2008). These incompatibilities will prevent applications written for IDNA2003 to access certain problematic domain names defined with IDNA2008, e.g., faß.de is translated to domain xn--fa-hia.de with IDNA2008, while in IDNA2003 it is translated to fass.de domain. That not only causes incompatibility problems, but may be used as an attack vector to redirect users to different web sites. v2: - keep libidn support - require libidn2 >= 2.0.0 v3: - keep dns_name_apply_idna caller dumb, and keep the #ifdefs inside of the function. - use both ±IDN and ±IDN2 in the version string
2017-05-10 04:56:34 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
resolved: support libidn2 in addition to libidn libidn2 2.0.0 supports IDNA2008, in contrast to libidn which supports IDNA2003. https://bugzilla.redhat.com/show_bug.cgi?id=1449145 From that bug report: Internationalized domain names exist for quite some time (IDNA2003), although the protocols describing them have evolved in an incompatible way (IDNA2008). These incompatibilities will prevent applications written for IDNA2003 to access certain problematic domain names defined with IDNA2008, e.g., faß.de is translated to domain xn--fa-hia.de with IDNA2008, while in IDNA2003 it is translated to fass.de domain. That not only causes incompatibility problems, but may be used as an attack vector to redirect users to different web sites. v2: - keep libidn support - require libidn2 >= 2.0.0 v3: - keep dns_name_apply_idna caller dumb, and keep the #ifdefs inside of the function. - use both ±IDN and ±IDN2 in the version string
2017-05-10 04:56:34 +03:00
libidn = []
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_LIBIDN', have)
m4_defines += have ? ['-DHAVE_LIBIDN'] : []
if not have and want_libidn2 != 'false'
NEWS: say that libidn2 is experimental (#6335) Handling of "_" and some other details requires more thought: https://gitlab.com/libidn/libidn2/issues/30 Let's switch the default back to libidn and add a note in NEWS.
2017-07-12 10:25:59 +03:00
# libidn is used for both libidn and libidn2 objects
libidn = dependency('libidn2',
required : want_libidn2 == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libidn.found()
else
have = false
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_LIBIDN2', have)
m4_defines += have ? ['-DHAVE_LIBIDN2'] : []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_libiptc = get_option('libiptc')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_libiptc != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libiptc = dependency('libiptc',
required : want_libiptc == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libiptc.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libiptc = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_LIBIPTC', have)
m4_defines += have ? ['-DHAVE_LIBIPTC'] : []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_qrencode = get_option('qrencode')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_qrencode != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libqrencode = dependency('libqrencode',
required : want_qrencode == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libqrencode.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libqrencode = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_QRENCODE', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
meson.build: reorder tests to match order in meson_options.txt This makes it easier to edit both files.
2017-07-26 21:08:46 +03:00
want_gcrypt = get_option('gcrypt')
if want_gcrypt != 'false'
libgcrypt = cc.find_library('gcrypt', required : want_gcrypt == 'true')
libgpg_error = cc.find_library('gpg-error', required : want_gcrypt == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libgcrypt.found() and libgpg_error.found()
meson.build: reorder tests to match order in meson_options.txt This makes it easier to edit both files.
2017-07-26 21:08:46 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
endif
if not have
# link to neither of the libs if one is not found
meson.build: reorder tests to match order in meson_options.txt This makes it easier to edit both files.
2017-07-26 21:08:46 +03:00
libgcrypt = []
libgpg_error = []
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_GCRYPT', have)
meson.build: reorder tests to match order in meson_options.txt This makes it easier to edit both files.
2017-07-26 21:08:46 +03:00
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_gnutls = get_option('gnutls')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_gnutls != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libgnutls = dependency('gnutls',
version : '>= 3.1.4',
required : want_gnutls == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libgnutls.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libgnutls = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_GNUTLS', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_elfutils = get_option('elfutils')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_elfutils != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libdw = dependency('libdw',
required : want_elfutils == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libdw.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libdw = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_ELFUTILS', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_zlib = get_option('zlib')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_zlib != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libz = dependency('zlib',
required : want_zlib == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libz.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libz = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_ZLIB', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_bzip2 = get_option('bzip2')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_bzip2 != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libbzip2 = cc.find_library('bz2',
required : want_bzip2 == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libbzip2.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libbzip2 = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_BZIP2', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_xz = get_option('xz')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_xz != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libxz = dependency('liblzma',
required : want_xz == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libxz.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libxz = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_XZ', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_lz4 = get_option('lz4')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_lz4 != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
liblz4 = dependency('liblz4',
required : want_lz4 == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = liblz4.found()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
liblz4 = []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_LZ4', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
meson.build: reorder tests to match order in meson_options.txt This makes it easier to edit both files.
2017-07-26 21:08:46 +03:00
want_xkbcommon = get_option('xkbcommon')
if want_xkbcommon != 'false'
libxkbcommon = dependency('xkbcommon',
version : '>= 0.3.0',
required : want_xkbcommon == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libxkbcommon.found()
meson.build: reorder tests to match order in meson_options.txt This makes it easier to edit both files.
2017-07-26 21:08:46 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson.build: reorder tests to match order in meson_options.txt This makes it easier to edit both files.
2017-07-26 21:08:46 +03:00
libxkbcommon = []
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_XKBCOMMON', have)
meson.build: reorder tests to match order in meson_options.txt This makes it easier to edit both files.
2017-07-26 21:08:46 +03:00
meson: define tests Tests can be run with 'ninja-build test' or using 'mesontest'. '-Dtests=unsafe' can be used to include the "unsafe" tests in the test suite, same as with autotools. v2: - use more conf.get guards are optional components - declare deps on generated headers for test-{af,arphrd,cap}-list v3: - define environment for tests Most test don't need this, but to be consistent with autotools-based build, and to avoid questions which tests need it and which don't, set the same environment for all tests. v4: - rework test generation Use a list of lists to define each test. This way we can reduce the boilerplate somewhat, although the test listings are still pretty verbose. We can also move the definitions of the tests to the subdirs. Unfortunately some subdirs are included earlier than some of the libraries that test binaries are linked to. So just dump all definitions of all tests that cannot be defined earlier into src/test. The `executable` definitions are still at the top level, so the binaries are compiled into the build root. v5: - tag test-dnssec-complex as manual v6: - fix HAVE_LIBZ typo - add missing libgobject/libgio defs - mark test-qcow2 as manual
2017-04-07 07:19:09 +03:00
want_glib = get_option('glib')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_glib != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libglib = dependency('glib-2.0',
version : '>= 2.22.0',
required : want_glib == 'true')
libgobject = dependency('gobject-2.0',
version : '>= 2.22.0',
required : want_glib == 'true')
libgio = dependency('gio-2.0',
required : want_glib == 'true')
meson: use booleans for conf.set and drop unecessary conditionals Using conf.set() with a boolean argument does the right thing: either #ifdef or #undef. This means that conf.set can be used unconditionally. Previously I used '1' as the placeholder value, and that needs to be changed to 'true' for consistency (under meson 1 cannot be used in boolean context). All checks need to be adjusted.
2017-04-28 04:13:08 +03:00
have = libglib.found() and libgobject.found() and libgio.found()
meson: define tests Tests can be run with 'ninja-build test' or using 'mesontest'. '-Dtests=unsafe' can be used to include the "unsafe" tests in the test suite, same as with autotools. v2: - use more conf.get guards are optional components - declare deps on generated headers for test-{af,arphrd,cap}-list v3: - define environment for tests Most test don't need this, but to be consistent with autotools-based build, and to avoid questions which tests need it and which don't, set the same environment for all tests. v4: - rework test generation Use a list of lists to define each test. This way we can reduce the boilerplate somewhat, although the test listings are still pretty verbose. We can also move the definitions of the tests to the subdirs. Unfortunately some subdirs are included earlier than some of the libraries that test binaries are linked to. So just dump all definitions of all tests that cannot be defined earlier into src/test. The `executable` definitions are still at the top level, so the binaries are compiled into the build root. v5: - tag test-dnssec-complex as manual v6: - fix HAVE_LIBZ typo - add missing libgobject/libgio defs - mark test-qcow2 as manual
2017-04-07 07:19:09 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libglib = []
libgobject = []
libgio = []
meson: define tests Tests can be run with 'ninja-build test' or using 'mesontest'. '-Dtests=unsafe' can be used to include the "unsafe" tests in the test suite, same as with autotools. v2: - use more conf.get guards are optional components - declare deps on generated headers for test-{af,arphrd,cap}-list v3: - define environment for tests Most test don't need this, but to be consistent with autotools-based build, and to avoid questions which tests need it and which don't, set the same environment for all tests. v4: - rework test generation Use a list of lists to define each test. This way we can reduce the boilerplate somewhat, although the test listings are still pretty verbose. We can also move the definitions of the tests to the subdirs. Unfortunately some subdirs are included earlier than some of the libraries that test binaries are linked to. So just dump all definitions of all tests that cannot be defined earlier into src/test. The `executable` definitions are still at the top level, so the binaries are compiled into the build root. v5: - tag test-dnssec-complex as manual v6: - fix HAVE_LIBZ typo - add missing libgobject/libgio defs - mark test-qcow2 as manual
2017-04-07 07:19:09 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_GLIB', have)
meson: define tests Tests can be run with 'ninja-build test' or using 'mesontest'. '-Dtests=unsafe' can be used to include the "unsafe" tests in the test suite, same as with autotools. v2: - use more conf.get guards are optional components - declare deps on generated headers for test-{af,arphrd,cap}-list v3: - define environment for tests Most test don't need this, but to be consistent with autotools-based build, and to avoid questions which tests need it and which don't, set the same environment for all tests. v4: - rework test generation Use a list of lists to define each test. This way we can reduce the boilerplate somewhat, although the test listings are still pretty verbose. We can also move the definitions of the tests to the subdirs. Unfortunately some subdirs are included earlier than some of the libraries that test binaries are linked to. So just dump all definitions of all tests that cannot be defined earlier into src/test. The `executable` definitions are still at the top level, so the binaries are compiled into the build root. v5: - tag test-dnssec-complex as manual v6: - fix HAVE_LIBZ typo - add missing libgobject/libgio defs - mark test-qcow2 as manual
2017-04-07 07:19:09 +03:00
want_dbus = get_option('dbus')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_dbus != 'false'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libdbus = dependency('dbus-1',
version : '>= 1.3.2',
required : want_dbus == 'true')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = libdbus.found()
meson: define tests Tests can be run with 'ninja-build test' or using 'mesontest'. '-Dtests=unsafe' can be used to include the "unsafe" tests in the test suite, same as with autotools. v2: - use more conf.get guards are optional components - declare deps on generated headers for test-{af,arphrd,cap}-list v3: - define environment for tests Most test don't need this, but to be consistent with autotools-based build, and to avoid questions which tests need it and which don't, set the same environment for all tests. v4: - rework test generation Use a list of lists to define each test. This way we can reduce the boilerplate somewhat, although the test listings are still pretty verbose. We can also move the definitions of the tests to the subdirs. Unfortunately some subdirs are included earlier than some of the libraries that test binaries are linked to. So just dump all definitions of all tests that cannot be defined earlier into src/test. The `executable` definitions are still at the top level, so the binaries are compiled into the build root. v5: - tag test-dnssec-complex as manual v6: - fix HAVE_LIBZ typo - add missing libgobject/libgio defs - mark test-qcow2 as manual
2017-04-07 07:19:09 +03:00
else
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = false
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libdbus = []
meson: define tests Tests can be run with 'ninja-build test' or using 'mesontest'. '-Dtests=unsafe' can be used to include the "unsafe" tests in the test suite, same as with autotools. v2: - use more conf.get guards are optional components - declare deps on generated headers for test-{af,arphrd,cap}-list v3: - define environment for tests Most test don't need this, but to be consistent with autotools-based build, and to avoid questions which tests need it and which don't, set the same environment for all tests. v4: - rework test generation Use a list of lists to define each test. This way we can reduce the boilerplate somewhat, although the test listings are still pretty verbose. We can also move the definitions of the tests to the subdirs. Unfortunately some subdirs are included earlier than some of the libraries that test binaries are linked to. So just dump all definitions of all tests that cannot be defined earlier into src/test. The `executable` definitions are still at the top level, so the binaries are compiled into the build root. v5: - tag test-dnssec-complex as manual v6: - fix HAVE_LIBZ typo - add missing libgobject/libgio defs - mark test-qcow2 as manual
2017-04-07 07:19:09 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('HAVE_DBUS', have)
meson: define tests Tests can be run with 'ninja-build test' or using 'mesontest'. '-Dtests=unsafe' can be used to include the "unsafe" tests in the test suite, same as with autotools. v2: - use more conf.get guards are optional components - declare deps on generated headers for test-{af,arphrd,cap}-list v3: - define environment for tests Most test don't need this, but to be consistent with autotools-based build, and to avoid questions which tests need it and which don't, set the same environment for all tests. v4: - rework test generation Use a list of lists to define each test. This way we can reduce the boilerplate somewhat, although the test listings are still pretty verbose. We can also move the definitions of the tests to the subdirs. Unfortunately some subdirs are included earlier than some of the libraries that test binaries are linked to. So just dump all definitions of all tests that cannot be defined earlier into src/test. The `executable` definitions are still at the top level, so the binaries are compiled into the build root. v5: - tag test-dnssec-complex as manual v6: - fix HAVE_LIBZ typo - add missing libgobject/libgio defs - mark test-qcow2 as manual
2017-04-07 07:19:09 +03:00
resolved: ignore DNSSEC= option when resolved is built without gcrypt (#6084) Fixes #5583.
2017-06-17 23:22:32 +03:00
default_dnssec = get_option('default-dnssec')
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if default_dnssec != 'no' and conf.get('HAVE_GCRYPT') == 0
resolved: ignore DNSSEC= option when resolved is built without gcrypt (#6084) Fixes #5583.
2017-06-17 23:22:32 +03:00
message('default-dnssec cannot be set to yes or allow-downgrade when gcrypt is disabled. Setting default-dnssec to no.')
default_dnssec = 'no'
endif
conf.set('DEFAULT_DNSSEC_MODE',
'DNSSEC_' + default_dnssec.underscorify().to_upper())
substs.set('DEFAULT_DNSSEC_MODE', default_dnssec)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_importd = get_option('importd')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_importd != 'false'
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = (conf.get('HAVE_LIBCURL') == 1 and
conf.get('HAVE_ZLIB') == 1 and
conf.get('HAVE_BZIP2') == 1 and
conf.get('HAVE_XZ') == 1 and
conf.get('HAVE_GCRYPT') == 1)
if want_importd == 'true' and not have
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
error('importd support was requested, but dependencies are not available')
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
else
have = false
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('ENABLE_IMPORTD', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
want_remote = get_option('remote')
meson: replace yes/no/auto with true/false/auto This doesn't feel as natural, but is more consistent with the boolean options which require true/false, and allows setting of options without knowing of which type the option is.
2017-04-14 03:30:07 +03:00
if want_remote != 'false'
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have_deps = [conf.get('HAVE_MICROHTTPD') == 1,
conf.get('HAVE_LIBCURL') == 1]
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
# sd-j-remote requires µhttpd, and sd-j-upload requires libcurl, so
# it's possible to build one without the other. Complain only if
# support was explictly requested. The auxiliary files like sysusers
# config should be installed when any of the programs are built.
if want_remote == 'true' and not (have_deps[0] and have_deps[1])
error('remote support was requested, but dependencies are not available')
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = have_deps[0] or have_deps[1]
else
have = false
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('ENABLE_REMOTE', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
meson: generate ENABLE_* names automatically After previous changes, the naming of configuration options and internal defines is consistent.
2017-10-03 14:15:27 +03:00
foreach term : ['utmp',
'hibernate',
'environment-d',
'binfmt',
'coredump',
'resolve',
'logind',
'hostnamed',
'localed',
'machined',
'networkd',
'timedated',
'timesyncd',
'myhostname',
'firstboot',
'randomseed',
'backlight',
'vconsole',
'quotacheck',
'sysusers',
'tmpfiles',
'hwdb',
'rfkill',
'ldconfig',
'efi',
'tpm',
'ima',
'smack',
'gshadow',
'idn',
'nss-systemd']
have = get_option(term)
name = 'ENABLE_' + term.underscorify().to_upper()
conf.set10(name, have)
m4_defines += have ? ['-D' + name] : []
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endforeach
meson: define tests Tests can be run with 'ninja-build test' or using 'mesontest'. '-Dtests=unsafe' can be used to include the "unsafe" tests in the test suite, same as with autotools. v2: - use more conf.get guards are optional components - declare deps on generated headers for test-{af,arphrd,cap}-list v3: - define environment for tests Most test don't need this, but to be consistent with autotools-based build, and to avoid questions which tests need it and which don't, set the same environment for all tests. v4: - rework test generation Use a list of lists to define each test. This way we can reduce the boilerplate somewhat, although the test listings are still pretty verbose. We can also move the definitions of the tests to the subdirs. Unfortunately some subdirs are included earlier than some of the libraries that test binaries are linked to. So just dump all definitions of all tests that cannot be defined earlier into src/test. The `executable` definitions are still at the top level, so the binaries are compiled into the build root. v5: - tag test-dnssec-complex as manual v6: - fix HAVE_LIBZ typo - add missing libgobject/libgio defs - mark test-qcow2 as manual
2017-04-07 07:19:09 +03:00
want_tests = get_option('tests')
meson: add option to install tests Instead of a separate target as with make, it is decided at configure time whether to install tests as part of the install target. The effect should be the same.
2017-04-08 08:55:38 +03:00
install_tests = get_option('install-tests')
meson: define tests Tests can be run with 'ninja-build test' or using 'mesontest'. '-Dtests=unsafe' can be used to include the "unsafe" tests in the test suite, same as with autotools. v2: - use more conf.get guards are optional components - declare deps on generated headers for test-{af,arphrd,cap}-list v3: - define environment for tests Most test don't need this, but to be consistent with autotools-based build, and to avoid questions which tests need it and which don't, set the same environment for all tests. v4: - rework test generation Use a list of lists to define each test. This way we can reduce the boilerplate somewhat, although the test listings are still pretty verbose. We can also move the definitions of the tests to the subdirs. Unfortunately some subdirs are included earlier than some of the libraries that test binaries are linked to. So just dump all definitions of all tests that cannot be defined earlier into src/test. The `executable` definitions are still at the top level, so the binaries are compiled into the build root. v5: - tag test-dnssec-complex as manual v6: - fix HAVE_LIBZ typo - add missing libgobject/libgio defs - mark test-qcow2 as manual
2017-04-07 07:19:09 +03:00
tests = []
meson: add option to configure "slow" tests on/off This is just the meson part, no functional change. Use meson -D slow-tests=yes to set the default, or SYSTEMD_SLOW_TESTS=yes build/test-foobar for just one test. Setting the default is more useful for installed tests.
2017-07-13 00:25:17 +03:00
conf.set10('SYSTEMD_SLOW_TESTS_DEFAULT', get_option('slow-tests'))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
#####################################################################
if get_option('efi')
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
efi_arch = host_machine.cpu_family()
meson: add support for building efi modules This is a very straightforward conversion of the rules in Makefile.am. Generated objects (on arm64) are identical. The only difference in executed commands is that automake uses ld -m elf_x86_64, without us specifying the -m option anywhere. I suspect that using the default for the given linker should be OK, so it's fine to just skip it.
2017-04-12 22:05:55 +03:00
meson: define gnu_efi_arch for the arch efi name, fix ldsdir detection This fixes ldsdir detection under Debian. v2: - define gnu_efi_arch for the arch efi include directory name In the autotools naming convention, efi_arch and this directory always had the same name. But meson.cpu_family() uses a slightly different convention, so those two don't always match.
2017-04-20 05:57:52 +03:00
if efi_arch == 'x86'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
EFI_MACHINE_TYPE_NAME = 'ia32'
meson: define gnu_efi_arch for the arch efi name, fix ldsdir detection This fixes ldsdir detection under Debian. v2: - define gnu_efi_arch for the arch efi include directory name In the autotools naming convention, efi_arch and this directory always had the same name. But meson.cpu_family() uses a slightly different convention, so those two don't always match.
2017-04-20 05:57:52 +03:00
gnu_efi_arch = 'ia32'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
elif efi_arch == 'x86_64'
EFI_MACHINE_TYPE_NAME = 'x64'
meson: define gnu_efi_arch for the arch efi name, fix ldsdir detection This fixes ldsdir detection under Debian. v2: - define gnu_efi_arch for the arch efi include directory name In the autotools naming convention, efi_arch and this directory always had the same name. But meson.cpu_family() uses a slightly different convention, so those two don't always match.
2017-04-20 05:57:52 +03:00
gnu_efi_arch = 'x86_64'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
elif efi_arch == 'arm'
EFI_MACHINE_TYPE_NAME = 'arm'
meson: define gnu_efi_arch for the arch efi name, fix ldsdir detection This fixes ldsdir detection under Debian. v2: - define gnu_efi_arch for the arch efi include directory name In the autotools naming convention, efi_arch and this directory always had the same name. But meson.cpu_family() uses a slightly different convention, so those two don't always match.
2017-04-20 05:57:52 +03:00
gnu_efi_arch = 'arm'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
elif efi_arch == 'aarch64'
EFI_MACHINE_TYPE_NAME = 'aa64'
meson: define gnu_efi_arch for the arch efi name, fix ldsdir detection This fixes ldsdir detection under Debian. v2: - define gnu_efi_arch for the arch efi include directory name In the autotools naming convention, efi_arch and this directory always had the same name. But meson.cpu_family() uses a slightly different convention, so those two don't always match.
2017-04-20 05:57:52 +03:00
gnu_efi_arch = 'aarch64'
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
else
EFI_MACHINE_TYPE_NAME = ''
meson: define gnu_efi_arch for the arch efi name, fix ldsdir detection This fixes ldsdir detection under Debian. v2: - define gnu_efi_arch for the arch efi include directory name In the autotools naming convention, efi_arch and this directory always had the same name. But meson.cpu_family() uses a slightly different convention, so those two don't always match.
2017-04-20 05:57:52 +03:00
gnu_efi_arch = ''
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
endif
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
have = true
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
conf.set_quoted('EFI_MACHINE_TYPE_NAME', EFI_MACHINE_TYPE_NAME)
meson: add forgotten tpm support
2017-04-25 02:28:04 +03:00
conf.set('SD_TPM_PCR', get_option('tpm-pcrindex').to_int())
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
else
have = false
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
conf.set10('ENABLE_EFI', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
#####################################################################
config_h = configure_file(
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
output : 'config.h',
configuration : conf)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
includes = include_directories('src/basic',
'src/shared',
'src/systemd',
'src/journal',
'src/resolve',
'src/timesync',
'src/login',
'src/udev',
'src/libudev',
'src/core',
'src/libsystemd/sd-bus',
'src/libsystemd/sd-device',
'src/libsystemd/sd-hwdb',
'src/libsystemd/sd-id128',
'src/libsystemd/sd-netlink',
'src/libsystemd/sd-network',
'src/libsystemd-network',
meson: add . to include_directories (#6634)
2017-08-30 18:34:44 +03:00
'.',
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
)
add_project_arguments('-include', 'config.h', language : 'c')
gcrypt_util_sources = files('src/shared/gcrypt-util.h',
'src/shared/gcrypt-util.c')
subdir('po')
subdir('catalog')
subdir('src/systemd')
subdir('src/basic')
subdir('src/libsystemd')
subdir('src/libsystemd-network')
subdir('src/journal')
subdir('src/login')
libjournal_core = static_library(
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
'journal-core',
libjournal_core_sources,
journald_gperf_c,
include_directories : includes,
install : false)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
meson: add test-libsystemd-sym, fix linking of libsystemd This is quite messy. I think libtool might have been using something like -Wl,--whole-archive, but I don't think meson has support for that. For now, just recompile all the sources for linking into libsystemd directly. This should not matter much for efficiency, since it's a few small files.
2017-04-10 21:13:40 +03:00
libsystemd_sym_path = '@0@/@1@'.format(meson.current_source_dir(), libsystemd_sym)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
libsystemd = shared_library(
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
'systemd',
libsystemd_internal_sources,
journal_internal_sources,
meson: move library version defines to the top (#6939)
2017-09-28 20:24:16 +03:00
version : libsystemd_version,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
include_directories : includes,
link_args : ['-shared',
'-Wl,--version-script=' + libsystemd_sym_path],
link_with : [libbasic],
dependencies : [threads,
meson: fix gcrypt config option Also detect libgpg-error. Require both to be present for HAVE_CRYPT, even though libgpg-error is only used in src/resolve. If one is available, the other should be too, so it doesn't seem worth the trouble to make two separate conditions.
2017-04-22 00:45:54 +03:00
libgcrypt,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
librt,
libxz,
liblz4],
link_depends : libsystemd_sym,
install : true,
install_dir : rootlibdir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
############################################################
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-15 03:10:28 +03:00
# binaries that have --help and are intended for use by humans,
# usually, but not always, installed in /bin.
public_programs = []
subdir('src/libudev')
subdir('src/shared')
subdir('src/core')
subdir('src/udev')
subdir('src/network')
subdir('src/analyze')
subdir('src/journal-remote')
subdir('src/coredump')
subdir('src/hostname')
subdir('src/import')
subdir('src/kernel-install')
subdir('src/locale')
subdir('src/machine')
subdir('src/nspawn')
subdir('src/resolve')
subdir('src/timedate')
subdir('src/timesync')
subdir('src/vconsole')
subdir('src/boot/efi')
subdir('src/test')
meson: re-attach rule-syntax-check.py test 39/248 rule-syntax-check OK 0.07 s --- command --- /home/zbyszek/src/systemd-work/test/rule-syntax-check.py \ /home/zbyszek/src/systemd-work/build/../rules/60-block.rules \ /home/zbyszek/src/systemd-work/build/../rules/60-cdrom_id.rules \ /home/zbyszek/src/systemd-work/build/../rules/60-drm.rules \ /home/zbyszek/src/systemd-work/build/../rules/60-evdev.rules \ /home/zbyszek/src/systemd-work/build/../rules/60-input-id.rules \ /home/zbyszek/src/systemd-work/build/../rules/60-persistent-alsa.rules \ /home/zbyszek/src/systemd-work/build/../rules/60-persistent-input.rules \ /home/zbyszek/src/systemd-work/build/../rules/60-persistent-storage.rules \ /home/zbyszek/src/systemd-work/build/../rules/60-persistent-storage-tape.rules \ /home/zbyszek/src/systemd-work/build/../rules/60-persistent-v4l.rules \ /home/zbyszek/src/systemd-work/build/../rules/60-sensor.rules \ /home/zbyszek/src/systemd-work/build/../rules/60-serial.rules \ /home/zbyszek/src/systemd-work/build/../rules/70-joystick.rules \ /home/zbyszek/src/systemd-work/build/../rules/70-mouse.rules \ /home/zbyszek/src/systemd-work/build/../rules/70-touchpad.rules \ /home/zbyszek/src/systemd-work/build/../rules/75-net-description.rules \ /home/zbyszek/src/systemd-work/build/../rules/75-probe_mtd.rules \ /home/zbyszek/src/systemd-work/build/../rules/78-sound-card.rules \ /home/zbyszek/src/systemd-work/build/../rules/80-drivers.rules \ /home/zbyszek/src/systemd-work/build/../rules/80-net-setup-link.rules \ /home/zbyszek/src/systemd-work/build/rules/50-udev-default.rules \ /home/zbyszek/src/systemd-work/build/rules/64-btrfs.rules \ /home/zbyszek/src/systemd-work/build/rules/99-systemd.rules --- stdout --- ... ------- It got dropped by mistake in 72cdb3e783174dcf9223a49f03e3b0e2ca95ddb8.
2017-11-22 14:42:28 +03:00
subdir('rules')
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-15 03:10:28 +03:00
subdir('test')
############################################################
# only static linking apart from libdl, to make sure that the
# module is linked to all libraries that it uses.
test_dlopen = executable(
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
'test-dlopen',
test_dlopen_c,
include_directories : includes,
link_with : [libbasic],
dependencies : [libdl])
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-15 03:10:28 +03:00
build-sys: s/HAVE_MYHOSTNAME/ENABLE_MYHOSTNAME/ Same justification as for HAVE_UTMP. HAVE_MYHOSTNAME was used before mysthostname was merged into systemd.
2017-10-03 13:23:55 +03:00
foreach tuple : [['myhostname', 'ENABLE_MYHOSTNAME'],
make nss-systemd support conditional (#6155) This allows the nss-systemd module to be disabled on minimal installations.
2017-06-24 20:30:26 +03:00
['systemd', 'ENABLE_NSS_SYSTEMD'],
nss-resolve: drop the internal fallback to libnss_dns (#5945) If we could not communicate with systemd-resolved, we would call into libnss_dns. libnss_dns would return NOTFOUND for stuff like "localhost" and other names resolved by nss-myhostname, which we would fall under the !UNAVAIL= condition and cause resolution to fail. So the following recommended configuration in nsswitch.conf would not work: hosts: resolve [!UNAVAIL=return] dns myhostname Remove the internal fallback code completely so that the fallback logic can be configured in nsswitch.conf. Tested with hosts: resolve [!UNAVAIL=return] myhostname and hosts: resolve [!UNAVAIL=return] dns myhostname Fixes #5742.
2017-05-12 15:31:46 +03:00
['mymachines', 'ENABLE_MACHINED'],
build-sys: s/ENABLE_RESOLVED/ENABLE_RESOLVE/ The configuration option was called -Dresolve, but the internal define was …RESOLVED. This options governs more than just resolved itself, so let's settle on the version without "d".
2017-10-03 14:12:29 +03:00
['resolve', 'ENABLE_RESOLVE']]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
condition = tuple[1] == '' or conf.get(tuple[1]) == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
if condition
module = tuple[0]
sym = 'src/nss-@0@/nss-@0@.sym'.format(module)
version_script_arg = join_paths(meson.current_source_dir(), sym)
nss = shared_library(
'nss_' + module,
'src/nss-@0@/nss-@0@.c'.format(module),
version : '2',
include_directories : includes,
meson: link NSS modules with -z nodelete (#7607) We might end up allocating mempools, and when we are unloaded we might orphan them, thus leaking them. Hence, let's just stick around for good, so the mempools remain referenced continously and for good, and thus no memory is leaked (though the memory isn't cleaned up either). Fixes: #7596
2017-12-12 22:13:16 +03:00
# Note that we link NSS modules with '-z nodelete' so that mempools never get orphaned
link_args : ['-Wl,-z,nodelete',
'-shared',
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
'-Wl,--version-script=' + version_script_arg,
'-Wl,--undefined'],
link_with : [libsystemd_internal,
libbasic],
dependencies : [threads,
nss-resolve: drop the internal fallback to libnss_dns (#5945) If we could not communicate with systemd-resolved, we would call into libnss_dns. libnss_dns would return NOTFOUND for stuff like "localhost" and other names resolved by nss-myhostname, which we would fall under the !UNAVAIL= condition and cause resolution to fail. So the following recommended configuration in nsswitch.conf would not work: hosts: resolve [!UNAVAIL=return] dns myhostname Remove the internal fallback code completely so that the fallback logic can be configured in nsswitch.conf. Tested with hosts: resolve [!UNAVAIL=return] myhostname and hosts: resolve [!UNAVAIL=return] dns myhostname Fixes #5742.
2017-05-12 15:31:46 +03:00
librt],
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
link_depends : sym,
install : true,
install_dir : rootlibdir)
# We cannot use shared_module because it does not support version suffix.
# Unfortunately shared_library insists on creating the symlink…
meson.add_install_script('sh', '-c',
'rm $DESTDIR@0@/libnss_@1@.so'
.format(rootlibdir, module))
test('dlopen-nss_' + module,
test_dlopen,
args : [nss.full_path()]) # path to dlopen must include a slash
endif
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endforeach
############################################################
executable('systemd',
systemd_sources,
include_directories : includes,
link_with : [libcore,
meson: drop implicit libudev link dependencies Executables which link against libshared do not need an explicit dependency on libudev, as libshared will make sure that those symbols are available.
2017-04-25 21:19:54 +03:00
libshared],
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
dependencies : [threads,
librt,
libseccomp,
libselinux,
meson: create a static version of libshared and link libcore and some tests to it This is what autoconf-based build does, and it makes test-bus-error and test-engine able to access the bus error mapping table. OTOH, this is a heavy price to pay: it would be excellent to link libcore.a to libsystemd-shared-NNN.so. Otherwise we duplicate the same code in 'systemd' and 'libsystemd-shared-NNN.so'. -rwxrwxr-x. 1 4075544 Apr 6 20:30 systemd* <-- libcore linked against libsystemd-shared.so -rwxrwxr-x. 1 5596504 Apr 9 14:07 systemd* <-- libcore linked against libsystemd-shared.a v2: - update for 6b5cf3ea621a5bfd218cd2249e066a52c0e73657
2017-04-09 21:08:53 +03:00
libmount,
libblkid],
meson: replace $ORIGIN with rootlibdir everywhere $ORIGIN does not work when /proc is not mounted. So the systemd binary cannot use $ORIGIN. Use the full path everywhere for consistency.
2017-04-12 19:02:30 +03:00
install_rpath : rootlibexecdir,
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
install : true,
install_dir : rootlibexecdir)
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
exe = executable('systemd-analyze',
systemd_analyze_sources,
include_directories : includes,
link_with : [libcore,
libshared],
dependencies : [threads,
librt,
libseccomp,
libselinux,
libmount,
libblkid],
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
executable('systemd-journald',
systemd_journald_sources,
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libjournal_core,
meson: drop implicit libudev link dependencies Executables which link against libshared do not need an explicit dependency on libudev, as libshared will make sure that those symbols are available.
2017-04-25 21:19:54 +03:00
libshared],
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
dependencies : [threads,
libxz,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
liblz4,
libselinux],
meson: replace $ORIGIN with rootlibdir everywhere $ORIGIN does not work when /proc is not mounted. So the systemd binary cannot use $ORIGIN. Use the full path everywhere for consistency.
2017-04-12 19:02:30 +03:00
install_rpath : rootlibexecdir,
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
install : true,
install_dir : rootlibexecdir)
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
exe = executable('systemd-cat',
systemd_cat_sources,
include_directories : includes,
link_with : [libjournal_core,
meson: drop implicit libudev link dependencies Executables which link against libshared do not need an explicit dependency on libudev, as libshared will make sure that those symbols are available.
2017-04-25 21:19:54 +03:00
libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
dependencies : [threads],
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
exe = executable('journalctl',
journalctl_sources,
include_directories : includes,
meson: drop implicit libudev link dependencies Executables which link against libshared do not need an explicit dependency on libudev, as libshared will make sure that those symbols are available.
2017-04-25 21:19:54 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
dependencies : [threads,
libqrencode,
libxz,
liblz4],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootbindir)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
executable('systemd-getty-generator',
'src/getty-generator/getty-generator.c',
include_directories : includes,
meson: specify rpath on all executables linking to libshared (i.e. pretty much all of them). libtools does this automatically for us. It would be great if meson did too. Otherwise, it would be nice to simply attach the information about necessary rpath to the library object, and have it used automatically for all users. Dunno, maybe there's already a way to do this. I used '$ORIGIN' for stuff installed into libexecdir, since that's where libshared ends up, and a full path for everything else. The advantage of using $ORIGIN is that those binaries can be started from $DESTDIR, which can sometimes be useful.
2017-04-11 01:13:00 +03:00
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : systemgeneratordir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
executable('systemd-debug-generator',
'src/debug-generator/debug-generator.c',
include_directories : includes,
meson: specify rpath on all executables linking to libshared (i.e. pretty much all of them). libtools does this automatically for us. It would be great if meson did too. Otherwise, it would be nice to simply attach the information about necessary rpath to the library object, and have it used automatically for all users. Dunno, maybe there's already a way to do this. I used '$ORIGIN' for stuff installed into libexecdir, since that's where libshared ends up, and a full path for everything else. The advantage of using $ORIGIN is that those binaries can be started from $DESTDIR, which can sometimes be useful.
2017-04-11 01:13:00 +03:00
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : systemgeneratordir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
executable('systemd-fstab-generator',
'src/fstab-generator/fstab-generator.c',
'src/core/mount-setup.c',
include_directories : includes,
meson: specify rpath on all executables linking to libshared (i.e. pretty much all of them). libtools does this automatically for us. It would be great if meson did too. Otherwise, it would be nice to simply attach the information about necessary rpath to the library object, and have it used automatically for all users. Dunno, maybe there's already a way to do this. I used '$ORIGIN' for stuff installed into libexecdir, since that's where libshared ends up, and a full path for everything else. The advantage of using $ORIGIN is that those binaries can be started from $DESTDIR, which can sometimes be useful.
2017-04-11 01:13:00 +03:00
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : systemgeneratordir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_ENVIRONMENT_D') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('30-systemd-environment-d-generator',
'src/environment-d-generator/environment-d-generator.c',
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : userenvgeneratordir)
meson: create various symlinks v2: - remove bashisms
2017-04-10 06:55:50 +03:00
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
meson.add_install_script(meson_make_symlink,
join_paths(sysconfdir, 'environment'),
join_paths(environmentdir, '99-environment.conf'))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_HIBERNATE') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-hibernate-resume-generator',
'src/hibernate-resume/hibernate-resume-generator.c',
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : systemgeneratordir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-hibernate-resume',
'src/hibernate-resume/hibernate-resume.c',
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('HAVE_BLKID') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-gpt-auto-generator',
'src/gpt-auto-generator/gpt-auto-generator.c',
'src/basic/blkid-util.h',
include_directories : includes,
meson: drop implicit libudev link dependencies Executables which link against libshared do not need an explicit dependency on libudev, as libshared will make sure that those symbols are available.
2017-04-25 21:19:54 +03:00
link_with : [libshared],
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
dependencies : libblkid,
install_rpath : rootlibexecdir,
install : true,
install_dir : systemgeneratordir)
exe = executable('systemd-dissect',
'src/dissect/dissect.c',
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: s/ENABLE_RESOLVED/ENABLE_RESOLVE/ The configuration option was called -Dresolve, but the internal define was …RESOLVED. This options governs more than just resolved itself, so let's settle on the version without "d".
2017-10-03 14:12:29 +03:00
if conf.get('ENABLE_RESOLVE') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-resolved',
systemd_resolved_sources,
meson: fix gcrypt config option Also detect libgpg-error. Require both to be present for HAVE_CRYPT, even though libgpg-error is only used in src/resolve. If one is available, the other should be too, so it doesn't seem worth the trouble to make two separate conditions.
2017-04-22 00:45:54 +03:00
gcrypt_util_sources,
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
dependencies : [threads,
meson: fix gcrypt config option Also detect libgpg-error. Require both to be present for HAVE_CRYPT, even though libgpg-error is only used in src/resolve. If one is available, the other should be too, so it doesn't seem worth the trouble to make two separate conditions.
2017-04-22 00:45:54 +03:00
libgcrypt,
libgpg_error,
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
libm,
libidn],
install_rpath : rootlibexecdir,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
install : true,
install_dir : rootlibexecdir)
exe = executable('systemd-resolve',
systemd_resolve_sources,
meson: fix gcrypt config option Also detect libgpg-error. Require both to be present for HAVE_CRYPT, even though libgpg-error is only used in src/resolve. If one is available, the other should be too, so it doesn't seem worth the trouble to make two separate conditions.
2017-04-22 00:45:54 +03:00
gcrypt_util_sources,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
include_directories : includes,
link_with : [libshared],
dependencies : [threads,
meson: fix gcrypt config option Also detect libgpg-error. Require both to be present for HAVE_CRYPT, even though libgpg-error is only used in src/resolve. If one is available, the other should be too, so it doesn't seem worth the trouble to make two separate conditions.
2017-04-22 00:45:54 +03:00
libgcrypt,
libgpg_error,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libm,
libidn],
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_LOGIND') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-logind',
systemd_logind_sources,
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
include_directories : includes,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
link_with : [liblogind_core,
meson: drop implicit libudev link dependencies Executables which link against libshared do not need an explicit dependency on libudev, as libshared will make sure that those symbols are available.
2017-04-25 21:19:54 +03:00
libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
dependencies : [threads,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libacl],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install_rpath : rootlibexecdir,
install : true,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
install_dir : rootlibexecdir)
exe = executable('loginctl',
loginctl_sources,
include_directories : includes,
meson: drop implicit libudev link dependencies Executables which link against libshared do not need an explicit dependency on libudev, as libshared will make sure that those symbols are available.
2017-04-25 21:19:54 +03:00
link_with : [libshared],
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
dependencies : [threads,
liblz4,
libxz],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootbindir)
public_programs += [exe]
exe = executable('systemd-inhibit',
'src/login/inhibit.c',
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootbindir)
public_programs += [exe]
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('HAVE_PAM') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
version_script_arg = join_paths(meson.current_source_dir(), pam_systemd_sym)
pam_systemd = shared_library(
'pam_systemd',
pam_systemd_c,
name_prefix : '',
include_directories : includes,
link_args : ['-shared',
'-Wl,--version-script=' + version_script_arg],
link_with : [libsystemd_internal,
libshared_static],
dependencies : [threads,
libpam,
libpam_misc],
link_depends : pam_systemd_sym,
install : true,
install_dir : pamlibdir)
test('dlopen-pam_systemd',
test_dlopen,
args : [pam_systemd.full_path()]) # path to dlopen must include a slash
endif
endif
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('HAVE_PAM') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-user-sessions',
'src/user-sessions/user-sessions.c',
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install_rpath : rootlibexecdir,
install : true,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
install_dir : rootlibexecdir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_EFI') == 1 and conf.get('HAVE_BLKID') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
exe = executable('bootctl',
'src/boot/bootctl.c',
include_directories : includes,
link_with : [libshared],
dependencies : [libblkid],
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
endif
exe = executable('systemd-socket-activate', 'src/activate/activate.c',
include_directories : includes,
link_with : [libshared],
dependencies : [threads],
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
exe = executable('systemctl', 'src/systemctl/systemctl.c',
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
dependencies : [threads,
libcap,
libselinux,
libxz,
liblz4],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootbindir)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_BACKLIGHT') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-backlight',
'src/backlight/backlight.c',
include_directories : includes,
meson: drop implicit libudev link dependencies Executables which link against libshared do not need an explicit dependency on libudev, as libshared will make sure that those symbols are available.
2017-04-25 21:19:54 +03:00
link_with : [libshared],
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_RFKILL') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-rfkill',
'src/rfkill/rfkill.c',
include_directories : includes,
meson: drop implicit libudev link dependencies Executables which link against libshared do not need an explicit dependency on libudev, as libshared will make sure that those symbols are available.
2017-04-25 21:19:54 +03:00
link_with : [libshared],
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
executable('systemd-system-update-generator',
'src/system-update-generator/system-update-generator.c',
include_directories : includes,
link_with : [libshared],
meson: specify rpath on all executables linking to libshared (i.e. pretty much all of them). libtools does this automatically for us. It would be great if meson did too. Otherwise, it would be nice to simply attach the information about necessary rpath to the library object, and have it used automatically for all users. Dunno, maybe there's already a way to do this. I used '$ORIGIN' for stuff installed into libexecdir, since that's where libshared ends up, and a full path for everything else. The advantage of using $ORIGIN is that those binaries can be started from $DESTDIR, which can sometimes be useful.
2017-04-11 01:13:00 +03:00
install_rpath : rootlibexecdir,
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
install : true,
install_dir : systemgeneratordir)
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('HAVE_LIBCRYPTSETUP') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-cryptsetup',
'src/cryptsetup/cryptsetup.c',
include_directories : includes,
link_with : [libshared],
dependencies : [libcryptsetup],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
executable('systemd-cryptsetup-generator',
'src/cryptsetup/cryptsetup-generator.c',
include_directories : includes,
link_with : [libshared],
dependencies : [libcryptsetup],
install_rpath : rootlibexecdir,
install : true,
install_dir : systemgeneratordir)
executable('systemd-veritysetup',
'src/veritysetup/veritysetup.c',
include_directories : includes,
link_with : [libshared],
dependencies : [libcryptsetup],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
executable('systemd-veritysetup-generator',
'src/veritysetup/veritysetup-generator.c',
include_directories : includes,
link_with : [libshared],
dependencies : [libcryptsetup],
install_rpath : rootlibexecdir,
install : true,
install_dir : systemgeneratordir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('HAVE_SYSV_COMPAT') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-sysv-generator',
'src/sysv-generator/sysv-generator.c',
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : systemgeneratordir)
executable('systemd-rc-local-generator',
'src/rc-local-generator/rc-local-generator.c',
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : systemgeneratordir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_HOSTNAMED') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-hostnamed',
'src/hostname/hostnamed.c',
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install_rpath : rootlibexecdir,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
install : true,
install_dir : rootlibexecdir)
exe = executable('hostnamectl',
'src/hostname/hostnamectl.c',
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_LOCALED') == 1
if conf.get('HAVE_XKBCOMMON') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
# logind will load libxkbcommon.so dynamically on its own
deps = [libdl]
else
deps = []
endif
executable('systemd-localed',
systemd_localed_sources,
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
dependencies : deps,
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install_rpath : rootlibexecdir,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
install : true,
install_dir : rootlibexecdir)
exe = executable('localectl',
localectl_sources,
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_TIMEDATED') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-timedated',
'src/timedate/timedated.c',
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
exe = executable('timedatectl',
'src/timedate/timedatectl.c',
include_directories : includes,
install_rpath : rootlibexecdir,
link_with : [libshared],
install : true)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_TIMESYNCD') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-timesyncd',
systemd_timesyncd_sources,
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
dependencies : [threads,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libm],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install_rpath : rootlibexecdir,
install : true,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
install_dir : rootlibexecdir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_MACHINED') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-machined',
systemd_machined_sources,
include_directories : includes,
link_with : [libmachine_core,
libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
exe = executable('machinectl',
'src/machine/machinectl.c',
include_directories : includes,
link_with : [libshared],
dependencies : [threads,
libxz,
liblz4],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootbindir)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_IMPORTD') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-importd',
systemd_importd_sources,
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
dependencies : [threads],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
systemd_pull = executable('systemd-pull',
systemd_pull_sources,
include_directories : includes,
link_with : [libshared],
dependencies : [libcurl,
libz,
libbzip2,
libxz,
libgcrypt],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
systemd_import = executable('systemd-import',
systemd_import_sources,
include_directories : includes,
link_with : [libshared],
dependencies : [libcurl,
libz,
libbzip2,
libxz],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
systemd_export = executable('systemd-export',
systemd_export_sources,
include_directories : includes,
link_with : [libshared],
dependencies : [libcurl,
libz,
libbzip2,
libxz],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
public_programs += [systemd_pull, systemd_import, systemd_export]
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_REMOTE') == 1 and conf.get('HAVE_LIBCURL') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
exe = executable('systemd-journal-upload',
systemd_journal_upload_sources,
include_directories : includes,
link_with : [libshared],
dependencies : [threads,
libcurl,
libgnutls,
libxz,
liblz4],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_REMOTE') == 1 and conf.get('HAVE_MICROHTTPD') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
s_j_remote = executable('systemd-journal-remote',
systemd_journal_remote_sources,
include_directories : includes,
link_with : [libshared],
dependencies : [threads,
libmicrohttpd,
libgnutls,
libxz,
liblz4],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
s_j_gatewayd = executable('systemd-journal-gatewayd',
systemd_journal_gatewayd_sources,
include_directories : includes,
link_with : [libshared],
dependencies : [threads,
libmicrohttpd,
libgnutls,
libxz,
liblz4],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
public_programs += [s_j_remote, s_j_gatewayd]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_COREDUMP') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-coredump',
systemd_coredump_sources,
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
dependencies : [threads,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
libacl,
libdw,
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
libxz,
liblz4],
install_rpath : rootlibexecdir,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
install : true,
install_dir : rootlibexecdir)
exe = executable('coredumpctl',
coredumpctl_sources,
include_directories : includes,
link_with : [libshared],
dependencies : [threads,
libxz,
liblz4],
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_BINFMT') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
exe = executable('systemd-binfmt',
'src/binfmt/binfmt.c',
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
public_programs += [exe]
meson.add_install_script('sh', '-c',
mkdir_p.format(binfmtdir))
meson.add_install_script('sh', '-c',
mkdir_p.format(join_paths(sysconfdir, 'binfmt.d')))
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_VCONSOLE') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-vconsole-setup',
'src/vconsole/vconsole-setup.c',
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_RANDOMSEED') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-random-seed',
'src/random-seed/random-seed.c',
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_FIRSTBOOT') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-firstboot',
'src/firstboot/firstboot.c',
include_directories : includes,
link_with : [libshared],
dependencies : [libcrypt],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootbindir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
executable('systemd-remount-fs',
'src/remount-fs/remount-fs.c',
'src/core/mount-setup.c',
'src/core/mount-setup.h',
include_directories : includes,
link_with : [libshared],
meson: specify rpath on all executables linking to libshared (i.e. pretty much all of them). libtools does this automatically for us. It would be great if meson did too. Otherwise, it would be nice to simply attach the information about necessary rpath to the library object, and have it used automatically for all users. Dunno, maybe there's already a way to do this. I used '$ORIGIN' for stuff installed into libexecdir, since that's where libshared ends up, and a full path for everything else. The advantage of using $ORIGIN is that those binaries can be started from $DESTDIR, which can sometimes be useful.
2017-04-11 01:13:00 +03:00
install_rpath : rootlibexecdir,
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
install : true,
install_dir : rootlibexecdir)
executable('systemd-machine-id-setup',
'src/machine-id-setup/machine-id-setup-main.c',
'src/core/machine-id-setup.c',
'src/core/machine-id-setup.h',
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: specify rpath on all executables linking to libshared (i.e. pretty much all of them). libtools does this automatically for us. It would be great if meson did too. Otherwise, it would be nice to simply attach the information about necessary rpath to the library object, and have it used automatically for all users. Dunno, maybe there's already a way to do this. I used '$ORIGIN' for stuff installed into libexecdir, since that's where libshared ends up, and a full path for everything else. The advantage of using $ORIGIN is that those binaries can be started from $DESTDIR, which can sometimes be useful.
2017-04-11 01:13:00 +03:00
install_rpath : rootlibexecdir,
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
install : true,
install_dir : rootbindir)
executable('systemd-fsck',
'src/fsck/fsck.c',
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: replace $ORIGIN with rootlibdir everywhere $ORIGIN does not work when /proc is not mounted. So the systemd binary cannot use $ORIGIN. Use the full path everywhere for consistency.
2017-04-12 19:02:30 +03:00
install_rpath : rootlibexecdir,
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
install : true,
install_dir : rootlibexecdir)
Add systemd-growfs tool
2017-10-23 14:40:38 +03:00
executable('systemd-growfs',
'src/partition/growfs.c',
include_directories : includes,
link_with : [libshared],
growfs: add support for resizing encrypted partitions
2017-11-21 20:56:52 +03:00
dependencies : [libcryptsetup],
Add systemd-growfs tool
2017-10-23 14:40:38 +03:00
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
Add mkfs wrapper which first checks if the partition is empty
2017-11-27 00:51:29 +03:00
executable('systemd-makefs',
'src/partition/makefs.c',
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
executable('systemd-sleep',
'src/sleep/sleep.c',
include_directories : includes,
link_with : [libshared],
meson: replace $ORIGIN with rootlibdir everywhere $ORIGIN does not work when /proc is not mounted. So the systemd binary cannot use $ORIGIN. Use the full path everywhere for consistency.
2017-04-12 19:02:30 +03:00
install_rpath : rootlibexecdir,
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
install : true,
install_dir : rootlibexecdir)
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
exe = executable('systemd-sysctl',
'src/sysctl/sysctl.c',
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
executable('systemd-ac-power',
'src/ac-power/ac-power.c',
include_directories : includes,
link_with : [libshared],
meson: replace $ORIGIN with rootlibdir everywhere $ORIGIN does not work when /proc is not mounted. So the systemd binary cannot use $ORIGIN. Use the full path everywhere for consistency.
2017-04-12 19:02:30 +03:00
install_rpath : rootlibexecdir,
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
install : true,
install_dir : rootlibexecdir)
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
exe = executable('systemd-detect-virt',
'src/detect-virt/detect-virt.c',
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
exe = executable('systemd-delta',
'src/delta/delta.c',
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
exe = executable('systemd-escape',
'src/escape/escape.c',
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootbindir)
public_programs += [exe]
exe = executable('systemd-notify',
'src/notify/notify.c',
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootbindir)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
executable('systemd-volatile-root',
'src/volatile-root/volatile-root.c',
include_directories : includes,
link_with : [libshared],
meson: replace $ORIGIN with rootlibdir everywhere $ORIGIN does not work when /proc is not mounted. So the systemd binary cannot use $ORIGIN. Use the full path everywhere for consistency.
2017-04-12 19:02:30 +03:00
install_rpath : rootlibexecdir,
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
install : true,
install_dir : rootlibexecdir)
executable('systemd-cgroups-agent',
'src/cgroups-agent/cgroups-agent.c',
include_directories : includes,
link_with : [libshared],
meson: replace $ORIGIN with rootlibdir everywhere $ORIGIN does not work when /proc is not mounted. So the systemd binary cannot use $ORIGIN. Use the full path everywhere for consistency.
2017-04-12 19:02:30 +03:00
install_rpath : rootlibexecdir,
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
install : true,
install_dir : rootlibexecdir)
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
exe = executable('systemd-path',
'src/path/path.c',
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
exe = executable('systemd-ask-password',
'src/ask-password/ask-password.c',
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install_rpath : rootlibexecdir,
install : true,
install_dir : rootbindir)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
executable('systemd-reply-password',
'src/reply-password/reply-password.c',
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: replace $ORIGIN with rootlibdir everywhere $ORIGIN does not work when /proc is not mounted. So the systemd binary cannot use $ORIGIN. Use the full path everywhere for consistency.
2017-04-12 19:02:30 +03:00
install_rpath : rootlibexecdir,
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
install : true,
install_dir : rootlibexecdir)
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
exe = executable('systemd-tty-ask-password-agent',
'src/tty-ask-password-agent/tty-ask-password-agent.c',
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install_rpath : rootlibexecdir,
install : true,
install_dir : rootbindir)
public_programs += [exe]
exe = executable('systemd-cgls',
'src/cgls/cgls.c',
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
exe = executable('systemd-cgtop',
'src/cgtop/cgtop.c',
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
executable('systemd-initctl',
'src/initctl/initctl.c',
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: replace $ORIGIN with rootlibdir everywhere $ORIGIN does not work when /proc is not mounted. So the systemd binary cannot use $ORIGIN. Use the full path everywhere for consistency.
2017-04-12 19:02:30 +03:00
install_rpath : rootlibexecdir,
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
install : true,
install_dir : rootlibexecdir)
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
exe = executable('systemd-mount',
'src/mount/mount-tool.c',
include_directories : includes,
meson: drop implicit libudev link dependencies Executables which link against libshared do not need an explicit dependency on libudev, as libshared will make sure that those symbols are available.
2017-04-25 21:19:54 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
meson: create various symlinks v2: - remove bashisms
2017-04-10 06:55:50 +03:00
meson.add_install_script(meson_make_symlink,
meson: use join_paths consistently With -Dsplit-usr=true, we set rootprefix to /. This leads to //lib/systemd or //lib/udev for various dir variables. Using join_paths() avoids this.
2017-04-13 17:30:56 +03:00
'systemd-mount', join_paths(bindir, 'systemd-umount'))
meson: create various symlinks v2: - remove bashisms
2017-04-10 06:55:50 +03:00
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
exe = executable('systemd-run',
'src/run/run.c',
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
exe = executable('systemd-stdio-bridge',
'src/stdio-bridge/stdio-bridge.c',
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
exe = executable('busctl',
'src/busctl/busctl.c',
'src/busctl/busctl-introspect.c',
'src/busctl/busctl-introspect.h',
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_SYSUSERS') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
exe = executable('systemd-sysusers',
'src/sysusers/sysusers.c',
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootbindir)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_TMPFILES') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
exe = executable('systemd-tmpfiles',
'src/tmpfiles/tmpfiles.c',
include_directories : includes,
link_with : [libshared],
dependencies : [libacl],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootbindir)
public_programs += [exe]
tmpfiles: add a special return code for syntax failures In this way, individual errors in files can be treated differently than a failure of the whole service. A test is added to check that the expected value is returned. Some parts are commented out, because it is not. This will be fixed in a subsequent commit.
2017-11-22 16:13:32 +03:00
test('test-systemd-tmpfiles',
test_systemd_tmpfiles_py,
args : exe.full_path())
# https://github.com/mesonbuild/meson/issues/2681
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_HWDB') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
exe = executable('systemd-hwdb',
'src/hwdb/hwdb.c',
'src/libsystemd/sd-hwdb/hwdb-internal.h',
include_directories : includes,
meson: allow static linking of systemd-hwdb Various distros split udev and its helpers into a separate binary package. In case of Debian this also includes systemd-hwdb. In Debian, the udev package is not supposed to depend on systemd, so a dependency on libsystemd-shared.so is not desired. Allow the systemd-hwdb binary to be linked statically if link-udev-shared is set to false. Follow-up for 3a30f21ff8ddbef5eae61d6e4305ec1cd2e4e041.
2017-04-21 19:32:14 +03:00
link_with : [libudev_internal],
install_rpath : udev_rpath,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
install : true,
install_dir : rootbindir)
public_programs += [exe]
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_QUOTACHECK') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-quotacheck',
'src/quotacheck/quotacheck.c',
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install_rpath : rootlibexecdir,
install : true,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
install_dir : rootlibexecdir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
exe = executable('systemd-socket-proxyd',
'src/socket-proxy/socket-proxyd.c',
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
dependencies : [threads],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
public_programs += [exe]
exe = executable('systemd-udevd',
systemd_udevd_sources,
include_directories : includes,
udev: use LOG_REALM_UDEV in all udev code Any call to set/query/use the log level in the code with LOG_REALM=LOG_REALM_UDEV refers to log_max_level[1]. In particular this means that systemd code using the libudev library uses does not set the log level for log calls done in libudev. Fixes #4525. v2: - also update meson's meson.build
2017-02-23 07:13:22 +03:00
c_args : ['-DLOG_REALM=LOG_REALM_UDEV'],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
link_with : [libudev_core,
libsystemd_network,
meson: add libblkid as libudev-core dep to fix include dir Follow-up for 6b5cf3ea62. v2: - also link udevadm and systemd-udev statically if requested v3: - fix install_rpath for udevadm (Michael Biebl)
2017-04-18 02:33:10 +03:00
libudev_internal],
meson: allow optional static linking of shared code in udev helpers This makes the helper binaries significantly bigger (in some cases, the final size depends on link options and optimization level), and is only useful for distributions which want to provide the option to install udev without systemd. As the linking is improved, the difference between the columns might shrink, but it's unlikely that linking libshared statically could ever be more efficient. E.g. with -O0, no -flto: (static) (shared) src/udev/ata_id 999176 85696 src/udev/cdrom_id 1024344 111656 src/udev/collect 990344 81280 src/udev/scsi_id 1023592 115656 src/udev/v4l_id 811736 17744 When linked dynamically, install_rpath must be specified, so add that.
2017-04-17 19:07:12 +03:00
dependencies : [threads,
libkmod,
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
libidn,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
libacl,
libblkid],
meson: add libblkid as libudev-core dep to fix include dir Follow-up for 6b5cf3ea62. v2: - also link udevadm and systemd-udev statically if requested v3: - fix install_rpath for udevadm (Michael Biebl)
2017-04-18 02:33:10 +03:00
install_rpath : udev_rpath,
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install : true,
install_dir : rootlibexecdir)
public_programs += [exe]
exe = executable('udevadm',
udevadm_sources,
meson: libudev_core and udevadm should have LOG_REALM=LOG_REALM_UDEV (#7666) Otherwise, setting udev_log=debug in /etc/udev/udev.conf has no effects since systemd-udevd is built with LOG_REALM=LOG_REALM_UDEV. However using LOG_REALM_UDEV (for libudev_core) reveals another similar bug for udevadm which should also define LOG_REALM_UDEV.
2017-12-16 11:36:36 +03:00
c_args : ['-DLOG_REALM=LOG_REALM_UDEV'],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
include_directories : includes,
link_with : [libudev_core,
libsystemd_network,
meson: add libblkid as libudev-core dep to fix include dir Follow-up for 6b5cf3ea62. v2: - also link udevadm and systemd-udev statically if requested v3: - fix install_rpath for udevadm (Michael Biebl)
2017-04-18 02:33:10 +03:00
libudev_internal],
meson: allow optional static linking of shared code in udev helpers This makes the helper binaries significantly bigger (in some cases, the final size depends on link options and optimization level), and is only useful for distributions which want to provide the option to install udev without systemd. As the linking is improved, the difference between the columns might shrink, but it's unlikely that linking libshared statically could ever be more efficient. E.g. with -O0, no -flto: (static) (shared) src/udev/ata_id 999176 85696 src/udev/cdrom_id 1024344 111656 src/udev/collect 990344 81280 src/udev/scsi_id 1023592 115656 src/udev/v4l_id 811736 17744 When linked dynamically, install_rpath must be specified, so add that.
2017-04-17 19:07:12 +03:00
dependencies : [threads,
libkmod,
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
libidn,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
libacl,
libblkid],
meson: add libblkid as libudev-core dep to fix include dir Follow-up for 6b5cf3ea62. v2: - also link udevadm and systemd-udev statically if requested v3: - fix install_rpath for udevadm (Michael Biebl)
2017-04-18 02:33:10 +03:00
install_rpath : udev_rpath,
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
install : true,
install_dir : rootbindir)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
executable('systemd-shutdown',
systemd_shutdown_sources,
include_directories : includes,
meson: drop implicit libudev link dependencies Executables which link against libshared do not need an explicit dependency on libudev, as libshared will make sure that those symbols are available.
2017-04-25 21:19:54 +03:00
link_with : [libshared],
meson: replace $ORIGIN with rootlibdir everywhere $ORIGIN does not work when /proc is not mounted. So the systemd binary cannot use $ORIGIN. Use the full path everywhere for consistency.
2017-04-12 19:02:30 +03:00
install_rpath : rootlibexecdir,
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
install : true,
install_dir : rootlibexecdir)
executable('systemd-update-done',
'src/update-done/update-done.c',
include_directories : includes,
link_with : [libshared],
meson: replace $ORIGIN with rootlibdir everywhere $ORIGIN does not work when /proc is not mounted. So the systemd binary cannot use $ORIGIN. Use the full path everywhere for consistency.
2017-04-12 19:02:30 +03:00
install_rpath : rootlibexecdir,
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
install : true,
install_dir : rootlibexecdir)
executable('systemd-update-utmp',
'src/update-utmp/update-utmp.c',
include_directories : includes,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
link_with : [libshared],
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
dependencies : [libaudit],
meson: replace $ORIGIN with rootlibdir everywhere $ORIGIN does not work when /proc is not mounted. So the systemd binary cannot use $ORIGIN. Use the full path everywhere for consistency.
2017-04-12 19:02:30 +03:00
install_rpath : rootlibexecdir,
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
install : true,
install_dir : rootlibexecdir)
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('HAVE_KMOD') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-modules-load',
'src/modules-load/modules-load.c',
include_directories : includes,
link_with : [libshared],
dependencies : [libkmod],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
meson: create dirs and touch /usr This is the equivalent of $(INSTALL_DIRS) and install-touch-usr-hook. I did not bother to create the directories into which we install files, since they will be created anyway. v2: - remove bashism
2017-04-10 06:55:05 +03:00
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
meson.add_install_script('sh', '-c',
mkdir_p.format(modulesloaddir))
meson.add_install_script('sh', '-c',
mkdir_p.format(join_paths(sysconfdir, 'modules-load.d')))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
endif
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
exe = executable('systemd-nspawn',
systemd_nspawn_sources,
'src/core/mount-setup.c', # FIXME: use a variable?
'src/core/mount-setup.h',
'src/core/loopback-setup.c',
'src/core/loopback-setup.h',
include_directories : [includes, include_directories('src/nspawn')],
Merge branch 'master' into meson-no-libudev
2017-04-27 20:39:54 +03:00
link_with : [libshared],
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
dependencies : [libacl,
libblkid,
libseccomp,
libselinux],
install_rpath : rootlibexecdir,
install : true)
public_programs += [exe]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_NETWORKD') == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
executable('systemd-networkd',
systemd_networkd_sources,
include_directories : includes,
link_with : [libnetworkd_core,
libsystemd_network,
libudev_internal,
libshared],
meson: add -pthreads and more libs when -Dlink-udev-shared=false (#6159) Three binaries would fail to link when ld.bfd was used and link-udev-shared was false. Add -pthreads (again) to the failing binaries and synchronize the dependency list between libsystemd-shared .a and .so versions. Apart from allowing the build to succeed, this shouldn't have much effect becuase systemd-networkd was already using pthreads. Fixes #5828.
2017-06-21 13:05:15 +03:00
dependencies : [threads],
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
executable('systemd-networkd-wait-online',
systemd_networkd_wait_online_sources,
include_directories : includes,
link_with : [libnetworkd_core,
libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
build-sys: don't build networkctl if networkd is disabled
2017-08-21 15:48:41 +03:00
exe = executable('networkctl',
networkctl_sources,
include_directories : includes,
link_with : [libsystemd_network,
meson: eliminate libsystemd_journal_internal and use libsystemd_internal less This simplifies things and leads to a smaller installation footprint. libsystemd_internal and libsystemd_journal_internal are linked into libystemd-shared and available to all programs linked to libsystemd-shared. libsystemd_journal_internal is not needed anymore, and libsystemd-shared is used everwhere. The few exceptions are: libsystemd.so, test-engine, test-bus-error, and various loadable modules.
2017-04-15 01:49:47 +03:00
libshared],
build-sys: don't build networkctl if networkd is disabled
2017-08-21 15:48:41 +03:00
install_rpath : rootlibexecdir,
install : true,
install_dir : rootbindir)
public_programs += [exe]
endif
meson: place systemd-sulogin-shell in build/ We do that will all executables so that it's easy to call them.
2017-12-07 12:44:43 +03:00
executable('systemd-sulogin-shell',
['src/sulogin-shell/sulogin-shell.c'],
include_directories : includes,
link_with : [libshared],
install_rpath : rootlibexecdir,
install : true,
install_dir : rootlibexecdir)
meson: define tests Tests can be run with 'ninja-build test' or using 'mesontest'. '-Dtests=unsafe' can be used to include the "unsafe" tests in the test suite, same as with autotools. v2: - use more conf.get guards are optional components - declare deps on generated headers for test-{af,arphrd,cap}-list v3: - define environment for tests Most test don't need this, but to be consistent with autotools-based build, and to avoid questions which tests need it and which don't, set the same environment for all tests. v4: - rework test generation Use a list of lists to define each test. This way we can reduce the boilerplate somewhat, although the test listings are still pretty verbose. We can also move the definitions of the tests to the subdirs. Unfortunately some subdirs are included earlier than some of the libraries that test binaries are linked to. So just dump all definitions of all tests that cannot be defined earlier into src/test. The `executable` definitions are still at the top level, so the binaries are compiled into the build root. v5: - tag test-dnssec-complex as manual v6: - fix HAVE_LIBZ typo - add missing libgobject/libgio defs - mark test-qcow2 as manual
2017-04-07 07:19:09 +03:00
############################################################
foreach tuple : tests
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
sources = tuple[0]
link_with = tuple[1].length() > 0 ? tuple[1] : [libshared]
dependencies = tuple[2]
condition = tuple.length() >= 4 ? tuple[3] : ''
type = tuple.length() >= 5 ? tuple[4] : ''
defs = tuple.length() >= 6 ? tuple[5] : []
incs = tuple.length() >= 7 ? tuple[6] : includes
timeout = 30
name = sources[0].split('/')[-1].split('.')[0]
if type.startswith('timeout=')
timeout = type.split('=')[1].to_int()
type = ''
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if condition == '' or conf.get(condition) == 1
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
exe = executable(
name,
sources,
include_directories : incs,
link_with : link_with,
dependencies : dependencies,
c_args : defs,
install_rpath : rootlibexecdir,
meson: install manual and unsafe tests (#6178) Always install all tests if install-tests is set to true, as they might be useful for CI. This includes manual and unsafe tests. Install those in subdirectories of /usr/lib/systemd/tests. Fixes #6163.
2017-06-23 04:23:30 +03:00
install : install_tests,
install_dir : join_paths(testsdir, type))
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
if type == 'manual'
message('@0@ is a manual test'.format(name))
elif type == 'unsafe' and want_tests != 'unsafe'
message('@0@ is an unsafe test'.format(name))
else
test(name, exe,
env : test_env,
timeout : timeout)
endif
else
message('Not compiling @0@ because @1@ is not true'.format(name, condition))
endif
meson: define tests Tests can be run with 'ninja-build test' or using 'mesontest'. '-Dtests=unsafe' can be used to include the "unsafe" tests in the test suite, same as with autotools. v2: - use more conf.get guards are optional components - declare deps on generated headers for test-{af,arphrd,cap}-list v3: - define environment for tests Most test don't need this, but to be consistent with autotools-based build, and to avoid questions which tests need it and which don't, set the same environment for all tests. v4: - rework test generation Use a list of lists to define each test. This way we can reduce the boilerplate somewhat, although the test listings are still pretty verbose. We can also move the definitions of the tests to the subdirs. Unfortunately some subdirs are included earlier than some of the libraries that test binaries are linked to. So just dump all definitions of all tests that cannot be defined earlier into src/test. The `executable` definitions are still at the top level, so the binaries are compiled into the build root. v5: - tag test-dnssec-complex as manual v6: - fix HAVE_LIBZ typo - add missing libgobject/libgio defs - mark test-qcow2 as manual
2017-04-07 07:19:09 +03:00
endforeach
meson: add test-libsystemd-sym, fix linking of libsystemd This is quite messy. I think libtool might have been using something like -Wl,--whole-archive, but I don't think meson has support for that. For now, just recompile all the sources for linking into libsystemd directly. This should not matter much for efficiency, since it's a few small files.
2017-04-10 21:13:40 +03:00
test_libsystemd_sym = executable(
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
'test-libsystemd-sym',
test_libsystemd_sym_c,
include_directories : includes,
link_with : [libsystemd],
install : install_tests,
install_dir : testsdir)
meson: add test-libsystemd-sym, fix linking of libsystemd This is quite messy. I think libtool might have been using something like -Wl,--whole-archive, but I don't think meson has support for that. For now, just recompile all the sources for linking into libsystemd directly. This should not matter much for efficiency, since it's a few small files.
2017-04-10 21:13:40 +03:00
test('test-libsystemd-sym',
test_libsystemd_sym)
meson: add test-libudev-sym This is slightly complicated by the fact that files('libudev.h') cannot be used as an argument in custom_target command (string is required). This restriction should be lifted in future versions of mesons, so this could be simplified.
2017-04-10 22:20:42 +03:00
test_libudev_sym = executable(
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
'test-libudev-sym',
test_libudev_sym_c,
include_directories : includes,
c_args : ['-Wno-deprecated-declarations'],
link_with : [libudev],
install : install_tests,
install_dir : testsdir)
meson: add test-libudev-sym This is slightly complicated by the fact that files('libudev.h') cannot be used as an argument in custom_target command (string is required). This restriction should be lifted in future versions of mesons, so this could be simplified.
2017-04-10 22:20:42 +03:00
test('test-libudev-sym',
test_libudev_sym)
meson: define tests Tests can be run with 'ninja-build test' or using 'mesontest'. '-Dtests=unsafe' can be used to include the "unsafe" tests in the test suite, same as with autotools. v2: - use more conf.get guards are optional components - declare deps on generated headers for test-{af,arphrd,cap}-list v3: - define environment for tests Most test don't need this, but to be consistent with autotools-based build, and to avoid questions which tests need it and which don't, set the same environment for all tests. v4: - rework test generation Use a list of lists to define each test. This way we can reduce the boilerplate somewhat, although the test listings are still pretty verbose. We can also move the definitions of the tests to the subdirs. Unfortunately some subdirs are included earlier than some of the libraries that test binaries are linked to. So just dump all definitions of all tests that cannot be defined earlier into src/test. The `executable` definitions are still at the top level, so the binaries are compiled into the build root. v5: - tag test-dnssec-complex as manual v6: - fix HAVE_LIBZ typo - add missing libgobject/libgio defs - mark test-qcow2 as manual
2017-04-07 07:19:09 +03:00
############################################################
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
make_directive_index_py = find_program('tools/make-directive-index.py')
make_man_index_py = find_program('tools/make-man-index.py')
meson: skip index generation when lxml is not available
2017-04-14 02:59:21 +03:00
xml_helper_py = find_program('tools/xml_helper.py')
meson: add hwdb/update target
2017-04-15 07:40:59 +03:00
hwdb_update_sh = find_program('tools/meson-hwdb-update.sh')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
subdir('units')
subdir('sysctl.d')
subdir('sysusers.d')
subdir('tmpfiles.d')
Rename "system-preset" source dir to "presets" I want to add presets/user/ later. This mirrors the layout for units: we have units/ and units/user. The advantage is that we avoid having yet another directory at the top level.
2017-11-23 15:23:42 +03:00
subdir('presets')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
subdir('hwdb')
subdir('network')
subdir('man')
subdir('shell-completion/bash')
subdir('shell-completion/zsh')
subdir('docs/sysvinit')
subdir('docs/var-log')
# FIXME: figure out if the warning is true:
# https://github.com/mesonbuild/meson/wiki/Reference-manual#install_subdir
install_subdir('factory/etc',
install_dir : factorydir)
install_data('xorg/50-systemd-user.sh',
install_dir : xinitrcdir)
modprobe.d: ship drop-in to set bonding max_bonds to 0 (#6448) This allows networkd to correctly manage bond0 using networkd, when requested by the user. Fixes #5971 #6184
2017-08-02 15:41:18 +03:00
install_data('modprobe.d/systemd.conf',
install_dir : modprobedir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
install_data('README',
'NEWS',
'CODING_STYLE',
'DISTRO_PORTING',
'ENVIRONMENT.md',
'LICENSE.GPL2',
'LICENSE.LGPL2.1',
build-sys: install TRANSIENT-SETTINGS.md and UIDS-GIDS.md (#7690)
2017-12-18 16:58:13 +03:00
'TRANSIENT-SETTINGS.md',
'UIDS-GIDS.md',
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-05 06:03:47 +03:00
'src/libsystemd/sd-bus/GVARIANT-SERIALIZATION',
install_dir : docdir)
meson: add TAGS target Quite necessary for all the emacs users out there ;) v2: - fix error with etags or git are not found v3: - move the definition of git and etags up so it's available for man/
2017-04-06 21:33:27 +03:00
meson: create dirs and touch /usr This is the equivalent of $(INSTALL_DIRS) and install-touch-usr-hook. I did not bother to create the directories into which we install files, since they will be created anyway. v2: - remove bashism
2017-04-10 06:55:05 +03:00
meson.add_install_script('sh', '-c', mkdir_p.format(systemdstatedir))
meson.add_install_script('sh', '-c', 'touch $DESTDIR@0@'.format(prefixdir))
meson: add TAGS target Quite necessary for all the emacs users out there ;) v2: - fix error with etags or git are not found v3: - move the definition of git and etags up so it's available for man/
2017-04-06 21:33:27 +03:00
############################################################
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
meson_check_help = find_program('tools/meson-check-help.sh')
foreach exec : public_programs
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
name = exec.full_path().split('/')[-1]
test('check-help-' + name,
meson_check_help,
args : [exec.full_path()])
meson: recreate dist-check-help
2017-04-13 18:52:05 +03:00
endforeach
############################################################
meson: add support for ctags This is a squash of casync commits https://github.com/systemd/casync/commit/02fbbdb2b9a926a695a3ede7f3e3c17b9779db1a (by Silvio Fricke) and https://github.com/systemd/casync/commit/b687a94b1e24df73d32f8cdcee29f3c00eae69c9. Instead of checking during every meson config whether etags are available, just try to call them and error out if not. This has the advantage that the target is always available (if git is installed), and the error message gives a hint what needs to be installed. The naming is confusing, but etags(1) is pretty clear: - emacs expects TAGS file in etags format - vi expects tags file in ctags format and automake docs are pretty clear too: - tags target generates TAGS file - ctags target generates tags file
2017-07-03 19:42:29 +03:00
if git.found()
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
all_files = run_command(
git,
meson: use current_source_dir() to scope git invocations (#6635)
2017-08-30 18:04:53 +03:00
['--git-dir=@0@/.git'.format(meson.current_source_dir()),
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
'ls-files',
':/*.[ch]'])
all_files = files(all_files.stdout().split())
meson: add TAGS target Quite necessary for all the emacs users out there ;) v2: - fix error with etags or git are not found v3: - move the definition of git and etags up so it's available for man/
2017-04-06 21:33:27 +03:00
build-sys: fix invalid args detected by meson 0.42 (#6561) some run_target() calls were using params from custom_target() example message: WARNING: Passed invalid keyword argument "input". This will become a hard error in the future. New way to call targets: ninja man/man ninja man/html ninja man/update-man-rules
2017-08-09 16:41:44 +03:00
custom_target(
meson: add support for ctags This is a squash of casync commits https://github.com/systemd/casync/commit/02fbbdb2b9a926a695a3ede7f3e3c17b9779db1a (by Silvio Fricke) and https://github.com/systemd/casync/commit/b687a94b1e24df73d32f8cdcee29f3c00eae69c9. Instead of checking during every meson config whether etags are available, just try to call them and error out if not. This has the advantage that the target is always available (if git is installed), and the error message gives a hint what needs to be installed. The naming is confusing, but etags(1) is pretty clear: - emacs expects TAGS file in etags format - vi expects tags file in ctags format and automake docs are pretty clear too: - tags target generates TAGS file - ctags target generates tags file
2017-07-03 19:42:29 +03:00
'tags',
build-sys: fix invalid args detected by meson 0.42 (#6561) some run_target() calls were using params from custom_target() example message: WARNING: Passed invalid keyword argument "input". This will become a hard error in the future. New way to call targets: ninja man/man ninja man/html ninja man/update-man-rules
2017-08-09 16:41:44 +03:00
output : 'tags',
meson: use current_source_dir() to scope git invocations (#6635)
2017-08-30 18:04:53 +03:00
command : ['env', 'etags', '-o', '@0@/TAGS'.format(meson.current_source_dir())] + all_files)
build-sys: fix invalid args detected by meson 0.42 (#6561) some run_target() calls were using params from custom_target() example message: WARNING: Passed invalid keyword argument "input". This will become a hard error in the future. New way to call targets: ninja man/man ninja man/html ninja man/update-man-rules
2017-08-09 16:41:44 +03:00
custom_target(
meson: add support for ctags This is a squash of casync commits https://github.com/systemd/casync/commit/02fbbdb2b9a926a695a3ede7f3e3c17b9779db1a (by Silvio Fricke) and https://github.com/systemd/casync/commit/b687a94b1e24df73d32f8cdcee29f3c00eae69c9. Instead of checking during every meson config whether etags are available, just try to call them and error out if not. This has the advantage that the target is always available (if git is installed), and the error message gives a hint what needs to be installed. The naming is confusing, but etags(1) is pretty clear: - emacs expects TAGS file in etags format - vi expects tags file in ctags format and automake docs are pretty clear too: - tags target generates TAGS file - ctags target generates tags file
2017-07-03 19:42:29 +03:00
'ctags',
build-sys: fix invalid args detected by meson 0.42 (#6561) some run_target() calls were using params from custom_target() example message: WARNING: Passed invalid keyword argument "input". This will become a hard error in the future. New way to call targets: ninja man/man ninja man/html ninja man/update-man-rules
2017-08-09 16:41:44 +03:00
output : 'ctags',
meson: use current_source_dir() to scope git invocations (#6635)
2017-08-30 18:04:53 +03:00
command : ['env', 'ctags', '-o', '@0@/tags'.format(meson.current_source_dir())] + all_files)
meson: add TAGS target Quite necessary for all the emacs users out there ;) v2: - fix error with etags or git are not found v3: - move the definition of git and etags up so it's available for man/
2017-04-06 21:33:27 +03:00
endif
meson: add git-contrib target
2017-04-15 07:16:23 +03:00
if git.found()
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
meson_git_contrib_sh = find_program('tools/meson-git-contrib.sh')
meson: use run_target instead of custom_target where appropriate This way a fake output does not need to be specified.
2017-04-18 02:48:20 +03:00
run_target(
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-18 02:25:00 +03:00
'git-contrib',
command : [meson_git_contrib_sh])
meson: add git-contrib target
2017-04-15 07:16:23 +03:00
endif
meson: add git-snapshot target The output is systemd-<sha1abbrev>.tar.gz in source root.
2017-04-25 02:28:05 +03:00
if git.found()
git_head = run_command(
git,
meson: use current_source_dir() to scope git invocations (#6635)
2017-08-30 18:04:53 +03:00
['--git-dir=@0@/.git'.format(meson.current_source_dir()),
meson: add git-snapshot target The output is systemd-<sha1abbrev>.tar.gz in source root.
2017-04-25 02:28:05 +03:00
'rev-parse', 'HEAD']).stdout().strip()
git_head_short = run_command(
git,
meson: use current_source_dir() to scope git invocations (#6635)
2017-08-30 18:04:53 +03:00
['--git-dir=@0@/.git'.format(meson.current_source_dir()),
meson: add git-snapshot target The output is systemd-<sha1abbrev>.tar.gz in source root.
2017-04-25 02:28:05 +03:00
'rev-parse', '--short=7', 'HEAD']).stdout().strip()
run_target(
'git-snapshot',
command : ['git', 'archive',
meson: use current_source_dir() to scope git invocations (#6635)
2017-08-30 18:04:53 +03:00
'-o', '@0@/systemd-@1@.tar.gz'.format(meson.current_source_dir(),
meson: add git-snapshot target The output is systemd-<sha1abbrev>.tar.gz in source root.
2017-04-25 02:28:05 +03:00
git_head_short),
'--prefix', 'systemd-@0@/'.format(git_head),
'HEAD'])
endif
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
############################################################
status = [
'@0@ @1@'.format(meson.project_name(), meson.project_version()),
meson: remove abbreviations in status
2017-11-25 14:35:24 +03:00
'prefix directory: @0@'.format(prefixdir),
'rootprefix directory: @0@'.format(rootprefixdir),
'sysconf directory: @0@'.format(sysconfdir),
'include directory: @0@'.format(includedir),
'lib directory: @0@'.format(libdir),
'rootlib directory: @0@'.format(rootlibdir),
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
'SysV init scripts: @0@'.format(sysvinit_path),
'SysV rc?.d directories: @0@'.format(sysvrcnd_path),
meson: remove abbreviations in status
2017-11-25 14:35:24 +03:00
'PAM modules directory: @0@'.format(pamlibdir),
'PAM configuration directory: @0@'.format(pamconfdir),
'RPM macros directory: @0@'.format(rpmmacrosdir),
'modprobe.d directory: @0@'.format(modprobedir),
'D-Bus policy directory: @0@'.format(dbuspolicydir),
'D-Bus session directory: @0@'.format(dbussessionservicedir),
'D-Bus system directory: @0@'.format(dbussystemservicedir),
'bash completions directory: @0@'.format(bashcompletiondir),
'zsh completions directory: @0@'.format(zshcompletiondir),
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
'extra start script: @0@'.format(get_option('rc-local')),
'extra stop script: @0@'.format(get_option('halt-local')),
'debug shell: @0@ @ @1@'.format(get_option('debug-shell'),
get_option('debug-tty')),
'TTY GID: @0@'.format(tty_gid),
sysusers: Provide meson argument to set gid for 'users' group (#7533) To allow better integration with distributions requiring an explicitly set gid for the `users` group, provide the new `-Dusers-gid` option to set to a new numeric value. In the absence of a specified gid, we'll fallback to the default existing behaviour of `-` as the gid value, to automatically assign the next available gid on the system.
2017-12-03 15:28:23 +03:00
'users GID: @0@'.format(users_gid),
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
'maximum system UID: @0@'.format(system_uid_max),
'maximum system GID: @0@'.format(system_gid_max),
build-sys: make the dynamic UID range, and the container UID range configurable Also, export these ranges in our pkg-config files.
2017-12-02 14:48:31 +03:00
'minimum dynamic UID: @0@'.format(dynamic_uid_min),
'maximum dynamic UID: @0@'.format(dynamic_uid_max),
'minimum container UID base: @0@'.format(container_uid_base_min),
'maximum container UID base: @0@'.format(container_uid_base_max),
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
'/dev/kvm access mode: @0@'.format(get_option('dev-kvm-mode')),
udev-rules: Permission changes for /dev/dri/renderD* - Remove the uaccess tag from /dev/dri/renderD*. - Change the owning group from video to render. - Change default mode to 0666. - Add an option to allow users to set the access mode for these devices at compile time.
2017-10-31 18:46:24 +03:00
'render group access mode: @0@'.format(get_option('group-render-mode')),
meson: remove abbreviations in status
2017-11-25 14:35:24 +03:00
'certificate root directory: @0@'.format(get_option('certificate-root')),
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
'support URL: @0@'.format(support_url),
meson: print warnings if the "nobody" user/group name is not compatible with the local system At least on Fedora and Debian systems this not obvious to get right, hence warn.
2017-12-05 13:00:24 +03:00
'nobody user name: @0@'.format(nobody_user),
'nobody group name: @0@'.format(nobody_group),
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
'fallback hostname: @0@'.format(get_option('fallback-hostname')),
resolved,nss-myhostname: use _gateway for the gateway This changes the symbolic name for the default gateway from "gateway" to "_gateway". A new configuration option -Dcompat-gateway-hostname=true|false is added. If it is set, the old name is also supported, but the new name is used as the canonical name in either case. This is intended as a temporary measure to make the transition easier, and the option should be removed after a few releases, at which point only the new name will be used. The old "gateway" name mostly works OK, but hasn't gained widespread acceptance because of the following (potential) conflicts: - it is completely legal to have a host called "gateway" - there is no guarantee that "gateway" will not be registered as a TLD, even though this currently seems unlikely. (Even then, there would be no conflict except for the case when the top-level domain itself was being resolved. The "gateway" or "_gateway" labels have only special meaning when the whole name consists of a single label, so resolution of any subdomain of the hypothetical gateway. TLD would still work OK. ) Moving to "_gateway" avoids those issues because underscores are not allowed in host names (RFC 1123, §2.1) and avoids potential conflicts with local or global names. v2: - simplify the logic to hardcode "_gateway" and allow -Dcompat-gateway-hostname=true as a temporary measure.
2017-07-11 09:15:08 +03:00
'symbolic gateway hostnames: @0@'.format(', '.join(gateway_hostnames)),
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
'default DNSSEC mode: @0@'.format(default_dnssec),
'default cgroup hierarchy: @0@'.format(default_hierarchy),
'default KillUserProcesses setting: @0@'.format(kill_user_processes)]
alt_dns_servers = '\n '.join(dns_servers.split(' '))
alt_ntp_servers = '\n '.join(ntp_servers.split(' '))
status += [
'default DNS servers: @0@'.format(alt_dns_servers),
'default NTP servers: @0@'.format(alt_ntp_servers)]
alt_time_epoch = run_command('date', '-Is', '-u', '-d',
'@@0@'.format(time_epoch)).stdout().strip()
status += [
'time epoch: @0@ (@1@)'.format(time_epoch, alt_time_epoch)]
# TODO:
# CFLAGS: ${OUR_CFLAGS} ${CFLAGS}
# CPPFLAGS: ${OUR_CPPFLAGS} ${CPPFLAGS}
# LDFLAGS: ${OUR_LDFLAGS} ${LDFLAGS}
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
if conf.get('ENABLE_EFI') == 1
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
status += [
'efi arch: @0@'.format(efi_arch)]
if have_gnu_efi
status += [
'EFI machine type: @0@'.format(EFI_MACHINE_TYPE_NAME),
'EFI CC @0@'.format(efi_cc),
meson: remove abbreviations in status
2017-11-25 14:35:24 +03:00
'EFI lib directory: @0@'.format(efi_libdir),
'EFI lds directory: @0@'.format(efi_ldsdir),
'EFI include directory: @0@'.format(efi_incdir)]
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
endif
endif
found = []
missing = []
foreach tuple : [
['libcryptsetup'],
['PAM'],
['AUDIT'],
['IMA'],
['AppArmor'],
['SELinux'],
['SECCOMP'],
['SMACK'],
['zlib'],
['xz'],
['lz4'],
['bzip2'],
['ACL'],
['gcrypt'],
['qrencode'],
['microhttpd'],
['gnutls'],
['libcurl'],
meson: add -Didn=yes|no option
2017-06-16 16:16:28 +03:00
['idn'],
resolved: support libidn2 in addition to libidn libidn2 2.0.0 supports IDNA2008, in contrast to libidn which supports IDNA2003. https://bugzilla.redhat.com/show_bug.cgi?id=1449145 From that bug report: Internationalized domain names exist for quite some time (IDNA2003), although the protocols describing them have evolved in an incompatible way (IDNA2008). These incompatibilities will prevent applications written for IDNA2003 to access certain problematic domain names defined with IDNA2008, e.g., faß.de is translated to domain xn--fa-hia.de with IDNA2008, while in IDNA2003 it is translated to fass.de domain. That not only causes incompatibility problems, but may be used as an attack vector to redirect users to different web sites. v2: - keep libidn support - require libidn2 >= 2.0.0 v3: - keep dns_name_apply_idna caller dumb, and keep the #ifdefs inside of the function. - use both ±IDN and ±IDN2 in the version string
2017-05-10 04:56:34 +03:00
['libidn2'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
['libidn'],
make nss-systemd support conditional (#6155) This allows the nss-systemd module to be disabled on minimal installations.
2017-06-24 20:30:26 +03:00
['nss-systemd'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
['libiptc'],
['elfutils'],
['binfmt'],
['vconsole'],
['quotacheck'],
['tmpfiles'],
['environment.d'],
['sysusers'],
['firstboot'],
['randomseed'],
['backlight'],
['rfkill'],
['logind'],
['machined'],
['importd'],
['hostnamed'],
['timedated'],
['timesyncd'],
['localed'],
['networkd'],
meson: fix feature list (#7011)
2017-10-06 10:33:21 +03:00
['resolve'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
['coredump'],
['polkit'],
['legacy pkla', install_polkit_pkla],
['efi'],
['gnu-efi', have_gnu_efi],
['kmod'],
['xkbcommon'],
['blkid'],
['dbus'],
['glib'],
build-sys: s/HAVE_MYHOSTNAME/ENABLE_MYHOSTNAME/ Same justification as for HAVE_UTMP. HAVE_MYHOSTNAME was used before mysthostname was merged into systemd.
2017-10-03 13:23:55 +03:00
['nss-myhostname', conf.get('ENABLE_MYHOSTNAME') == 1],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
['hwdb'],
['tpm'],
['man pages', want_man],
['html pages', want_html],
['man page indices', want_man and have_lxml],
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
['split /usr', conf.get('HAVE_SPLIT_USR') == 1],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
['SysV compat'],
['utmp'],
['ldconfig'],
['hibernate'],
['adm group', get_option('adm-group')],
['wheel group', get_option('wheel-group')],
sysusers: make group shadow support configurable Some distros (openSUSE) don't have group shadow support enabled. This can lead to the following error: # systemd-sysusers Creating group foofoo with gid 478. # systemd-sysusers # groupdel foofoo # systemd-sysusers Creating group foofoo with gid 478. Failed to write files: File exists This patch adds --disable-gshadow option to configure. If used, systemd-sysvusers won't consider /etc/gshadow.
2017-05-09 15:02:37 +03:00
['gshadow'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
['debug hashmap'],
['debug mmap cache'],
]
cond = tuple.get(1, '')
if cond == ''
ident1 = 'HAVE_' + tuple[0].underscorify().to_upper()
ident2 = 'ENABLE_' + tuple[0].underscorify().to_upper()
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 11:41:51 +03:00
cond = conf.get(ident1, 0) == 1 or conf.get(ident2, 0) == 1
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
endif
if cond
found += [tuple[0]]
else
missing += [tuple[0]]
endif
endforeach
status += [
meson: add empty lines before enabled/disabled status Those line are long enough to wrap around a few times, and they are much more legible with some whitespace.
2017-07-26 21:14:44 +03:00
'',
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
'enabled features: @0@'.format(', '.join(found)),
meson: add empty lines before enabled/disabled status Those line are long enough to wrap around a few times, and they are much more legible with some whitespace.
2017-07-26 21:14:44 +03:00
'',
'disabled features: @0@'.format(', '.join(missing)),
'']
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-28 03:54:52 +03:00
message('\n '.join(status))
meson: emit a warning if rootprefix is set It's printed at the end because it's the easiest to spot there. When meson is upgraded to 0.43.1, we'll be able to use warning() instead.
2017-11-28 23:46:53 +03:00
if rootprefixdir != rootprefix_default
message('WARNING:\n' +
' Note that the installation prefix was changed to "@0@".\n'.format(rootprefixdir) +
' systemd used fixed names for unit file directories and other paths, so anything\n' +
' except the default ("@0@") is strongly discouraged.'.format(rootprefix_default))
endif
Copy Permalink