IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
we do not complete them in bash copletion either and their just there
for backward compatibillity, so filter them out.
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
followup for previous commit to show the full short usage string for
the CLI tool in the case of an unknown command
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
Allow to return the original tree with all refs. We use this
with our new pveclient which needs the full api definition.
Keeping refs makes it possible to store the tree more efficiently.
Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
This contains 2 functional changes:
First: resolve_cmd no longer keeps a hash of which arguments
were expanded. This information is not required and not used
properly: For one it would conflict if the same word
appeared twice in a longer subcommand, and secondly we lose
the information when recursing into an alias anyway. And
lastly, we do not support tab completing multiple parameters
simultaneously anyway (as in, `pveum u a<tab>` does not
become `pveum user add`).
So now we simply return the expanded version of the last
command or undef if it was unknown in place of the hash we
returned previously.
The second change is how we use the new returned value:
Previously if resolve_cmd() returned a new subcommand in
$def we skipped over finishing the last word. Of course, if
the command was already fully specified (but no space put
after it), we already considered it complete and returned
the new $def.
This condition can be detected as in this case the $prev
command equals the $cur command. (Additionally, the $cur
command is either '' (after the space) or also the $prev
command (before the space), but checking this would only be
required when the same word can actually appear multiple
times in a row in a sub command chain...)
This case now takes precedence over looking through the
nested $def, so that bash will put the space after a full
command which requires another subcommand to be added.
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
We cannot properly encode blessed objects as json, so
instead, we should stringify them. This happened for
instance if a VM's systemd scope wasn't cleaned up as we
got an error as a Net::DBus::Error object causing a
"malformed json string" error to appear instead of the
actual message.
Additionally, add a 'must_stringify' helper: The above error
object implements a '""' operator for stringification (as
all error should), but in theory that could die as well, in
which case we just return a generic error string we'll
hopefully never see...
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
Not only because <=> is correct, but using 'cmp' also has
the side effect that it adds a string version to the
variable and the API's json output turns into a string as
well, and this only happens once a task has completed
(while it's an integer while it's still running...)
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
this implements those parts of draft-ietf-acme-acme-09 which are needed
to use Let's Encrypt's v2 API.
(based on an internal implement for the Let's Encrypt v1 API)
Co-Authored-By: Wolfgang Bumiller <w.bumiller@proxmox.com>
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
since we sometimes use their length in a format string for printf
Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
Co-authored-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
Instead of hardcoding 'password' as a special case in the
JSONSchema's getopt handling, extend the new parameter
mapping to allow defining a parameters as 'interactive'.
They also take an optional argument on the command line
directly.
This effectively deprecates the password special case which
should be replaced in pct/pveum/... and then dropped in
pve-common.
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
track our command string, i.e. everything which cannot be an argument
for a specific command, in resolve_cmd, as we go through the commando
definition there anyway and know if a ARGV element is part of the
command itself or its arguments.
Fixes a problem where a invalid command had all the passed parameter
attached in the resulting USAGE output.
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
the timegm(gmtime()) and timelocal(localtime(()) constructs are
problematic in the following case: - $last is such that $year gets set
to a two-digit value (e.g., the referred to timestamp is somewhere in
the range of 1900-1999) - the current date is such that the value of
$year gets interpreted wrongly (e.g., anything other than 1950).
the exact breakage depends on the actual current date AND value of
$last, since localtime/gmtime will interpret two-digit years as (perldoc
Time::Local):
[...] shorthand for years in the rolling "current century," defined
as 50 years on either side of the current year. Thus, today, in
1999, 0 would refer to 2000, and 45 to 2045, but 55 would refer to
1955. Twenty years from now, 55 would instead refer to 2055.
fix it by adding 1900 to force 4-digit $year values, as the localtime
documentation suggests.
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
We don't use them and iproute2 4.13.0 has an issue parsing
parameters after a policing description.
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
For sub commands we resolve the real $cmd, $def and its arguments,
thus we should also get the handler from the resolved $def, not the
global one.
No change for normal (consisting of only the first argument)
commands, for them $cmddef == $def.
This sneaked in in a respin/rebase of the series.
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
Improves usabillity by allowing to pass a sub copmmand unquoted to
the help command, e.g.:
# pveum help user delete
without this only a quoted version worked, e.g.:
# pveum help "user delete"
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
Acked-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
allow to use sub commands alá
# pveum user add
The new resolve_cmd traverses $cmddef, resolves one level of aliases
and returns the respective sub command, its cmddef, arguments and if
it was expanded (e.g., pveum u d ... => pveum user delete ...) which
allows quite easy integration in the usage/synopsis generator, bash
completion helper and command handler.
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
