sin/samba-mirror
1
0
Fork 0
mirror of https://github.com/samba-team/samba.git synced 2025-01-24 02:04:21 +03:00
Code
samba-mirror/source4/torture/basic/delete.c

1546 lines
51 KiB
C
Raw Normal View History

r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
/*
Unix SMB/CIFS implementation.
delete on close testing
Copyright (C) Andrew Tridgell 2003
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
This program 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 General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*/
#include "includes.h"
r12694: Move some headers to the directory of the subsystem they belong to. (This used to be commit c722f665c90103f3ed57621c460e32ad33e7a8a3)
2006-01-03 15:40:05 +00:00
#include "libcli/libcli.h"
r12693: Move core data structures out of smb.h into core.h torture prototypes in seperate header (This used to be commit 73610639b23ca3743077193fa0b1de7c7f65944d)
2006-01-03 13:41:17 +00:00
#include "torture/torture.h"
r14527: Fix build problems. (This used to be commit 863ca4014d9b821706ee90f58ab5d5cf3899a4c7)
2006-03-17 17:59:58 +00:00
#include "torture/util.h"
r5298: - got rid of pstring.h from includes.h. This at least makes it a bit less likely that anyone will use pstring for new code - got rid of winbind_client.h from includes.h. This one triggered a huge change, as winbind_client.h was including system/filesys.h and defining the old uint32 and uint16 types, as well as its own pstring and fstring. (This used to be commit 9db6c79e902ec538108d6b7d3324039aabe1704f)
2005-02-10 05:09:35 +00:00
#include "system/filesys.h"
r9431: Check an error code in BASE-DELETE. Volker (This used to be commit 71571fffc0493a5658c5980e6ebe4d8f9ada4699)
2005-08-20 21:28:30 +00:00
#include "libcli/raw/libcliraw.h"
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r13346: use private proto header files for the torture tests metze (This used to be commit 67837dbd2bcff8ec1917ba02884ee2eaa0776b46)
2006-02-04 14:08:24 +00:00
#include "torture/raw/proto.h"
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool check_delete_on_close(struct torture_context *tctx,
struct smbcli_state *cli, int fnum,
const char *fname, bool expect_it,
const char *where)
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
{
union smb_search_data data;
NTSTATUS status;
time_t c_time, a_time, m_time;
size_t size;
uint16_t mode;
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
status = torture_single_search(cli, tctx,
r16834: split the level's of smb_search_first/smb_search_next and the levels of smb_search_data metze (This used to be commit 78c201db8a47a71908698c4dda2add4cf85694d9)
2006-07-06 08:00:24 +00:00
fname,
RAW_SEARCH_TRANS2,
RAW_SEARCH_DATA_FULL_DIRECTORY_INFO,
r8185: Delete on close on directories: Creating a file in a directory with delete-on-close set returns DELETE_PENDING, and trying to set the flag on a non-empty directory returns DIRECTORY_NOT_EMPTY. Volker (This used to be commit 5680f34778b2f5291936f4d4fb937a7713696c52)
2005-07-06 14:56:45 +00:00
FILE_ATTRIBUTE_DIRECTORY,
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
&data);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, status,
talloc_asprintf(tctx, "single_search failed (%s)", where));
r8182: Little more testing delete-on-close: Check flag with qfileinfo. Volker (This used to be commit 47a9df946d5ff967289fba0ff4209711ead11e31)
2005-07-06 13:24:38 +00:00
if (fnum != -1) {
union smb_fileinfo io;
int nlink = expect_it ? 0 : 1;
io.all_info.level = RAW_FILEINFO_ALL_INFO;
r14256: - rename smb_file -> smb_handle - move it into the in/out substructs again - allow file.path only on smb_fileinfo/smb_setfileinfo metze (This used to be commit be6d5298a2cdb7e7c61d70471bad445645af5963)
2006-03-12 22:48:25 +00:00
io.all_info.in.file.fnum = fnum;
r8182: Little more testing delete-on-close: Check flag with qfileinfo. Volker (This used to be commit 47a9df946d5ff967289fba0ff4209711ead11e31)
2005-07-06 13:24:38 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
status = smb_raw_fileinfo(cli->tree, tctx, &io);
torture_assert_ntstatus_ok(tctx, status, talloc_asprintf(tctx,
"qfileinfo failed (%s)", where));
r8182: Little more testing delete-on-close: Check flag with qfileinfo. Volker (This used to be commit 47a9df946d5ff967289fba0ff4209711ead11e31)
2005-07-06 13:24:38 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, expect_it == io.all_info.out.delete_pending,
talloc_asprintf(tctx,
"%s - Expected del_on_close flag %d, qfileinfo/all_info gave %d",
where, expect_it, io.all_info.out.delete_pending));
r8182: Little more testing delete-on-close: Check flag with qfileinfo. Volker (This used to be commit 47a9df946d5ff967289fba0ff4209711ead11e31)
2005-07-06 13:24:38 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, nlink == io.all_info.out.nlink,
talloc_asprintf(tctx,
"%s - Expected nlink %d, qfileinfo/all_info gave %d",
where, nlink, io.all_info.out.nlink));
r8182: Little more testing delete-on-close: Check flag with qfileinfo. Volker (This used to be commit 47a9df946d5ff967289fba0ff4209711ead11e31)
2005-07-06 13:24:38 +00:00
io.standard_info.level = RAW_FILEINFO_STANDARD_INFO;
r14256: - rename smb_file -> smb_handle - move it into the in/out substructs again - allow file.path only on smb_fileinfo/smb_setfileinfo metze (This used to be commit be6d5298a2cdb7e7c61d70471bad445645af5963)
2006-03-12 22:48:25 +00:00
io.standard_info.in.file.fnum = fnum;
r8182: Little more testing delete-on-close: Check flag with qfileinfo. Volker (This used to be commit 47a9df946d5ff967289fba0ff4209711ead11e31)
2005-07-06 13:24:38 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
status = smb_raw_fileinfo(cli->tree, tctx, &io);
torture_assert_ntstatus_ok(tctx, status, talloc_asprintf(tctx, "qpathinfo failed (%s)", where));
r8182: Little more testing delete-on-close: Check flag with qfileinfo. Volker (This used to be commit 47a9df946d5ff967289fba0ff4209711ead11e31)
2005-07-06 13:24:38 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, expect_it == io.standard_info.out.delete_pending,
talloc_asprintf(tctx, "%s - Expected del_on_close flag %d, qfileinfo/standard_info gave %d\n",
where, expect_it, io.standard_info.out.delete_pending));
r8182: Little more testing delete-on-close: Check flag with qfileinfo. Volker (This used to be commit 47a9df946d5ff967289fba0ff4209711ead11e31)
2005-07-06 13:24:38 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, nlink == io.standard_info.out.nlink,
talloc_asprintf(tctx, "%s - Expected nlink %d, qfileinfo/standard_info gave %d",
where, nlink, io.all_info.out.nlink));
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
}
status = smbcli_qpathinfo(cli->tree, fname,
&c_time, &a_time, &m_time,
&size, &mode);
if (expect_it) {
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_equal(tctx, status, NT_STATUS_DELETE_PENDING,
"qpathinfo did not give correct error code");
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
} else {
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, status,
talloc_asprintf(tctx, "qpathinfo failed (%s)", where));
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
}
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
return true;
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
}
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
#define CHECK_STATUS(_cli, _expected) \
torture_assert_ntstatus_equal(tctx, _cli->tree->session->transport->error.e.nt_status, _expected, \
"Incorrect status")
r9431: Check an error code in BASE-DELETE. Volker (This used to be commit 71571fffc0493a5658c5980e6ebe4d8f9ada4699)
2005-08-20 21:28:30 +00:00
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
static const char *fname = "\\delete.file";
static const char *fname_new = "\\delete.new";
static const char *dname = "\\delete.dir";
r13275: With a liberal sprinkling of smb_raw_exit this now passes for me for W2K and W2K3... booting the XP box to test it... Jeremy. (This used to be commit 0133ba78c15b10e925f4e26e163656aa7cd11476)
2006-02-01 04:41:54 +00:00
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
static void del_clean_area(struct smbcli_state *cli1, struct smbcli_state *cli2)
{
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
smb_raw_exit(cli1->session);
smb_raw_exit(cli2->session);
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
smbcli_deltree(cli1->tree, dname);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
smbcli_setatr(cli1->tree, fname, 0, 0);
smbcli_unlink(cli1->tree, fname);
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
smbcli_setatr(cli1->tree, fname_new, 0, 0);
smbcli_unlink(cli1->tree, fname_new);
}
/* Test 1 - this should delete the file on close. */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest1(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
del_clean_area(cli1, cli2);
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
r4035: more effort on consistent naming of the access mask bits. This removes the duplicate named SEC_RIGHTS_MAXIMUM_ALLOWED and SEC_RIGHTS_FULL_CONTROL, which are just other names for SEC_FLAG_MAXIMUM_ALLOWED and SEC_RIGHTS_FILE_ALL. The latter names match the new naming conventions in security.idl Also added names for the generic->specific mappings for files are directories (This used to be commit 17a4e0b3aca227b40957ed1e0c57e498debc6ddf)
2004-12-02 04:37:36 +00:00
SEC_RIGHTS_FILE_ALL,
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_DELETE, NTCREATEX_DISP_OVERWRITE_IF,
NTCREATEX_OPTIONS_DELETE_ON_CLOSE, 0);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_close(cli1->tree, fnum1),
talloc_asprintf(tctx, "close failed (%s)", smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
fnum1 = smbcli_open(cli1->tree, fname, O_RDWR, DENY_NONE);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 == -1, talloc_asprintf(tctx, "open of %s succeeded (should fail)",
fname));
r13707: expanded the delete on close test some more, and make it easier to track down where a error is happening. The semantics sure are strange! (This used to be commit f722aed66d19edc6b49ecde927853aaa7895ab91)
2006-02-27 03:23:28 +00:00
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return True;
}
/* Test 2 - this should delete the file on close. */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest2(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
del_clean_area(cli1, cli2);
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
r4035: more effort on consistent naming of the access mask bits. This removes the duplicate named SEC_RIGHTS_MAXIMUM_ALLOWED and SEC_RIGHTS_FULL_CONTROL, which are just other names for SEC_FLAG_MAXIMUM_ALLOWED and SEC_RIGHTS_FILE_ALL. The latter names match the new naming conventions in security.idl Also added names for the generic->specific mappings for files are directories (This used to be commit 17a4e0b3aca227b40957ed1e0c57e498debc6ddf)
2004-12-02 04:37:36 +00:00
SEC_RIGHTS_FILE_ALL,
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
FILE_ATTRIBUTE_NORMAL, NTCREATEX_SHARE_ACCESS_NONE,
NTCREATEX_DISP_OVERWRITE_IF, 0, 0);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1,
talloc_asprintf(tctx, "open of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_nt_delete_on_close(cli1->tree, fnum1, True),
talloc_asprintf(tctx, "setting delete_on_close failed (%s)",
smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_close(cli1->tree, fnum1),
talloc_asprintf(tctx, "close failed (%s)",
smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
fnum1 = smbcli_open(cli1->tree, fname, O_RDONLY, DENY_NONE);
if (fnum1 != -1) {
r3150: printing __location__ is more useful than a operation number (This used to be commit 022b21460a572803b86ef5c11f6fe0b6fa1dcae1)
2004-10-24 10:27:38 +00:00
printf("(%s) open of %s succeeded should have been deleted on close !\n",
__location__, fname);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
if (NT_STATUS_IS_ERR(smbcli_close(cli1->tree, fnum1))) {
r3150: printing __location__ is more useful than a operation number (This used to be commit 022b21460a572803b86ef5c11f6fe0b6fa1dcae1)
2004-10-24 10:27:38 +00:00
printf("(%s) close failed (%s)\n",
__location__, smbcli_errstr(cli1->tree));
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return False;
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
}
smbcli_unlink(cli1->tree, fname);
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
}
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
return true;
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
}
r13275: With a liberal sprinkling of smb_raw_exit this now passes for me for W2K and W2K3... booting the XP box to test it... Jeremy. (This used to be commit 0133ba78c15b10e925f4e26e163656aa7cd11476)
2006-02-01 04:41:54 +00:00
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
/* Test 3 - ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest3(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
int fnum2 = -1;
del_clean_area(cli1, cli2);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r3150: printing __location__ is more useful than a operation number (This used to be commit 022b21460a572803b86ef5c11f6fe0b6fa1dcae1)
2004-10-24 10:27:38 +00:00
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
r4035: more effort on consistent naming of the access mask bits. This removes the duplicate named SEC_RIGHTS_MAXIMUM_ALLOWED and SEC_RIGHTS_FULL_CONTROL, which are just other names for SEC_FLAG_MAXIMUM_ALLOWED and SEC_RIGHTS_FILE_ALL. The latter names match the new naming conventions in security.idl Also added names for the generic->specific mappings for files are directories (This used to be commit 17a4e0b3aca227b40957ed1e0c57e498debc6ddf)
2004-12-02 04:37:36 +00:00
SEC_RIGHTS_FILE_ALL,
r3150: printing __location__ is more useful than a operation number (This used to be commit 022b21460a572803b86ef5c11f6fe0b6fa1dcae1)
2004-10-24 10:27:38 +00:00
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|NTCREATEX_SHARE_ACCESS_WRITE,
NTCREATEX_DISP_OVERWRITE_IF, 0, 0);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open - 1 of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
/* This should fail with a sharing violation - open for delete is only compatible
with SHARE_DELETE. */
r3150: printing __location__ is more useful than a operation number (This used to be commit 022b21460a572803b86ef5c11f6fe0b6fa1dcae1)
2004-10-24 10:27:38 +00:00
fnum2 = smbcli_nt_create_full(cli1->tree, fname, 0,
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
SEC_RIGHTS_FILE_READ,
r3150: printing __location__ is more useful than a operation number (This used to be commit 022b21460a572803b86ef5c11f6fe0b6fa1dcae1)
2004-10-24 10:27:38 +00:00
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|NTCREATEX_SHARE_ACCESS_WRITE,
NTCREATEX_DISP_OPEN, 0, 0);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum2 == -1,
talloc_asprintf(tctx, "open - 2 of %s succeeded - should have failed.",
fname));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
/* This should succeed. */
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
fnum2 = smbcli_nt_create_full(cli1->tree, fname, 0,
SEC_RIGHTS_FILE_READ,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|NTCREATEX_SHARE_ACCESS_WRITE|NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN, 0, 0);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum2 != -1, talloc_asprintf(tctx, "open - 2 of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx,
smbcli_nt_delete_on_close(cli1->tree, fnum1, True),
talloc_asprintf(tctx, "setting delete_on_close failed (%s)",
smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_close(cli1->tree, fnum1),
talloc_asprintf(tctx, "close 1 failed (%s)",
smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_close(cli1->tree, fnum2),
talloc_asprintf(tctx, "close 2 failed (%s)",
smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
/* This should fail - file should no longer be there. */
fnum1 = smbcli_open(cli1->tree, fname, O_RDONLY, DENY_NONE);
if (fnum1 != -1) {
r3150: printing __location__ is more useful than a operation number (This used to be commit 022b21460a572803b86ef5c11f6fe0b6fa1dcae1)
2004-10-24 10:27:38 +00:00
printf("(%s) open of %s succeeded should have been deleted on close !\n",
__location__, fname);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
if (NT_STATUS_IS_ERR(smbcli_close(cli1->tree, fnum1))) {
r3150: printing __location__ is more useful than a operation number (This used to be commit 022b21460a572803b86ef5c11f6fe0b6fa1dcae1)
2004-10-24 10:27:38 +00:00
printf("(%s) close failed (%s)\n",
__location__, smbcli_errstr(cli1->tree));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
}
smbcli_unlink(cli1->tree, fname);
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return False;
}
return True;
}
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
/* Test 4 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest4(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
int fnum2 = -1;
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
bool correct = True;
r13275: With a liberal sprinkling of smb_raw_exit this now passes for me for W2K and W2K3... booting the XP box to test it... Jeremy. (This used to be commit 0133ba78c15b10e925f4e26e163656aa7cd11476)
2006-02-01 04:41:54 +00:00
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
del_clean_area(cli1, cli2);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
SEC_FILE_READ_DATA |
SEC_FILE_WRITE_DATA |
SEC_STD_DELETE,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|NTCREATEX_SHARE_ACCESS_WRITE,
NTCREATEX_DISP_OVERWRITE_IF, 0, 0);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
/* This should succeed. */
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
fnum2 = smbcli_nt_create_full(cli1->tree, fname, 0,
SEC_RIGHTS_FILE_READ,
r3152: reformatted some of the delete test code (This used to be commit d7b0dece6fbc2b5f8216b6a6a41e3db76b8627aa)
2004-10-24 12:38:32 +00:00
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ |
NTCREATEX_SHARE_ACCESS_WRITE |
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN, 0, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum2 != -1, talloc_asprintf(tctx, "open - 2 of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx,
smbcli_close(cli1->tree, fnum2),
talloc_asprintf(tctx, "close - 1 failed (%s)",
smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx,
smbcli_nt_delete_on_close(cli1->tree, fnum1, True),
talloc_asprintf(tctx, "setting delete_on_close failed (%s)",
smbcli_errstr(cli1->tree)));
r23029: fixed formatting (This used to be commit 72ee4707725e9bbc9ab564315c7d78bdc564256f)
2007-05-21 05:40:04 +00:00
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
/* This should fail - no more opens once delete on close set. */
r3152: reformatted some of the delete test code (This used to be commit d7b0dece6fbc2b5f8216b6a6a41e3db76b8627aa)
2004-10-24 12:38:32 +00:00
fnum2 = smbcli_nt_create_full(cli1->tree, fname, 0,
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
SEC_RIGHTS_FILE_READ,
r3152: reformatted some of the delete test code (This used to be commit d7b0dece6fbc2b5f8216b6a6a41e3db76b8627aa)
2004-10-24 12:38:32 +00:00
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|NTCREATEX_SHARE_ACCESS_WRITE|NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN, 0, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum2 == -1,
talloc_asprintf(tctx, "open - 3 of %s succeeded ! Should have failed.",
fname ));
r9431: Check an error code in BASE-DELETE. Volker (This used to be commit 71571fffc0493a5658c5980e6ebe4d8f9ada4699)
2005-08-20 21:28:30 +00:00
CHECK_STATUS(cli1, NT_STATUS_DELETE_PENDING);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_close(cli1->tree, fnum1),
talloc_asprintf(tctx, "close - 2 failed (%s)",
smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return correct;
}
/* Test 5 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest5(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
del_clean_area(cli1, cli2);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
fnum1 = smbcli_open(cli1->tree, fname, O_RDWR|O_CREAT, DENY_NONE);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
/* This should fail - only allowed on NT opens with DELETE access. */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, !NT_STATUS_IS_OK(smbcli_nt_delete_on_close(cli1->tree, fnum1, True)),
"setting delete_on_close on OpenX file succeeded - should fail !");
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_close(cli1->tree, fnum1),
talloc_asprintf(tctx, "close - 2 failed (%s)", smbcli_errstr(cli1->tree)));
r13275: With a liberal sprinkling of smb_raw_exit this now passes for me for W2K and W2K3... booting the XP box to test it... Jeremy. (This used to be commit 0133ba78c15b10e925f4e26e163656aa7cd11476)
2006-02-01 04:41:54 +00:00
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return True;
}
/* Test 6 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest6(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
del_clean_area(cli1, cli2);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
SEC_FILE_READ_DATA | SEC_FILE_WRITE_DATA,
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ |
NTCREATEX_SHARE_ACCESS_WRITE |
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OVERWRITE_IF, 0, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
/* This should fail - only allowed on NT opens with DELETE access. */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx,
!NT_STATUS_IS_OK(smbcli_nt_delete_on_close(cli1->tree, fnum1, True)),
"setting delete_on_close on file with no delete access succeeded - should fail !");
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx,
smbcli_close(cli1->tree, fnum1),
talloc_asprintf(tctx, "close - 2 failed (%s)",
smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return True;
}
/* Test 7 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest7(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
bool correct = True;
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
del_clean_area(cli1, cli2);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
SEC_FILE_READ_DATA |
SEC_FILE_WRITE_DATA |
SEC_STD_DELETE,
FILE_ATTRIBUTE_NORMAL, 0,
NTCREATEX_DISP_OVERWRITE_IF, 0, 0);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_nt_delete_on_close(cli1->tree, fnum1, True),
"setting delete_on_close on file failed !");
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum1, fname, True, __location__);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx,
smbcli_nt_delete_on_close(cli1->tree, fnum1, False),
"unsetting delete_on_close on file failed !");
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum1, fname, False, __location__);
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_close(cli1->tree, fnum1),
talloc_asprintf(tctx, "close - 2 failed (%s)", smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
/* This next open should succeed - we reset the flag. */
fnum1 = smbcli_open(cli1->tree, fname, O_RDONLY, DENY_NONE);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_close(cli1->tree, fnum1),
talloc_asprintf(tctx, "close - 2 failed (%s)",
smbcli_errstr(cli1->tree)));
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return correct;
}
/* Test 8 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest8(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
int fnum2 = -1;
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
bool correct = True;
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
del_clean_area(cli1, cli2);
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
SEC_FILE_READ_DATA|
SEC_FILE_WRITE_DATA|
SEC_STD_DELETE,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|NTCREATEX_SHARE_ACCESS_WRITE|NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OVERWRITE_IF, 0, 0);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1,
talloc_asprintf(tctx, "open of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
fnum2 = smbcli_nt_create_full(cli2->tree, fname, 0,
SEC_FILE_READ_DATA|
SEC_FILE_WRITE_DATA|
SEC_STD_DELETE,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|NTCREATEX_SHARE_ACCESS_WRITE|NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN, 0, 0);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum2 != -1, talloc_asprintf(tctx, "open of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx,
smbcli_nt_delete_on_close(cli1->tree, fnum1, True),
"setting delete_on_close on file failed !");
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum1, fname, True, __location__);
correct &= check_delete_on_close(tctx, cli2, fnum2, fname, True, __location__);
r8177: More explorations. The share mode db is actually checked on qpathinfo even before the delete-on-close is executed. Volker (This used to be commit 124f3b74ca2ece9ba73737c3ccb75e1730973f19)
2005-07-06 08:13:11 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_close(cli1->tree, fnum1),
talloc_asprintf(tctx, "close - 1 failed (%s)",
smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, -1, fname, True, __location__);
correct &= check_delete_on_close(tctx, cli2, fnum2, fname, True, __location__);
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_close(cli2->tree, fnum2),
talloc_asprintf(tctx, "close - 2 failed (%s)", smbcli_errstr(cli2->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
/* This should fail.. */
fnum1 = smbcli_open(cli1->tree, fname, O_RDONLY, DENY_NONE);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 == -1,
talloc_asprintf(tctx, "open of %s succeeded should have been deleted on close !\n", fname));
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return correct;
}
/* Test 9 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest9(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
del_clean_area(cli1, cli2);
r13275: With a liberal sprinkling of smb_raw_exit this now passes for me for W2K and W2K3... booting the XP box to test it... Jeremy. (This used to be commit 0133ba78c15b10e925f4e26e163656aa7cd11476)
2006-02-01 04:41:54 +00:00
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
/* This should fail - we need to set DELETE_ACCESS. */
r3152: reformatted some of the delete test code (This used to be commit d7b0dece6fbc2b5f8216b6a6a41e3db76b8627aa)
2004-10-24 12:38:32 +00:00
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
SEC_FILE_READ_DATA|SEC_FILE_WRITE_DATA,
r3152: reformatted some of the delete test code (This used to be commit d7b0dece6fbc2b5f8216b6a6a41e3db76b8627aa)
2004-10-24 12:38:32 +00:00
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_NONE,
NTCREATEX_DISP_OVERWRITE_IF,
NTCREATEX_OPTIONS_DELETE_ON_CLOSE, 0);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 == -1,
talloc_asprintf(tctx, "open of %s succeeded should have failed!",
fname));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return True;
}
/* Test 10 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest10(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
del_clean_area(cli1, cli2);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r3152: reformatted some of the delete test code (This used to be commit d7b0dece6fbc2b5f8216b6a6a41e3db76b8627aa)
2004-10-24 12:38:32 +00:00
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
SEC_FILE_READ_DATA|
SEC_FILE_WRITE_DATA|
SEC_STD_DELETE,
r3152: reformatted some of the delete test code (This used to be commit d7b0dece6fbc2b5f8216b6a6a41e3db76b8627aa)
2004-10-24 12:38:32 +00:00
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_NONE,
NTCREATEX_DISP_OVERWRITE_IF,
NTCREATEX_OPTIONS_DELETE_ON_CLOSE, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1,
talloc_asprintf(tctx, "open of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
/* This should delete the file. */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_close(cli1->tree, fnum1),
talloc_asprintf(tctx, "close failed (%s)",
smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
/* This should fail.. */
fnum1 = smbcli_open(cli1->tree, fname, O_RDONLY, DENY_NONE);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 == -1,
talloc_asprintf(tctx, "open of %s succeeded should have been deleted on close !",
fname));
return true;
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
}
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
/* Test 11 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest11(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
NTSTATUS status;
del_clean_area(cli1, cli2);
/* test 11 - does having read only attribute still allow delete on close. */
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
r3155: reformat a delete test (This used to be commit 739c9e401cfbe04f2596e5b4b178243263218c04)
2004-10-24 12:55:12 +00:00
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
r4035: more effort on consistent naming of the access mask bits. This removes the duplicate named SEC_RIGHTS_MAXIMUM_ALLOWED and SEC_RIGHTS_FULL_CONTROL, which are just other names for SEC_FLAG_MAXIMUM_ALLOWED and SEC_RIGHTS_FILE_ALL. The latter names match the new naming conventions in security.idl Also added names for the generic->specific mappings for files are directories (This used to be commit 17a4e0b3aca227b40957ed1e0c57e498debc6ddf)
2004-12-02 04:37:36 +00:00
SEC_RIGHTS_FILE_ALL,
r3155: reformat a delete test (This used to be commit 739c9e401cfbe04f2596e5b4b178243263218c04)
2004-10-24 12:55:12 +00:00
FILE_ATTRIBUTE_READONLY,
NTCREATEX_SHARE_ACCESS_NONE,
NTCREATEX_DISP_OVERWRITE_IF, 0, 0);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
status = smbcli_nt_delete_on_close(cli1->tree, fnum1, True);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_equal(tctx, status, NT_STATUS_CANNOT_DELETE,
talloc_asprintf(tctx, "setting delete_on_close should fail with NT_STATUS_CANNOT_DELETE. Got %s instead)", smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_close(cli1->tree, fnum1),
talloc_asprintf(tctx, "close failed (%s)",
smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return True;
}
r13275: With a liberal sprinkling of smb_raw_exit this now passes for me for W2K and W2K3... booting the XP box to test it... Jeremy. (This used to be commit 0133ba78c15b10e925f4e26e163656aa7cd11476)
2006-02-01 04:41:54 +00:00
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
/* Test 12 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest12(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
NTSTATUS status;
r13275: With a liberal sprinkling of smb_raw_exit this now passes for me for W2K and W2K3... booting the XP box to test it... Jeremy. (This used to be commit 0133ba78c15b10e925f4e26e163656aa7cd11476)
2006-02-01 04:41:54 +00:00
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
del_clean_area(cli1, cli2);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r12150: Reformatting (This used to be commit aaa21b7132ec81008ad2ec7b2aafc4604d13093d)
2005-12-09 19:36:40 +00:00
/* test 12 - does having read only attribute still allow delete on
* close at time of open. */
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
r4035: more effort on consistent naming of the access mask bits. This removes the duplicate named SEC_RIGHTS_MAXIMUM_ALLOWED and SEC_RIGHTS_FULL_CONTROL, which are just other names for SEC_FLAG_MAXIMUM_ALLOWED and SEC_RIGHTS_FILE_ALL. The latter names match the new naming conventions in security.idl Also added names for the generic->specific mappings for files are directories (This used to be commit 17a4e0b3aca227b40957ed1e0c57e498debc6ddf)
2004-12-02 04:37:36 +00:00
SEC_RIGHTS_FILE_ALL,
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
FILE_ATTRIBUTE_READONLY,
r12150: Reformatting (This used to be commit aaa21b7132ec81008ad2ec7b2aafc4604d13093d)
2005-12-09 19:36:40 +00:00
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OVERWRITE_IF,
r4011: get rid of rpc_secdes.h and replace it with a single sane set of definitions for security access masks, in security.idl The previous definitions were inconsistently named, and contained many duplicate and misleading entries. I kept finding myself tripping up while using them. (This used to be commit 01c0fa722f80ceeb3f81f01987de95f365a2ed3d)
2004-11-30 04:33:27 +00:00
NTCREATEX_OPTIONS_DELETE_ON_CLOSE, 0);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 == -1,
talloc_asprintf(tctx, "open of %s succeeded. Should fail with "
"NT_STATUS_CANNOT_DELETE.\n", fname));
status = smbcli_nt_error(cli1->tree);
torture_assert_ntstatus_equal(tctx, status, NT_STATUS_CANNOT_DELETE,
talloc_asprintf(tctx, "setting delete_on_close on open should "
r12150: Reformatting (This used to be commit aaa21b7132ec81008ad2ec7b2aafc4604d13093d)
2005-12-09 19:36:40 +00:00
"fail with NT_STATUS_CANNOT_DELETE. Got %s "
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
"instead)",
smbcli_errstr(cli1->tree)));
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
return true;
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
}
/* Test 13 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest13(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
int fnum2 = -1;
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
bool correct = True;
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
del_clean_area(cli1, cli2);
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
/* Test 13: Does resetting the delete on close flag affect a second
* fd? */
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
SEC_FILE_READ_DATA|
SEC_FILE_WRITE_DATA|
SEC_STD_DELETE,
FILE_ATTRIBUTE_NORMAL,
r12150: Reformatting (This used to be commit aaa21b7132ec81008ad2ec7b2aafc4604d13093d)
2005-12-09 19:36:40 +00:00
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
NTCREATEX_DISP_OVERWRITE_IF,
0, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1,
talloc_asprintf(tctx, "open of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
fnum2 = smbcli_nt_create_full(cli2->tree, fname, 0,
SEC_FILE_READ_DATA|
SEC_FILE_WRITE_DATA|
SEC_STD_DELETE,
FILE_ATTRIBUTE_NORMAL,
r12150: Reformatting (This used to be commit aaa21b7132ec81008ad2ec7b2aafc4604d13093d)
2005-12-09 19:36:40 +00:00
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
NTCREATEX_DISP_OPEN, 0, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum2 != -1, talloc_asprintf(tctx,
"open of %s failed (%s)",
fname, smbcli_errstr(cli2->tree)));
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx,
smbcli_nt_delete_on_close(cli1->tree, fnum1,
True),
"setting delete_on_close on file failed !");
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum1, fname, True, __location__);
correct &= check_delete_on_close(tctx, cli2, fnum2, fname, True, __location__);
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_nt_delete_on_close(cli2->tree, fnum2,
False),
"setting delete_on_close on file failed !");
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum1, fname, False, __location__);
correct &= check_delete_on_close(tctx, cli2, fnum2, fname, False, __location__);
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_close(cli1->tree, fnum1),
talloc_asprintf(tctx, "close - 1 failed (%s)",
smbcli_errstr(cli1->tree)));
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, smbcli_close(cli2->tree, fnum2),
talloc_asprintf(tctx, "close - 2 failed (%s)",
smbcli_errstr(cli2->tree)));
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
fnum1 = smbcli_open(cli1->tree, fname, O_RDONLY, DENY_NONE);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open of %s failed!",
fname));
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
smbcli_close(cli1->tree, fnum1);
r13275: With a liberal sprinkling of smb_raw_exit this now passes for me for W2K and W2K3... booting the XP box to test it... Jeremy. (This used to be commit 0133ba78c15b10e925f4e26e163656aa7cd11476)
2006-02-01 04:41:54 +00:00
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return correct;
}
/* Test 14 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest14(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int dnum1 = -1;
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
bool correct = true;
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
del_clean_area(cli1, cli2);
r8179: Delete-on-close is really a shared DB. Setting it on one connection(!) and resetting it on another resets it for both. Volker (This used to be commit 30bd7e36669dbb2fd7d85a1cd72927370267d616)
2005-07-06 09:50:31 +00:00
r8185: Delete on close on directories: Creating a file in a directory with delete-on-close set returns DELETE_PENDING, and trying to set the flag on a non-empty directory returns DIRECTORY_NOT_EMPTY. Volker (This used to be commit 5680f34778b2f5291936f4d4fb937a7713696c52)
2005-07-06 14:56:45 +00:00
/* Test 14 -- directory */
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
dnum1 = smbcli_nt_create_full(cli1->tree, dname, 0,
r8185: Delete on close on directories: Creating a file in a directory with delete-on-close set returns DELETE_PENDING, and trying to set the flag on a non-empty directory returns DIRECTORY_NOT_EMPTY. Volker (This used to be commit 5680f34778b2f5291936f4d4fb937a7713696c52)
2005-07-06 14:56:45 +00:00
SEC_FILE_READ_DATA|
SEC_FILE_WRITE_DATA|
SEC_STD_DELETE,
FILE_ATTRIBUTE_DIRECTORY,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_CREATE, 0, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, dnum1 != -1, talloc_asprintf(tctx, "open of %s failed: %s!",
dname, smbcli_errstr(cli1->tree)));
r8185: Delete on close on directories: Creating a file in a directory with delete-on-close set returns DELETE_PENDING, and trying to set the flag on a non-empty directory returns DIRECTORY_NOT_EMPTY. Volker (This used to be commit 5680f34778b2f5291936f4d4fb937a7713696c52)
2005-07-06 14:56:45 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, dnum1, dname, False, __location__);
torture_assert_ntstatus_ok(tctx, smbcli_nt_delete_on_close(cli1->tree, dnum1, True),
"setting delete_on_close on file failed !");
correct &= check_delete_on_close(tctx, cli1, dnum1, dname, True, __location__);
r8185: Delete on close on directories: Creating a file in a directory with delete-on-close set returns DELETE_PENDING, and trying to set the flag on a non-empty directory returns DIRECTORY_NOT_EMPTY. Volker (This used to be commit 5680f34778b2f5291936f4d4fb937a7713696c52)
2005-07-06 14:56:45 +00:00
smbcli_close(cli1->tree, dnum1);
/* Now it should be gone... */
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
dnum1 = smbcli_nt_create_full(cli1->tree, dname, 0,
r8185: Delete on close on directories: Creating a file in a directory with delete-on-close set returns DELETE_PENDING, and trying to set the flag on a non-empty directory returns DIRECTORY_NOT_EMPTY. Volker (This used to be commit 5680f34778b2f5291936f4d4fb937a7713696c52)
2005-07-06 14:56:45 +00:00
SEC_FILE_READ_DATA|
SEC_FILE_WRITE_DATA|
SEC_STD_DELETE,
FILE_ATTRIBUTE_DIRECTORY,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN, 0, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, dnum1 == -1, "setting delete_on_close on file succeeded !");
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return correct;
}
/* Test 15 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest15(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
bool correct = true;
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
int fnum2 = -1;
NTSTATUS status;
del_clean_area(cli1, cli2);
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
/* Test 15: delete on close under rename */
r12154: Torture test for bug # 3303. Jeremy, to run this against Samba3 at all you need to insert a "goto line 957" in line 548. Without this we fail some tests before # 16 and bail out. While looking at it, you wanted to fix the directory-based ones a while ago.... :-)) Volker (This used to be commit 45cd224102f21364c4f6ca056417f956f21eb02e)
2005-12-09 21:49:11 +00:00
smbcli_setatr(cli1->tree, fname, 0, 0);
smbcli_unlink(cli1->tree, fname);
smbcli_unlink(cli1->tree, fname_new);
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
SEC_FILE_READ_DATA,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OVERWRITE_IF,
0, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1,
talloc_asprintf(tctx, "open - 1 of %s failed (%s)", fname, smbcli_errstr(cli1->tree)));
r12154: Torture test for bug # 3303. Jeremy, to run this against Samba3 at all you need to insert a "goto line 957" in line 548. Without this we fail some tests before # 16 and bail out. While looking at it, you wanted to fix the directory-based ones a while ago.... :-)) Volker (This used to be commit 45cd224102f21364c4f6ca056417f956f21eb02e)
2005-12-09 21:49:11 +00:00
status = smbcli_rename(cli2->tree, fname, fname_new);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, status, "renaming failed!");
r12154: Torture test for bug # 3303. Jeremy, to run this against Samba3 at all you need to insert a "goto line 957" in line 548. Without this we fail some tests before # 16 and bail out. While looking at it, you wanted to fix the directory-based ones a while ago.... :-)) Volker (This used to be commit 45cd224102f21364c4f6ca056417f956f21eb02e)
2005-12-09 21:49:11 +00:00
fnum2 = smbcli_nt_create_full(cli2->tree, fname_new, 0,
SEC_GENERIC_ALL,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OVERWRITE_IF,
0, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum2 != -1,
talloc_asprintf(tctx, "open - 1 of %s failed (%s)",
fname_new, smbcli_errstr(cli1->tree)));
r12154: Torture test for bug # 3303. Jeremy, to run this against Samba3 at all you need to insert a "goto line 957" in line 548. Without this we fail some tests before # 16 and bail out. While looking at it, you wanted to fix the directory-based ones a while ago.... :-)) Volker (This used to be commit 45cd224102f21364c4f6ca056417f956f21eb02e)
2005-12-09 21:49:11 +00:00
status = smbcli_nt_delete_on_close(cli2->tree, fnum2, True);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, status,
"setting delete_on_close on file failed !");
r12154: Torture test for bug # 3303. Jeremy, to run this against Samba3 at all you need to insert a "goto line 957" in line 548. Without this we fail some tests before # 16 and bail out. While looking at it, you wanted to fix the directory-based ones a while ago.... :-)) Volker (This used to be commit 45cd224102f21364c4f6ca056417f956f21eb02e)
2005-12-09 21:49:11 +00:00
smbcli_close(cli2->tree, fnum2);
/* The file should be around under the new name, there's a second
* handle open */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum1, fname_new, True, __location__);
r12154: Torture test for bug # 3303. Jeremy, to run this against Samba3 at all you need to insert a "goto line 957" in line 548. Without this we fail some tests before # 16 and bail out. While looking at it, you wanted to fix the directory-based ones a while ago.... :-)) Volker (This used to be commit 45cd224102f21364c4f6ca056417f956f21eb02e)
2005-12-09 21:49:11 +00:00
fnum2 = smbcli_nt_create_full(cli2->tree, fname, 0,
SEC_GENERIC_ALL,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OVERWRITE_IF,
0, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum2 != -1, talloc_asprintf(tctx, "open - 1 of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r12154: Torture test for bug # 3303. Jeremy, to run this against Samba3 at all you need to insert a "goto line 957" in line 548. Without this we fail some tests before # 16 and bail out. While looking at it, you wanted to fix the directory-based ones a while ago.... :-)) Volker (This used to be commit 45cd224102f21364c4f6ca056417f956f21eb02e)
2005-12-09 21:49:11 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli2, fnum2, fname, False, __location__);
r13707: expanded the delete on close test some more, and make it easier to track down where a error is happening. The semantics sure are strange! (This used to be commit f722aed66d19edc6b49ecde927853aaa7895ab91)
2006-02-27 03:23:28 +00:00
r12154: Torture test for bug # 3303. Jeremy, to run this against Samba3 at all you need to insert a "goto line 957" in line 548. Without this we fail some tests before # 16 and bail out. While looking at it, you wanted to fix the directory-based ones a while ago.... :-)) Volker (This used to be commit 45cd224102f21364c4f6ca056417f956f21eb02e)
2005-12-09 21:49:11 +00:00
smbcli_close(cli2->tree, fnum2);
smbcli_close(cli1->tree, fnum1);
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
SEC_FILE_READ_EA,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
0, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open - 1 of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r12154: Torture test for bug # 3303. Jeremy, to run this against Samba3 at all you need to insert a "goto line 957" in line 548. Without this we fail some tests before # 16 and bail out. While looking at it, you wanted to fix the directory-based ones a while ago.... :-)) Volker (This used to be commit 45cd224102f21364c4f6ca056417f956f21eb02e)
2005-12-09 21:49:11 +00:00
smbcli_close(cli1->tree, fnum1);
fnum1 = smbcli_nt_create_full(cli1->tree, fname_new, 0,
SEC_FILE_READ_EA,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
0, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 == -1,
"smbcli_open succeeded, should have "
"failed");
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return correct;
}
/* Test 16 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest16(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
int fnum2 = -1;
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
bool correct = true;
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
del_clean_area(cli1, cli2);
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
/* Test 16. */
r13270: Add tests for even more insane delete-on-close semantics. Jeremy. (This used to be commit ae0851ce667c0559e786c3a83389ccfddce1a813)
2006-02-01 02:30:57 +00:00
/* Ensure the file doesn't already exist. */
smbcli_close(cli1->tree, fnum1);
smbcli_close(cli1->tree, fnum2);
smbcli_setatr(cli1->tree, fname, 0, 0);
smbcli_unlink(cli1->tree, fname);
/* Firstly create with all access, but delete on close. */
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
SEC_RIGHTS_FILE_ALL,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_CREATE,
NTCREATEX_OPTIONS_DELETE_ON_CLOSE, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert (tctx, fnum1 != -1, talloc_asprintf(tctx, "open - 1 of %s failed (%s)", fname, smbcli_errstr(cli1->tree)));
r13270: Add tests for even more insane delete-on-close semantics. Jeremy. (This used to be commit ae0851ce667c0559e786c3a83389ccfddce1a813)
2006-02-01 02:30:57 +00:00
/* The delete on close bit is *not* reported as being set. */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum1, fname, False, __location__);
r13707: expanded the delete on close test some more, and make it easier to track down where a error is happening. The semantics sure are strange! (This used to be commit f722aed66d19edc6b49ecde927853aaa7895ab91)
2006-02-27 03:23:28 +00:00
/* The delete on close bit is *not* reported as being set. */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, -1, fname, False, __location__);
correct &= check_delete_on_close(tctx, cli2, -1, fname, False, __location__);
r13270: Add tests for even more insane delete-on-close semantics. Jeremy. (This used to be commit ae0851ce667c0559e786c3a83389ccfddce1a813)
2006-02-01 02:30:57 +00:00
/* Now try opening again for read-only. */
r13707: expanded the delete on close test some more, and make it easier to track down where a error is happening. The semantics sure are strange! (This used to be commit f722aed66d19edc6b49ecde927853aaa7895ab91)
2006-02-27 03:23:28 +00:00
fnum2 = smbcli_nt_create_full(cli2->tree, fname, 0,
r13270: Add tests for even more insane delete-on-close semantics. Jeremy. (This used to be commit ae0851ce667c0559e786c3a83389ccfddce1a813)
2006-02-01 02:30:57 +00:00
SEC_RIGHTS_FILE_READ,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
0, 0);
/* Should work. */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum2 != -1, talloc_asprintf(tctx, "open - 1 of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r13270: Add tests for even more insane delete-on-close semantics. Jeremy. (This used to be commit ae0851ce667c0559e786c3a83389ccfddce1a813)
2006-02-01 02:30:57 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum1, fname, False, __location__);
correct &= check_delete_on_close(tctx, cli1, -1, fname, False, __location__);
correct &= check_delete_on_close(tctx, cli2, fnum2, fname, False, __location__);
correct &= check_delete_on_close(tctx, cli2, -1, fname, False, __location__);
r13707: expanded the delete on close test some more, and make it easier to track down where a error is happening. The semantics sure are strange! (This used to be commit f722aed66d19edc6b49ecde927853aaa7895ab91)
2006-02-27 03:23:28 +00:00
r13270: Add tests for even more insane delete-on-close semantics. Jeremy. (This used to be commit ae0851ce667c0559e786c3a83389ccfddce1a813)
2006-02-01 02:30:57 +00:00
smbcli_close(cli1->tree, fnum1);
r13707: expanded the delete on close test some more, and make it easier to track down where a error is happening. The semantics sure are strange! (This used to be commit f722aed66d19edc6b49ecde927853aaa7895ab91)
2006-02-27 03:23:28 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli2, fnum2, fname, True, __location__);
correct &= check_delete_on_close(tctx, cli2, -1, fname, True, __location__);
r13707: expanded the delete on close test some more, and make it easier to track down where a error is happening. The semantics sure are strange! (This used to be commit f722aed66d19edc6b49ecde927853aaa7895ab91)
2006-02-27 03:23:28 +00:00
smbcli_close(cli2->tree, fnum2);
r13270: Add tests for even more insane delete-on-close semantics. Jeremy. (This used to be commit ae0851ce667c0559e786c3a83389ccfddce1a813)
2006-02-01 02:30:57 +00:00
/* And the file should be deleted ! */
fnum1 = smbcli_open(cli1->tree, fname, O_RDWR, DENY_NONE);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 == -1, talloc_asprintf(tctx, "open of %s succeeded (should fail)",
fname));
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return correct;
}
/* Test 17 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest17(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
int fnum2 = -1;
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
bool correct = True;
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
del_clean_area(cli1, cli2);
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
/* Test 17. */
r13270: Add tests for even more insane delete-on-close semantics. Jeremy. (This used to be commit ae0851ce667c0559e786c3a83389ccfddce1a813)
2006-02-01 02:30:57 +00:00
/* Ensure the file doesn't already exist. */
smbcli_close(cli1->tree, fnum1);
smbcli_close(cli1->tree, fnum2);
smbcli_setatr(cli1->tree, fname, 0, 0);
smbcli_unlink(cli1->tree, fname);
/* Firstly open and create with all access */
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
SEC_RIGHTS_FILE_ALL,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_CREATE,
0, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open - 1 of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r13270: Add tests for even more insane delete-on-close semantics. Jeremy. (This used to be commit ae0851ce667c0559e786c3a83389ccfddce1a813)
2006-02-01 02:30:57 +00:00
/* And close - just to create the file. */
smbcli_close(cli1->tree, fnum1);
/* Next open with all access, but add delete on close. */
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
SEC_RIGHTS_FILE_ALL,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
NTCREATEX_OPTIONS_DELETE_ON_CLOSE, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open - 1 of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r13270: Add tests for even more insane delete-on-close semantics. Jeremy. (This used to be commit ae0851ce667c0559e786c3a83389ccfddce1a813)
2006-02-01 02:30:57 +00:00
/* The delete on close bit is *not* reported as being set. */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum1, fname, False, __location__);
r13270: Add tests for even more insane delete-on-close semantics. Jeremy. (This used to be commit ae0851ce667c0559e786c3a83389ccfddce1a813)
2006-02-01 02:30:57 +00:00
/* Now try opening again for read-only. */
fnum2 = smbcli_nt_create_full(cli1->tree, fname, 0,
r13370: Added deltest21 - pull the rug out from a connection by socket close after setting delete on close flag. Jeremy. (This used to be commit fbea18e78f8a3c6dbb36aa935b7044c0fcf61da4)
2006-02-06 19:43:24 +00:00
SEC_RIGHTS_FILE_READ|
SEC_STD_DELETE,
r13270: Add tests for even more insane delete-on-close semantics. Jeremy. (This used to be commit ae0851ce667c0559e786c3a83389ccfddce1a813)
2006-02-01 02:30:57 +00:00
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
0, 0);
/* Should work. */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum2 != -1, talloc_asprintf(tctx, "open - 1 of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r13270: Add tests for even more insane delete-on-close semantics. Jeremy. (This used to be commit ae0851ce667c0559e786c3a83389ccfddce1a813)
2006-02-01 02:30:57 +00:00
r13707: expanded the delete on close test some more, and make it easier to track down where a error is happening. The semantics sure are strange! (This used to be commit f722aed66d19edc6b49ecde927853aaa7895ab91)
2006-02-27 03:23:28 +00:00
/* still not reported as being set on either */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum1, fname, False, __location__);
correct &= check_delete_on_close(tctx, cli1, fnum2, fname, False, __location__);
r13707: expanded the delete on close test some more, and make it easier to track down where a error is happening. The semantics sure are strange! (This used to be commit f722aed66d19edc6b49ecde927853aaa7895ab91)
2006-02-27 03:23:28 +00:00
r13270: Add tests for even more insane delete-on-close semantics. Jeremy. (This used to be commit ae0851ce667c0559e786c3a83389ccfddce1a813)
2006-02-01 02:30:57 +00:00
smbcli_close(cli1->tree, fnum1);
r13707: expanded the delete on close test some more, and make it easier to track down where a error is happening. The semantics sure are strange! (This used to be commit f722aed66d19edc6b49ecde927853aaa7895ab91)
2006-02-27 03:23:28 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum2, fname, False, __location__);
r13707: expanded the delete on close test some more, and make it easier to track down where a error is happening. The semantics sure are strange! (This used to be commit f722aed66d19edc6b49ecde927853aaa7895ab91)
2006-02-27 03:23:28 +00:00
r13270: Add tests for even more insane delete-on-close semantics. Jeremy. (This used to be commit ae0851ce667c0559e786c3a83389ccfddce1a813)
2006-02-01 02:30:57 +00:00
smbcli_close(cli1->tree, fnum2);
/* See if the file is deleted - shouldn't be.... */
fnum1 = smbcli_open(cli1->tree, fname, O_RDWR, DENY_NONE);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open of %s failed (should succeed) - %s",
fname, smbcli_errstr(cli1->tree)));
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return correct;
}
/* Test 18 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest18(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
int fnum2 = -1;
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
bool correct = True;
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
del_clean_area(cli1, cli2);
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
/* Test 18. With directories. */
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
/* Ensure the file doesn't already exist. */
smbcli_close(cli1->tree, fnum1);
smbcli_close(cli1->tree, fnum2);
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
smbcli_deltree(cli1->tree, dname);
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
/* Firstly create with all access, but delete on close. */
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
fnum1 = smbcli_nt_create_full(cli1->tree, dname, 0,
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
SEC_FILE_READ_DATA|
SEC_FILE_WRITE_DATA|
SEC_STD_DELETE,
FILE_ATTRIBUTE_DIRECTORY,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_CREATE,
NTCREATEX_OPTIONS_DIRECTORY|NTCREATEX_OPTIONS_DELETE_ON_CLOSE, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open - 1 of %s failed (%s)",
dname, smbcli_errstr(cli1->tree)));
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
/* The delete on close bit is *not* reported as being set. */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum1, dname, False, __location__);
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
/* Now try opening again for read-only. */
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
fnum2 = smbcli_nt_create_full(cli1->tree, dname, 0,
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
SEC_RIGHTS_FILE_READ,
FILE_ATTRIBUTE_DIRECTORY,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
NTCREATEX_OPTIONS_DIRECTORY, 0);
/* Should work. */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum2 != -1, talloc_asprintf(tctx, "open - 1 of %s failed (%s)",
dname, smbcli_errstr(cli1->tree)));
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum1, dname, False, __location__);
correct &= check_delete_on_close(tctx, cli1, fnum2, dname, False, __location__);
r13707: expanded the delete on close test some more, and make it easier to track down where a error is happening. The semantics sure are strange! (This used to be commit f722aed66d19edc6b49ecde927853aaa7895ab91)
2006-02-27 03:23:28 +00:00
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
smbcli_close(cli1->tree, fnum1);
r13707: expanded the delete on close test some more, and make it easier to track down where a error is happening. The semantics sure are strange! (This used to be commit f722aed66d19edc6b49ecde927853aaa7895ab91)
2006-02-27 03:23:28 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum2, dname, True, __location__);
r13707: expanded the delete on close test some more, and make it easier to track down where a error is happening. The semantics sure are strange! (This used to be commit f722aed66d19edc6b49ecde927853aaa7895ab91)
2006-02-27 03:23:28 +00:00
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
smbcli_close(cli1->tree, fnum2);
/* And the directory should be deleted ! */
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
fnum1 = smbcli_nt_create_full(cli1->tree, dname, 0,
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
SEC_RIGHTS_FILE_READ,
FILE_ATTRIBUTE_DIRECTORY,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
NTCREATEX_OPTIONS_DIRECTORY, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 == -1, talloc_asprintf(tctx, "open of %s succeeded (should fail)",
dname));
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return correct;
}
/* Test 19 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest19(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
int fnum2 = -1;
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
bool correct = True;
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
del_clean_area(cli1, cli2);
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
/* Test 19. */
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
smbcli_deltree(cli1->tree, dname);
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
/* Firstly open and create with all access */
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
fnum1 = smbcli_nt_create_full(cli1->tree, dname, 0,
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
SEC_FILE_READ_DATA|
SEC_FILE_WRITE_DATA|
SEC_STD_DELETE,
FILE_ATTRIBUTE_DIRECTORY,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_CREATE,
NTCREATEX_OPTIONS_DIRECTORY, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open - 1 of %s failed (%s)",
dname, smbcli_errstr(cli1->tree)));
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
/* And close - just to create the directory. */
smbcli_close(cli1->tree, fnum1);
/* Next open with all access, but add delete on close. */
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
fnum1 = smbcli_nt_create_full(cli1->tree, dname, 0,
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
SEC_FILE_READ_DATA|
SEC_FILE_WRITE_DATA|
SEC_STD_DELETE,
FILE_ATTRIBUTE_DIRECTORY,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
NTCREATEX_OPTIONS_DIRECTORY|NTCREATEX_OPTIONS_DELETE_ON_CLOSE, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1,
talloc_asprintf(tctx, "open - 1 of %s failed (%s)", fname, smbcli_errstr(cli1->tree)));
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
/* The delete on close bit is *not* reported as being set. */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum1, dname, False, __location__);
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
/* Now try opening again for read-only. */
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
fnum2 = smbcli_nt_create_full(cli1->tree, dname, 0,
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
SEC_RIGHTS_FILE_READ,
FILE_ATTRIBUTE_DIRECTORY,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
NTCREATEX_OPTIONS_DIRECTORY, 0);
/* Should work. */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum2 != -1, talloc_asprintf(tctx, "open - 1 of %s failed (%s)",
dname, smbcli_errstr(cli1->tree)));
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
smbcli_close(cli1->tree, fnum1);
r13707: expanded the delete on close test some more, and make it easier to track down where a error is happening. The semantics sure are strange! (This used to be commit f722aed66d19edc6b49ecde927853aaa7895ab91)
2006-02-27 03:23:28 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum2, dname, True, __location__);
r13707: expanded the delete on close test some more, and make it easier to track down where a error is happening. The semantics sure are strange! (This used to be commit f722aed66d19edc6b49ecde927853aaa7895ab91)
2006-02-27 03:23:28 +00:00
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
smbcli_close(cli1->tree, fnum2);
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
/* See if the file is deleted - for a directory this seems to be true ! */
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
fnum1 = smbcli_nt_create_full(cli1->tree, dname, 0,
r13271: Do the same tests with directories. Jeremy. (This used to be commit 7986af2ece2e707ee3d98dadb37af9fc5126d138)
2006-02-01 03:36:04 +00:00
SEC_RIGHTS_FILE_READ,
FILE_ATTRIBUTE_DIRECTORY,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
NTCREATEX_OPTIONS_DIRECTORY, 0);
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
CHECK_STATUS(cli1, NT_STATUS_OBJECT_NAME_NOT_FOUND);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 == -1,
talloc_asprintf(tctx, "open of %s succeeded (should fail)", dname));
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return correct;
}
/* Test 20 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest20(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
int fnum1 = -1;
int dnum1 = -1;
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
bool correct = True;
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
NTSTATUS status;
del_clean_area(cli1, cli2);
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
/* Test 20 -- non-empty directory hardest to get right... */
r21134: Attempt to quieten the build farm for Samba3 (This used to be commit 966c30ca591c410a4a229396c7919fe8f98bd066)
2007-02-03 17:46:32 +00:00
if (torture_setting_bool(tctx, "samba3", False)) {
return True;
}
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
smbcli_deltree(cli1->tree, dname);
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
dnum1 = smbcli_nt_create_full(cli1->tree, dname, 0,
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
SEC_FILE_READ_DATA|
SEC_FILE_WRITE_DATA|
SEC_STD_DELETE,
FILE_ATTRIBUTE_DIRECTORY,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_CREATE,
NTCREATEX_OPTIONS_DIRECTORY, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, dnum1 != -1, talloc_asprintf(tctx, "open of %s failed: %s!",
dname, smbcli_errstr(cli1->tree)));
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, dnum1, dname, False, __location__);
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
status = smbcli_nt_delete_on_close(cli1->tree, dnum1, True);
{
char *fullname;
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
asprintf(&fullname, "\\%s%s", dname, fname);
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
fnum1 = smbcli_open(cli1->tree, fullname, O_CREAT|O_RDWR,
DENY_NONE);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 == -1,
"smbcli_open succeeded, should have "
"failed with NT_STATUS_DELETE_PENDING"
);
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_equal(tctx,
smbcli_nt_error(cli1->tree),
NT_STATUS_DELETE_PENDING,
"smbcli_open failed");
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
}
status = smbcli_nt_delete_on_close(cli1->tree, dnum1, False);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx, status,
"setting delete_on_close on file failed !");
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
{
char *fullname;
r15865: using dirname for a variable isn't that good, but making it a global symbol is really bad... fix linking on sun1 in the build-farm metze (This used to be commit d073320f642ceeb49b11060aa958608248f3aff5)
2006-05-24 14:45:07 +00:00
asprintf(&fullname, "\\%s%s", dname, fname);
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
fnum1 = smbcli_open(cli1->tree, fullname, O_CREAT|O_RDWR,
DENY_NONE);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1,
talloc_asprintf(tctx, "smbcli_open failed: %s\n",
smbcli_errstr(cli1->tree)));
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
smbcli_close(cli1->tree, fnum1);
}
status = smbcli_nt_delete_on_close(cli1->tree, dnum1, True);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_equal(tctx, status, NT_STATUS_DIRECTORY_NOT_EMPTY,
"setting delete_on_close failed");
r13272: Re-arrange so all the normal tests we can pass come first. Jeremy. (This used to be commit 0b61a8df3c1b5b1a73bf6afd0404871286d3a2fc)
2006-02-01 04:09:02 +00:00
smbcli_close(cli1->tree, dnum1);
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
return correct;
}
r23099: New interesting delete on close tests :-). This will break all delete tests on the build farm until I check in my server patch. Jeremy. (This used to be commit c31f775f1c8f6f7e50d038d093898a1f4280be23)
2007-05-23 21:14:05 +00:00
/* Test 20a ... */
static bool deltest20a(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
{
int fnum1 = -1;
int fnum2 = -1;
bool correct = True;
del_clean_area(cli1, cli2);
/* Test 20a. */
/* Ensure the file doesn't already exist. */
smbcli_close(cli1->tree, fnum1);
smbcli_close(cli1->tree, fnum2);
smbcli_setatr(cli1->tree, fname, 0, 0);
smbcli_unlink(cli1->tree, fname);
/* Firstly open and create with all access */
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
SEC_RIGHTS_FILE_ALL,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_CREATE,
0, 0);
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open - 1 of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
/* Next open with all access, but add delete on close. */
fnum2 = smbcli_nt_create_full(cli2->tree, fname, 0,
SEC_RIGHTS_FILE_ALL,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
NTCREATEX_OPTIONS_DELETE_ON_CLOSE, 0);
torture_assert(tctx, fnum2 != -1, talloc_asprintf(tctx, "open - 2 of %s failed (%s)",
fname, smbcli_errstr(cli2->tree)));
/* The delete on close bit is *not* reported as being set. */
correct &= check_delete_on_close(tctx, cli1, fnum1, fname, False, __location__);
correct &= check_delete_on_close(tctx, cli2, fnum2, fname, False, __location__);
smbcli_close(cli1->tree, fnum1);
correct &= check_delete_on_close(tctx, cli2, fnum2, fname, False, __location__);
smbcli_close(cli2->tree, fnum2);
/* See if the file is deleted - should be.... */
fnum1 = smbcli_open(cli1->tree, fname, O_RDWR, DENY_NONE);
torture_assert(tctx, fnum1 == -1, talloc_asprintf(tctx, "open of %s succeeded (should fail) - %s",
fname, smbcli_errstr(cli1->tree)));
return correct;
}
r23103: Add a torture test for the insane semantics Steve's cifsfs client uses so we don't regress :-). Jeremy. (This used to be commit 5ce41ae40c5413bb9f15d1c47a095f63d7f167e0)
2007-05-23 23:11:16 +00:00
/* Test 20b ... */
r23104: Add comment explaining the new test. Jeremy. (This used to be commit 7b9b5c59a4ac2afca151ba12eba7c808057a677e)
2007-05-23 23:14:48 +00:00
/* This is the delete semantics that the cifsfs client depends on when
* trying to delete an open file on a Windows server. It
* opens a file with initial delete on close set, renames it then closes
* all open handles. The file goes away on Windows.
*/
r23103: Add a torture test for the insane semantics Steve's cifsfs client uses so we don't regress :-). Jeremy. (This used to be commit 5ce41ae40c5413bb9f15d1c47a095f63d7f167e0)
2007-05-23 23:11:16 +00:00
static bool deltest20b(struct torture_context *tctx, struct smbcli_state *cli1, struct smbcli_state *cli2)
{
int fnum1 = -1;
int fnum2 = -1;
bool correct = True;
del_clean_area(cli1, cli2);
r23104: Add comment explaining the new test. Jeremy. (This used to be commit 7b9b5c59a4ac2afca151ba12eba7c808057a677e)
2007-05-23 23:14:48 +00:00
/* Test 20b. */
r23103: Add a torture test for the insane semantics Steve's cifsfs client uses so we don't regress :-). Jeremy. (This used to be commit 5ce41ae40c5413bb9f15d1c47a095f63d7f167e0)
2007-05-23 23:11:16 +00:00
/* Ensure the file doesn't already exist. */
smbcli_close(cli1->tree, fnum1);
smbcli_close(cli1->tree, fnum2);
smbcli_setatr(cli1->tree, fname, 0, 0);
smbcli_unlink(cli1->tree, fname);
smbcli_setatr(cli1->tree, fname_new, 0, 0);
smbcli_unlink(cli1->tree, fname_new);
/* Firstly open and create with all access */
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
SEC_RIGHTS_FILE_ALL,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_CREATE,
0, 0);
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open - 1 of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
/* And close - just to create the file. */
smbcli_close(cli1->tree, fnum1);
/* Firstly open and create with all access */
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
SEC_RIGHTS_FILE_ALL,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
0, 0);
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open - 1 of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
/* Next open with all access, but add delete on close. */
fnum2 = smbcli_nt_create_full(cli2->tree, fname, 0,
SEC_RIGHTS_FILE_ALL,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
NTCREATEX_OPTIONS_DELETE_ON_CLOSE, 0);
torture_assert(tctx, fnum2 != -1, talloc_asprintf(tctx, "open - 2 of %s failed (%s)",
fname, smbcli_errstr(cli2->tree)));
/* The delete on close bit is *not* reported as being set. */
correct &= check_delete_on_close(tctx, cli1, fnum1, fname, False, __location__);
correct &= check_delete_on_close(tctx, cli2, fnum2, fname, False, __location__);
smbcli_close(cli1->tree, fnum1);
correct &= check_delete_on_close(tctx, cli2, fnum2, fname, False, __location__);
/* Rename the file by handle. */
{
union smb_setfileinfo sfinfo;
NTSTATUS status;
memset(&sfinfo, '\0', sizeof(sfinfo));
sfinfo.generic.level = RAW_SFILEINFO_RENAME_INFORMATION;
sfinfo.generic.in.file.fnum = fnum2;
sfinfo.rename_information.in.root_fid = 0;
/* Don't start the filename with '\\', we get NT_STATUS_NOT_SUPPORTED if so. */
sfinfo.rename_information.in.new_name = fname_new + 1;
sfinfo.rename_information.in.overwrite = 1;
status = smb_raw_setfileinfo(cli2->tree, &sfinfo);
torture_assert_ntstatus_equal(tctx,status,NT_STATUS_OK,talloc_asprintf(tctx, "rename of %s to %s failed (%s)",
fname, fname_new, smbcli_errstr(cli2->tree)));
}
correct &= check_delete_on_close(tctx, cli2, fnum2, fname_new, False, __location__);
smbcli_close(cli2->tree, fnum2);
/* See if the file is deleted - should be.... */
fnum1 = smbcli_open(cli1->tree, fname, O_RDWR, DENY_NONE);
torture_assert(tctx, fnum1 == -1, talloc_asprintf(tctx, "open of %s succeeded (should fail) - %s",
fname, smbcli_errstr(cli1->tree)));
fnum1 = smbcli_open(cli1->tree, fname_new, O_RDWR, DENY_NONE);
torture_assert(tctx, fnum1 == -1, talloc_asprintf(tctx, "open of %s succeeded (should fail) - %s",
fname_new, smbcli_errstr(cli1->tree)));
return correct;
}
r13370: Added deltest21 - pull the rug out from a connection by socket close after setting delete on close flag. Jeremy. (This used to be commit fbea18e78f8a3c6dbb36aa935b7044c0fcf61da4)
2006-02-06 19:43:24 +00:00
/* Test 21 ... */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
static bool deltest21(struct torture_context *tctx)
r13370: Added deltest21 - pull the rug out from a connection by socket close after setting delete on close flag. Jeremy. (This used to be commit fbea18e78f8a3c6dbb36aa935b7044c0fcf61da4)
2006-02-06 19:43:24 +00:00
{
int fnum1 = -1;
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
struct smbcli_state *cli1;
struct smbcli_state *cli2;
bool correct = True;
if (!torture_open_connection(&cli1, 0))
return False;
if (!torture_open_connection(&cli2, 1))
return False;
r13370: Added deltest21 - pull the rug out from a connection by socket close after setting delete on close flag. Jeremy. (This used to be commit fbea18e78f8a3c6dbb36aa935b7044c0fcf61da4)
2006-02-06 19:43:24 +00:00
del_clean_area(cli1, cli2);
/* Test 21 -- Test removal of file after socket close. */
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
SEC_RIGHTS_FILE_ALL,
FILE_ATTRIBUTE_NORMAL, NTCREATEX_SHARE_ACCESS_NONE,
NTCREATEX_DISP_OVERWRITE_IF, 0, 0);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert(tctx, fnum1 != -1, talloc_asprintf(tctx, "open of %s failed (%s)",
fname, smbcli_errstr(cli1->tree)));
r13370: Added deltest21 - pull the rug out from a connection by socket close after setting delete on close flag. Jeremy. (This used to be commit fbea18e78f8a3c6dbb36aa935b7044c0fcf61da4)
2006-02-06 19:43:24 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_assert_ntstatus_ok(tctx,
smbcli_nt_delete_on_close(cli1->tree, fnum1, True),
talloc_asprintf(tctx, "setting delete_on_close failed (%s)",
smbcli_errstr(cli1->tree)));
r13370: Added deltest21 - pull the rug out from a connection by socket close after setting delete on close flag. Jeremy. (This used to be commit fbea18e78f8a3c6dbb36aa935b7044c0fcf61da4)
2006-02-06 19:43:24 +00:00
/* Ensure delete on close is set. */
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
correct &= check_delete_on_close(tctx, cli1, fnum1, fname, True, __location__);
r13370: Added deltest21 - pull the rug out from a connection by socket close after setting delete on close flag. Jeremy. (This used to be commit fbea18e78f8a3c6dbb36aa935b7044c0fcf61da4)
2006-02-06 19:43:24 +00:00
/* Now yank the rug from under cli1. */
r17020: pass the real error to the failing requests metze (This used to be commit 49b96ac44a883c020c69df7a12df154dc4faa4d5)
2006-07-13 17:37:45 +00:00
smbcli_transport_dead(cli1->transport, NT_STATUS_LOCAL_DISCONNECT);
r13370: Added deltest21 - pull the rug out from a connection by socket close after setting delete on close flag. Jeremy. (This used to be commit fbea18e78f8a3c6dbb36aa935b7044c0fcf61da4)
2006-02-06 19:43:24 +00:00
fnum1 = -1;
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
if (!torture_open_connection(&cli1, 0)) {
r13370: Added deltest21 - pull the rug out from a connection by socket close after setting delete on close flag. Jeremy. (This used to be commit fbea18e78f8a3c6dbb36aa935b7044c0fcf61da4)
2006-02-06 19:43:24 +00:00
return False;
}
r16756: Some machines on the build farms sporadically fail the test /* Test 21 -- Test removal of file after socket close. */ I think it might be because they are too slow to delete the file. Jeremy, can you check this test does not change semantics in a way you don't want it? Volker (This used to be commit 92aa95f8206364a832bd6ad8cebc030bc001d940)
2006-07-02 08:53:49 +00:00
/* On slow build farm machines it might happen that they are not fast
* enogh to delete the file for this test */
msleep(200);
r13370: Added deltest21 - pull the rug out from a connection by socket close after setting delete on close flag. Jeremy. (This used to be commit fbea18e78f8a3c6dbb36aa935b7044c0fcf61da4)
2006-02-06 19:43:24 +00:00
/* File should not be there. */
fnum1 = smbcli_nt_create_full(cli1->tree, fname, 0,
SEC_RIGHTS_FILE_READ,
FILE_ATTRIBUTE_NORMAL,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
0, 0);
CHECK_STATUS(cli1, NT_STATUS_OBJECT_NAME_NOT_FOUND);
return correct;
}
r20525: Closing a dir with del-on-close set requires the same as files: Don't actually unlink/rmdir if another process still has it open. Jeremy, this is a potential merger to 3.0.24. Volker (This used to be commit d0550a01af219344ec2e4c16d5253814afc89974)
2007-01-04 16:26:15 +00:00
/* Test 22 ... */
/*
* Test whether a second *directory* handle inhibits delete if the first has
* del-on-close set and is closed
*/
static bool deltest22(struct torture_context *tctx)
{
int dnum1 = -1;
int dnum2 = -1;
struct smbcli_state *cli1;
bool correct = True;
if (!torture_open_connection(&cli1, 0))
return False;
smbcli_deltree(cli1->tree, dname);
torture_assert_ntstatus_ok(
tctx, smbcli_mkdir(cli1->tree, dname),
talloc_asprintf(tctx, "smbcli_mdir failed: (%s)\n",
smbcli_errstr(cli1->tree)));
dnum1 = smbcli_nt_create_full(cli1->tree, dname, 0,
SEC_FILE_READ_DATA|
SEC_FILE_WRITE_DATA|
SEC_STD_DELETE,
FILE_ATTRIBUTE_DIRECTORY,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
NTCREATEX_OPTIONS_DIRECTORY, 0);
torture_assert(tctx, dnum1 != -1,
talloc_asprintf(tctx, "open of %s failed: %s!",
dname, smbcli_errstr(cli1->tree)));
dnum2 = smbcli_nt_create_full(cli1->tree, dname, 0,
SEC_FILE_READ_DATA|
SEC_FILE_WRITE_DATA,
FILE_ATTRIBUTE_DIRECTORY,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
NTCREATEX_OPTIONS_DIRECTORY, 0);
torture_assert(tctx, dnum2 != -1,
talloc_asprintf(tctx, "open of %s failed: %s!",
dname, smbcli_errstr(cli1->tree)));
torture_assert_ntstatus_ok(
tctx, smbcli_nt_delete_on_close(cli1->tree, dnum1, True),
talloc_asprintf(tctx, "setting delete_on_close failed (%s)",
smbcli_errstr(cli1->tree)));
smbcli_close(cli1->tree, dnum1);
dnum1 = smbcli_nt_create_full(cli1->tree, dname, 0,
SEC_FILE_READ_DATA|
SEC_FILE_WRITE_DATA|
SEC_STD_DELETE,
FILE_ATTRIBUTE_DIRECTORY,
NTCREATEX_SHARE_ACCESS_READ|
NTCREATEX_SHARE_ACCESS_WRITE|
NTCREATEX_SHARE_ACCESS_DELETE,
NTCREATEX_DISP_OPEN,
NTCREATEX_OPTIONS_DIRECTORY, 0);
torture_assert(tctx, dnum1 == -1,
talloc_asprintf(tctx, "open of %s succeeded!\n",
dname));
CHECK_STATUS(cli1, NT_STATUS_DELETE_PENDING);
return correct;
}
r23099: New interesting delete on close tests :-). This will break all delete tests on the build farm until I check in my server patch. Jeremy. (This used to be commit c31f775f1c8f6f7e50d038d093898a1f4280be23)
2007-05-23 21:14:05 +00:00
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
/*
Test delete on close semantics.
*/
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
struct torture_suite *torture_test_delete(void)
r13297: It's a good thing the shipment of function headers tridge sent me arrived on time... :-). Refactor this code to make it comprehensible. Tested against W2K3 SP 1 and W2K SP 4. Test 19 is different from what I thought. Turns out delete on close on "open" of a directory (not create) does have an effect - even if not reported in the flag bit. trige please test against Vista (my XP box is refusing to serve at the moment - have to reinstall). Jeremy. (This used to be commit 2b708e26185bfc0a909a33e74e67dd2101c3bbbe)
2006-02-03 02:07:22 +00:00
{
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
struct torture_suite *suite = torture_suite_create(
r23029: fixed formatting (This used to be commit 72ee4707725e9bbc9ab564315c7d78bdc564256f)
2007-05-21 05:40:04 +00:00
talloc_autofree_context(),
"DELETE");
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_suite_add_2smb_test(suite, "deltest1", deltest1);
torture_suite_add_2smb_test(suite, "deltest2", deltest2);
torture_suite_add_2smb_test(suite, "deltest3", deltest3);
torture_suite_add_2smb_test(suite, "deltest4", deltest4);
torture_suite_add_2smb_test(suite, "deltest5", deltest5);
torture_suite_add_2smb_test(suite, "deltest6", deltest6);
torture_suite_add_2smb_test(suite, "deltest7", deltest7);
torture_suite_add_2smb_test(suite, "deltest8", deltest8);
torture_suite_add_2smb_test(suite, "deltest9", deltest9);
torture_suite_add_2smb_test(suite, "deltest10", deltest10);
torture_suite_add_2smb_test(suite, "deltest11", deltest11);
torture_suite_add_2smb_test(suite, "deltest12", deltest12);
torture_suite_add_2smb_test(suite, "deltest13", deltest13);
torture_suite_add_2smb_test(suite, "deltest14", deltest14);
torture_suite_add_2smb_test(suite, "deltest15", deltest15);
r19389: Remove samba3-specific checks: instead, always simply mark these tests as known failing. (This used to be commit 3855ec9dcaa694fffa1f2a24205d490708d12750)
2006-10-18 12:23:35 +00:00
torture_suite_add_2smb_test(suite, "deltest16", deltest16);
torture_suite_add_2smb_test(suite, "deltest17", deltest17);
torture_suite_add_2smb_test(suite, "deltest18", deltest18);
torture_suite_add_2smb_test(suite, "deltest19", deltest19);
torture_suite_add_2smb_test(suite, "deltest20", deltest20);
r23099: New interesting delete on close tests :-). This will break all delete tests on the build farm until I check in my server patch. Jeremy. (This used to be commit c31f775f1c8f6f7e50d038d093898a1f4280be23)
2007-05-23 21:14:05 +00:00
torture_suite_add_2smb_test(suite, "deltest20a", deltest20a);
r23103: Add a torture test for the insane semantics Steve's cifsfs client uses so we don't regress :-). Jeremy. (This used to be commit 5ce41ae40c5413bb9f15d1c47a095f63d7f167e0)
2007-05-23 23:11:16 +00:00
torture_suite_add_2smb_test(suite, "deltest20b", deltest20b);
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
torture_suite_add_simple_test(suite, "deltest21", deltest21);
r20525: Closing a dir with del-on-close set requires the same as files: Don't actually unlink/rmdir if another process still has it open. Jeremy, this is a potential merger to 3.0.24. Volker (This used to be commit d0550a01af219344ec2e4c16d5253814afc89974)
2007-01-04 16:26:15 +00:00
torture_suite_add_simple_test(suite, "deltest22", deltest22);
r13370: Added deltest21 - pull the rug out from a connection by socket close after setting delete on close flag. Jeremy. (This used to be commit fbea18e78f8a3c6dbb36aa935b7044c0fcf61da4)
2006-02-06 19:43:24 +00:00
r19339: Merge my 4.0-unittest branch. This adds an API for more fine-grained output in the testsuite rather than just True or False for a set of tests. The aim is to use this for: * known failure lists (run all tests and detect tests that started working or started failing). This would allow us to get rid of the RPC-SAMBA3-* tests * nicer torture output * simplification of the testsuite system * compatibility with other unit testing systems * easier usage of smbtorture (being able to run one test and automatically set up the environment for that) This is still a work-in-progress; expect more updates over the next couple of days. (This used to be commit 0eb6097305776325c75081356309115f445a7218)
2006-10-16 13:06:41 +00:00
return suite;
r3149: separate the delete on close test into torture/basic/delete.c (This used to be commit 50379a0a58d9eade3e1390713ef89473c66e65fc)
2004-10-24 09:08:52 +00:00
}
Copy Permalink