]> granicus.if.org Git - php/log
php
15 years agoBFN
Scott MacVicar [Tue, 9 Jun 2009 02:00:46 +0000 (02:00 +0000)]
BFN

15 years agoMFH Fix bug #48493 - spl_autoload_register can leave the HT in an inconsistent way.
Scott MacVicar [Tue, 9 Jun 2009 01:58:07 +0000 (01:58 +0000)]
MFH Fix bug #48493 - spl_autoload_register can leave the HT in an inconsistent way.
Need to point the second elements previous item to head so we can traverse upwards.

15 years agoMove entry to 5.2
Scott MacVicar [Tue, 9 Jun 2009 00:42:51 +0000 (00:42 +0000)]
Move entry to 5.2

15 years ago #41874, separate STDOUT and STDERR in exec functions
Pierre Joye [Tue, 9 Jun 2009 00:25:37 +0000 (00:25 +0000)]
 #41874, separate STDOUT and STDERR in exec functions

15 years ago- typo
Pierre Joye [Mon, 8 Jun 2009 21:27:05 +0000 (21:27 +0000)]
- typo

15 years ago- Updated to version 2009.9 (2009i)
Derick Rethans [Mon, 8 Jun 2009 13:53:34 +0000 (13:53 +0000)]
- Updated to version 2009.9 (2009i)

15 years agoFixed test
Ilia Alshanetsky [Mon, 8 Jun 2009 12:31:08 +0000 (12:31 +0000)]
Fixed test

15 years agoNew tests for operators. These tests were auto generated to verify correctness for...
Robert Nicholson [Mon, 8 Jun 2009 11:59:32 +0000 (11:59 +0000)]
New tests for operators. These tests were auto generated to verify correctness for a number of significant values.  The work was done by Dave Kelsey.

15 years agoMerge with HEAD. Someone committed changes to HEAD and did not merge back to
Andrey Hristov [Mon, 8 Jun 2009 10:20:27 +0000 (10:20 +0000)]
Merge with HEAD. Someone committed changes to HEAD and did not merge back to
the branch.
Also switch off the zval cache, for now.

15 years ago- sync with 1.0.2, improve perf x86-64 (patch by Solar Designer)
Pierre Joye [Mon, 8 Jun 2009 08:44:15 +0000 (08:44 +0000)]
- sync with 1.0.2, improve perf x86-64 (patch by Solar Designer)

15 years agoMFH make SQLite3::enableExceptions() return the previous value.
Scott MacVicar [Mon, 8 Jun 2009 02:15:54 +0000 (02:15 +0000)]
MFH make SQLite3::enableExceptions() return the previous value.

15 years agoFixed bug #44827 (define() is missing error checks for class constants)
Ilia Alshanetsky [Mon, 8 Jun 2009 01:26:03 +0000 (01:26 +0000)]
Fixed bug #44827 (define() is missing error checks for class constants)

15 years agoMFH: Removed credit
Patrick Allaert [Sun, 7 Jun 2009 21:27:19 +0000 (21:27 +0000)]
MFH: Removed credit

15 years agoNew tests for foreach loops. These were written by another member of the Projectzero...
Robert Nicholson [Sun, 7 Jun 2009 21:03:48 +0000 (21:03 +0000)]
New tests for foreach loops. These were written by another member of the Projectzero team.

15 years agoMFH: Fixed bug #47836 (array operator [] inconsistency when the array has PHP_INT_MAX...
Matt Wilmas [Sun, 7 Jun 2009 19:28:15 +0000 (19:28 +0000)]
MFH: Fixed bug #47836 (array operator [] inconsistency when the array has PHP_INT_MAX index value)
Also simplified related array_push() test

15 years agoMFH: Changed error messages to use "cannot" instead of "can not" (meaning "also can")
Matt Wilmas [Sun, 7 Jun 2009 15:46:54 +0000 (15:46 +0000)]
MFH: Changed error messages to use "cannot" instead of "can not" (meaning "also can")

15 years agoMFH: Fixed bug #48469 (ldap_get_entries() memory leaks on empty search results)
Patrick Allaert [Sun, 7 Jun 2009 13:07:58 +0000 (13:07 +0000)]
MFH: Fixed bug #48469 (ldap_get_entries() memory leaks on empty search results)

15 years agoMFH: Fixed error message grammar:
Matt Wilmas [Sat, 6 Jun 2009 02:40:49 +0000 (02:40 +0000)]
MFH: Fixed error message grammar:
 - "cannot" instead of "can not" (meaning "also can")
 - "than" instead of "then" (Hint: "then" should hardly be needed;
     you're not telling the order in which to do something)

... plus removed a couple ending. dots

15 years agoMFH: Make pass_two visible to extensions with ZEND_API (required for optimizer).
Brian Shire [Fri, 5 Jun 2009 23:20:59 +0000 (23:20 +0000)]
MFH: Make pass_two visible to extensions with ZEND_API (required for optimizer).

15 years agoThere is no always_inline attribute in gcc2
Rasmus Lerdorf [Fri, 5 Jun 2009 22:14:21 +0000 (22:14 +0000)]
There is no always_inline attribute in gcc2

15 years agoMFH: Added zend_eval_stringl and made create_function(), etc. binary-safe
Matt Wilmas [Fri, 5 Jun 2009 18:50:32 +0000 (18:50 +0000)]
MFH: Added zend_eval_stringl and made create_function(), etc. binary-safe

15 years agoMFH: Fixed tests again after double->long conversion change (to match 5.2)
Matt Wilmas [Fri, 5 Jun 2009 17:15:33 +0000 (17:15 +0000)]
MFH: Fixed tests again after double->long conversion change (to match 5.2)

15 years agoMFH: better fix for #48409 , #48428 , #48228
Arnaud Le Blanc [Fri, 5 Jun 2009 11:21:31 +0000 (11:21 +0000)]
MFH: better fix for #48409 , #48428 , #48228

15 years agoMFH: fixed isset() on sub-directories (isset("blah") if file "blah/foo.php" exists)
Greg Beaver [Fri, 5 Jun 2009 04:46:49 +0000 (04:46 +0000)]
MFH: fixed isset() on sub-directories (isset("blah") if file "blah/foo.php" exists)

15 years agoMFPECL: fix security vulnerability in phar's handling of long tar filenames
Greg Beaver [Thu, 4 Jun 2009 19:59:09 +0000 (19:59 +0000)]
MFPECL: fix security vulnerability in phar's handling of long tar filenames

15 years agoMFH: Updated tests after double->long conversion change (basically to match 5.2)
Matt Wilmas [Thu, 4 Jun 2009 18:25:06 +0000 (18:25 +0000)]
MFH: Updated tests after double->long conversion change (basically to match 5.2)

# Let me know about other failing tests I missed

15 years agoMFH:
Matt Wilmas [Thu, 4 Jun 2009 18:20:45 +0000 (18:20 +0000)]
MFH:
Restored double->long conversion behavior to that of PHP 5.2 (on most platforms) and prior:
 * Out-of-range numbers overflow/preserve least significant bits (no LONG_MAX/MIN limit)
 * See bug #42868 (presumably-rare platform with different results in 5.2)
 * On 32-bit platforms with 64-bit long type, a zend_long64 cast has been added,
    otherwise it's the same as 5.2
 * Use this conversion method everywhere instead of some plain (long) casts

Added 'L' parameter parsing specifier to ensure a LONG_MAX/MIN limit:
 * Essentially what 5.3's new conversion was doing in most cases
 * Functions with "limit" or "length" type params could be updated to use this,
    and prevent confusing overflow behavior with huge numbers (*also* in 5.2)
  - See bug #47854, for example; or even #42868 again

# Test updates coming

15 years agoMFB
Andrei Zmievski [Thu, 4 Jun 2009 18:17:43 +0000 (18:17 +0000)]
MFB

15 years agoMFH: Include other flags as well
Etienne Kneuss [Thu, 4 Jun 2009 14:46:26 +0000 (14:46 +0000)]
MFH: Include other flags as well

15 years agoMFH: Fixed creating directory structure when extracting phars
Hannes Magnusson [Thu, 4 Jun 2009 12:22:32 +0000 (12:22 +0000)]
MFH: Fixed creating directory structure when extracting phars

15 years agoMFH:- Fixed bug #48419 (non-portable STDIN fileno in cgi_main.c)
Jani Taskinen [Thu, 4 Jun 2009 11:51:43 +0000 (11:51 +0000)]
MFH:- Fixed bug #48419 (non-portable STDIN fileno in cgi_main.c)

15 years agoMFH: CS + WS changes
Jani Taskinen [Thu, 4 Jun 2009 11:47:27 +0000 (11:47 +0000)]
MFH: CS + WS changes

15 years ago- typo
Pierre Joye [Thu, 4 Jun 2009 07:41:01 +0000 (07:41 +0000)]
- typo

15 years ago- new constnats
Pierre Joye [Thu, 4 Jun 2009 07:01:47 +0000 (07:01 +0000)]
- new constnats

15 years ago- [DOC] add PHP_WINDOWS_NT_* and PHP_WINDOWS_VERSION_* constants (usefull for testing...
Pierre Joye [Thu, 4 Jun 2009 06:59:46 +0000 (06:59 +0000)]
- [DOC] add PHP_WINDOWS_NT_* and PHP_WINDOWS_VERSION_* constants (usefull for testing and features detections)
   see http://msdn.microsoft.com/en-us/library/ms724451(VS.85).aspx for the explanation

15 years ago- Fixed tests
Felipe Pena [Thu, 4 Jun 2009 01:15:12 +0000 (01:15 +0000)]
- Fixed tests

15 years agoMFH: Add test for Bug #47859 parse_ini_file() does not like asterisk (*) in key in...
Brian Shire [Wed, 3 Jun 2009 19:37:26 +0000 (19:37 +0000)]
MFH: Add test for Bug #47859 parse_ini_file() does not like asterisk (*) in key in the beginning

15 years agoFixed bug #46386 (Digest authentication with SOAP module fails against MSSQL
Ilia Alshanetsky [Wed, 3 Jun 2009 12:39:50 +0000 (12:39 +0000)]
Fixed bug #46386 (Digest authentication with SOAP module fails against MSSQL
SOAP services)

#original patch by lordelph at gmail dot com

15 years agoCommiting testfest tests from tek
Zoe Slattery [Wed, 3 Jun 2009 12:30:40 +0000 (12:30 +0000)]
Commiting testfest tests from tek

15 years ago- fix title
Pierre Joye [Wed, 3 Jun 2009 08:31:24 +0000 (08:31 +0000)]
- fix title

15 years ago- fix parameter order and return value check in windows (Venkat Raman Don, Pierre...
Pierre Joye [Wed, 3 Jun 2009 07:56:53 +0000 (07:56 +0000)]
- fix parameter order and return value check in windows (Venkat Raman Don, Pierre)' link

15 years agoMFH: config.h is in this directory, not in ext/bcmath root, fixes re-compilation...
Kalle Sommer Nielsen [Wed, 3 Jun 2009 01:19:33 +0000 (01:19 +0000)]
MFH: config.h is in this directory, not in ext/bcmath root, fixes re-compilation for me

15 years agoMFH:
Kalle Sommer Nielsen [Wed, 3 Jun 2009 01:14:58 +0000 (01:14 +0000)]
MFH:
* We dont need to wrap to two lines here, its small enough to show with a standard sized cmd window
* Bump copyright year in the .rc

15 years agoMFH:- Fixed bug #48456 (CPPFLAGS not restored properly in phpize.m4)
Jani Taskinen [Tue, 2 Jun 2009 19:54:03 +0000 (19:54 +0000)]
MFH:- Fixed bug #48456 (CPPFLAGS not restored properly in phpize.m4)

15 years agoFixed test
Ilia Alshanetsky [Tue, 2 Jun 2009 12:19:34 +0000 (12:19 +0000)]
Fixed test

15 years agofix test
Jani Taskinen [Tue, 2 Jun 2009 08:03:07 +0000 (08:03 +0000)]
fix test

15 years agofix compiler warnings by moving all std includes after include of zend_config.h.
Nuno Lopes [Mon, 1 Jun 2009 17:51:53 +0000 (17:51 +0000)]
fix compiler warnings by moving all std includes after include of zend_config.h.
this also fixes the build with stricter compilers than gcc (like clang)

15 years agoMFH: New tests
Arnaud Le Blanc [Mon, 1 Jun 2009 16:10:35 +0000 (16:10 +0000)]
MFH: New tests
# These tests were failing on different configurations, so adding them
# all to get more chances of seeing one failing in case of problem.

15 years agoMFH: better fix for #48409 , #48428 , #48228
Arnaud Le Blanc [Mon, 1 Jun 2009 15:07:42 +0000 (15:07 +0000)]
MFH: better fix for #48409 , #48428 , #48228

15 years agoFixed bug #48273 (snmp*_real_walk() returns SNMP errors as values)
Ilia Alshanetsky [Mon, 1 Jun 2009 13:10:18 +0000 (13:10 +0000)]
Fixed bug #48273 (snmp*_real_walk() returns SNMP errors as values)

15 years agoEngine execution order tests
Zoe Slattery [Mon, 1 Jun 2009 13:01:04 +0000 (13:01 +0000)]
Engine execution order tests

15 years agoFixes to work on Windows as well as Linux
Zoe Slattery [Mon, 1 Jun 2009 09:01:06 +0000 (09:01 +0000)]
Fixes to work on Windows as well as Linux

15 years agoMFH: Add test for the timeout inside shutdown function
Etienne Kneuss [Sun, 31 May 2009 21:32:42 +0000 (21:32 +0000)]
MFH: Add test for the timeout inside shutdown function

15 years agofix for #48247
Stanislav Malyshev [Sun, 31 May 2009 21:28:38 +0000 (21:28 +0000)]
fix for #48247

15 years agoFix the build.
Andrei Zmievski [Sun, 31 May 2009 18:55:10 +0000 (18:55 +0000)]
Fix the build.

15 years agoFixed bug #48359 (Script hangs on snmprealwalk if OID is not increasing)
Ilia Alshanetsky [Sun, 31 May 2009 14:11:04 +0000 (14:11 +0000)]
Fixed bug #48359 (Script hangs on snmprealwalk if OID is not increasing)

# Original patch by simonov at gmail dot com

15 years agoMFH: no dots in errors
Jani Taskinen [Sun, 31 May 2009 13:51:08 +0000 (13:51 +0000)]
MFH: no dots in errors

15 years agofix news
Jani Taskinen [Sun, 31 May 2009 13:20:22 +0000 (13:20 +0000)]
fix news

15 years agoThis shouldn't have been in there
Kalle Sommer Nielsen [Sun, 31 May 2009 11:16:46 +0000 (11:16 +0000)]
This shouldn't have been in there

15 years agoExpose encode/decode API.
Andrei Zmievski [Sun, 31 May 2009 01:44:07 +0000 (01:44 +0000)]
Expose encode/decode API.

15 years agoMFH: Take small blocks cache into account in memory_get_usage()
Arnaud Le Blanc [Sat, 30 May 2009 16:42:13 +0000 (16:42 +0000)]
MFH: Take small blocks cache into account in memory_get_usage()

15 years agoMFH: Fix aliasing issue (fixes #48409 , #48428 , #48228)
Arnaud Le Blanc [Sat, 30 May 2009 16:31:12 +0000 (16:31 +0000)]
MFH: Fix aliasing issue (fixes #48409 , #48428 , #48228)

15 years agoMFH: Make sure the timeout is checked in the shutdown function, patch from Etienne
Scott MacVicar [Sat, 30 May 2009 09:48:16 +0000 (09:48 +0000)]
MFH: Make sure the timeout is checked in the shutdown function, patch from Etienne

15 years agoMFH - Fixing tests - messages/warnings can vary by version
Ulf Wendel [Fri, 29 May 2009 13:37:08 +0000 (13:37 +0000)]
MFH - Fixing tests - messages/warnings can vary by version

15 years agoMFH - Updating test to reflect API changes.
Ulf Wendel [Fri, 29 May 2009 13:23:01 +0000 (13:23 +0000)]
MFH - Updating test to reflect API changes.

15 years agoMFH - Skip test for libmysql. The MySQL C-API does not support it, mysqlnd does...
Ulf Wendel [Fri, 29 May 2009 13:12:39 +0000 (13:12 +0000)]
MFH - Skip test for libmysql. The MySQL C-API does not support it, mysqlnd does...

15 years agoNEWS
Andrey Hristov [Fri, 29 May 2009 13:11:10 +0000 (13:11 +0000)]
NEWS

15 years agoMFH:Add support for mysql_stmt_store_result() from libmysql 6.0.8+ and 5.4.x
Andrey Hristov [Fri, 29 May 2009 13:09:47 +0000 (13:09 +0000)]
MFH:Add support for mysql_stmt_store_result() from libmysql 6.0.8+ and 5.4.x

15 years agoMFH - 1) The wording of a message can differ by version and by libmysql vs. mysqlnd...
Ulf Wendel [Fri, 29 May 2009 13:02:43 +0000 (13:02 +0000)]
MFH - 1) The wording of a message can differ by version and by libmysql vs. mysqlnd. 2) %i is better than %s

15 years agoMFH - Behaviour is undefined with libmysql, accept whatever libmysql may return....
Ulf Wendel [Fri, 29 May 2009 12:55:55 +0000 (12:55 +0000)]
MFH - Behaviour is undefined with libmysql, accept whatever libmysql may return. With mysqlnd you get the defined value of -1 to indicate 'no value'.

15 years agoMFH:Fix for the failing mysql_stmt_execute() test. Data was cleaned after
Andrey Hristov [Fri, 29 May 2009 12:19:40 +0000 (12:19 +0000)]
MFH:Fix for the failing mysql_stmt_execute() test. Data was cleaned after
stmt_reset() but it should stay and be freed as later as next stmt_execute()

15 years agoMFH - 'Fixing' test to reflect a libmysql vs. mysqlnd difference. With libmysql it...
Ulf Wendel [Fri, 29 May 2009 11:59:26 +0000 (11:59 +0000)]
MFH - 'Fixing' test to reflect a libmysql vs. mysqlnd difference. With libmysql it is not defined what mysql_stmt_store_result() shall return when one tries to store the (non-existing) result set of a failed SQL statement, http://dev.mysql.com/doc/refman/6.0/en/mysql-stmt-store-result.html . mysqlnd will detect the SQL failure and return false. libmysql will return true. If you want libmysql to change, file a bug report at MySQL... I don't think there is a good way to work around this in the extension itself.

15 years agobetter fix
Zoe Slattery [Fri, 29 May 2009 11:46:42 +0000 (11:46 +0000)]
better fix

15 years agoMFH - Another detail where mysqlnd is superior than libmysql. The mysql_info() from...
Ulf Wendel [Fri, 29 May 2009 11:01:06 +0000 (11:01 +0000)]
MFH - Another detail where mysqlnd is superior than libmysql. The mysql_info() from libmysql does not support SELECT: http://dev.mysql.com/doc/refman/6.0/en/mysql-info.html . mysqlnd does support it. Libmysql feature request filed http://bugs.mysql.com/bug.php?id=45189

15 years agoMFH: Weaker test - libmysql/server error messages can differ by version
Ulf Wendel [Fri, 29 May 2009 10:34:23 +0000 (10:34 +0000)]
MFH: Weaker test - libmysql/server error messages can differ by version

15 years agoMFH - Weaker test: functions are only available if HAVE_EMBEDDED_MYSQLI is set. There...
Ulf Wendel [Fri, 29 May 2009 10:25:49 +0000 (10:25 +0000)]
MFH - Weaker test: functions are only available if HAVE_EMBEDDED_MYSQLI is set. There is probably no proper way to check this condition in the user land.

15 years agoMFH - Skip when using libmysql. Likely a regression bug - http://bugs.mysql.com/bug...
Ulf Wendel [Fri, 29 May 2009 10:15:50 +0000 (10:15 +0000)]
MFH - Skip when using libmysql. Likely a regression bug - http://bugs.mysql.com/bug.php?id=45184

15 years agofixing to work with older levels of libcurl
Zoe Slattery [Fri, 29 May 2009 09:46:11 +0000 (09:46 +0000)]
fixing to work with older levels of libcurl

15 years agoMFH:The option has changed the name, hence we change the test
Andrey Hristov [Fri, 29 May 2009 08:58:32 +0000 (08:58 +0000)]
MFH:The option has changed the name, hence we change the test

15 years ago- MF52: #42143, the constant NAN is reported as 0 on Windows
Pierre Joye [Fri, 29 May 2009 08:11:36 +0000 (08:11 +0000)]
- MF52: #42143, the constant NAN is reported as 0 on Windows

15 years agoMFH:Fix a very well hidden error because of not being careful with CPP.
Andrey Hristov [Fri, 29 May 2009 08:05:38 +0000 (08:05 +0000)]
MFH:Fix a very well hidden error because of not being careful with CPP.
I usually don't make this mistake :)

15 years agoEngine execution order tests
Zoe Slattery [Fri, 29 May 2009 07:52:04 +0000 (07:52 +0000)]
Engine execution order tests

15 years agoMFH: If we don't have a comment (for the 3rd argument in ARG_[ENABLE|WITH], then...
Kalle Sommer Nielsen [Fri, 29 May 2009 07:43:07 +0000 (07:43 +0000)]
MFH: If we don't have a comment (for the 3rd argument in ARG_[ENABLE|WITH], then don't print a comment say "undefined"

15 years agoMFH: Fix bug #48416 - Force a cache size for ereg to stop it getting out of control...
Scott MacVicar [Thu, 28 May 2009 21:57:24 +0000 (21:57 +0000)]
MFH: Fix bug #48416 - Force a cache size for ereg to stop it getting out of control. Lazy LRU here.

15 years agoMFH - Asynchronous queries are mysqlnd only. If you want them with libmysql, ping...
Ulf Wendel [Thu, 28 May 2009 18:57:34 +0000 (18:57 +0000)]
MFH - Asynchronous queries are mysqlnd only. If you want them with libmysql, ping JimW who works on Connector/C.

15 years agoMFH - Andrey needs to have a look at this one. The bug report itself is bogus. Howeve...
Ulf Wendel [Thu, 28 May 2009 18:28:08 +0000 (18:28 +0000)]
MFH - Andrey needs to have a look at this one. The bug report itself is bogus. However, the bug report shows a result set and that is wrong.

15 years agoMFH:
Andrey Hristov [Thu, 28 May 2009 18:10:40 +0000 (18:10 +0000)]
MFH:
Forgot to commit this one, of the ANSI_QUOTES fame

15 years agoMFH:Fix again the tests for ANSI_QUOTES
Andrey Hristov [Thu, 28 May 2009 18:00:28 +0000 (18:00 +0000)]
MFH:Fix again the tests for ANSI_QUOTES

15 years agoMFH:Fix a valgrind warning as well as more trace log information
Andrey Hristov [Thu, 28 May 2009 17:49:29 +0000 (17:49 +0000)]
MFH:Fix a valgrind warning as well as more trace log information

15 years agoMFH:Fix a problem with cursors, which did not happen with unbuffered PS for
Andrey Hristov [Thu, 28 May 2009 16:35:41 +0000 (16:35 +0000)]
MFH:Fix a problem with cursors, which did not happen with unbuffered PS for
some reason. Double free of the data, which led to valgrind warnigns.
The fix actually optimizes the code in this cases because the old code
used copy_ctor while the new one skips it because it is not needed.
Transferring data ownership and nulling works best, for PS where we
always copy the string from the result set, unlike the text protocol.

15 years agoMFH - I had accidently removed Andrey's path to work with ANSI QUOTES... fixing what...
Ulf Wendel [Thu, 28 May 2009 15:49:14 +0000 (15:49 +0000)]
MFH - I had accidently removed Andrey's path to work with ANSI QUOTES... fixing what I broke (part 1)

15 years agoMFH - what a wonderful word. After the change in HEAD its easy to do a merge. The...
Ulf Wendel [Thu, 28 May 2009 14:33:43 +0000 (14:33 +0000)]
MFH - what a wonderful word. After the change in HEAD its easy to do a merge. The merge removes all UEXPECTF sections from the old days. Yes, there have been UEXPECTFs in 5_3. I hope PHP 6 won't change again over time... :-)

15 years agofix test
Antony Dovgal [Thu, 28 May 2009 13:56:44 +0000 (13:56 +0000)]
fix test

15 years ago- MFH: #48378, exif_read_data() segfaults on certain corrupted .jpeg files
Pierre Joye [Thu, 28 May 2009 13:48:32 +0000 (13:48 +0000)]
- MFH: #48378, exif_read_data() segfaults on certain corrupted .jpeg files

15 years agoMFH: New test
Arnaud Le Blanc [Thu, 28 May 2009 13:22:28 +0000 (13:22 +0000)]
MFH: New test

15 years agoMoved to 5.2
Arnaud Le Blanc [Thu, 28 May 2009 13:21:08 +0000 (13:21 +0000)]
Moved to 5.2

15 years agoMFH:
Andrey Hristov [Thu, 28 May 2009 11:47:48 +0000 (11:47 +0000)]
MFH:
Fix a bug with mysqlnd_fetch_field(_direct()). With mysqlnd the optimised
function was called, which however, doesn't respect that during store the
raw data is not unpacked, to be lazy. The data is unpacked to zvals later,
during every row fetch. However, this way max_length won't be calculated
correctly. So, if a mysqlnd_fetch_field(_direct) call comes we need to
unpack everything and then calculate max_length...and that is expensive,
defies our lazy unpacking optimisation.

15 years agoMFH:Fix a test, UNSIGNED is expected
Andrey Hristov [Thu, 28 May 2009 10:16:41 +0000 (10:16 +0000)]
MFH:Fix a test, UNSIGNED is expected

15 years agoMFB52: Fix "request_startup() for zlib module failed" when client does not
Arnaud Le Blanc [Thu, 28 May 2009 10:15:14 +0000 (10:15 +0000)]
MFB52: Fix "request_startup() for zlib module failed" when client does not
accepts compression and zlib.output_compression is On
# (not in HEAD)

15 years agoMFH: Fix failures when the server is running with ANSI_QUOTES
Andrey Hristov [Thu, 28 May 2009 09:58:36 +0000 (09:58 +0000)]
MFH: Fix failures when the server is running with ANSI_QUOTES