Andrey Hristov
1aad525c6e
Move definition
2009-10-22 15:17:03 +00:00
Andrey Hristov
a692479d21
Count number of affected_rows for normal and PS queries
2009-10-22 14:30:51 +00:00
Ulf Wendel
d356459971
There subtle differences between Linux and Win. mysqlnd is a bit more verbose on Win... silencing to make the test pass
2009-10-21 18:57:45 +00:00
Pierre Joye
26270b4fd7
- fix leak in DL on error (windows)
2009-10-21 06:42:08 +00:00
Ulf Wendel
8df115c25e
Making test a bit more portable
2009-10-20 17:54:57 +00:00
Ulf Wendel
03c3c1edb3
Fixing test
2009-10-20 17:03:32 +00:00
Ulf Wendel
3258a6572d
libmysql is still 'buggy' - skip test with libmysql
2009-10-20 14:11:43 +00:00
Ulf Wendel
3fb67347ea
Well, hacking around to cope with mysql version specific behaviour if using libmysql...
2009-10-20 13:58:57 +00:00
Ulf Wendel
87a4f09cbf
Skipping tests if InnoDB is available but disabled. I wonder what InnoDB status comes next...
2009-10-20 12:41:58 +00:00
Ulf Wendel
57c1b63c2d
Fixing typo in SKIPIF and partly skipping tests if using remote MySQL because LOAD LOCAL INFILE will fail
2009-10-20 12:31:04 +00:00
Ulf Wendel
0efa2445ad
Making test pass when MySQL runs on Win*
2009-10-20 08:41:11 +00:00
Ulf Wendel
9bb4db8781
Bail if connect.inc gets included twice. Including twice is most likely a 'bug'. See also http://news.php.net/php.cvs/60720
2009-10-20 08:23:06 +00:00
Johannes Schlüter
f5a5d321ed
Fix test to remove temporary files
2009-10-19 21:50:08 +00:00
Ulf Wendel
eb0e09f5dc
Using require[_once] instead of include. Adding logic to connect.inc to detect multiple inclusions of connect.inc which leads to a Fatal error because of my_mysqli_connect() function redefinition. Some tests did include connect.inc twice -> run-tests - even the one from trunk - did not detect Fatal error in SKIPIF -> false-positive test result...
2009-10-19 20:07:25 +00:00
Derick Rethans
afaa4ed8ad
- Updated to version 2009.15 (2009o)
2009-10-19 16:29:13 +00:00
Ulf Wendel
344d6803ec
mysqlnd is a bit more verbose than libmysql - silence mysqlnd here for easier comparison of test results.
2009-10-19 16:07:10 +00:00
Jani Taskinen
fc4352e2d8
- It is not const in PHP_5_2 but it is in HEAD and PHP_5_3
2009-10-19 15:34:50 +00:00
Ulf Wendel
f610cdd9a2
MySQL server and metadata - a sad story... trying to make the test pass with a few more server versions
2009-10-19 12:55:11 +00:00
Ulf Wendel
d01d398bad
Fixing host info for TCP/IP and aligning mysqlnd and libmysql
2009-10-19 12:09:48 +00:00
Ulf Wendel
3eeaf566e0
Extending mysqlnd statistics/monitoring. Now counting COM_* commands.
2009-10-16 13:02:01 +00:00
Scott MacVicar
cd5e3b1632
Add metadata support to the bundled version of libsqlite on Windows, part of bug #49206 .
...
Will commit the new feature part separately.
2009-10-16 02:22:48 +00:00
Scott MacVicar
76d875278c
Part fix for bug #49224
2009-10-16 02:10:52 +00:00
Scott MacVicar
5af006d640
Update libsqlite to 3.6.19
2009-10-16 01:56:36 +00:00
Ulf Wendel
d8dc47d58f
Making mysqlnd emit no warnings when fetching pooled persistent connections that have timed out or are unusable for any other reason - bug #49761
2009-10-15 20:46:08 +00:00
Ulf Wendel
ab4670ee53
Fixing problems when calling connect (again and again) on a valid connection handle. Most of the patch comes from Andrey.
2009-10-15 15:49:40 +00:00
Ilia Alshanetsky
ef1731590a
Fixed bug #49855 (import_request_variables() always returns NULL).
2009-10-15 12:26:54 +00:00
Ilia Alshanetsky
31729738c7
Fixed bug #49847 (exec() fails to return data inside 2nd parameter, given output lines >4095 bytes).
2009-10-14 01:32:07 +00:00
Derick Rethans
5c9ac94eab
- Updated to version 2009.14 (2009n)
2009-10-13 08:16:35 +00:00
Moriyoshi Koizumi
20737bac6a
- Bug #49785 : take 5. What the hell happened to me...
2009-10-13 05:18:37 +00:00
Felipe Pena
16be0343d7
- Fixed possible crash when calling SQLiteDatabase::{ArrayQuery, SingleQuery} methods using Reflection
2009-10-12 19:04:00 +00:00
Felipe Pena
b394a75ad5
- Fixed PDORow and PDOStatement crashes when instantiating throught Reflection
2009-10-12 17:09:11 +00:00
Moriyoshi Koizumi
884cf3f1c0
- Bug #49785 : take 4 - typo. this flaw is unharmful since the return value of get_next_char() is only used when UTF-8 is specified to the third argument.
2009-10-12 14:29:45 +00:00
Pierre Joye
ed3e96197d
- check max colors while loading gd2 palette image (patch by Tomas Hoger)
2009-10-12 10:01:37 +00:00
Moriyoshi Koizumi
1835a63dfd
- A couple more fix for my previous fix.
...
(one of the fix by Arnaud Le Blanc. Thanks!)
2009-10-11 23:52:33 +00:00
Felipe Pena
8952155898
- Fix compiler warning
2009-10-11 15:23:09 +00:00
Moriyoshi Koizumi
8a22597b65
- More tests to go.
2009-10-09 17:13:05 +00:00
Moriyoshi Koizumi
9d19866476
- Fixed bug #49785 (insufficient input string validation of htmlspecialchars()).
2009-10-09 10:02:38 +00:00
Pierre Joye
f88c96a7f3
- missing define
2009-10-08 20:59:50 +00:00
Pierre Joye
878ba27e60
- typo
2009-10-08 20:40:48 +00:00
Pierre Joye
02a046b1d9
- [DOC] add enchant_broket_set_dict_path and enchant_broker_get_dict_path
...
- add constants ENCHANT_MYSPELL and ENCHANT_ISPELL
2009-10-08 20:28:21 +00:00
Ulf Wendel
a1d3a3224e
Avoid buffer resizing and let the initial buffer allocation happen a couple of lines later with conn->m->set_client_option(conn, MYSQLND_OPT_NET_CMD_BUFFER_SIZE, ...). Andrey, I leave it to you to change the default size to 128k.
2009-10-08 16:33:17 +00:00
Ulf Wendel
4e6a1c2754
Fixing pconnect statistics.
2009-10-08 11:40:16 +00:00
Matteo Beccati
9ee8dd90a3
- Fixed bug #48764 (PDO_pgsql::query always uses implicit prepared statements if v3 proto available)
...
# original patch by Mark Kirkwood
2009-10-07 17:40:16 +00:00
Andrey Hristov
309fced64d
Increase the minimal size of the command buffer, which is used
...
for sending queries and in the future commpressed patch for reading
almost all packets from the wire instead of using stack buffers.
2009-10-07 17:33:51 +00:00
Ilia Alshanetsky
b6b56ba87a
Fixed bug #49800 (SimpleXML allow (un)serialize() calls without warning).
...
# original patch by wmeler at wp-sa dot pl
2009-10-07 12:46:29 +00:00
Christopher Jones
264d6db5a6
bump ver
2009-10-06 23:21:11 +00:00
Christopher Jones
2769ae0444
1. Introduce connection attribute functions:
...
oci_set_module_name
oci_set_action
oci_set_client_info
oci_set_client_identifier
These functions set values that are visible and used by the
database. They aid tracing, authentication and auditing.
2. Introduce connection attribute function:
oci_set_edition
Oracle 11g R2 "editions" allow multiple versions of DB objects
to exist at one time. By setting different editions, two
different versions of an application can run concurrently,
making upgrades or A/B testing easier.
3. Introduce OCI_NO_AUTO_COMMIT as an alias for the OCI_DEFAULT
constant (which is not the default value) used by oci_execute().
4. Allow the oci_set_prefetch value to be 0. This is important in
some cases using REF CURSORS in Oracle 11gR2.
5. Set the DRIVER_NAME attribute of Oracle Database 11gR2
connections to aid application tracing. The value used is to
"PHP OCI8" followed by the OCI8 version number. Note the
version number may get truncated in DB views such as
v$session_connect_info.
6. Generate an error if an invalid resource type is used in
oci_bind_by_name
[DOC] Documentation will be added for the changes
2009-10-06 22:36:32 +00:00
Etienne Kneuss
ef2b799a28
Fix bug #49263 (Offset error when unserializing self-references in SplObjectStorage)
2009-10-06 13:34:56 +00:00
Andrey Hristov
06523621dd
revert the last change
2009-10-06 12:53:01 +00:00
Andrey Hristov
d9382399fc
count the queries
2009-10-06 10:33:23 +00:00
Christopher Jones
76b321c350
Fix odd little diff in some environments. No need for trunk merge
2009-10-05 22:57:48 +00:00
Christopher Jones
7c7cc5d785
Fix diff when full connection string used. Clean up helper files
2009-10-05 22:51:21 +00:00
Ilia Alshanetsky
a3430a6b24
Fixed bug #49757 (long2ip() can return wrong value in a multi-threaded applications).
...
# original patch by Florian Anderiasch
2009-10-05 14:45:54 +00:00
Dmitry Stogov
ffb07732f7
Fixed bug #49627 (error_log to specified file does not log time according to date.timezone)
2009-10-05 13:56:49 +00:00
Felipe Pena
8172c388b0
- Fix some arginfos (thanks Philip)
...
- Added arginfo info. for void params
2009-10-03 20:45:29 +00:00
Pierre Joye
f86cde26fb
- drop unused var and fix example
2009-10-03 10:19:06 +00:00
Christopher Jones
2d64d1ff72
Oracle 11.2 (and beyond) compat changes
2009-10-02 22:39:28 +00:00
Christopher Jones
de0592a3a3
refactor tests
2009-10-02 20:16:59 +00:00
Christopher Jones
460a4ceaff
Fix skipifs to accept Oracle 11.2
2009-10-02 04:19:48 +00:00
Sriram Natarajan
0f57b15e23
- Fixed bug #49738 (calling mcrypt after mcrypt_generic_deinit crashes).
2009-10-02 00:13:53 +00:00
Pierre Joye
e0b392f028
- fix #49732 , fix crash when timestamp conversion fails
2009-10-01 13:55:16 +00:00
Ulf Wendel
5cb8e9b22a
Updating test because of http://news.php.net/php.cvs/60561
2009-10-01 08:56:28 +00:00
Christopher Jones
8f65525ce6
Add xmltype_02.phpt. Add XFAIL to xmltype_01 for PHP 6
2009-10-01 06:22:32 +00:00
Christopher Jones
8fca6e46eb
Eliminate type warnings. This change is not needed in other branches
2009-10-01 06:06:41 +00:00
Christopher Jones
609804cd44
PECL Bug #16842 (oci_error return false when NO_DATA_FOUND is raised)
2009-10-01 05:51:11 +00:00
Andrey Hristov
2f81bcd44b
Fix ZTS build
2009-09-30 23:34:56 +00:00
Andrey Hristov
ba5c40b2ad
Use mnd_ wrappers wherever possible. Needed for
...
statistics.
2009-09-30 23:13:43 +00:00
Andrey Hristov
b9dcc28c24
Fix trace output. DBG_RETURN should be used in traced
...
functions.
2009-09-30 22:26:19 +00:00
Ulf Wendel
9408303390
Fixing a leak in mysqlnd when passing invalid fetch modes to mysqlnd.
2009-09-30 14:39:33 +00:00
Dmitry Stogov
7a535b807c
pcntl_exec() should inherit environment in case it's not provided as an argument
2009-09-30 13:04:28 +00:00
Ilia Alshanetsky
809dbcda08
Fixed bug #49517 (cURL's CURLOPT_FILE prevents file from being deleted after fclose).
2009-09-30 02:34:17 +00:00
Ulf Wendel
e56e60c3a0
Fixing test
2009-09-29 16:37:50 +00:00
Rafael Machado Dohms
e91fffb24a
Fixing tests for head and wrong SKIPIF blocks
2009-09-29 14:34:06 +00:00
Rasmus Lerdorf
80bdbc8f66
Fixed a safe_mode bypass in tempnam() identified by Grzegorz Stachowiak.
2009-09-29 14:14:02 +00:00
Rasmus Lerdorf
8c15aa2314
Add missing open_basedir check in posix_mkfifo
2009-09-29 14:02:50 +00:00
Rasmus Lerdorf
99f30c1b6a
Fix date_sunrise/date_sunset tests
2009-09-29 13:56:51 +00:00
Rafael Machado Dohms
9ba46557e9
Adding tests for bug #49692
2009-09-28 22:57:41 +00:00
Rasmus Lerdorf
b3de24c9b8
Fix for bug #49698
2009-09-28 13:29:53 +00:00
Ulf Wendel
91315a9307
Fixing test
2009-09-28 10:42:36 +00:00
Ulf Wendel
fbfd13310b
Allow setting of default connection flags through the environment variable MYSQL_TEST_CONNECT_FLAGS
2009-09-28 10:39:42 +00:00
Ilia Alshanetsky
5edef9d2dd
Fixed test
2009-09-28 03:14:46 +00:00
Felipe Pena
b6a1b37889
- Fixed tests
2009-09-26 17:59:33 +00:00
Felipe Pena
5132913dd3
- Fixed tests
2009-09-26 16:54:08 +00:00
Rasmus Lerdorf
5afa8a23bb
Refix bug #49558 as per bug #49676
2009-09-26 00:53:59 +00:00
Ulf Wendel
8621a06eed
Trying to respect time limit to please Andrey who must be using slooow hardware...
2009-09-25 13:21:44 +00:00
Andrey Hristov
3828ca158e
Fix mysqli_stmt_attr_set in libmysql mode.
...
mysql_stmt_attr_set returns my_bool instead of int.
If it was int, then 0 is success and !0 is failure, but
for my_bool 0 (FALSE) is failure and !0 is success.
2009-09-25 12:25:54 +00:00
Andrey Hristov
5d3152b181
Fix a crash in the rare case when a persistent connection
...
is found dead, after CHNG_USER. mysql->mysql will be NULL
and segfault will follow.
2009-09-25 11:38:19 +00:00
Ulf Wendel
bd69506171
Large package to check how compression handles package queues.
2009-09-25 10:56:17 +00:00
Andrey Hristov
9b44cbb490
Remove unused variable
2009-09-25 10:55:06 +00:00
Ulf Wendel
a5e4c6f795
Replacing MYSQL_TEST_COMPRESS env var with more generic MYSQL_TEST_CONNECT_FLAGS to be able to test all connection flags. Fixing some test.
2009-09-25 10:54:16 +00:00
Andrey Hristov
eea0660def
Fix error code checking for mysql_stmt_attr_set. There
...
is a test that checks this.
2009-09-25 10:52:29 +00:00
Andrey Hristov
80f629ed9f
Fix possible crash when conn is NULL
2009-09-25 10:37:44 +00:00
Felipe Pena
22876972a5
- Fix test
2009-09-25 02:07:41 +00:00
Felipe Pena
20c5df8cce
- Fixed mem leak
2009-09-25 01:46:58 +00:00
Ilia Alshanetsky
eb2d54822b
Make tempnam() return FALSE on error as documented
2009-09-24 18:59:28 +00:00
Sriram Natarajan
92c08b1224
- Fixed bug #49571 (CURLOPT_POSTREDIR not implemented).
2009-09-24 18:20:49 +00:00
Jani Taskinen
2c90fb8ea2
- Fix the build issue with php_mysqlnd_config.h ending up in wrong build directory
2009-09-24 18:01:17 +00:00
Rob Richards
c3d66dbfed
Fixed bug #49647 (DOMUserData does not exist)
2009-09-24 13:18:22 +00:00
Ulf Wendel
c01b505a44
Proxying as many connect calls as possible to make testing of the compression protocol @ mysqlnd easier. By setting the environment variable MYSQL_TEST_COMPRESS you can tell the tests to set the MYSQLI_CLIENT_COMPRESS flag upon connect.
2009-09-24 12:51:03 +00:00
Rob Richards
9b923dd6ce
fix memleak
2009-09-24 12:40:59 +00:00
Ulf Wendel
7b50b0794d
Fixing test
2009-09-24 09:40:24 +00:00
Ulf Wendel
057a62fc52
Coverage for the OO variants of some functions
2009-09-24 08:19:56 +00:00
Ilia Alshanetsky
7eeff10cc0
Fixed unused var
2009-09-23 23:14:17 +00:00
Moriyoshi Koizumi
0c974164e2
- Fixed bug #49354 (mb_strcut() cuts wrong length when offset is within a
...
multibyte character).
(This bug was introduced by the commit by r202895. Please double-check the
specification of the function you are going to *fix*.)
2009-09-23 15:22:47 +00:00
Dmitry Stogov
6f32199f75
Fixed compilation error
2009-09-23 08:33:23 +00:00
Felipe Pena
788031f093
- Fixed bug #49531 (CURLOPT_INFILESIZE sometimes causes warning "CURLPROTO_FILE cannot be set ...")
2009-09-23 02:08:19 +00:00
Ilia Alshanetsky
420a3abc6a
Fixed compiler warning
2009-09-23 00:18:32 +00:00
Felipe Pena
5596b05449
- Fixed bug #49630 (imap_listscan function missing)
...
# Missing PHP_FE(), though nowdays it is used through of two aliases. (imap_scan() and imap_scanmailbox())
2009-09-22 18:18:57 +00:00
Ulf Wendel
381ebe751b
mysqlnd supports INIT_COMMAND and so we can add it to PDO_MYSQL @ mysqlnd
2009-09-22 15:31:35 +00:00
Andrey Hristov
ac93b60f68
MFH:Fix handling of BIT fields for non-PS. We need
...
macros from mysqlnd to be able to convert a bit
stream to a number.
2009-09-22 15:08:11 +00:00
Dmitry Stogov
38cdb16dd2
Fixed tests which fail from time to time because of race conditions ("echo" command didn't wait for all the data php tried to send and as result php got a SIGPIPE)
2009-09-22 14:52:47 +00:00
Andrey Hristov
452aca97e3
MFH:Check for the different types and create the
...
corresponding configuration file, so it can be used
by mysqli, even if mysqlnd is not enabled at all.
2009-09-22 14:44:11 +00:00
Andrey Hristov
dcb5c5b0a4
MFH: Fix handling of BIT fields in mysqli, when libmysql is used
...
We need to use macros from mysqlnd to be able to read the
bit fields, as they are specially encoded. mysqlnd is always
there, 5.3+, so its macros can be used, even if mysqlnd is
not compiled as library of choice.
2009-09-22 13:59:29 +00:00
Nuno Lopes
24090104ef
allow tests to pass when a random username is used.
...
patch by Olivier Doucet
2009-09-22 12:30:21 +00:00
Ulf Wendel
26b4248799
Fixing test
2009-09-22 11:58:46 +00:00
Dmitry Stogov
bb81ef5d71
Fixed test (it failed from time to time because of very small timeouts)
2009-09-22 10:15:10 +00:00
Jani Taskinen
4cce432eeb
- Move tests in the right location
2009-09-22 09:41:39 +00:00
Ulf Wendel
75d415efbe
Fixing test
2009-09-22 08:42:44 +00:00
Dmitry Stogov
17aff0eb25
Fixed test (removed "bad" locale)
2009-09-22 08:22:29 +00:00
Dmitry Stogov
42b62e7ca6
Fixed ext/spl/tests/arrayObject_magicMethods2.phpt
2009-09-22 07:54:06 +00:00
Ulf Wendel
77f44b7926
Fixing test. There is nothing we can do about the MySQL Client Library allocate huge 4GB buffers for some SQL types.
2009-09-22 07:11:50 +00:00
Ulf Wendel
5c67cdf62b
Fixing test. Can't wait to see the new run-tests materialize for detecting skip section parse errors.
2009-09-22 06:59:04 +00:00
Ilia Alshanetsky
d223d65aad
Fixed build with older version of libcurl
2009-09-22 01:17:16 +00:00
Ulf Wendel
6ef632d129
Fixing test
2009-09-21 15:59:22 +00:00
Ulf Wendel
e94d475ad0
Fixing test
2009-09-21 15:24:47 +00:00
Dmitry Stogov
883b992ea7
Fixed tests (removed irrelevant check which makes test to fail from time to time)
2009-09-21 15:23:25 +00:00
Dmitry Stogov
d12c66770f
Fixed tests ("none" is an alias of "false" and it can't be used as a string value of ini option)
2009-09-21 14:32:20 +00:00
Ulf Wendel
e4f41b1f9f
Moving test code around to get test failures down without testing less. Users of the MySQL Client Lib should always call stmt_bind_result after execute - not before.
2009-09-21 14:20:05 +00:00
Ulf Wendel
9afd024715
Fixing test by calling it a known limit. We keept the test failing as a reminder to think about how to handle BIT(32) and up but none of us has an idea. Thinks work fine up to BIT(31) - the limitation should be documented.
2009-09-21 12:02:30 +00:00
Ulf Wendel
47b209a3cd
Fixing test
2009-09-21 11:54:15 +00:00
Dmitry Stogov
d49abc2736
Fixed tests
2009-09-21 11:38:14 +00:00
Dmitry Stogov
623fe9aaf9
Fixed tests
2009-09-21 11:30:38 +00:00
Ulf Wendel
dc1e7d6d6d
Not nice but neither the MySQL Client Library not mysqlnd are supposed to detect invalid flags. Therefore: fixing test
2009-09-21 10:16:44 +00:00
Ulf Wendel
1a3b7774e6
Fixing mysqli_result_references_mysqlnd.php
2009-09-21 10:06:13 +00:00
Dmitry Stogov
7bfe056174
Fixed tests
2009-09-21 09:52:19 +00:00
Ulf Wendel
a0a8c5832f
Fixing mysqli_get_client_stats.phpt. More testing for MYSQLI_INIT_COMMAND
2009-09-21 09:35:23 +00:00
Ulf Wendel
025e670f54
Fixing test: references has to be collected even if the users disables the collection of statistics. It is needed for memory management.
2009-09-21 08:59:00 +00:00
Ulf Wendel
a77ee71cf0
Fixing test - the standard test table will not accept NULL values for column id
2009-09-21 08:36:14 +00:00
Hannes Magnusson
f7096da972
And now use proper colspanning
2009-09-18 23:06:21 +00:00
Hannes Magnusson
07618b1edd
Fix typo in comment (danbrown)
2009-09-18 22:26:44 +00:00
Hannes Magnusson
e3134c3139
Throw some credit around
2009-09-18 22:01:58 +00:00
Andrey Hristov
967239727e
MFH:Fix for bug#48909 Segmentation fault in mysqli_stmt_execute
2009-09-18 14:33:08 +00:00
Andrey Hristov
8aae3baa45
MFH:Fix for bug#48754 mysql_close() crash php when no handle specified
2009-09-18 10:49:31 +00:00
Andrey Hristov
c2e66884e1
MFH, a fix not included by the fix committed by Ulf.
...
pemalloc and then efree - too bad!
Sorry for committing in two passes, my big tree is already patched
with compressed.
2009-09-17 15:13:14 +00:00
Ulf Wendel
1388912ded
better coverage for mysqli_info
2009-09-17 07:11:25 +00:00
Felipe Pena
fff0a24656
- Fixed ZTS build
2009-09-16 18:01:10 +00:00
Sebastian Bergmann
ceaf590530
Merge ReflectionMethod::setAccessible() to PHP 5.3.2, approved by Johannes.
2009-09-16 17:24:46 +00:00
Ulf Wendel
e6cf6693e6
Fix (by Andrey) and test for bug #49442 . Don't use efree() for memory allocated with malloc()... If a connection gets created by mysqli_init(), mysqlnd makes it 'persistent'. 'Persistent' means that mysqlnd uses malloc(). mysqlnd does use malloc() instead of ealloc() because it is unknown if the connection will become a true persistent connection in the sense of ext/mysqli. It is unknown if the user wants a persistent connection or not until the user calls mysqli_real_connect(). To avoid tricky conversions mysqlnd uses malloc(), which sets a private persistent flag in the mysqlnd structures. A precondition for the crash to happen was that the private persistent flag is set. The flag is also set when creating a real persistent connection (in the sense of ext/mysqli) and so the bug can happen with mysql_init()/mysqli_real_connect() and mysql_connect('p:<host>', ...). Therefore we test both cases. Note the (tricky?) difference between the implementation detail'mysqlnd private persistent flag = use malloc()' and persistent connections from a user perspective. Although mysqlnd will always set its private persistent flag and use malloc() for connections created with mysqli_init() it is still up to the user to decide in mysqli_real_connect() if the connection shall become a (true) persistent connection or not.
2009-09-16 17:03:44 +00:00
Ulf Wendel
20005db2a0
Fix and tests for bug #49511 . mysqlnd and the MySQL Client Library (libmysql) use different networking APIs. mysqlnd does use PHP streams whereas libmysql uses its own wrapper of the operating level network calls. PHP sets by default a read timeout of 60s for streams - php.ini, default_socket_timeout. This default applies to all streams that set no other timeout value. mysqlnd has not set any other value and therefore it connections of long running queries can have been cut off after default_socket_timeout seconds resulting in a 2006 - MySQL Server has gone away error message. The MySQL Client Library sets a default timeout of 365 * 24 * 3600 seconds (1year) and waits for other timeouts to happen, e.g. TCP/IP timeouts. mysqlnd now uses the same very long timeout. The value is configurable through a new php.ini setting: mysqlnd.net_read_timeout. mysqlnd.net_read_timeout gets used by any extension (ext/mysql, ext/mysqli, PDO_MySQL) that uses mysqlnd. mysqlnd tells PHP Streams to use mysqlnd.net_read_timeout. Please note that there may be subtle differences between MYSQL_OPT_READ_TIMEOUT from the MySQL Client Library and PHP Streams. E.g. MYSQL_OPT_READ_TIMEOUT is documented to work only for TCP/IP connections and, prior to MySQL 5.1.2, only for Windows. PHP streams may not have this limitation. Please check the streams documentation, if in doubt.
2009-09-16 15:00:54 +00:00
andy wharmby
e509cc4283
Fix SKIPIF sections
2009-09-15 21:44:08 +00:00