]>
granicus.if.org Git - php/log
Reeze Xia [Fri, 10 Aug 2012 16:34:37 +0000 (00:34 +0800)]
Skip tests reply on Zend MM turned on
Those two tests have expect memory exhausted fatal error
Daniel Veillard [Thu, 9 Aug 2012 18:02:33 +0000 (14:02 -0400)]
Update to work with libxml 2.9.0
Matt Ficken [Mon, 6 Aug 2012 12:07:42 +0000 (14:07 +0200)]
Added the intl tests extracted from Symfony. These are making the intl ext crash with ICU at least 4.6, but probably with earlier versions too.
Stanislav Malyshev [Mon, 6 Aug 2012 03:31:08 +0000 (20:31 -0700)]
fix for #62460
Reeze Xia [Wed, 18 Jul 2012 16:27:34 +0000 (00:27 +0800)]
Fixed bug #62460 (php binaries installed as binary.dSYM)
See http://marc.info/?l=php-cvs&m=
125961714419896
Mac OS X 10.7&10.8 affected too.
Xinchen Hui [Sun, 5 Aug 2012 04:40:59 +0000 (12:40 +0800)]
Fix 126 tests failed of phar when --enable-zend-multibyte
Xinchen Hui [Sat, 4 Aug 2012 03:03:21 +0000 (11:03 +0800)]
This becomes useless, since we have set that in the INIT macro
Xinchen Hui [Sat, 4 Aug 2012 02:41:26 +0000 (10:41 +0800)]
Fixed bug #62744 (dangling pointers made by zend_disable_class)
the test will be added while commit the fix for #62737
Xinchen Hui [Thu, 2 Aug 2012 14:16:46 +0000 (22:16 +0800)]
Fixed bug that can not get default value of parameter if it's not `optional`
Xinchen Hui [Thu, 2 Aug 2012 11:31:34 +0000 (19:31 +0800)]
Fix test, wrong exepct rule used
Xinchen Hui [Thu, 2 Aug 2012 11:14:06 +0000 (19:14 +0800)]
fix test due to float value
Xinchen Hui [Thu, 2 Aug 2012 04:52:11 +0000 (12:52 +0800)]
Fixed bug #62685 (Wrong return datatype in PDO::inTransaction())
Xinchen Hui [Thu, 2 Aug 2012 04:30:07 +0000 (12:30 +0800)]
Fixed bug #62716 (munmap() is called with the incorrect length)
Xinchen Hui [Wed, 1 Aug 2012 12:23:30 +0000 (20:23 +0800)]
Fixed bug #62715 (ReflectionParameter::isDefaultValueAvailable() wrong result)
Xinchen Hui [Tue, 31 Jul 2012 02:49:13 +0000 (10:49 +0800)]
Test for bug #62680
Gustavo André dos Santos Lopes [Mon, 30 Jul 2012 08:27:41 +0000 (10:27 +0200)]
Limit test to ICU 49
Gustavo André dos Santos Lopes [Mon, 30 Jul 2012 08:25:17 +0000 (10:25 +0200)]
Remove executable bit from files
Xinchen Hui [Sun, 29 Jul 2012 06:30:01 +0000 (14:30 +0800)]
Sike test while there is no zend mm max_size guard
Xinchen Hui [Sun, 29 Jul 2012 05:25:31 +0000 (13:25 +0800)]
Skip test while zend mm is disabled
Xinchen Hui [Sun, 29 Jul 2012 04:18:07 +0000 (12:18 +0800)]
Merge branch 'PHP-5.3' of git.php.net:php-src into PHP-5.3
Xinchen Hui [Sun, 29 Jul 2012 04:17:43 +0000 (12:17 +0800)]
Skip test while zend_mm is disabled
Felipe Pena [Sat, 28 Jul 2012 22:05:13 +0000 (19:05 -0300)]
- Fixed bug #57933 (Wrong table type used in phpinfo output) patch by: selsky at columbia dot edu
Xinchen Hui [Sat, 28 Jul 2012 09:00:05 +0000 (17:00 +0800)]
Fixed bug #60194 for eavl
same reason here
Xinchen Hui [Fri, 27 Jul 2012 16:50:45 +0000 (00:50 +0800)]
Fixed bug #60194 (--with-zend-multibyte and --enable-debug reports LEAK with run-test.php)
It's not a big deal, just because lexer will read the char after cursor before leaving
Rasmus Lerdorf [Tue, 24 Jul 2012 23:28:49 +0000 (16:28 -0700)]
Fix bug #62654
Johannes Schlüter [Sun, 22 Jul 2012 20:45:39 +0000 (22:45 +0200)]
merge 5.3.15 NEWS
Xinchen Hui [Sat, 21 Jul 2012 15:30:32 +0000 (23:30 +0800)]
missed the test script
Xinchen Hui [Sat, 21 Jul 2012 15:27:09 +0000 (23:27 +0800)]
Fix test failed
Xinchen Hui [Sat, 21 Jul 2012 15:26:18 +0000 (23:26 +0800)]
Fixed bug #62616 (ArrayIterator::count() from IteratorIterator instance gives Segmentation fault)
Xinchen Hui [Fri, 20 Jul 2012 16:46:32 +0000 (00:46 +0800)]
Anatoliy Belsky [Tue, 17 Jul 2012 15:50:05 +0000 (17:50 +0200)]
Fix for bug #62379 was missing in 5.3
Hannes Magnusson [Tue, 17 Jul 2012 09:04:56 +0000 (10:04 +0100)]
Do not unload shared extensions when checking for leaks
Jille Timmermans [Thu, 12 Jul 2012 14:41:07 +0000 (16:41 +0200)]
Typofix in test-description
Felipe Pena [Sat, 14 Jul 2012 18:15:11 +0000 (15:15 -0300)]
- Fixed bug #62525 (sigabrt while converting floating point to string)
Felipe Pena [Sat, 14 Jul 2012 16:13:31 +0000 (13:13 -0300)]
- Fixed memory leak when extending DOMXPath
Anatoliy Belsky [Fri, 13 Jul 2012 09:39:57 +0000 (11:39 +0200)]
Fixed bug #61677 ext\zlib\tests\bug_52944.phpt fails
The test is known to fail on windows with zlib version < 1.2.7 (current dep is 1.2.5),
with 1.2.7 it works. As it's primarily a zlib 1.2.5 issue on windows, skip it for now.
Anatoliy Belsky [Thu, 12 Jul 2012 08:54:14 +0000 (10:54 +0200)]
reverted changes for #62477
Anatoliy Belsky [Wed, 11 Jul 2012 20:33:42 +0000 (22:33 +0200)]
updated NEWS
Anatoliy Belsky [Wed, 11 Jul 2012 20:25:31 +0000 (22:25 +0200)]
Fixed bug #62477 LimitIterator int overflow
Xinchen Hui [Tue, 10 Jul 2012 10:43:11 +0000 (18:43 +0800)]
Fixed Bug #62500 (Segfault in DateInterval class when extended)
Xinchen Hui [Mon, 9 Jul 2012 00:32:40 +0000 (08:32 +0800)]
Fixed bug #62499 (curl_setopt($ch, CURLOPT_COOKIEFILE, "") returns false)
this bc break is caused by the fix for #61948
Nuno Lopes [Sun, 8 Jul 2012 19:19:41 +0000 (15:19 -0400)]
appease MSVC (doesnt like unary minus of unsigned ints)
Anthony Ferrara [Sat, 7 Jul 2012 02:37:50 +0000 (22:37 -0400)]
Fix two issues with run-tests.php
1. E_STRICT error due to passing return of array_intersect() into reset() directly
2. Details in junit output can produce invalid UTF-8 and XML due to unescaped characters
Nikita Popov [Thu, 5 Jul 2012 18:31:58 +0000 (20:31 +0200)]
Fix potential integer overflow in nl2br
The buffer size was calculated manually, thus creating integer overflows
for very large inputs, e.g. nl2br(str_repeat("\n",
613566757 )).
The code now uses safe_emalloc, thus making the code throw an error
instead of crashing.
Nikita Popov [Thu, 5 Jul 2012 18:14:49 +0000 (20:14 +0200)]
Fix potential integer overflow in bin2hex
The code was already using safe_emalloc but did the multiplication in
the first argument, thus making the use of safe_emalloc pretty useless.
The *2 is now moved to the second argument.
Johannes Schlüter [Wed, 4 Jul 2012 23:15:27 +0000 (01:15 +0200)]
This wil be PHP 5.3.16
johannes [Wed, 4 Jul 2012 22:55:47 +0000 (00:55 +0200)]
Revert change
3f3ad30c50 : There shouldn't be new features in 5.3, especially
not if they aren't in 5.4, too.
Nuno Lopes [Mon, 2 Jul 2012 05:31:40 +0000 (01:31 -0400)]
fix (signed) integer overflow (part of bug #52550
Nikita Popov [Sun, 1 Jul 2012 14:28:57 +0000 (16:28 +0200)]
Revert JSON changes to PHP 5.3
This reverts the following commits:
974324676b2436f159f42d9241c569f813471684
4662151ea7d7b6920d115cf2a2d6e9d4232727a3
84fe2cc890e49f40bac7c3ba74b3cfc6dc4cef2f
This does not revert the JSON changes released in PHP 5.3.14.
Stanislav Malyshev [Sun, 1 Jul 2012 00:36:22 +0000 (17:36 -0700)]
remove duplicates, fix order
Felipe Pena [Sat, 30 Jun 2012 19:29:30 +0000 (16:29 -0300)]
- Fixed bug #62415 (Undefined variable: diff in run-tests.php on line 2093 for successful XFAIL)
andrey [Fri, 29 Jun 2012 11:42:36 +0000 (14:42 +0300)]
fix Bug #62273 Segmentation Fault in Mysqli/Mysqlnd
andrey [Fri, 29 Jun 2012 11:40:41 +0000 (14:40 +0300)]
Merge branch 'PHP-5.3' of ssh://git.php.net/php-src into PHP-5.3
Nikita Popov [Fri, 29 Jun 2012 10:47:37 +0000 (12:47 +0200)]
Fix some lengths in crypt()
Use salt_len_in instead of strlen(salt) or PHP_MAX_SALT_LEN, otherwise too
much memory will be allocated.
sha512 has a 86 character checksum, not 43. That probably was a copy&paste
from the sha256 code which indeed has 43.
The allocation also was using sizeof(char *), thus allocating 4 or 8 times
as much memory as necessary. The sizeof(char *) was removed in the 5.4
branch in
b7a92c9 but forgotten on 5.3.
The memset 0 call was using PHP_MAX_SALT_LEN which can be smaller than the
output buffer and thus not zeroing out everything. Use the size of the
output buffer (needed) instead.
andrey [Fri, 29 Jun 2012 09:42:54 +0000 (12:42 +0300)]
new charsets
Anthony Ferrara [Fri, 29 Jun 2012 00:00:03 +0000 (20:00 -0400)]
Fixed bug #62443 (Crypt SHA256/512 Segfaults With Malformed Salt)
Fixed a memory allocation bug in crypt() SHA256/512 that can
cause segmentation faults when passed in salts with a null byte
early.
Nikita Popov [Thu, 28 Jun 2012 18:09:47 +0000 (20:09 +0200)]
Add json_last_error_msg() function
This replaces json_last_error(true) and is consistent with other custom
error handling functions.
Johannes Schlüter [Wed, 27 Jun 2012 21:26:33 +0000 (23:26 +0200)]
Fix #62432 ReflectionMethod random corrupt memory on high concurrent
This fixes the same issue in multiple extensions. This isn't needed
in later branches as 5.4 introduced object_properties_init()
Nikita Popov [Wed, 27 Jun 2012 10:19:41 +0000 (12:19 +0200)]
Improve JSON error handling
json_encode() no longer throws warnings. Instead only the error code for
json_last_error() is set.
As it is hard to debug the error from just an error code an optional
$as_string parameter was added to json_last_error(), which returns an
error message instead of an error code.
Johannes Schlüter [Wed, 27 Jun 2012 09:42:43 +0000 (11:42 +0200)]
Fix memleak in CLI
Stanislav Malyshev [Tue, 26 Jun 2012 22:22:29 +0000 (15:22 -0700)]
ws fix
Moriyoshi Koizumi [Mon, 25 Jun 2012 10:20:38 +0000 (19:20 +0900)]
BFN
Moriyoshi Koizumi [Mon, 25 Jun 2012 10:13:23 +0000 (19:13 +0900)]
Fix bug #62373 (serialize() generates wrong reference to the object)
Nikita Popov [Sun, 24 Jun 2012 21:32:50 +0000 (23:32 +0200)]
Fix memory allocation checks for base64 encode
base64_encode used safe_emalloc, but one of the arguments was derived from a
multiplication, thus making the allocation unsafe again.
There was a size check in place, but it was off by a factor of two as it
didn't account for the signedness of the integer type.
The unsafe allocation is not exploitable, but still causes funny behavior
when the sized overflows into a negative number.
To fix the issue the *4 factor is moved into the size argument (where it is
known to be safe), so safe_emalloc can carry out the multiplication.
The size check is removed as it doesn't really make sense once safe_emalloc
works correctly. (Would only cause base64_encode to silently return false
instead of throwing an error. Also could cause problems with other uses of
the base64 encoding API, which all don't check for a NULL return value.)
Furthermore the (length + 2) < 0 check is replaced with just length < 0.
Allowing lengths -2 and -1 doesn't make sense semantically and also is not
honored in the following code (negative length would access unallocated
memory.)
Actually the length < 0 check doesn't make sense altogether, but I left it
there just to be safe.
Nikita Popov [Sat, 23 Jun 2012 18:46:27 +0000 (20:46 +0200)]
Improve json_encode error handling
json_encode() now returns bool(false) for all possible errors, throws the
respective warning and also sets the respective json_last_error() error
code. Three new error codes have been added:
* JSON_ERROR_RECURSION
* JSON_ERROR_INF_OR_NAN
* JSON_ERROR_UNSUPPORTED_TYPE
To get a partial JSON output instead of bool(false) the option
JSON_PARTIAL_OUTPUT_ON_ERROR can be specified. In this case the invalid
segments will be replaced either by null (for recursion, unsupported type
and invalid JSON) or 0 (for Inf and NaN).
The warning for invalid UTF-8 stays intact and is thrown also with
display_errors = On. If this behavior is undesired this can be remedied
later.
Felipe Pena [Sat, 23 Jun 2012 18:21:20 +0000 (15:21 -0300)]
- BFN
Felipe Pena [Sat, 23 Jun 2012 18:10:47 +0000 (15:10 -0300)]
- Improved fix for #62384
Johannes Schlüter [Fri, 22 Jun 2012 18:00:20 +0000 (20:00 +0200)]
Remove unneeded and outdated ChangeLog file
Johannes Schlüter [Fri, 22 Jun 2012 15:32:46 +0000 (17:32 +0200)]
Remove extra ;
Felipe Pena [Fri, 22 Jun 2012 15:05:29 +0000 (12:05 -0300)]
- Fixed bug #62384 (Attempting to invoke a Closure more than once causes segfaul)
Johannes Schlüter [Fri, 22 Jun 2012 10:48:39 +0000 (12:48 +0200)]
Improve check for :memory: pseudo-filename in SQlite
Johannes Schlüter [Wed, 20 Jun 2012 17:46:32 +0000 (19:46 +0200)]
Remove outdated and user-specific files
Nikita Popov [Tue, 19 Jun 2012 16:40:46 +0000 (18:40 +0200)]
Add NEWS for bug #62262
Nikita Popov [Tue, 19 Jun 2012 15:45:04 +0000 (17:45 +0200)]
Fixed bug RecursiveArrayIterator does not implement Countable
ArrayIterator implemented Countable only after it was already inherited by
RecursiveArrayIterator. Thus the interface was missing in RAI.
Anatoliy Belsky [Tue, 19 Jun 2012 15:15:56 +0000 (17:15 +0200)]
sync zip ext version with pecl
Anatoliy Belsky [Tue, 19 Jun 2012 12:42:59 +0000 (14:42 +0200)]
one more correction for COM upgrading notes
Anatoliy Belsky [Tue, 19 Jun 2012 12:08:34 +0000 (14:08 +0200)]
split gzgetc_basic.phpt for zlib 1.2.7
Anatoliy Belsky [Mon, 18 Jun 2012 20:07:19 +0000 (22:07 +0200)]
com ext upgrading correction
Anatoliy Belsky [Mon, 18 Jun 2012 19:56:17 +0000 (21:56 +0200)]
com ext upgrading info
Anatoliy Belsky [Sun, 17 Jun 2012 14:04:36 +0000 (16:04 +0200)]
Reverted the BC fix regarding to #57905, test adopted
New 0.10.1 has a stronger archive integrity check. Restoring
the old behaviour of libzip < 0.10.1 makes no sense at this
place.
Johannes Schlüter [Wed, 13 Jun 2012 08:35:58 +0000 (10:35 +0200)]
Merge PHP 5.3.14 NEWS
Stanislav Malyshev [Tue, 12 Jun 2012 18:49:35 +0000 (11:49 -0700)]
re-add 61755 to NEWS
David Soria Parra [Sun, 10 Jun 2012 14:15:34 +0000 (18:15 +0400)]
Make travis silent
Travis will always build all branches. As we just have a .travis.yml
on master, travis will go ahead and checkout PHP-5.3. It fails and
then sends mails. We really don't want to get spammed, so we add
a .travis.yml that is just silent.
michelangelo [Sat, 9 Jun 2012 16:07:11 +0000 (16:07 +0000)]
Adding a test for ext/posix/tests/posix_getegid_basic.phpt
Stanislav Malyshev [Fri, 8 Jun 2012 08:24:49 +0000 (01:24 -0700)]
typo
Stanislav Malyshev [Fri, 8 Jun 2012 08:21:37 +0000 (01:21 -0700)]
improve overflow checks
Stanislav Malyshev [Fri, 8 Jun 2012 06:05:23 +0000 (23:05 -0700)]
fix potential overflow in _php_stream_scandir
Anatoliy Belsky [Thu, 7 Jun 2012 20:32:03 +0000 (22:32 +0200)]
set current versions for libzip and zip ext
Anatoliy Belsky [Thu, 7 Jun 2012 19:43:07 +0000 (21:43 +0200)]
updated NEWS
Anatoliy Belsky [Thu, 7 Jun 2012 15:59:57 +0000 (17:59 +0200)]
zip windows fixes
Anatoliy Belsky [Thu, 7 Jun 2012 14:27:47 +0000 (16:27 +0200)]
fixed bc break related to #57905
Anatoliy Belsky [Tue, 5 Jun 2012 22:57:12 +0000 (00:57 +0200)]
fixed a bit overlooked from the last libzip port
Anatoliy Belsky [Tue, 5 Jun 2012 21:22:27 +0000 (23:22 +0200)]
brought the fix for #47667 back
Anatoliy Belsky [Tue, 5 Jun 2012 20:49:31 +0000 (22:49 +0200)]
fixed zip entry freeing
Anatoliy Belsky [Tue, 5 Jun 2012 20:11:07 +0000 (22:11 +0200)]
rechecked the merged libzip, took also the indents from the original to
avoid confusion next time
Anatoliy Belsky [Mon, 4 Jun 2012 20:51:31 +0000 (22:51 +0200)]
fixed a double freeing crash
Anatoliy Belsky [Mon, 4 Jun 2012 20:51:00 +0000 (22:51 +0200)]
fixed header include
Anatoliy Belsky [Mon, 4 Jun 2012 19:30:04 +0000 (21:30 +0200)]
initial libzip upgrade patch to 0.10.1
Felipe Pena [Thu, 7 Jun 2012 17:23:17 +0000 (14:23 -0300)]
- BFN
Felipe Pena [Thu, 7 Jun 2012 17:18:34 +0000 (14:18 -0300)]