]> granicus.if.org Git - php/log
php
10 years agoAdd to NEWS
Keyur Govande [Tue, 7 Oct 2014 21:22:54 +0000 (21:22 +0000)]
Add to NEWS

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Keyur Govande [Tue, 7 Oct 2014 21:22:07 +0000 (21:22 +0000)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  Add to NEWS
  Fix for bug #68087 (ODBC not reading DATE columns correctly)

10 years agoAdd to NEWS
Keyur Govande [Tue, 7 Oct 2014 21:21:23 +0000 (21:21 +0000)]
Add to NEWS

10 years agoFix for bug #68087 (ODBC not reading DATE columns correctly)
Keyur Govande [Tue, 7 Oct 2014 21:17:36 +0000 (21:17 +0000)]
Fix for bug #68087 (ODBC not reading DATE columns correctly)

Temporary variable indicating column field type ID should be
reset to default for loop iteration (i.e. every column in the
record set. The old buggy code made it persist across all columns
leading to invalid reads from the buffer, if for example a DATE
column was preceded by a VARCHAR column.

10 years agoAdd to NEWS
Keyur Govande [Tue, 7 Oct 2014 21:09:47 +0000 (21:09 +0000)]
Add to NEWS

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Keyur Govande [Tue, 7 Oct 2014 21:08:18 +0000 (21:08 +0000)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  Add to NEWS
  Fix for bug #68114 (Build fails on OS X due to undefined symbols)

10 years agoAdd to NEWS
Keyur Govande [Tue, 7 Oct 2014 21:07:19 +0000 (21:07 +0000)]
Add to NEWS

10 years agoFix for bug #68114 (Build fails on OS X due to undefined symbols)
Keyur Govande [Tue, 7 Oct 2014 20:58:43 +0000 (20:58 +0000)]
Fix for bug #68114 (Build fails on OS X due to undefined symbols)

gcc (i686-apple-darwin10-gcc-4.2.1) on OS X cannot link fixed-width
decimals and fails with undefined symbols errors like ___extendsddf.
If configure used gcc for compiling it would notice and mark the
feature HAVE_DECIMAL_FP_SUPPORT as unsupported.
But configure seems to use cc (i686-apple-darwin10-llvm-gcc-4.2)
instead, which doesn't support fixed-width decimals either, but the
code compiles and links just fine. I suspect it may have something
to do with the llvm backend printed in the version.
Lacking the time to debug this further, the patch fixes the issue by
checking the expected output when fixed-width decimal support is
present and correctly implemented.

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Johannes Schlüter [Sat, 4 Oct 2014 18:06:44 +0000 (20:06 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

10 years agoFix accidental edit in previous commit
Johannes Schlüter [Sat, 4 Oct 2014 18:06:02 +0000 (20:06 +0200)]
Fix accidental edit in previous commit

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Johannes Schlüter [Sat, 4 Oct 2014 18:00:10 +0000 (20:00 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

10 years agoFix arginfo
Johannes Schlüter [Sat, 4 Oct 2014 17:59:21 +0000 (19:59 +0200)]
Fix arginfo

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Derick Rethans [Sat, 4 Oct 2014 14:59:17 +0000 (15:59 +0100)]
Merge branch 'PHP-5.5' into PHP-5.6

10 years agoMerge branch 'PHP-5.4' into PHP-5.5
Derick Rethans [Sat, 4 Oct 2014 14:58:40 +0000 (15:58 +0100)]
Merge branch 'PHP-5.4' into PHP-5.5

10 years agoDateTimeZone::getOffset() now accepts a DateTimeInterface
Florian MARGAINE [Mon, 22 Sep 2014 21:35:29 +0000 (23:35 +0200)]
DateTimeZone::getOffset() now accepts a DateTimeInterface

Fixes #68062

10 years agoMerge branch 'PHP-5.6' of git.php.net:php-src into PHP-5.6
George Wang [Fri, 3 Oct 2014 20:43:35 +0000 (16:43 -0400)]
Merge branch 'PHP-5.6' of git.php.net:php-src into PHP-5.6

10 years agoMerge branch 'PHP-5.5' of git.php.net:php-src into PHP-5.5
George Wang [Fri, 3 Oct 2014 20:43:08 +0000 (16:43 -0400)]
Merge branch 'PHP-5.5' of git.php.net:php-src into PHP-5.5

10 years agoFixed a bug that causes crash when environment variable is access while parsing php.ini
George Wang [Fri, 3 Oct 2014 20:41:32 +0000 (16:41 -0400)]
Fixed a bug that causes crash when environment variable is access while parsing php.ini

10 years agoFixed a bug that causes crash when environment variable is access while parsing php.ini
George Wang [Fri, 3 Oct 2014 20:41:32 +0000 (16:41 -0400)]
Fixed a bug that causes crash when environment variable is access while parsing php.ini

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Nikita Popov [Fri, 3 Oct 2014 19:42:11 +0000 (21:42 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

10 years agoFix bug number
Nikita Popov [Fri, 3 Oct 2014 19:41:58 +0000 (21:41 +0200)]
Fix bug number

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Nikita Popov [Fri, 3 Oct 2014 19:26:51 +0000 (21:26 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

10 years agoFix bug #68188
Nikita Popov [Fri, 3 Oct 2014 19:24:28 +0000 (21:24 +0200)]
Fix bug #68188

10 years agoNEWS entry for previous commit
Ferenc Kovacs [Fri, 3 Oct 2014 09:39:53 +0000 (11:39 +0200)]
NEWS entry for previous commit

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Ferenc Kovacs [Fri, 3 Oct 2014 09:39:09 +0000 (11:39 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  NEWS entry for previous commit
  add IPv6 support to php-fpm

10 years agoNEWS entry for previous commit
Ferenc Kovacs [Fri, 3 Oct 2014 09:38:32 +0000 (11:38 +0200)]
NEWS entry for previous commit

10 years agoadd IPv6 support to php-fpm
Robin Gloster [Fri, 3 Oct 2014 09:35:27 +0000 (11:35 +0200)]
add IPv6 support to php-fpm

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Sara Golemon [Fri, 3 Oct 2014 01:55:29 +0000 (18:55 -0700)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  Add hash to EXTENSIONS file

10 years agoMerge branch 'PHP-5.4' into PHP-5.5
Sara Golemon [Fri, 3 Oct 2014 01:55:14 +0000 (18:55 -0700)]
Merge branch 'PHP-5.4' into PHP-5.5

* PHP-5.4:
  Add hash to EXTENSIONS file

10 years agoAdd hash to EXTENSIONS file
Sara Golemon [Fri, 3 Oct 2014 01:54:46 +0000 (18:54 -0700)]
Add hash to EXTENSIONS file

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Ferenc Kovacs [Thu, 2 Oct 2014 20:32:18 +0000 (22:32 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  we also have xz release tarballs since 5.5
  5.5.19 now

Conflicts:
configure.in
main/php_version.h

10 years agowe also have xz release tarballs since 5.5
Ferenc Kovacs [Thu, 2 Oct 2014 20:28:19 +0000 (22:28 +0200)]
we also have xz release tarballs since 5.5

10 years ago5.5.19 now
Julien Pauli [Wed, 1 Oct 2014 14:16:05 +0000 (16:16 +0200)]
5.5.19 now

10 years agoupdate NEWS about the release date of 5.6.1
Ferenc Kovacs [Wed, 1 Oct 2014 09:15:59 +0000 (11:15 +0200)]
update NEWS about the release date of 5.6.1

10 years agoonly no_NO.ISO-8859-1 have to be ignored
Remi Collet [Wed, 1 Oct 2014 08:14:14 +0000 (10:14 +0200)]
only no_NO.ISO-8859-1 have to be ignored

10 years agoremove bogus locale use from test
krakjoe [Thu, 21 Aug 2014 09:03:48 +0000 (10:03 +0100)]
remove bogus locale use from test

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Remi Collet [Wed, 1 Oct 2014 07:26:46 +0000 (09:26 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  use $PHP_LIBDIR instread of "lib" in skel example

10 years agouse $PHP_LIBDIR instread of "lib" in skel example
Remi Collet [Wed, 1 Oct 2014 07:26:21 +0000 (09:26 +0200)]
use $PHP_LIBDIR instread of "lib" in skel example

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Anatol Belski [Mon, 29 Sep 2014 18:30:43 +0000 (20:30 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  fix test filename

10 years agofix test filename
Anatol Belski [Mon, 29 Sep 2014 18:30:18 +0000 (20:30 +0200)]
fix test filename

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Anatol Belski [Mon, 29 Sep 2014 18:19:31 +0000 (20:19 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  one more test to illustrate transfer of an arbitrary data amount throug pipes

10 years agoone more test to illustrate transfer of an arbitrary data amount throug pipes
Anatol Belski [Mon, 29 Sep 2014 18:18:33 +0000 (20:18 +0200)]
one more test to illustrate transfer of an arbitrary data amount throug pipes

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Anatol Belski [Mon, 29 Sep 2014 16:12:56 +0000 (18:12 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  fix tests on linux
  better test cleanup

10 years agofix tests on linux
Anatol Belski [Mon, 29 Sep 2014 16:12:25 +0000 (18:12 +0200)]
fix tests on linux

10 years agobetter test cleanup
Anatol Belski [Mon, 29 Sep 2014 16:11:06 +0000 (18:11 +0200)]
better test cleanup

and even force the XFAIL to fail as it would take too long anyway, even if it passes

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Anatol Belski [Mon, 29 Sep 2014 15:28:27 +0000 (17:28 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  increase the polling period to not to break existing behaviours

10 years agoincrease the polling period to not to break existing behaviours
Anatol Belski [Mon, 29 Sep 2014 15:27:24 +0000 (17:27 +0200)]
increase the polling period to not to break existing behaviours

10 years agoupdated NEWS
Anatol Belski [Mon, 29 Sep 2014 14:28:12 +0000 (16:28 +0200)]
updated NEWS

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Anatol Belski [Mon, 29 Sep 2014 14:27:33 +0000 (16:27 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  Fixed bug #51800 proc_open on Windows hangs forever

10 years agoFixed bug #51800 proc_open on Windows hangs forever
Anatol Belski [Mon, 29 Sep 2014 14:24:34 +0000 (16:24 +0200)]
Fixed bug #51800 proc_open on Windows hangs forever

This loop can block for some minutes, theoretically. Practially
however, this is a 99% non issue for a normal use case. This is
required because read() is synchronous. The PHP streams API wants
to fill its internal buffers, therefore it might try to read some
more data than user has demanded. Also, for a case where we want
to read X bytes, but neither enough data nor EOF arrives, read()
will block until it could fill the buffer. If a counterpart station
runs slowly or delivers not all the data at once, read() would
still be waiting. If we quit too early, we possibly could loose
some data from the pipe. Thus it has to emulate the read()
behaviour, but obviously not completely, just to some grade.

Reading big data amount is for sure an issue on any platforms, it
depends on the pipe buffer size, which is controlled by the system.
On Windows, the buffer size seems to be way too small, which causes
buffer congestion and a dead lock. It is essential to read the pipe
descriptors simultaneously and possibly in the same order as the
opposite writes them.

Thus, this will work with smaller buffer data sizes passed through
pipes. As MSDN states, anonymous pipes don't support asynchronous
operations. Neither anonymous pipes do support select() as they are
not SOCKETs but file descriptors. Consequently - bigger data sizes
will need a better solution based on threads. However it is much
more expencive. Maybe a better solution could be exporting a part
of the internal doing as a userspace function which could perform
some kind of lookahead operation on the pipe descriptor.

This is just the first stone, depending on the user feedback we
might go for further improvements in this area.

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Stanislav Malyshev [Sun, 28 Sep 2014 19:25:49 +0000 (12:25 -0700)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  Set an LDAP error code when failing ldap_bind due to null bytes

10 years agoMerge branch 'PHP-5.4' into PHP-5.5
Stanislav Malyshev [Sun, 28 Sep 2014 19:25:24 +0000 (12:25 -0700)]
Merge branch 'PHP-5.4' into PHP-5.5

* PHP-5.4:
  Set an LDAP error code when failing ldap_bind due to null bytes

10 years agoSet an LDAP error code when failing ldap_bind due to null bytes
Matthew Daley [Sun, 28 Sep 2014 04:49:35 +0000 (17:49 +1300)]
Set an LDAP error code when failing ldap_bind due to null bytes

Some applications check a LDAP link's error code after seeing ldap_bind
fail due to a null byte bind attempt and hence incorrectly receive the
last set error code.

Fix by setting an LDAP error code before returning in this case.

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Johannes Schlüter [Sat, 27 Sep 2014 00:18:29 +0000 (02:18 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

10 years agoMerge branch 'PHP-5.4' into PHP-5.5
Johannes Schlüter [Sat, 27 Sep 2014 00:17:52 +0000 (02:17 +0200)]
Merge branch 'PHP-5.4' into PHP-5.5

10 years agoThis test should pass
Johannes Schlüter [Sat, 27 Sep 2014 00:17:26 +0000 (02:17 +0200)]
This test should pass

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Derick Rethans [Fri, 26 Sep 2014 15:27:05 +0000 (16:27 +0100)]
Merge branch 'PHP-5.5' into PHP-5.6

10 years agoMerge branch 'PHP-5.4' into PHP-5.5
Derick Rethans [Fri, 26 Sep 2014 15:27:00 +0000 (16:27 +0100)]
Merge branch 'PHP-5.4' into PHP-5.5

10 years ago- Updated to version 2014.8 (2014h)
Derick Rethans [Fri, 26 Sep 2014 15:26:59 +0000 (16:26 +0100)]
- Updated to version 2014.8 (2014h)

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Stanislav Malyshev [Fri, 26 Sep 2014 08:01:39 +0000 (01:01 -0700)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  fix date
  update NEWS
  Revert xp_ssl.c to the state of 5.4.32 due to regressions

10 years agoMerge branch 'PHP-5.4' into PHP-5.5
Stanislav Malyshev [Fri, 26 Sep 2014 08:00:20 +0000 (01:00 -0700)]
Merge branch 'PHP-5.4' into PHP-5.5

* PHP-5.4:
  update NEWS
  Revert xp_ssl.c to the state of 5.4.32 due to regressions

Conflicts:
ext/openssl/xp_ssl.c

10 years agofix date
Stanislav Malyshev [Fri, 26 Sep 2014 07:56:39 +0000 (00:56 -0700)]
fix date

10 years agoupdate NEWS
Stanislav Malyshev [Fri, 26 Sep 2014 07:55:49 +0000 (00:55 -0700)]
update NEWS

10 years agoRevert xp_ssl.c to the state of 5.4.32 due to regressions
Stanislav Malyshev [Fri, 26 Sep 2014 07:44:24 +0000 (00:44 -0700)]
Revert xp_ssl.c to the state of 5.4.32 due to regressions

10 years agoNEWS
Remi Collet [Fri, 26 Sep 2014 07:07:12 +0000 (09:07 +0200)]
NEWS

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Remi Collet [Fri, 26 Sep 2014 07:06:51 +0000 (09:06 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  NEWS
  Fixed bug #68103 Dupplicate entry in Reflection

10 years agoNEWS
Remi Collet [Fri, 26 Sep 2014 07:06:12 +0000 (09:06 +0200)]
NEWS

10 years agoFixed bug #68103 Dupplicate entry in Reflection
Remi Collet [Fri, 26 Sep 2014 06:56:42 +0000 (08:56 +0200)]
Fixed bug #68103 Dupplicate entry in Reflection

$ php -r '$r=new ReflectionExtension("pthreads"); print_r($r->getClassNames());'
Array
(
    [0] => Threaded
    [1] => stackable
    [2] => Thread
    [3] => Worker
    [4] => Mutex
    [5] => Cond
    [6] => Collectable
    [7] => Pool
)

In getClasses() output, it is possible to compare key (ex "stackable")
with $obj->name (ex "Threaded") to detect class alias.

...
    [Threaded] => ReflectionClass Object
        (
            [name] => Threaded
        )
    [stackable] => ReflectionClass Object
        (
            [name] => Threaded
        )
...

10 years agoupdate the certificate used for the test, as it expired recently
Ferenc Kovacs [Wed, 24 Sep 2014 11:40:52 +0000 (13:40 +0200)]
update the certificate used for the test, as it expired recently

10 years agotypo
Remi Collet [Wed, 24 Sep 2014 08:42:20 +0000 (10:42 +0200)]
typo

10 years agoNEWS
Remi Collet [Wed, 24 Sep 2014 08:36:29 +0000 (10:36 +0200)]
NEWS

10 years agoFix bug #68074 Allow to use system cipher list instead of hardcoded value
Remi Collet [Wed, 24 Sep 2014 08:34:55 +0000 (10:34 +0200)]
Fix bug #68074 Allow to use system cipher list instead of hardcoded value

10 years agothe order of the blocks should be Core, then exts in alphabetical order
Ferenc Kovacs [Wed, 24 Sep 2014 07:40:52 +0000 (09:40 +0200)]
the order of the blocks should be Core, then exts in alphabetical order

10 years agoadd missing NEWS entry for the phpdbg compilation fix
Ferenc Kovacs [Wed, 24 Sep 2014 07:39:14 +0000 (09:39 +0200)]
add missing NEWS entry for the phpdbg compilation fix

10 years agoadd NEWS entry for #68088
Ferenc Kovacs [Wed, 24 Sep 2014 07:09:34 +0000 (09:09 +0200)]
add NEWS entry for #68088

10 years agoPR 837: Fixed typo in php.ini-development and php.ini-production
Tjerk Meesters [Tue, 23 Sep 2014 10:10:32 +0000 (18:10 +0800)]
PR 837: Fixed typo in php.ini-development and php.ini-production

Fixed typo in comment for 'mbstring.detect_order'

10 years agoFixed typo
Sebastian Zartner [Tue, 23 Sep 2014 05:45:03 +0000 (07:45 +0200)]
Fixed typo

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Ard Biesheuvel [Sat, 20 Sep 2014 23:42:12 +0000 (16:42 -0700)]
Merge branch 'PHP-5.5' into PHP-5.6

10 years agoMerge branch 'PHP-5.4' into PHP-5.5
Ard Biesheuvel [Sat, 20 Sep 2014 23:40:02 +0000 (16:40 -0700)]
Merge branch 'PHP-5.4' into PHP-5.5

10 years agoFixed bug #66242 (don't assume char is signed)
Ard Biesheuvel [Sat, 20 Sep 2014 23:32:31 +0000 (16:32 -0700)]
Fixed bug #66242 (don't assume char is signed)

This fixes a bug in libmagic where a cast to 'char' is assumed to result
in sign extension to occur. However, unqualified 'char' is unsigned on
architectures such as ARM, so the cast needs to be to 'signed char'
explicitly.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Nikita Popov [Sat, 20 Sep 2014 19:47:16 +0000 (21:47 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

10 years agoFixed bug #67633
Nikita Popov [Sat, 20 Sep 2014 19:46:25 +0000 (21:46 +0200)]
Fixed bug #67633

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Anatol Belski [Fri, 19 Sep 2014 18:13:46 +0000 (20:13 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  fix possible array underflow

10 years agofix possible array underflow
Anatol Belski [Fri, 19 Sep 2014 18:12:24 +0000 (20:12 +0200)]
fix possible array underflow

there are multiple issues with this code

- php_stream_read() returns an unsigned val, so is >= 0
- if it read less than sizeof(a) bytes, the function operates on garbage
- result->channels is an unsigned val, so >= 0

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Matteo Beccati [Tue, 16 Sep 2014 08:22:49 +0000 (10:22 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  Fixed freetype test on multiple environments

10 years agoMerge branch 'PHP-5.4' into PHP-5.5
Matteo Beccati [Tue, 16 Sep 2014 08:22:34 +0000 (10:22 +0200)]
Merge branch 'PHP-5.4' into PHP-5.5

* PHP-5.4:
  Fixed freetype test on multiple environments

10 years agoFixed freetype test on multiple environments
Matteo Beccati [Tue, 16 Sep 2014 08:17:25 +0000 (10:17 +0200)]
Fixed freetype test on multiple environments

Some environments, apparently regardless to the freetype version, output 155, while others 156. I guess we can accept both ;)

This reverts commit 592df890276604c790d02be26b1b3c3710b3b888.

10 years agoNEWS
Remi Collet [Mon, 15 Sep 2014 11:32:13 +0000 (13:32 +0200)]
NEWS

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Remi Collet [Mon, 15 Sep 2014 11:31:26 +0000 (13:31 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  NEWS
  Fixed bug #65641 PHP-FPM incorrectly defines the SCRIPT_NAME variable when using Apache

10 years agoNEWS
Remi Collet [Mon, 15 Sep 2014 11:31:17 +0000 (13:31 +0200)]
NEWS

10 years agoFixed bug #65641 PHP-FPM incorrectly defines the SCRIPT_NAME variable when using...
Remi Collet [Mon, 15 Sep 2014 11:29:55 +0000 (13:29 +0200)]
Fixed bug #65641 PHP-FPM incorrectly defines the SCRIPT_NAME variable when using Apache

ProxyPass is unable to provide correct PATH_INFO
as it is not aware of file path (while SetHandler is).

As we can extract PATH_INFO from PATH_TRANSLATED,
we also need to check if present in SCRIPT_NAME
and remove it.

After applying this patch.
With mod_php
_SERVER["REQUEST_URI"]     /info.php/foo/bar?q=1
_SERVER["SCRIPT_NAME"]     /info.php
_SERVER["PATH_INFO"]       /foor/bar
_SERVER["PHP_SELF"]        /info.php/foo/bar
_SERVER["QUERY_STRING"]    q=1

With mod_proxy_fcgi + SetHandler
_SERVER["REQUEST_URI"]     /info.php/foo/bar?q=1
_SERVER["SCRIPT_NAME"]     /info.php
_SERVER["PATH_INFO"]       /foo/bar
_SERVER["PHP_SELF"]        /info.php/foo/bar
_SERVER["QUERY_STRING"]    q=1

With mod_proxy_fcgi + ProxyPass
_SERVER["REQUEST_URI"]     /info.php/foo/bar?q=1
_SERVER["SCRIPT_NAME"]     /info.php
_SERVER["PATH_INFO"]       /foo/bar
_SERVER["PHP_SELF"]        /info.php/foo/bar
_SERVER["QUERY_STRING"]    q=1

10 years agoupdated NEWS
Anatol Belski [Mon, 15 Sep 2014 08:29:00 +0000 (10:29 +0200)]
updated NEWS

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Anatol Belski [Mon, 15 Sep 2014 08:28:22 +0000 (10:28 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  updated NEWS

10 years agoupdated NEWS
Anatol Belski [Mon, 15 Sep 2014 08:27:52 +0000 (10:27 +0200)]
updated NEWS

10 years agoMerge branch 'PHP-5.5' into PHP-5.6
Remi Collet [Mon, 15 Sep 2014 06:24:20 +0000 (08:24 +0200)]
Merge branch 'PHP-5.5' into PHP-5.6

* PHP-5.5:
  Fix NEWS - #65641 mod_proxy-fcgi is not fixed (still open) - #67606 mod_fastcgi is fixed in ee275e34c8b303945945c650d4bc90dcc2ac0b17
  Fix NEWS - #65641 mod_proxy-fcgi is not fixed (still open) - #67606 mod_fastcgi is fixed in ee275e34c8b303945945c650d4bc90dcc2ac0b17

10 years agoFix NEWS
Remi Collet [Mon, 15 Sep 2014 06:24:10 +0000 (08:24 +0200)]
Fix NEWS
- #65641 mod_proxy-fcgi is not fixed (still open)
- #67606 mod_fastcgi is fixed in ee275e34c8b303945945c650d4bc90dcc2ac0b17

10 years agoMerge branch 'PHP-5.4' into PHP-5.5
Remi Collet [Mon, 15 Sep 2014 06:23:40 +0000 (08:23 +0200)]
Merge branch 'PHP-5.4' into PHP-5.5

* PHP-5.4:
  Fix NEWS - #65641 mod_proxy-fcgi is not fixed (still open) - #67606 mod_fastcgi is fixed in ee275e34c8b303945945c650d4bc90dcc2ac0b17

10 years agoFix NEWS
Remi Collet [Mon, 15 Sep 2014 06:23:25 +0000 (08:23 +0200)]
Fix NEWS
- #65641 mod_proxy-fcgi is not fixed (still open)
- #67606 mod_fastcgi is fixed in ee275e34c8b303945945c650d4bc90dcc2ac0b17

10 years agoMerge phpdbg into PHP-5.6
Bob Weinand [Mon, 15 Sep 2014 03:48:09 +0000 (05:48 +0200)]
Merge phpdbg into PHP-5.6

10 years agoFix bug #111 (compile error without ZEND_SIGNALS)
Bob Weinand [Sun, 14 Sep 2014 10:24:56 +0000 (12:24 +0200)]
Fix bug #111 (compile error without ZEND_SIGNALS)