]> granicus.if.org Git - php/log
php
7 years agofixed bug #50989 (DOM support for LIBXML_NOXMLDECL)
jhdxr [Sat, 28 Jan 2017 06:56:03 +0000 (14:56 +0800)]
fixed bug #50989 (DOM support for LIBXML_NOXMLDECL)

8 years agoFix 00version.phpt when commit message has multiple lines.
Mitch Hagstrand [Thu, 26 Jan 2017 10:24:07 +0000 (02:24 -0800)]
Fix 00version.phpt when commit message has multiple lines.

The environment variable "TRAVIS_COMMIT_MESSAGE" can contain multiple lines.

An example can be seen here https://travis-ci.org/php/php-src/jobs/195465293

8 years agoFixed #73973 - debug_zval_dump() assertion error for resource consts with --enable...
andrewnester [Tue, 24 Jan 2017 13:38:30 +0000 (16:38 +0300)]
Fixed #73973 - debug_zval_dump() assertion error for resource consts with --enable-debug

8 years ago[ci skip] news entry for PR #2337
Joe Watkins [Thu, 26 Jan 2017 05:50:46 +0000 (05:50 +0000)]
[ci skip] news entry for PR #2337

8 years agoFix bug #54382 (getAttributeNodeNS doesn't get xmlns* attributes)
Arnout Boks [Wed, 25 Jan 2017 20:09:03 +0000 (21:09 +0100)]
Fix bug #54382 (getAttributeNodeNS doesn't get xmlns* attributes)

The fix is based on the same strategy for handling namespace
declarations as used by getAttributeNode. Note that this strategy makes
these methods not return a DOMAttr for xmlns* attributes, but an
instance of the (undocumented) class DOMNameSpaceNode. This is not
really ideal, but at least this fix makes the behavior of
getAttributeNode and getAttributeNodeNS consistent.

A follow-up action would be to investigate whether DOMNameSpaceNode can
be made into a subclass of DOMAttr (which may be hard due to the way
libxml treats namespace declarations) or document this deviating return
value for xmlns* attributes.

8 years agoAdded (failing) testcase for bug #54382
Arnout Boks [Wed, 25 Jan 2017 20:03:50 +0000 (21:03 +0100)]
Added (failing) testcase for bug #54382

8 years agoFix copying too much leading to potential segfault
Bob Weinand [Wed, 25 Jan 2017 12:00:48 +0000 (13:00 +0100)]
Fix copying too much leading to potential segfault

8 years agoFix Bug #69865 php-fpm does not close stderr when using syslog
Michael Wallner [Wed, 22 Jul 2015 11:44:21 +0000 (13:44 +0200)]
Fix Bug #69865 php-fpm does not close stderr when using syslog

8 years agomove decl to correct place
Joe Watkins [Mon, 23 Jan 2017 05:33:58 +0000 (05:33 +0000)]
move decl to correct place

8 years agoFixed #73969 - Fixed segmentation fault when debug_print_backtrace called
andrewnester [Sun, 22 Jan 2017 15:59:59 +0000 (18:59 +0300)]
Fixed #73969 - Fixed segmentation fault when debug_print_backtrace called

8 years agoAdd regression test for bug #47946
Christoph M. Becker [Sun, 22 Jan 2017 14:01:01 +0000 (15:01 +0100)]
Add regression test for bug #47946

This bug had already been fixed, but apparently there's no regression
test yet, so we add one.

Note that the expected image has black pixel artifacts, which are
another issue (perhaps bug #40158), and would have to be adressed
separately.

8 years agoFix intermittent segfault in GD library
Mitch Hagstrand [Sat, 21 Jan 2017 20:40:15 +0000 (12:40 -0800)]
Fix intermittent segfault in GD library

The gdIOCtx struct should be zero filled with ecalloc.
emalloc does not zero fill the struct.

8 years agoAdd myself as primary maintainer of GD
Christoph M. Becker [Sat, 21 Jan 2017 17:52:34 +0000 (18:52 +0100)]
Add myself as primary maintainer of GD

8 years agoFix #73968: Premature failing of XBM reading
Christoph M. Becker [Sat, 21 Jan 2017 00:21:03 +0000 (01:21 +0100)]
Fix #73968: Premature failing of XBM reading

We must take into account the line padding, when we're reading XBM
files.

We deliberately ignore the potential integer overflow here, because
that would be caught by gdImageCreate() or even earlier if `bytes==0`,
what happens in libgd00094.phpt which we adapt accordingly.

8 years agobackport 51e1da6ea1e663908302e162ced1b7a8cb5aee05 into 7.0
Anatol Belski [Sat, 21 Jan 2017 01:56:15 +0000 (02:56 +0100)]
backport 51e1da6ea1e663908302e162ced1b7a8cb5aee05 into 7.0

8 years agoFixed #73959 - lastInsertId fails to throw an exception in pdsql
andrewnester [Fri, 20 Jan 2017 18:39:45 +0000 (21:39 +0300)]
Fixed #73959 - lastInsertId fails to throw an exception in pdsql

8 years agoNEWS
Remi Collet [Fri, 20 Jan 2017 16:46:11 +0000 (17:46 +0100)]
NEWS

8 years agoFixed bug #73965 DTrace reported as enabled when disabled
Remi Collet [Fri, 20 Jan 2017 16:45:15 +0000 (17:45 +0100)]
Fixed bug #73965 DTrace reported as enabled when disabled

8 years agoFixed another part in ldap (bug #73933)
Xinchen Hui [Thu, 19 Jan 2017 04:23:23 +0000 (12:23 +0800)]
Fixed another part in ldap (bug #73933)

8 years agoMerge branch 'PHP-5.6' into PHP-7.0
Ferenc Kovacs [Thu, 19 Jan 2017 00:20:57 +0000 (01:20 +0100)]
Merge branch 'PHP-5.6' into PHP-7.0

8 years agoupdate NEWS
Ferenc Kovacs [Thu, 19 Jan 2017 00:16:31 +0000 (01:16 +0100)]
update NEWS

8 years agoRemoving zmemcheck from .gdbinit
Mitch Hagstrand [Tue, 17 Jan 2017 20:39:07 +0000 (12:39 -0800)]
Removing zmemcheck from .gdbinit

It doesn't seem to make sense anymore

8 years agoFix bug #73949 leak in mysqli_fetch_object
Joe Watkins [Wed, 18 Jan 2017 18:22:57 +0000 (18:22 +0000)]
Fix bug #73949 leak in mysqli_fetch_object

8 years agoRemove remnants of TSMRLS_* macro calls
Thomas Punt [Wed, 18 Jan 2017 09:48:23 +0000 (09:48 +0000)]
Remove remnants of TSMRLS_* macro calls

8 years agoDNS CAA record type implementation and tests for https://bugs.php.net/bug.php?id...
Marcus Bointon [Tue, 3 Jan 2017 22:32:22 +0000 (23:32 +0100)]
DNS CAA record type implementation and tests for https://bugs.php.net/bug.php?id=73850

8 years agoChange free to OPENSSL_free
Leigh [Wed, 18 Jan 2017 14:45:41 +0000 (14:45 +0000)]
Change free to OPENSSL_free

8 years agoNEWS
Remi Collet [Wed, 18 Jan 2017 12:02:03 +0000 (13:02 +0100)]
NEWS

8 years agoFix #73956 Link use CC instead of CXX
Remi Collet [Wed, 18 Jan 2017 12:00:49 +0000 (13:00 +0100)]
Fix #73956 Link use CC instead of CXX

8 years agoFix ce comparison
Thomas Punt [Tue, 17 Jan 2017 18:10:33 +0000 (18:10 +0000)]
Fix ce comparison

8 years agoupdates to release docs
Joe Watkins [Tue, 17 Jan 2017 20:33:25 +0000 (20:33 +0000)]
updates to release docs

8 years agoFixes in .gdbinit for ZE 3
Mitch Hagstrand [Tue, 17 Jan 2017 05:55:03 +0000 (21:55 -0800)]
Fixes in .gdbinit for ZE 3

1. Fixed print_pi function
2. Fixed printing properties in ____printzv_contents
3. Added optional max length parameter in ____print_str

8 years ago[ci skip] sync NEWS
Anatol Belski [Tue, 17 Jan 2017 09:54:53 +0000 (10:54 +0100)]
[ci skip] sync NEWS

8 years agoMerge branch 'PHP-5.6' into PHP-7.0
Anatol Belski [Tue, 17 Jan 2017 08:33:51 +0000 (09:33 +0100)]
Merge branch 'PHP-5.6' into PHP-7.0

* PHP-5.6:
  Fix #73869: Signed Integer Overflow gd_io.c
  Fix #73868: DOS vulnerability in gdImageCreateFromGd2Ctx()

8 years agoFix #73869: Signed Integer Overflow gd_io.c
Christoph M. Becker [Sat, 17 Dec 2016 16:06:58 +0000 (17:06 +0100)]
Fix #73869: Signed Integer Overflow gd_io.c

GD2 stores the number of horizontal and vertical chunks as words (i.e. 2
byte unsigned). These values are multiplied and assigned to an int when
reading the image, what can cause integer overflows. We have to avoid
that, and also make sure that either chunk count is actually greater
than zero. If illegal chunk counts are detected, we bail out from
reading the image.

(cherry picked from commit 5b5d9db3988b829e0b121b74bb3947f01c2796a1)

8 years agoFix #73868: DOS vulnerability in gdImageCreateFromGd2Ctx()
Christoph M. Becker [Tue, 16 Aug 2016 16:23:36 +0000 (18:23 +0200)]
Fix #73868: DOS vulnerability in gdImageCreateFromGd2Ctx()

We must not pretend that there are image data if there are none. Instead
we fail reading the image file gracefully.

(cherry picked from commit cdb648dc4115ce0722f3cc75e6a65115fc0e56ab)

8 years agoFixed bug #73933 (error/segfault with ldap_mod_replace and opcache)
Xinchen Hui [Tue, 17 Jan 2017 07:34:19 +0000 (15:34 +0800)]
Fixed bug #73933 (error/segfault with ldap_mod_replace and opcache)

8 years agoMerge branch 'PHP-5.6' into PHP-7.0
Nikita Popov [Mon, 16 Jan 2017 13:11:41 +0000 (14:11 +0100)]
Merge branch 'PHP-5.6' into PHP-7.0

8 years agoAdd additional serialize tests for fixed bugs
Nikita Popov [Mon, 16 Jan 2017 12:24:13 +0000 (13:24 +0100)]
Add additional serialize tests for fixed bugs

These have been fixed as a side-effect of the delayed __wakeup
patch.

8 years agoFix typo
Stanislav Malyshev [Mon, 16 Jan 2017 08:23:06 +0000 (00:23 -0800)]
Fix typo

8 years agoAdd test ini_set with invalid time zone
Niklas Keller [Sun, 15 Jan 2017 22:06:08 +0000 (23:06 +0100)]
Add test ini_set with invalid time zone

Wasn't covered before according to gcov.php.net.

8 years agoMerge branch 'PHP-5.6' into PHP-7.0
Stanislav Malyshev [Mon, 16 Jan 2017 02:42:33 +0000 (18:42 -0800)]
Merge branch 'PHP-5.6' into PHP-7.0

* PHP-5.6:
  Fix test

8 years agoFix test
Stanislav Malyshev [Mon, 16 Jan 2017 02:42:22 +0000 (18:42 -0800)]
Fix test

8 years agoMerge branch 'PHP-5.6' into PHP-7.0
Stanislav Malyshev [Mon, 16 Jan 2017 01:32:26 +0000 (17:32 -0800)]
Merge branch 'PHP-5.6' into PHP-7.0

* PHP-5.6:
  Update more functions with path check

8 years agoUpdate more functions with path check
Stanislav Malyshev [Mon, 16 Jan 2017 01:31:08 +0000 (17:31 -0800)]
Update more functions with path check

8 years agofix AppVeyor failures with Opcache enabled
Anatol Belski [Sun, 15 Jan 2017 20:13:15 +0000 (21:13 +0100)]
fix AppVeyor failures with Opcache enabled

opcache_reset() only schedules the restart. Under circumstances,
the follow up requests might run uncached, until the restart
condition is met. To mitigate the false positives caused by this
behavior, any tests using opcache_reset() should not be put in
between other tests. Thus, moving the corresponding test to be
executed last.

8 years agoFixed bug #69993
Yordan Gigov [Fri, 12 Aug 2016 14:52:18 +0000 (17:52 +0300)]
Fixed bug #69993

8 years agoAdded test for "opcache: print restart reason" to basic_logging.phpt
Mitch Hagstrand [Sat, 14 Jan 2017 00:33:05 +0000 (16:33 -0800)]
Added test for "opcache: print restart reason" to basic_logging.phpt

8 years agoopcache: print restart reason
Alexander Polyakov [Fri, 13 Jan 2017 11:55:10 +0000 (14:55 +0300)]
opcache: print restart reason

8 years agofix remaining tests for Opcache runs
Anatol Belski [Fri, 13 Jan 2017 16:16:15 +0000 (17:16 +0100)]
fix remaining tests for Opcache runs

The fail reason here is the TMP change while both top and test
run same binary with opcache enabled.

8 years agoensure the end path wouldn't contain double slashes
Anatol Belski [Fri, 13 Jan 2017 15:50:05 +0000 (16:50 +0100)]
ensure the end path wouldn't contain double slashes

8 years agotemplatize the resource number
Anatol Belski [Fri, 13 Jan 2017 14:45:35 +0000 (15:45 +0100)]
templatize the resource number

8 years agoIncrease wait time for tests involving sockets to improve Travis relability
Mitch Hagstrand [Fri, 13 Jan 2017 06:09:46 +0000 (22:09 -0800)]
Increase wait time for tests involving sockets to improve Travis relability

1. Increased sleep time from 25ms to 50ms
2. Added sleep to shutdown handlers

8 years agoFixed bug #73916 (zend_print_flat_zval_r doesn't consider reference)
Xinchen Hui [Fri, 13 Jan 2017 05:04:37 +0000 (13:04 +0800)]
Fixed bug #73916 (zend_print_flat_zval_r doesn't consider reference)

8 years agoFix bug #69899
rfussenegger [Fri, 30 Sep 2016 08:49:08 +0000 (10:49 +0200)]
Fix bug #69899

8 years agoFixed bug #72680 gdbinit print_cvs not prepared for PHP 7
Mitch Hagstrand [Thu, 12 Jan 2017 04:57:33 +0000 (20:57 -0800)]
Fixed bug #72680 gdbinit print_cvs not prepared for PHP 7

8 years agoadded a note to opache.fast_shutdown why its not enabled by default in php.ini
Markus Staab [Wed, 11 Jan 2017 10:51:25 +0000 (11:51 +0100)]
added a note to opache.fast_shutdown why its not enabled by default in php.ini

8 years agoMerge branch 'PHP-7.0' of git.php.net:/php-src into PHP-7.0
Joe Watkins [Wed, 11 Jan 2017 08:12:51 +0000 (08:12 +0000)]
Merge branch 'PHP-7.0' of git.php.net:/php-src into PHP-7.0

* 'PHP-7.0' of git.php.net:/php-src:
  Fixed bug #67707 IV not needed for ECB encryption mode, but it returns a warning
  Fixed #73907 (nextSibling property not included in var_dump of DOMNode)
  Fixed bug #61858 (DOMAttr debug info generates E_WARNING)
  Fix glob-wrapper.phpt to not fail in Windows
  Fix #73893: A hidden danger of death cycle in a function of gd

8 years agoFixed bug #71219 configure script incorrectly checks for ttyname_r
Joe Watkins [Wed, 11 Jan 2017 08:09:10 +0000 (08:09 +0000)]
Fixed bug #71219 configure script incorrectly checks for ttyname_r

8 years agoFixed bug #67707 IV not needed for ECB encryption mode, but it returns a warning
Robrecht Plaisier [Fri, 14 Aug 2015 19:16:09 +0000 (19:16 +0000)]
Fixed bug #67707 IV not needed for ECB encryption mode, but it returns a warning

8 years agoFixed #73907 (nextSibling property not included in var_dump of DOMNode)
Arnout Boks [Tue, 10 Jan 2017 19:48:45 +0000 (20:48 +0100)]
Fixed #73907 (nextSibling property not included in var_dump of DOMNode)

8 years agoFixed bug #61858 (DOMAttr debug info generates E_WARNING)
Arnout Boks [Tue, 10 Jan 2017 20:19:37 +0000 (21:19 +0100)]
Fixed bug #61858 (DOMAttr debug info generates E_WARNING)

It seems fair to remove this warning, given that:

* it is not documented in the official documentation
* the $specified property, which has a similar 'not implemented' status,
  also does not trigger a warning
* it apparently hinders quite a lot of people during debugging, judging by
  the number of votes on the bug

8 years agoMerge branch 'PHP-5.6' into PHP-7.0
Nikita Popov [Tue, 10 Jan 2017 22:19:19 +0000 (23:19 +0100)]
Merge branch 'PHP-5.6' into PHP-7.0

8 years agoFix glob-wrapper.phpt to not fail in Windows
Mitch Hagstrand [Tue, 10 Jan 2017 19:51:55 +0000 (11:51 -0800)]
Fix glob-wrapper.phpt to not fail in Windows

8 years agoFix #73893: A hidden danger of death cycle in a function of gd
Christoph M. Becker [Tue, 10 Jan 2017 18:31:33 +0000 (19:31 +0100)]
Fix #73893: A hidden danger of death cycle in a function of gd

We remove the unused, but potentially dangerous functions.

8 years agoMerge branch 'pull-request/1807' into PHP-7.0
Joe Watkins [Tue, 10 Jan 2017 13:56:26 +0000 (13:56 +0000)]
Merge branch 'pull-request/1807' into PHP-7.0

* pull-request/1807:
  test for dns_get_record

8 years agopull-request/1807:
marcosptf [Sun, 13 Mar 2016 04:29:04 +0000 (01:29 -0300)]
pull-request/1807:
test for dns_get_record

8 years agoPrevent fd leak in readdir_r that causes false negative on readdir_r implementation...
Ondřej Surý [Tue, 10 Jan 2017 09:21:54 +0000 (10:21 +0100)]
Prevent fd leak in readdir_r that causes false negative on readdir_r implementation when compiled with AddressSanitizer

8 years agoFixed #73902 mention router support in usage output for php-cli
Joe Watkins [Tue, 10 Jan 2017 06:19:34 +0000 (06:19 +0000)]
Fixed #73902 mention router support in usage output for php-cli

8 years agoMerge branch 'PHP-5.6' into PHP-7.0
Sara Golemon [Tue, 10 Jan 2017 00:07:33 +0000 (16:07 -0800)]
Merge branch 'PHP-5.6' into PHP-7.0

* PHP-5.6:
  Fix open_basedir check for glob:// opendir wrapper

8 years agoUn-XFAIL serialization test
Nikita Popov [Mon, 9 Jan 2017 23:07:38 +0000 (00:07 +0100)]
Un-XFAIL serialization test

8 years agoFix check for linux getrandom syscall
Leigh [Wed, 23 Dec 2015 21:54:15 +0000 (21:54 +0000)]
Fix check for linux getrandom syscall

8 years agoAdd print zend string function to gdbinit
Mitch Hagstrand [Mon, 9 Jan 2017 21:10:57 +0000 (13:10 -0800)]
Add print zend string function to gdbinit

8 years agoFix open_basedir check for glob:// opendir wrapper
Sara Golemon [Mon, 9 Jan 2017 19:02:50 +0000 (11:02 -0800)]
Fix open_basedir check for glob:// opendir wrapper

php_check_open_basedir() expects a local filesystem path,
but we're handing it a `glob://...` URI instead.

Move the check to after the path trim so that we're checking
a meaningful pathspec.

8 years agoRevert "set more suitable defaults for opcache in ini and code"
Joe Watkins [Mon, 9 Jan 2017 17:08:07 +0000 (17:08 +0000)]
Revert "set more suitable defaults for opcache in ini and code"

This reverts commit b655f2d6325c64e5da1f016a2742a6abf669ecaf.

This has been deemed unsuitable for 7.0

8 years agoupdate release process notes
Joe Watkins [Mon, 9 Jan 2017 16:20:48 +0000 (16:20 +0000)]
update release process notes

8 years agoMerge branch 'PHP-7.0' of git.php.net:/php-src into PHP-7.0
Joe Watkins [Mon, 9 Jan 2017 15:59:21 +0000 (15:59 +0000)]
Merge branch 'PHP-7.0' of git.php.net:/php-src into PHP-7.0

* 'PHP-7.0' of git.php.net:/php-src:
  Fixed bug #73896 (spl_autoload() crashes when calls magic _call())

8 years agofix stub to accept user properly
Joe Watkins [Mon, 9 Jan 2017 15:57:46 +0000 (15:57 +0000)]
fix stub to accept user properly

8 years agoFixed bug #73896 (spl_autoload() crashes when calls magic _call())
Dmitry Stogov [Mon, 9 Jan 2017 12:11:33 +0000 (15:11 +0300)]
Fixed bug #73896 (spl_autoload() crashes when calls magic _call())

8 years agoMerge branch 'pull-request/2291' into PHP-7.0
Joe Watkins [Mon, 9 Jan 2017 10:54:57 +0000 (10:54 +0000)]
Merge branch 'pull-request/2291' into PHP-7.0

* pull-request/2291:
  set more suitable defaults for opcache in ini and code

8 years agoset more suitable defaults for opcache in ini and code
Markus Staab [Mon, 9 Jan 2017 10:33:05 +0000 (11:33 +0100)]
set more suitable defaults for opcache in ini and code

8 years agoFixed bug #73854 (Always true condition in php_openssl_sockop_io)
Xinchen Hui [Mon, 9 Jan 2017 08:30:00 +0000 (16:30 +0800)]
Fixed bug #73854 (Always true condition in php_openssl_sockop_io)

8 years agoFixed bug #72979 money_format stores wrong length on AIX
Joe Watkins [Mon, 9 Jan 2017 05:31:41 +0000 (05:31 +0000)]
Fixed bug #72979 money_format stores wrong length on AIX

8 years agoupdate ip2long_varation2 test to skip for AIX, see bug #72973
Joe Watkins [Mon, 9 Jan 2017 05:24:44 +0000 (05:24 +0000)]
update ip2long_varation2 test to skip for AIX, see bug #72973

8 years agoFixed #72974 imap is undefined service on AIX
Joe Watkins [Mon, 9 Jan 2017 05:13:46 +0000 (05:13 +0000)]
Fixed #72974 imap is undefined service on AIX

8 years agoFix FTP tests for Travis CI.
Mitch Hagstrand [Sun, 8 Jan 2017 21:25:46 +0000 (13:25 -0800)]
Fix FTP tests for Travis CI.

This suppresses the errors from stream_socket_server() until
server.inc will not make anymore attempts.

8 years agoUse proper file extension for the README
Markus Staab [Sun, 8 Jan 2017 21:02:50 +0000 (22:02 +0100)]
Use proper file extension for the README

8 years agoFix 404 links in Litespeed README - Bug 73890
Peter Kokot [Sun, 8 Jan 2017 20:24:25 +0000 (21:24 +0100)]
Fix 404 links in Litespeed README - Bug 73890

8 years agoupdate NEWS
Anatol Belski [Sun, 8 Jan 2017 19:07:16 +0000 (20:07 +0100)]
update NEWS

8 years agoimprove FPM tests (mostly for systemd build)
Remi Collet [Sun, 8 Jan 2017 18:55:30 +0000 (19:55 +0100)]
improve FPM tests (mostly for systemd build)

8 years ago[ast] Fix exporting **= in expansion of assign op
Sara Golemon [Sat, 6 Aug 2016 22:12:09 +0000 (15:12 -0700)]
[ast] Fix exporting **= in expansion of assign op

(cherry picked from commit 9c3865eb6a72a6f369ba524c5a87b4896a3a9acb)

8 years agodeflate_add(): Don't add unnecessary +1 to size
Nikita Popov [Sun, 8 Jan 2017 12:47:20 +0000 (13:47 +0100)]
deflate_add(): Don't add unnecessary +1 to size

zend_string_realloc will do this by itself.

8 years agoFix OOB read introduced by previous commit
Nikita Popov [Sat, 7 Jan 2017 23:01:05 +0000 (00:01 +0100)]
Fix OOB read introduced by previous commit

Ooops...

8 years agohttp_fopen_wrapper.c - Handle HTTP headers with varying white space
Rowan Collins [Wed, 30 Mar 2016 22:12:03 +0000 (22:12 +0000)]
http_fopen_wrapper.c - Handle HTTP headers with varying white space

The stream handler assumed all HTTP headers contained exactly one space,
but the standard says there may be zero or more. Should fix Bug #47021,
and any other edge cases caused by a web server sending unusual spacing,
e.g. the MIME type discovered from Content-Type: can no longer contain
leading whitespace.

We strip trailing whitespace from the headers added into
$http_response_header as well.

8 years agoFixed bug #67583
Nikita Popov [Sat, 7 Jan 2017 21:51:18 +0000 (22:51 +0100)]
Fixed bug #67583

As fcgi_request is an opaque struct as of PHP 7, expose a new API
function fcgi_end() which does fcgi_flush() with end=1 and checks/
sets the ->ended flag.

8 years agoRename fcgi_request.closed to .ended
Nikita Popov [Sat, 7 Jan 2017 21:42:45 +0000 (22:42 +0100)]
Rename fcgi_request.closed to .ended

"closed" refers to whether FCGI_END_REQUEST has been sent, while
the "close" operation does something entirely different. It gets
extra confusing when fcgi_is_closed() does not actually return
fcgi_request.closed...

8 years agoMerge branch '1391' into PHP-7.0
Joe Watkins [Sat, 7 Jan 2017 16:45:33 +0000 (16:45 +0000)]
Merge branch '1391' into PHP-7.0

* 1391:
  Adapts the readme to the new test-possibility

8 years agoAdapts the readme to the new test-possibility
Andreas Heigl [Mon, 6 Jul 2015 08:54:07 +0000 (10:54 +0200)]
Adapts the readme to the new test-possibility

This commit introduces a way to use a vagrant setup to test against a
known LDAP without the need to setup an LDAP on your own machine

8 years agoImplement github PR #1736
Anatol Belski [Sat, 7 Jan 2017 14:36:14 +0000 (15:36 +0100)]
Implement github PR #1736

This fixes leak issues on *BSD systems, as described in the PR.

8 years agofix wrong version
Anatol Belski [Sat, 7 Jan 2017 11:43:56 +0000 (12:43 +0100)]
fix wrong version

8 years agoextend UPGRADING
Anatol Belski [Sat, 7 Jan 2017 11:43:19 +0000 (12:43 +0100)]
extend UPGRADING