]> granicus.if.org Git - php/log
php
7 years agoFixed bug #74625 (Integer overflow in oci_bind_array_by_name).
Ingmar Runge [Mon, 22 May 2017 06:58:18 +0000 (08:58 +0200)]
Fixed bug #74625 (Integer overflow in oci_bind_array_by_name).

7 years agoRemove use of obsolete create_function()
Christopher Jones [Wed, 21 Jun 2017 05:02:58 +0000 (15:02 +1000)]
Remove use of obsolete create_function()

7 years agoUpdate OCI8 version test and package.xml
Christopher Jones [Wed, 21 Jun 2017 04:29:30 +0000 (14:29 +1000)]
Update OCI8 version test and package.xml

7 years agoFix bug #73634 - grapheme_strpos illegal memory access
Stanislav Malyshev [Mon, 16 Jan 2017 09:18:16 +0000 (01:18 -0800)]
Fix bug #73634 - grapheme_strpos illegal memory access

7 years agoFix bug #73648 - integer overflow in substr
Stanislav Malyshev [Mon, 16 Jan 2017 20:11:50 +0000 (12:11 -0800)]
Fix bug #73648 - integer overflow in substr

7 years agoSplit requested size as expected by CreateFileMapping.
Anatol Belski [Tue, 20 Jun 2017 13:38:25 +0000 (15:38 +0200)]
Split requested size as expected by CreateFileMapping.

Thus fixing ext\opcache\tests\log_verbosity_bug.phpt fails on 64-bit.

7 years agomove next
Anatol Belski [Tue, 20 Jun 2017 12:05:50 +0000 (14:05 +0200)]
move next

7 years agoPDO_OCI: support IC 12.2 in configure option
Christopher Jones [Tue, 20 Jun 2017 06:00:22 +0000 (16:00 +1000)]
PDO_OCI: support IC 12.2 in configure option

7 years agoBump OCI8 version
Christopher Jones [Tue, 20 Jun 2017 04:54:45 +0000 (14:54 +1000)]
Bump OCI8 version

7 years agoMerge branch 'pull-request/1569' into PHP-7.0
Christopher Jones [Tue, 20 Jun 2017 04:43:50 +0000 (14:43 +1000)]
Merge branch 'pull-request/1569' into PHP-7.0

* pull-request/1569:
  add test for bug #70700

7 years agoadd test for bug #70700
Chuck Burgess [Mon, 12 Oct 2015 19:56:59 +0000 (14:56 -0500)]
add test for bug #70700

7 years agoUpdate NEWS
Christopher Jones [Tue, 20 Jun 2017 04:37:16 +0000 (14:37 +1000)]
Update NEWS

7 years agoMerge branch 'pull-request/2459' into PHP-7.0
Christopher Jones [Tue, 20 Jun 2017 04:33:06 +0000 (14:33 +1000)]
Merge branch 'pull-request/2459' into PHP-7.0

* pull-request/2459:
  oci8 - Implementation of Oracle TAF Callback

7 years agooci8 - Implementation of Oracle TAF Callback
KoenigsKind [Fri, 7 Apr 2017 18:40:40 +0000 (20:40 +0200)]
oci8 - Implementation of Oracle TAF Callback

Adds support for the Transparent Application Failover Callback.
The php_oci_connection struct got a char* added which will contain the
callback function, it should be set to PHP_OCI_TAF_DISABLE_CALLBACK at
the end of a php request for permanent connections so that, if a TAF
callback occurs, no userspace function will be called.
Maybe add support for registering object functions (via array),
currently the register function only accepts a string. I didn't know how
to implement it correctly. As a failover occurs very rarely it might be
better to not keep the cache when saving the zend_fcall_info.

Things to do
[ ] config.m4 needs to compile oci8_failover.c
[ ] Check if correctly implemented (especially for multithreading)
[ ] Add support for registering callback function via array

7 years agoMerge branch 'pull-request/1566' into PHP-7.0
Christopher Jones [Tue, 20 Jun 2017 04:23:20 +0000 (14:23 +1000)]
Merge branch 'pull-request/1566' into PHP-7.0

* pull-request/1566:
  add test for bug #60994

7 years agoadd test for bug #60994
Chuck Burgess [Mon, 12 Oct 2015 15:50:09 +0000 (10:50 -0500)]
add test for bug #60994

7 years agoopenssl: Add openssl_pkcs12_(read, export) tests
Jelle van der Waa [Thu, 8 Jun 2017 12:31:48 +0000 (14:31 +0200)]
openssl: Add openssl_pkcs12_(read, export) tests

Add basic tests for openssl_pkcs12_read and openssl_pkcs12_export.
openssl_pkcs12_export has two tests disable which currently do not work
as intended and should be fixed.

7 years agoRegenerate and bump re2c version to 0.16
Jakub Zelenka [Sun, 18 Jun 2017 17:26:48 +0000 (18:26 +0100)]
Regenerate and bump re2c version to 0.16

It also matches the re2c version for Zend language scanner

7 years agoNext round on AppVeyor reliability.
Anatol Belski [Fri, 16 Jun 2017 23:57:47 +0000 (01:57 +0200)]
Next round on AppVeyor reliability.

7 years agoopenssl: Fix spkstr and spki leak in openssl_spki_new
Jelle van der Waa [Tue, 6 Jun 2017 14:45:39 +0000 (16:45 +0200)]
openssl: Fix spkstr and spki leak in openssl_spki_new

spkstr is not free'd when running the test and valgrind reports a leak
of 32,318 bytes. The free condition is not met since keyresource is not
NULL, apart from that OPENSSL_free actually free's the allocated memory
by OPENSSL_malloc inside OpenSSL.

Valgrind reports 768 bytes leaked in openssl_spki_new when running the
testsuite. Remove the keyresource check to always free spki.

7 years agoFixed bug #74679 (Incorrect conversion array with WSDL_CACHE_MEMORY)
Dmitry Stogov [Tue, 13 Jun 2017 14:15:47 +0000 (17:15 +0300)]
Fixed bug #74679 (Incorrect conversion array with WSDL_CACHE_MEMORY)

7 years agoNEWS
Remi Collet [Tue, 13 Jun 2017 06:50:02 +0000 (08:50 +0200)]
NEWS

7 years agoFix bug #74705 for collator_get_sort_key
Remi Collet [Tue, 13 Jun 2017 06:48:17 +0000 (08:48 +0200)]
Fix bug #74705 for collator_get_sort_key

7 years agoFixes bug #74705 Wrong ReflectionInfo for Collator::getSortKey()
Tyson Andre [Thu, 8 Jun 2017 04:32:57 +0000 (21:32 -0700)]
Fixes bug #74705 Wrong ReflectionInfo for Collator::getSortKey()

https://secure.php.net/manual/en/collator.getsortkey.php

7 years agoNEWS
Remi Collet [Tue, 13 Jun 2017 06:14:34 +0000 (08:14 +0200)]
NEWS

7 years agoFixes bug #74708 reflection signatures for random_bytes+random_int
Tyson Andre [Thu, 8 Jun 2017 06:50:20 +0000 (23:50 -0700)]
Fixes bug #74708 reflection signatures for random_bytes+random_int

They have 1 and 2 required parameters, respectively
See https://secure.php.net/manual/en/function.random-int.php

7 years agoForgotten NEWs update
Xinchen Hui [Tue, 13 Jun 2017 04:20:42 +0000 (12:20 +0800)]
Forgotten NEWs update

7 years agoMerge branch 'bug74600' of https://github.com/manuelm/php-src into PHP-7.0
Xinchen Hui [Mon, 12 Jun 2017 09:12:15 +0000 (17:12 +0800)]
Merge branch 'bug74600' of https://github.com/manuelm/php-src into PHP-7.0

* 'bug74600' of https://github.com/manuelm/php-src:
  Add simple cli test for PATH/HOST ini sections
  Fixed bug #74600

7 years agoAdd simple cli test for PATH/HOST ini sections
Manuel Mausz [Fri, 9 Jun 2017 21:28:19 +0000 (23:28 +0200)]
Add simple cli test for PATH/HOST ini sections

7 years agoUse reasonable SQLite3 module version
Christoph M. Becker [Wed, 7 Jun 2017 13:22:20 +0000 (15:22 +0200)]
Use reasonable SQLite3 module version

Since PECL/sqlite3 has been moved to ext/sqlite3 in 2008, the SQLite3
module version is hardcoded to `0.7-dev`. It doesn't appear to make
much sense to have an own module version for a bundled extensions, but
as other code might rely on it, we don't remove the constant but rather
make it an alias of the PHP version.

7 years agoFixed bug #74600
Manuel Mausz [Thu, 8 Jun 2017 23:43:08 +0000 (01:43 +0200)]
Fixed bug #74600

Make sure the hash entry is an array.

The origin fix broke support for HOST/PATH ini sections. Only the
beginning of the string has to match. Revert this check but use
zend_binary_strncasecmp instead of strncasecmp.

7 years agoenforce exit status propagation and up the sdk version
Anatol Belski [Thu, 8 Jun 2017 15:14:52 +0000 (17:14 +0200)]
enforce exit status propagation and up the sdk version

7 years agoReplace ASN1_STRING_data with ASN1_STRING_get0_data
Jakub Zelenka [Sun, 4 Jun 2017 17:08:51 +0000 (18:08 +0100)]
Replace ASN1_STRING_data with ASN1_STRING_get0_data

This is a slightly modified version of the patch from Jelle van der Waa
( @jelly ) so full credit to him.

7 years agoFix leak in WDDX serialization
Nikita Popov [Sat, 3 Jun 2017 13:08:42 +0000 (15:08 +0200)]
Fix leak in WDDX serialization

7 years agoFixed bug #69373
Tim Toohey [Fri, 2 Jun 2017 22:38:02 +0000 (00:38 +0200)]
Fixed bug #69373

xmlNodeSetContentLen() calls xmlFreeNode() on node->children. This
causes problems if there are other references around to those children.

7 years agoFixed bug #73473: Stack Buffer Overflow in msgfmt_parse_message
libnex [Mon, 29 May 2017 13:13:52 +0000 (13:13 +0000)]
Fixed bug #73473: Stack Buffer Overflow in msgfmt_parse_message

7 years agoFixed bug #74673 (Segfault when cast Reflection object to string with undefined constant)
Xinchen Hui [Wed, 31 May 2017 04:39:26 +0000 (12:39 +0800)]
Fixed bug #74673 (Segfault when cast Reflection object to string with undefined constant)

7 years agoFixed wrong condition
Dmitry Stogov [Tue, 30 May 2017 21:40:33 +0000 (00:40 +0300)]
Fixed wrong condition

7 years agoMerge branch 'pull-request/2536' into PHP-7.0
Côme Bernigaud [Tue, 30 May 2017 14:45:27 +0000 (16:45 +0200)]
Merge branch 'pull-request/2536' into PHP-7.0

* pull-request/2536:
  ext/ldap/test: Test that ldap_connect() uses defaults from ldap.conf (openldap)
  ext/ldap: Allow default host from ldap.conf to work.

7 years agoNEWS
Remi Collet [Tue, 30 May 2017 13:45:09 +0000 (15:45 +0200)]
NEWS

7 years agoPatch from the upstream git
Remi Collet [Tue, 30 May 2017 13:40:32 +0000 (15:40 +0200)]
Patch from the upstream git
https://github.com/kkos/oniguruma/issues/60 (CVE-2017-9228)

Thanks to Mamoru TASAKA <mtasaka@fedoraproject.org>

7 years agoPatch from the upstream git
Remi Collet [Tue, 30 May 2017 13:39:21 +0000 (15:39 +0200)]
Patch from the upstream git
https://github.com/kkos/oniguruma/issues/59 (CVE-2017-9229)
b690371bbf97794b4a1d3f295d4fb9a8b05d402d Modified for onig 5.9.6

Thanks to Mamoru TASAKA <mtasaka@fedoraproject.org>

7 years agoPatch from the upstream git
Remi Collet [Tue, 30 May 2017 13:38:17 +0000 (15:38 +0200)]
Patch from the upstream git
https://github.com/kkos/oniguruma/issues/58 (CVE-2017-9227)

Thanks to Mamoru TASAKA <mtasaka@fedoraproject.org>

7 years agoPatch from the upstream git
Remi Collet [Tue, 30 May 2017 13:37:11 +0000 (15:37 +0200)]
Patch from the upstream git
https://github.com/kkos/oniguruma/issues/57 (CVE-2017-9224)

Thanks to Mamoru TASAKA <mtasaka@fedoraproject.org>

7 years agoPatch from the upstream git
Remi Collet [Tue, 30 May 2017 13:35:42 +0000 (15:35 +0200)]
Patch from the upstream git
https://github.com/kkos/oniguruma/issues/55 (CVE-2017-9226)
b4bf968ad52afe14e60a2dc8a95d3555c543353a Modified for onig 5.9.6
f015fbdd95f76438cd86366467bb2b39870dd7c6 Modified for onig 5.9.6

Thanks to Mamoru TASAKA <mtasaka@fedoraproject.org>

7 years agoFix Bug #74386 Phar::__construct(): wrong number of parameters by reflection
Fabien Villepinte [Wed, 10 May 2017 10:40:21 +0000 (12:40 +0200)]
Fix Bug #74386 Phar::__construct(): wrong number of parameters by reflection

7 years agoBugfix #74556 stream_socket_get_name() returns empty string
Sara Golemon [Sun, 28 May 2017 14:23:57 +0000 (07:23 -0700)]
Bugfix #74556 stream_socket_get_name() returns empty string

The original bug report had it returning '\0',
but with a fix to abstract name handling (6d2d0bbda7)
it now actually returns ''.

Neither of these are good, as per unix(7)
an empty socket name indicates an unbound name
and "should not be inspected".

7 years agoFix abstract name handling to be binary safe
Sara Golemon [Sun, 28 May 2017 15:20:21 +0000 (08:20 -0700)]
Fix abstract name handling to be binary safe

Per unix(7):
abstract: an abstract socket address is distinguished (from a
pathname socket) by the fact that sun_path[0] is a null byte
('\0').  The socket's address in this namespace is given by the
additional bytes in sun_path that are covered by the specified
length of the address structure.  (Null bytes in the name have no
special significance.)  The name has no connection with filesystem
pathnames.  When the address of an abstract socket is returned,
the returned addrlen is greater than sizeof(sa_family_t) (i.e.,
greater than 2), and the name of the socket is contained in the
first (addrlen - sizeof(sa_family_t)) bytes of sun_path.

The existing implementation was assuming significance in null bytes
contained in the abstract address identifier.

7 years agoBugfix #74598 ftp:// ignores context
Sara Golemon [Sun, 28 May 2017 04:12:49 +0000 (21:12 -0700)]
Bugfix #74598 ftp:// ignores context

7 years agoFixed bug #74663 (Segfault with opcache.memory_protect and validate_timestamp)
Xinchen Hui [Sat, 27 May 2017 14:42:17 +0000 (22:42 +0800)]
Fixed bug #74663 (Segfault with opcache.memory_protect and validate_timestamp)

7 years agoCorrect the location
Xinchen Hui [Sat, 27 May 2017 04:11:53 +0000 (12:11 +0800)]
Correct the location

7 years agoFixed bug #74657 (Undefined constants in array properties result in broken properties)
Xinchen Hui [Sat, 27 May 2017 04:06:43 +0000 (12:06 +0800)]
Fixed bug #74657 (Undefined constants in array properties result in broken properties)

7 years agoext/ldap/test: Test that ldap_connect() uses defaults from ldap.conf (openldap)
David Caldwell [Thu, 25 May 2017 21:27:11 +0000 (21:27 +0000)]
ext/ldap/test: Test that ldap_connect() uses defaults from ldap.conf (openldap)

7 years agoext/ldap: Allow default host from ldap.conf to work.
David Caldwell [Thu, 25 May 2017 00:41:12 +0000 (00:41 +0000)]
ext/ldap: Allow default host from ldap.conf to work.

This fixes an regression introduced in
e7af0fe1eb89e40671e86a588aa1b78607b85461. Previously, calling
ldap_connect() with no parameters would pass NULL to ldap_init(),
which causes it to use the default host specified in
/etc/ldap/ldap.conf (on Ubuntu).

When the code changed to use ldap_initialize(), it initialized a uri,
even if there were no parameters passed to ldap_connect(). Because of
this, there's no way to pass a NULL into ldap_initialize(), making it
impossible to use the default uri from ldap.conf.

This commit bypasses the uri creation when there is no host argument,
passing on a NULL to ldap_initialize() which restores the old PHP 5.5
behavior.

7 years agoFixed bug #74478
jhdxr [Wed, 24 May 2017 16:00:48 +0000 (18:00 +0200)]
Fixed bug #74478

7 years agotime to move next
Anatol Belski [Tue, 23 May 2017 10:31:03 +0000 (12:31 +0200)]
time to move next

7 years agomove to the next sdk version for AppVeyor
Anatol Belski [Sat, 20 May 2017 11:48:35 +0000 (13:48 +0200)]
move to the next sdk version for AppVeyor

7 years agofollow up fix on bug #74022
Anatol Belski [Thu, 18 May 2017 10:03:13 +0000 (12:03 +0200)]
follow up fix on bug #74022

7 years agoAdded cleanup
Xinchen Hui [Wed, 17 May 2017 15:24:58 +0000 (23:24 +0800)]
Added cleanup

7 years agoAdded test for bug #74600
Xinchen Hui [Wed, 17 May 2017 15:21:18 +0000 (23:21 +0800)]
Added test for bug #74600

7 years agoAdded test for bug #74596
Xinchen Hui [Wed, 17 May 2017 14:42:18 +0000 (22:42 +0800)]
Added test for bug #74596

7 years agoMerge branch 'PHP-7.0' of git.php.net:/php-src into PHP-7.0
Xinchen Hui [Tue, 16 May 2017 11:32:33 +0000 (19:32 +0800)]
Merge branch 'PHP-7.0' of git.php.net:/php-src into PHP-7.0

* 'PHP-7.0' of git.php.net:/php-src:
  variation of icu libraries causes failure

7 years agoFixed bug #74600 (crash (SIGSEGV) in _zend_hash_add_or_update_i)
Xinchen Hui [Tue, 16 May 2017 11:32:13 +0000 (19:32 +0800)]
Fixed bug #74600 (crash (SIGSEGV) in _zend_hash_add_or_update_i)

7 years agovariation of icu libraries causes failure
Steve Walk [Fri, 5 May 2017 18:45:27 +0000 (11:45 -0700)]
variation of icu libraries causes failure

7 years agoFixed bug #74596 (SIGSEGV with opcache.revalidate_path enabled)
Xinchen Hui [Tue, 16 May 2017 05:22:08 +0000 (13:22 +0800)]
Fixed bug #74596 (SIGSEGV with opcache.revalidate_path enabled)

Yeah, no test script is provided.. I got some troubles to make a one

7 years agoswitch to Postgres 9.6 for appveyor
Anatol Belski [Mon, 15 May 2017 15:47:41 +0000 (17:47 +0200)]
switch to Postgres 9.6 for appveyor

7 years agoRemove typo from test header
Sara Golemon [Wed, 10 May 2017 23:50:19 +0000 (16:50 -0700)]
Remove typo from test header

7 years agoFixed tests when using cumstom php.ini(session.save_handler)
Xinchen Hui [Tue, 9 May 2017 03:13:41 +0000 (11:13 +0800)]
Fixed tests when using cumstom php.ini(session.save_handler)

7 years agoMerge branch 'fix_74541' of https://github.com/villfa/php-src into PHP-7.0
Xinchen Hui [Tue, 9 May 2017 03:04:39 +0000 (11:04 +0800)]
Merge branch 'fix_74541' of https://github.com/villfa/php-src into PHP-7.0

* 'fix_74541' of https://github.com/villfa/php-src:
  Fix Bug #74541 Wrong reflection on session_start()

7 years agoMerge branch 'patch-1' of https://github.com/ifdattic/php-src into PHP-7.0
Xinchen Hui [Tue, 9 May 2017 03:01:33 +0000 (11:01 +0800)]
Merge branch 'patch-1' of https://github.com/ifdattic/php-src into PHP-7.0

* 'patch-1' of https://github.com/ifdattic/php-src:
  Fix typo: exectute => execute

7 years agoFix Bug #74541 Wrong reflection on session_start()
Fabien Villepinte [Mon, 8 May 2017 20:42:09 +0000 (22:42 +0200)]
Fix Bug #74541 Wrong reflection on session_start()

7 years agosync bin sdk version
Anatol Belski [Mon, 8 May 2017 17:38:48 +0000 (19:38 +0200)]
sync bin sdk version

7 years ago[ci skip] update libs versions file
Anatol Belski [Mon, 8 May 2017 11:09:15 +0000 (13:09 +0200)]
[ci skip] update libs versions file

7 years agoFix typo: exectute => execute
Andrew Marcinkevičius [Mon, 8 May 2017 10:59:05 +0000 (13:59 +0300)]
Fix typo: exectute => execute

7 years agoFixed another potential dangling pointer
Xinchen Hui [Mon, 8 May 2017 03:36:07 +0000 (11:36 +0800)]
Fixed another potential dangling pointer

7 years agoFixed bug #74546 (SIGILL in ZEND_FETCH_CLASS_CONSTANT_SPEC_CONST_CONST_HANDLER())
Xinchen Hui [Mon, 8 May 2017 03:32:08 +0000 (11:32 +0800)]
Fixed bug #74546 (SIGILL in ZEND_FETCH_CLASS_CONSTANT_SPEC_CONST_CONST_HANDLER())

7 years ago[ci skip] update NEWS
Anatol Belski [Sun, 7 May 2017 11:04:21 +0000 (13:04 +0200)]
[ci skip] update NEWS

7 years agoFixed bug #74547 mysqli::change_user() doesn't accept null as $database argument...
Anatol Belski [Sat, 6 May 2017 13:06:34 +0000 (15:06 +0200)]
Fixed bug #74547 mysqli::change_user() doesn't accept null as $database argument w/strict_types

7 years ago[ci skip] update NEWS
Anatol Belski [Tue, 2 May 2017 12:53:32 +0000 (14:53 +0200)]
[ci skip] update NEWS

7 years agoFixed bug #51918 Phar::webPhar() does not handle requests sent through PUT and DELETE...
Christian Weiske [Fri, 21 Apr 2017 19:20:00 +0000 (21:20 +0200)]
Fixed bug #51918 Phar::webPhar() does not handle requests sent through PUT and DELETE method

phar: Support DELETE, HEAD and PUT HTTP methods in Phar::webPhar

Up to now only GET and POST requests could be handled with Phar::webPhar(),
which is insufficient for today's REST APIs.
This patch expands the list of supported HTTP methods.

7 years ago[ci skip] update NEWS
Anatol Belski [Tue, 2 May 2017 10:38:04 +0000 (12:38 +0200)]
[ci skip] update NEWS

7 years agoFixed bug #74510 win32/sendmail.c anchors CC header but not BCC
Anatol Belski [Tue, 2 May 2017 10:31:00 +0000 (12:31 +0200)]
Fixed bug #74510 win32/sendmail.c anchors CC header but not BCC

7 years agoFix bug #74457 Wrong reflection on XMLReader::expand
Fabien Villepinte [Mon, 17 Apr 2017 09:20:47 +0000 (11:20 +0200)]
Fix bug #74457 Wrong reflection on XMLReader::expand

7 years agoFix bug #74468 wrong reflection on Collator::sortWithSortKeys
Fabien Villepinte [Tue, 18 Apr 2017 15:03:42 +0000 (17:03 +0200)]
Fix bug #74468 wrong reflection on Collator::sortWithSortKeys

7 years agoTests to ensure BC with undocumented behavior used in hacks
Anatol Belski [Mon, 1 May 2017 16:01:05 +0000 (18:01 +0200)]
Tests to ensure BC with undocumented behavior used in hacks

Should become obsolete, once behaviors change.

7 years agoFix typo in SplFixedArray has_dimension implementation
Nikita Popov [Mon, 1 May 2017 10:14:53 +0000 (12:14 +0200)]
Fix typo in SplFixedArray has_dimension implementation

7 years agoFix compiler warnings when compiled against musl libc
Michael Heimpold [Mon, 17 Apr 2017 22:22:34 +0000 (00:22 +0200)]
Fix compiler warnings when compiled against musl libc

musl libc is complaining when <sys/poll.h> is used instead of <poll.h>
so change this.

This issue was reported for OpenWrt/LEDE where musl libc is the standard
C library instead of e.g. glibc, see the following link for the original PR:
https://github.com/openwrt/packages/pull/4263

Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
--

v3: refined checks/fallback paths as suggested by @bukka

v2: rebased to resolve merge conflict in main/php_network.h

v1: initial PR

7 years agoFix possible memory leak in openssl_encrypt and openssl_decrypt
Jakub Zelenka [Thu, 27 Apr 2017 14:44:26 +0000 (15:44 +0100)]
Fix possible memory leak in openssl_encrypt and openssl_decrypt

7 years agoPrefer using DIRECTORY_SEPARATOR in OpenSSL config path
Jakub Zelenka [Thu, 27 Apr 2017 14:10:55 +0000 (15:10 +0100)]
Prefer using DIRECTORY_SEPARATOR in OpenSSL config path

7 years agoAdd oid_section to the testing openssl.cnf
Jakub Zelenka [Thu, 27 Apr 2017 14:09:58 +0000 (15:09 +0100)]
Add oid_section to the testing openssl.cnf

It tests repeated OID addition

7 years ago[ci skip] update NEWS
Anatol Belski [Thu, 27 Apr 2017 09:35:00 +0000 (11:35 +0200)]
[ci skip] update NEWS

7 years agoRevert "Detect invalid port in xp_socket parse ip address"
Anatol Belski [Thu, 27 Apr 2017 00:49:12 +0000 (02:49 +0200)]
Revert "Detect invalid port in xp_socket parse ip address"

This reverts commit bab0b99f376dac9170ac81382a5ed526938d595a.

7 years agoRevert "Follow up patch regarding bug #74216, see bug #74429"
Anatol Belski [Thu, 27 Apr 2017 00:49:01 +0000 (02:49 +0200)]
Revert "Follow up patch regarding bug #74216, see bug #74429"

This reverts commit cda7dcf4cacef3346f9dc2a4dc947e6a74769259.

7 years agosync bin sdk version
Anatol Belski [Wed, 26 Apr 2017 15:04:29 +0000 (17:04 +0200)]
sync bin sdk version

7 years agoDo not add already added object to the internal OpenSSL table
Jakub Zelenka [Wed, 3 Aug 2016 19:01:41 +0000 (20:01 +0100)]
Do not add already added object to the internal OpenSSL table

This fixes OpenSSL 1.1 where adding object with OID that has been
already added causes an error - preventing of duplication.

7 years agorely on bundle config, instead of system one
Remi Collet [Wed, 26 Apr 2017 12:41:40 +0000 (14:41 +0200)]
rely on bundle config, instead of system one

7 years agoskip the test on OpenSSL < 1.1 for now
Anatol Belski [Thu, 23 Mar 2017 02:09:30 +0000 (03:09 +0100)]
skip the test on OpenSSL < 1.1 for now

7 years agomove next
Anatol Belski [Tue, 25 Apr 2017 12:06:17 +0000 (14:06 +0200)]
move next

7 years ago[ci skip] update NEWS
Anatol Belski [Tue, 25 Apr 2017 11:59:20 +0000 (13:59 +0200)]
[ci skip] update NEWS

7 years agoFollow up patch regarding bug #74216, see bug #74429
Sara Golemon [Tue, 25 Apr 2017 10:52:48 +0000 (12:52 +0200)]
Follow up patch regarding bug #74216, see bug #74429

While the case in bug #74429 is not documented and is only worky due to
an implementation bug, the strength seems to breach some real world
apps. Given this patch doesn't impact the initial security fix for
bug #74216, it is reasonable to let the apps keep working. As mentioned
in the ticket, this behavior is a subject to change in future versions
and should not be abused.