Upon unserialize of the acl, the acl is currently set to the id field which should be a string. Currently it passes the acl object into the id field which causes the following error.
Warning: Illegal offset type in isset or empty in Symfony/Component/Security/Acl/Dbal/AclProvider.php line 404
Commits
-------
5395c6c Few RO translation tweaks
Discussion
----------
Few RO translation tweaks
---------------------------------------------------------------------------
by lsmith77 at 2011/07/15 02:48:01 -0700
you seem to have done something to the line endings ..
---------------------------------------------------------------------------
by schniper at 2011/07/15 03:13:32 -0700
The file was edited on the site and because of some 404 error I couldn't save it so I copy/pasted it entirely, after I have corrected the problem.
Want me to retry?
Commits
-------
71cfb56 Thrown a \RuntimeException in RequestMatcher::checkIp6() if PHP is compiled with the option "disable-ipv6"
Discussion
----------
[HttpFoundation] Problem with RequestMatcher if PHP is compiled with the option "disable-ipv6"
Thrown a \RuntimeException in RequestMatcher::checkIp6() if PHP is compiled with the option "disable-ipv6".
Commits
-------
af6d4e4 changed test
6d1eb48 fixes a bug during session initialization in test environment
Discussion
----------
fixes a bug during session initialization in test environment
Commits
-------
05cc24c [Yaml] Wrap numeric strings in quotes when dumping
Discussion
----------
[Yaml] Wrap numeric strings in quotes when dumping
This addresses an obscure case where a hash string (actually a commit-ish, "686e444") was dumped to YAML as an unquoted string value. It was later parsed from YAML as an exponential numeric and changed to ".Inf".
This commit should not change the existing behavior when dumping non-string numerics. It also doesn't appear to disturb any of the other test cases. I realize it's a huge edge case, so I'm open to discussion.
The alternative to this fix was an ugly `preg_replace()` to apply quoting around the commit-ish after dumping. I would look forward to removing that :)
Commits
-------
5803a8e [Yaml] Do not skip consistency checks in InlineTest::testDump()
Discussion
----------
[Yaml] Do not skip consistency checks in InlineTest::testDump()
These conditionals were excluding 12.30e+02 from being checked. Removing the conditionals does not break any tests.
I investigated the conditionals with git blame, but that stops at previous commits for code formatting changes. These appear to be very old. Perhaps they were added for test failures on certain platforms? I don't spot any failures on Linux.
This addresses an obscure case where a hash string (actually a commit-ish, "686e444") was dumped to YAML as an unquoted string value. It was later parsed from YAML as an exponential numeric and changed to ".Inf".
Commits
-------
29e4063 [Security] changed order of checks to check for more specific things first
Discussion
----------
[Security] changed order of checks
Commits
-------
61de80d [FrameworkBundle] Updated the Dutch validator translations for the changes in 95f7eedd63
Discussion
----------
[FrameworkBundle] Updated the Dutch validator translations
[FrameworkBundle] Updated the Dutch validator translations for the changes in 95f7eedd63
Commits
-------
64e9263 Updated UPDATE.md
7cf891a Renamed variable returned and used self in place of static for constants
f91f4dd Added the possibility to set cookies with the same name for different domains and paths for Symfony\Component\HttpFoundation\ResponseHeaderBag
f08eeb4 Moved managing cookies of HeaderBag in ResponseHeaderBag
Discussion
----------
[HttpFoundation] Cookies management in ResponseHeaderBag
Fixed cookies management in `Symfony\Component\HttpFoundation\HeaderBag` and `Symfony\Component\HttpFoundation\ResponseHeaderBag`