Edit report at https://bugs.php.net/bug.php?id=60923&edit=1
ID: 60923 Updated by: m...@php.net Reported by: robertbasic dot com at gmail dot com Summary: Failing tests for sapi/cli -Status: Critical +Status: Closed Type: Bug Package: CGI/CLI related Operating System: GNU/Linux (Fedora 16) PHP Version: 5.4SVN-2012-01-29 (SVN) Assigned To: mike Block user comment: N Private report: N New Comment: This bug has been fixed in SVN. Snapshots of the sources are packaged every three hours; this change will be in the next snapshot. You can grab the snapshot at http://snaps.php.net/. For Windows: http://windows.php.net/snapshots/ Thank you for the report, and for helping us make PHP better. Previous Comments: ------------------------------------------------------------------------ [2012-01-30 12:17:49] m...@php.net Automatic comment from SVN on behalf of mike Revision: http://svn.php.net/viewvc/?view=revision&revision=322931 Log: use stderr before, stdout after php_output_startup() and sapi_module.ub_write() after php_output_activate() - fixes Bug #60920 CLI: php -v on STDERR - fixes Bug #60923 Failing tests for sapi/cli ------------------------------------------------------------------------ [2012-01-29 19:05:18] ras...@php.net Yeah, looks to be the same as bug #60920 ------------------------------------------------------------------------ [2012-01-29 17:39:06] robertbasic dot com at gmail dot com Think I narrowed it down even more. Seems to me the problem is in the new php_output_stderr function - when I comment it out, the tests pass again. I'll attach a diff to show what I did. ------------------------------------------------------------------------ [2012-01-29 16:21:26] robertbasic dot com at gmail dot com Hello again! I did a svn log -r {2012-01-10}:HEAD against the PHP54 branch and went through those commits. What I've managed to do so far is to find what commit is causing the tests to fail (at least I think it's this one): http://svn.php.net/viewvc?view=revision&revision=322743 The r322743 of the code has the tests failing, while the one prior to it (r322678) has the tests passing. Sadly my C skills are almost non-existent, so I can't promise a patch, but will continue to poke around it. ------------------------------------------------------------------------ [2012-01-29 14:24:04] robertbasic dot com at gmail dot com Will do. By looking at the expected and the actual outputs of sapi/cli/tests/001.phpt, looks like that var_dump(`$php -n -v`); first prints out the PHP version (the result from php -n -v) and then var_dumps a NULL, whereas it should var_dump the result from php -n -v. Actually, all these failing tests are this kind of thing: printing the result, var_dumping NULL, instead of var_dumping the result. What's strange, is that, for example sapi/cli/tests/007.phpt, from 3 var_dumps in that test, only one does this (the 2nd), the other 2 work as expected. Given that this works for RC6, I'm looking at commits since then, to see if anything from there is causing this. Any hints on what should I do to check is my system affecting this somehow? I'm using the test instructions you posted on codepad some time ago. Thanks! ------------------------------------------------------------------------ The remainder of the comments for this report are too long. To view the rest of the comments, please view the bug report online at https://bugs.php.net/bug.php?id=60923 -- Edit this bug report at https://bugs.php.net/bug.php?id=60923&edit=1