Commit Graph

42 Commits

Author SHA1 Message Date
Nikita Popov
1b29e0cacd Switch zend_print_zval_r to use smart_str
Instead of directly writing to stdout. This allows doing a print_r
into a string, without using output buffering.

The motivation for this is bug #67467: print_r() in return mode
will still dump the string to stdout (causing a potential information
leak) if a fatal error occurs.
2016-06-13 18:04:22 +02:00
Xinchen Hui
0e91e5fb90 Fixed test 2015-11-26 12:00:25 +08:00
Xinchen Hui
4a7e83f54a Fixed bug #70970 (Segfault when combining error handler with output buffering)
of course we can try to refactor the current flow to make this error can
be catched safly.

but as 7.0.0 is releasing,  I don't think a refactor is safe now. and
actually I don't see any gain to make this catchable.

so let's keep this be consistent with 5.6 and safe for now
2015-11-25 08:00:20 -08:00
Dmitry Stogov
ed9c8a23ba More accurate reference counting on closures 2015-03-10 23:04:41 +03:00
Reeze Xia
31d5b65c74 Fix test 2015-03-03 22:56:23 +08:00
Michael Wallner
6fed9960a9 fix test 2015-02-18 14:05:45 +01:00
Michael Wallner
84fad268d0 Merge branch 'PHP-5.6'
* PHP-5.6:
  fix test
2015-02-18 14:05:10 +01:00
Michael Wallner
8b6919c77c fix test 2015-02-18 14:04:53 +01:00
Michael Wallner
b36581aa75 Merge branch 'PHP-5.6'
* PHP-5.6:
  Fixed bug #65593 (Segfault when calling ob_start from output buffering callback)
2015-02-18 14:04:19 +01:00
Michael Wallner
225af964c0 Fixed bug #65593 (Segfault when calling ob_start from output buffering callback) 2015-02-18 13:58:31 +01:00
Kalle Sommer Nielsen
334b9718d5 Fix tests I broke with the E_ERROR > E_RECOVERABLE_ERROR change 2014-12-29 07:54:44 +01:00
Andrea Faulds
e5eb9530ab Use "float" and "integer" in typehint and zpp errors 2014-12-21 13:23:02 +00:00
Andrea Faulds
ebaa42302a Skip buncha tests on 32-bit 2014-11-29 01:05:19 +00:00
Stanislav Malyshev
e74776266f Merge branch 'PHP-5.4' into PHP-5.5
* PHP-5.4:
  fix test for bug #60322
2013-06-13 10:30:18 -07:00
Stanislav Malyshev
10b9579395 fix test for bug #60322 2013-06-13 10:30:00 -07:00
Dmitry Stogov
e9641141aa Made test compatible with O+ 2013-03-16 16:11:00 +04:00
Michael Wallner
575790e842 fix bug #61272 ob_start callback get passed empty string 2012-11-28 10:19:44 +01:00
Stanislav Malyshev
e36a5a1197 fix test 2012-05-15 23:09:46 -07:00
Olivier DOUCET
e613889d50 more verbose skip reason in test files with not so obvious extension requirements 2012-02-25 12:10:41 +00:00
Michael Wallner
6b31413b13 Fix bug #60768 Output buffer not discarded
in php_output_handler_op():
  * if appending to buffer succeeds, just return HANDLER_NO_DATA
    and do nothing else
  * if a zero sized string or true is returned from the handler
    function, reset the context as well as the handler's buffer
2012-01-16 17:51:35 +00:00
Stanislav Malyshev
f863d486d1 fix test 2011-12-26 05:08:09 +00:00
Stanislav Malyshev
87b9f7bb3a fix bug #60322 (ob_get_clean() now raises an E_NOTICE if no buffers exist) 2011-12-26 05:01:46 +00:00
Pierrick Charron
8e2dcfd8b6 Fixed bug #60321 (ob_get_status(true) no longer returns an array when buffer is empty). 2011-11-18 00:16:58 +00:00
Lars Strojny
c2ad166fd5 Reverting r319136 2011-11-17 09:24:15 +00:00
Lars Strojny
7d0fe66ebe Adjusting error messages 2011-11-14 00:26:37 +00:00
Xinchen Hui
9b091f8c34 add skipif 2011-11-13 14:31:30 +00:00
Xinchen Hui
89feaefce1 Fixed bug #60282 (Segfault when using ob_gzhandler() with open buffers) 2011-11-13 04:11:57 +00:00
Jani Taskinen
72ae65bc9d - CS and "de-facto" style of returning NULL for failing parse-params, sync partially with PHP_5_3 too (reordering part :) 2010-12-03 15:34:24 +00:00
Jani Taskinen
07535b7986 - Fix test 2010-12-03 15:30:54 +00:00
Jani Taskinen
41a1de45fe - Add test for the ob_start($foo); leak/crash bug 2010-12-03 15:29:53 +00:00
Felipe Pena
42a57f7656 - Fix tests 2010-06-16 02:14:32 +00:00
Michael Wallner
11d24c1593 * implement new output API, fixing some bugs and implementing some feature
requests--let's see what I can dig out of the bugtracker for NEWS--
  and while crossing the road:
   * implemented new zlib API
   * fixed up ext/tidy (what was "s&" in zend_parse_parameters() supposed to do?)

Thanks to Jani and Felipe for pioneering.
2010-05-31 10:29:43 +00:00
Jani Taskinen
af49e58f51 - Reverted r296062 and r296065 2010-03-12 10:28:59 +00:00
Jani Taskinen
06f072cb5e MFH: Improved / fixed output buffering (Michael Wallner) 2010-03-11 10:24:29 +00:00
Jani Taskinen
c4d350632b - Use correct revision 2009-11-28 00:47:42 +00:00
Jani Taskinen
e8e213ca01 - No dots here 2009-11-28 00:35:56 +00:00
andy wharmby
1e119743a5 Backport ob test to php 5.2 and 53. 2009-06-18 21:03:00 +00:00
andy wharmby
3fd133bcf3 Backport ob test to php 5.2 and 53. 2009-06-14 16:47:11 +00:00
Sebastian Schürmann
7942cd2c3c - Initial commit 2009-04-25 11:33:26 +00:00
Hannes Magnusson
d800809da4 This is apparently fixed 2009-04-06 13:16:05 +00:00
David Coallier
2e5f11b841 - Fixed bug #46897: ob_flush() should fail to flush unerasable buffers
- Added test for bug46897
- A memleak has been identified while fixing this bug but the changes
  made by this commit do not affect ob_get_clean where the memleak (which
  is fixed in HEAD with the new handler) resides.
- [DOC] New noticed is outputted in ob_flush when the buffer is unerasable
2009-02-13 11:48:17 +00:00
Robin Fernandes
bd7f18ab4f Additional output buffering tests. 2008-12-18 15:20:07 +00:00