On 13/08/14 13:48, Jan Ehrhardt wrote: > Michael Wallner in php.pecl.dev (Wed, 13 Aug 2014 12:56:42 +0200): >> Hey, I'm not sure I can follow this discussion :-/ Is there anything for >> me to fix before RC3? > > Probably not. I ran the tests for PHP 5.4 and PHP 5.5 with version > 2.0.7. They gave the same results, so the compilation on PHP 5.5 was > fixed and did not cause (additional) problems. > > Test results: > > ===================================================================== > Number of tests : 137 137 > Tests skipped : 0 ( 0.0%) -------- > Tests warned : 0 ( 0.0%) ( 0.0%) > Tests failed : 8 ( 5.8%) ( 5.8%) > Expected fail : 0 ( 0.0%) ( 0.0%) > Tests passed : 129 ( 94.2%) ( 94.2%) > --------------------------------------------------------------------- > Time taken : 40 seconds > ===================================================================== > > ===================================================================== > FAILED TEST SUMMARY > --------------------------------------------------------------------- > Bug #66388 (Crash on POST with Content-Length:0 and untouched body) > [N:\php-sdk\pecl-share\http\tests\bug66388.phpt] > client ssl [N:\php-sdk\pecl-share\http\tests\client012.phpt] > env request json > [N:\php-sdk\pecl-share\http\tests\envrequestjson001.phpt] > env request json > [N:\php-sdk\pecl-share\http\tests\envrequestjson002.phpt] > etags with hash [N:\php-sdk\pecl-share\http\tests\etag001.phpt] > chunked filter [N:\php-sdk\pecl-share\http\tests\filterchunked.phpt] > zlib filter [N:\php-sdk\pecl-share\http\tests\filterzlib.phpt] > env request message [N:\php-sdk\pecl-share\http\tests\message002.phpt] > ===================================================================== > > The 'etags with hash' failure could have something to do with > FOREACH_HASH_KEYVAL (wild guess, just because of the keyword hash) but > it fails under PHP 5.4 as well.
Nope, the test is about etag generation with ext/hash. FOREACH_HASH_KEYVAL is for iteration over a HashTable. Could you upload or mail me the failing tests, so I could have a look how severe the failures are? Thanks! -- Regards, Mike -- PECL development discussion Mailing List (http://pecl.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
