[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Ruby test failure on 1.6.x

From: Joe Swatosh <joe.swatosh_at_gmail.com>
Date: Mon, 11 Jan 2010 19:15:28 -0800

On Mon, Jan 11, 2010 at 12:01 PM, Hyrum K. Wright
<hyrum_wright_at_mail.utexas.edu> wrote:
>
> On Jan 11, 2010, at 12:12 PM, Philip Martin wrote:
>
>> "Hyrum K. Wright" <hyrum_wright_at_mail.utexas.edu> writes:
>>
>>> After merging the pending fixes into 1.6.x, I reran the tests.  The only failure was in the ruby bindings:
>>>
>>> [[[
>>> hwright_at_orac:~/dev/svn-1.6.x$ make check-swig-rb
>>> cd /home/hwright/dev/svn-1.6.x/subversion/bindings/swig/ruby; \
>>>        /usr/bin/ruby -I /home/hwright/dev/svn-1.6.x/subversion/bindings/swig/ruby \
>>>          /home/hwright/dev/svn-1.6.x/subversion/bindings/swig/ruby/test/run-test.rb \
>>>          --verbose=normal
>>> Loaded suite .
>>> Started
>>> ...................................................................................................................E.......................................................................................................
>>> Finished in 18.84964 seconds.
>>>
>>>  1) Error:
>>> test_apply(SvnDeltaTest):
>>> ArgumentError: NULL pointer given
>>> /home/hwright/dev/svn-1.6.x/subversion/bindings/swig/ruby/svn/util.rb:86:in `svn_txdelta_apply_wrapper'
>>> /home/hwright/dev/svn-1.6.x/subversion/bindings/swig/ruby/svn/util.rb:86:in `txdelta_apply_wrapper'
>>> /home/hwright/dev/svn-1.6.x/subversion/bindings/swig/ruby/svn/delta.rb:54:in `apply'
>>> /home/hwright/dev/svn-1.6.x/subversion/bindings/swig/ruby/test/test_delta.rb:128:in `test_apply'
>>>
>>> 219 tests, 1490 assertions, 0 failures, 1 errors
>>> make: *** [check-swig-rb] Error 1
>>> ]]]
>>
>> On my debian stable box using r897912 with ruby-1.8.7, swig-1.3.36 and
>> gcc-4.3.2 I get no failures.  It takes 528 seconds rather than 18.8.
>
> If Joe can not reproduce, we'll chalk it up to local issues and I'll proceed with the release.
>

All the tests are passing for me.

But this sounded kinda familiar. The wc-ng work on trunk has exposed
a lot of places where the bindings or the tests of the bindings have
had pool lifetime related issues. I'd be curious to know if this
failure is intermittent for you, and if merging r876245 from trunk
helps?

--
Joe
Received on 2010-01-12 04:16:03 CET

This is an archived mail posted to the Subversion Dev mailing list.