Re: [PATCH] ruby test race

Hi,

In <20060915092039.GG3212@p12n.org>
  "[PATCH] ruby test race" on Fri, 15 Sep 2006 04:20:39 -0500,
  Peter Samuelson <peter@p12n.org> wrote:

> On a really slow machine [50 MHz 68060], I got the following in the
> ruby testsuite:
>
> 2) Failure:
> test_commit_editor(SvnRaTest)
> [svn140]/bindings/swig/ruby/test/test_ra.rb:228:in `test_commit_editor'
> [svn140]/bindings/swig/ruby/test/util.rb:57:in `change_gc_status'
> [svn140]/bindings/swig/ruby/test/util.rb:68:in `gc_disable'
> [svn140]/bindings/swig/ruby/test/test_ra.rb:216:in `test_commit_editor']:
> <[1, "Fri Sep 15 01:34:39 CEST 2006", "buildd"]> expected but was
> <[1, "Fri Sep 15 01:34:40 CEST 2006", "buildd"]>.
>
> This could happen on _any_ machine, just more frequently on slow ones.
> The following patch is untested, so feel free to treat it as a mere
> clarification of the problem.

Thanks for reporting. I updated test cases in r21518 using
your strategy but not using your implementation. I used
(start_time..end_time).include?(target_time) for testing.

Thanks,

--
kou
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org