Spurious failures from CVS tests

[Mads Kiilerich]

  Martin Geisler wrote, On 09/02/2010 05:39 PM:
> Hey everybody,
>
> I get a lot of mails from apycot about failures caused by the CVS
> convert tests. Failures such as this one:
>
>    http://apycot.hg-scm.org/projectenvironment/hg-crew/full/367764?tab=hgtest
>
> Should we blacklist the test in apycot or can we somehow make it more
> robust?

I added this new test that apparently lacked the workarounds we use in 
other places. I have pushed a fix. Please let me know if you see other 
failures.

/Mads