[MERGE] Fixed bug related to test_iter_changes_new breakage on Windows [version 2]

[Dmitry Vasiliev]

Alexander Belchenko wrote:
> Dmitry Vasiliev пишет:
> OK. But I'm curious why this test was not fails before WT4 is introduced
> and only starts to fails now with WT4. I'm hope there is not because
> of some latent bug in workingtree_4.py.

Aaron answered the question but I think it would be useful to have some 
buildbot installation to run bzr tests on different platforms.

> Well, looking on your patch (that quite trivial) I'm think
> it's probably better to use inline conditional expression, like this:
> 
>             if not supports_executable():
> - -                executable = entry.executable
> +                executable = bool(entry) and entry.executable
>             else:
>                 executable = bool(stat.S_ISREG(mode) and stat.S_IEXEC & mode)

In most cases I don't like such expressions because very often they are 
abused, but in this case it seems a reasonable approach. The new patch 
attached.

-- 
Dmitry Vasiliev <dima at hlabs.spb.ru>
http://hlabs.spb.ru
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: workingtree.bundle
Url: https://lists.ubuntu.com/archives/bazaar/attachments/20070413/0c1346d9/attachment.diff