Extra ignore files with relative paths

[Martin Geisler]

Mads Kiilerich <mads at kiilerich.com> writes:

> Roger Pate wrote, On 06/01/2010 02:56 PM:
>> On Tue, Jun 1, 2010 at 6:37 AM, Mads Kiilerich<mads at kiilerich.com>  wrote:
>>    
>> I applied your change, cloned the hg repo, and spent some time rooting
>> around in there, but I think I got the test correct: it works with
>> run-tests.py test-hgignore.  I don't know the preferred form for
>> submitting changes, but here's this:
>>    
>
> http://mercurial.selenic.com/wiki/ContributingChanges says a lot, but
> FWIW the patch mostly looks fine to me.

Mads, thanks for the review!

> You could add a descriptive commit message and use "patchbomb" to send
> the patch to the list without extra linebreaks. Posting to
> mercurial-devel might also get more developers attention.

Those are good suggestions, though we also read this list. I would have
included the patch if it had not been for the things you point out
below.

>> --- a/tests/test-hgignore
>> +++ b/tests/test-hgignore
>> @@ -69,3 +69,16 @@
>>
>>   cd dir
>>   echo "--" ; hg status .
>> +
>> +
>> +# test that additional ignore files are resolved correctly
>> +# if hg status doesn't show a "not found" error, that is enough to
>> determine that they are found
>> +cd ..
>> +touch .hg/ignore-relative
>> +touch .hg/ignore-absolute
>>    
>
> Perhaps you could put something in the files to show that they
> actually are used - and that both glob and regexp works as expected.

We must have some other test that checks that both types of patterns
work, so that should not be the focus. But I agree that it's more
convincing to see a test with content :-)

-- 
Martin Geisler

See my Mercurial presentation: http://vimeo.com/11497288
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
URL: <http://lists.mercurial-scm.org/pipermail/mercurial/attachments/20100601/b3e443da/attachment.asc>