[PATCH] Reproduce crash where synthetic revs break merge detection (issue1578)

[Patrick Mézard]

Martin Geisler a écrit :
> Greg Ward <greg-hg at gerg.ca> writes:
> 
>> [my patch, from Apr 18]
>>> # HG changeset patch
>>> # User Greg Ward <greg-hg at gerg.ca>
>>> # Date 1240075788 14400
>>> # Node ID 3d1f4648305f0cca07b01b8094dffc5d7bb9db86
>>> # Parent  81aba9e79054e5af5a17cecd90178dcd35b76b8b
>>> Reproduce crash where synthetic revs break merge detection (issue1578).
>>>
>>> (The fix for this was committed as 9bbcfa898cd3.)
>> [Martin Geisler plowing through the backlog]
>>> I just tried this patch and it works here.
>> "Works" as in "applies cleanly" or "applies cleanly and passes".
>> Actually, I expect both to be true on current crew.
> 
> The latter, of course :-)
> 
>> Correct.  And *that* patch is in crew:
>>
>> $ hg log -k1577
>> changeset:   7870:6c3b8132078e
>> user:        Greg Ward <greg-hg at gerg.ca>
>> date:        Sat Mar 28 12:24:53 2009 -0400
>> summary:     issue1577: fix broken test by assuming less about CVS output.
>>
>> So the above patch (reproduce issue1578) should be fine to go in.
>> Nothing is blocking it apart from code review.
> 
> Super, Patrick just said on IRC that he'll review it.

Pushed: http://hg.intevation.org/mercurial/crew/rev/2c7c973c2abd

I replaced the rm -rf by another conversion target.

Thanks!

--
Patrick Mézard