[PATCH 1 of 1] [mq] addchangegroupe finalize mq patches if changeset children are added

[Martin Geisler]

Matt Mackall <mpm at selenic.com> writes:

> On Tue, 2011-03-15 at 12:29 +0100, Martin Geisler wrote:
>> Matt Mackall <mpm at selenic.com> writes:
>> 
>> > On Sun, 2011-03-13 at 12:08 +0100, Pierre-Yves David wrote:
>> >> # HG changeset patch
>> >> # User Pierre-Yves David <pierre-yves.david at ens-lyon.org>
>> >> # Date 1300014459 -3600
>> >> # Branch stable
>> >> # Node ID 37831a0e8aa5d553e9951894b737e38f00c4ac6c
>> >> # Parent  38c9837b1f75f53cd3e5414e4283d21da1eb125f
>> >> [mq] addchangegroupe finalize mq patches if changeset children are added.
>> >
>> > I think I've been swayed by the arguments for aborting.
>> 
>> What do you think of the idea of leaving local tags behind with the
>> mq patch names when they are qfinished? That way no information is
>> lost.
>
> Ew.
>
> If you want to do a clever hack, you should have mq internally
> remember the names of changesets it's qfinished so they can be
> restored by qimport.

Please look at the idea -- the exact form of storage of the patch names
is of course not the important part.

> I actually run into this about once a week, prematurely qfinishing
> some joker's patch only to remember that I've forgotten to run tests
> that it turns out not to pass. So I qimport it and get some numeric
> name. This isn't really much of a problem for me, as it's not likely
> to live as an mq patch much longer.

I do the same and the numeric names are normally not a problem, unless
they happen to collide with another patch with the same numeric name.
Then I have to come up with a name for the patch when I qimport.

-- 
Martin Geisler

Mercurial links: http://mercurial.ch/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
URL: <http://selenic.com/pipermail/mercurial-devel/attachments/20110315/f01a0ff7/attachment.pgp>