Should we use subrepos ourselves?

[Martin Geisler]

Matt Mackall <mpm at selenic.com> writes:

> On Wed, 2010-08-18 at 16:51 +0200, Martin Geisler wrote:
>> Hi guys,
>> 
>> I had a meeting with my customer today where we dicussed subrepos (as
>> usual) and half-joking, they suggested that we try using them ourselves
>> for the Mercurial repository...
>
> It might indeed help us improve subrepos. On the other hand, it's
> definitely not a good use case for them and other people might decide
> to follow our bad example.

Yeah, I tend to agree... unfortunately.

However, there is one case where subrepos would make things nicer for
us: when the i18n repository is merged into main, we get these big
"junctions" when 4 lines of development is merged using 3 merges:

  http://selenic.com/hg/graph/b9d316bcc013

Using subrepos, we would effectively decouple the two repositories and
so there would be a single changeset in main when the .hgsubstate file
is updated. Overall this means less clutter.

I had not considered this use case before: even for strongly dependant
repositories, subrepos will make it easy to decouple development in
different parts of the overall repository.

Another place where it could make sense is if we moved some extensions
to separate repositories. The keyword extension is a good example as it
is mostly Christian that maintains it -- eol could be another example,
though I don't want to be the sole maintainer of that.

We clearly don't *have* to do this since the current model works, but I
think we should consider it.

-- 
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/20100820/a2c45197/attachment.pgp>