Double merge (was: Re: First impressions of Mercurial)

[Thomas Arendsen Hein]

* Matt Mackall <mpm at selenic.com> [20050616 08:19]:
> On Thu, Jun 16, 2005 at 08:13:46AM +0200, Thomas Arendsen Hein wrote:
> > Why is this double merge needed this time?
> 
> A merge should happen if:
> 
> a) the working dir is not pristine and is therefore effectively a
>    branch off its parent
> 
> or
> 
> b) what you're checking out is not a direct descendent of what's in
>    your working dir
> 
> Or at least that's the theory.

Ok, now for the real thing:

mkdir double && cd double && \
hg init http://hg.intevation.org/tah1 && \
hg co && \
hg pull http://hg.intevation.org/upstream1 && \
hg co

requesting all changes
adding changesets
adding manifests
adding file revisions
modified 88 files, added 379 changesets and 733 new revisions
searching for changes
adding changesets
adding manifests
adding file revisions
modified 2 files, added 2 changesets and 3 new revisions
this update spans a branch affecting the following files:
 TODO (resolve)
 templates/map
aborting update spanning branches!
(use update -m to perform a branch merge)

But:

mkdir single && cd single && \
hg init http://hg.intevation.org/tah1 && \
hg pull http://hg.intevation.org/upstream1 && \
hg co

requesting all changes
adding changesets
adding manifests
adding file revisions
modified 88 files, added 379 changesets and 733 new revisions
searching for changes
adding changesets
adding manifests
adding file revisions
modified 2 files, added 2 changesets and 3 new revisions

(nothing more here)

Why does this 'hg co' make such a big difference?

Thomas

-- 
Email: thomas at intevation.de
http://intevation.de/~thomas/