[PATCH] Improved named branch support when pushing changesets

[Matt Mackall]

On Mon, May 18, 2009 at 12:48:54PM +0200, Henrik Stuart wrote:
> Henrik Stuart skrev:
> > Matt Mackall wrote:
> >> On Thu, 2009-03-26 at 22:13 +0100, Sune Foldager wrote:
> >>> On 26/03/2009, at 20.15, Matt Mackall wrote:
> >>>
> >>>> Very interesting.
> >>> Thanks :-)
> >>>
> >>>> Did you consider using lookup to lookup individual heads?
> >>> Yes, we did consider using the already available information, as well  
> >>> as already available wire-commands to get the desired information  
> >>> across. The problem is that a branch head isn't necessarily a head in  
> >>> the graph, so remote.heads() won't always return it. Even if it could  
> >>> work with lookup, we would get quite a few roundtrips for complex  
> >>> repositories. Extending with a 'branchmap' command, we get all the  
> >>> information (all heads on all branches) in a single chunk :-).
> >> The idea would be to lookup the remote revisions for local new branch
> >> heads. ie I push stable which is inactive in the remote repo, so I
> >> detect push would add one new head, namely my 'stable' head. I look up
> >> remote 'stable' and discover that it's an ancestor of my local 'stable'
> >> head, so everything's copacetic.
> >>
> >> Not terribly expensive and works with servers going back quite some
> >> time. Not sure it's sufficient to do everything right though.
> 
> This is a follow-up mail to the one we sent in March, in an effort to
> get the named branch support into crew.
> 
> BACKGROUND

Excellent! Please fold this discussion into appropriate patch in your
series and resend for code review.

-- 
Mathematics is the supreme nostalgia of our time.