Faster "searching for changes" on push and pull? - a shortcut

[Arne Babenhauserheide]

Hi, 

I'm currently pushing my repo to sf.net, and it sits at "searching for 
changes" for quite some time. 

Since the remote repo has no changesets yet, I don't quite understand why that 
takes so long. Logically (and if I understand it correctly) it would suffice 
to say 

	"give me your head IDs"

and then check, if I have all those IDs in my local repo. If yes, I know 
exactly what I can push, since the remote repo contains no new changesets. 

When pulling, the opposite is true: It would suffice to just send the heads 
and let the remote site figure out which changesets I need. 


We should only have to resort to more powerful means, if the remote repo has 
changes I don't have AND I have changes the remote repo doesn't have. Then the 
heads could have branched off, so no repo would know the common places in 
history. 


For "hg outgoing" the effect would be even bigger than for push, since that 
head ID transfer would be the only network activity we'd need (except if both 
repos have changed too far). 

Is this a possible future improvement or is there an important reason not to 
pursue that path? 

Or is the information "searching for changes" misleading? 

Or are there better ways to archieve the same? 

Best wishes, 
Arne
-- 
-- Ein Würfel System: http://1w6.org - einfach saubere (Rollenspiel-) Regeln.
-- Infinite Hands: http://infinite-hands.draketo.de - singing a part of the 
history of free software.
-- My stuff: http://draketo.de - stories, songs, poems, programs and stuff :)

-- PGP/GnuPG: http://draketo.de/inhalt/ich/pubkey.txt
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
Url : http://selenic.com/pipermail/mercurial/attachments/20090324/a061d356/attachment.pgp