Issue83

classification
Title Branches and Mercurial?
Type defect Module repository
Severity inconvenience Platform N/A
Keywords host config Nosy List
explanation
process
These controls should only be changed by committers and tracker administrators.
Status chatting   Reason
Superseder   Submitted 2007-12-10.19:39:47
Priority normal   Assigned To

Created on 2008-01-19.04:58:04 by stephen, last changed 2008-02-27.06:57:44 by stephen.

Files
File name Uploaded Type Edit Remove
unnamed sperber, 2008-01-19.06:43:07 text/plain
Messages
msg278 [hidden] ([hidden]) Date: 2008-01-19.06:43:07
  Message-ID: <y9lodcxa793.fsf@dyn5.specs.de>
"Stephen J. Turnbull" <stephen@xemacs.org> writes:

> Mike, I know that this issue is going to come up; maybe you can
> publish a repo containing the script and any advice you have on using
> it to graft branches into the Mercurial tree?

I did use Tailor, and I've attached the config file that I used to use
to mirror from CVS to Mercurial.  Note the use of `start-revision' in
both cases: I.e. you can start with a pre-populated Mercurial repo, but
you have to make sure that Tailor doesn't reply the entire CVS history
from the beginning, but only from the point where CVS and Mercurial
already agree.

> Specifically, do we want to proliferate heads inside of the "xemacs"
> repo, or should we create external repos?

Definitely external repos---Alioth allows us to create new ones at will.
Alternatively, named branches within xemacs, but I see no payoff there
at this point.  (It's easy to pull an external repo into a named branch
later, should we so desire.)
msg101 [hidden] ([hidden]) Date: 2008-01-19.04:58:04
  Message-ID: <877ijme3uk.fsf@uwakimon.sk.tsukuba.ac.jp>
skip@pobox.com writes:
 > 
 >     Aidan> The Mercurial version of the Carbon2 branch doesn't yet
 >     Aidan> exist. The Red Book is excellent, if you want to look into making
 >     Aidan> one: http://hgbook.red-bean.com/hgbookch8.html#x12-1630008.5
 > 
 > Are you suggesting that I should just create an empty hg repository for the
 > Carbon2 branch and check in the latest files from that branch on CVS?

Actually, that branch should have a rather short history in XEmacs, so
better yet, you could just grab Choi's latest patch and start from
there.  As Aidan points out, our history is simply tracking Choi's
patches so far.

 > I was hoping there would be some way to suck the info out of CVS
 > into Mercurial.  I messed around a bit with tailor over the weekend
 > but got nowhere with it.

"Mess" is indeed an appropriate word regarding tailor's Mercurial
support.  Mercurial has its own CVS-to-Hg script, and I believe that a
modified version is what Mike used.

Mike, I know that this issue is going to come up; maybe you can
publish a repo containing the script and any advice you have on using
it to graft branches into the Mercurial tree?

Specifically, do we want to proliferate heads inside of the "xemacs"
repo, or should we create external repos?

_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta@xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
History
Date User Action Args
2008-02-27 06:57:44stephensetseverity: inconvenience
keyword: + host config
module: + repository
priority: normal
platform: + N/A
type: defect
2008-02-27 06:56:03stephensetfiles: - unnamed
2008-01-19 06:43:07sperbersetfiles: + unnamed, unnamed
status: new -> chatting
messages: + msg278
2008-01-19 04:58:04stephencreate