Releasing from Mercurial repo -- procedures?

[Stephen J. Turnbull]

Michael Sperber writes:

 > I think it meets all of your desiderata, but as I said, no problem with
 > holding off.  It even works without a named branch: If you just pull the
 > version you want, make your changes, tag the version you want, make the
 > tarballs, pull again, merge, and push, then (a) (b) and (c) are all met.

Not AFAICS.  True, I'll see what I want to see in my repo.  But the
*default* output of hg log *at the public repo* will make it appear
that the concurrent commit is included in the release.  In any case,
unless we freeze around the release, our developers and users need to
become graph theory/concurrency experts to interpret the output of "hg
log".  I don't think that's a good idea.

 > > No, I wanted to know if anybody objected to a normally short freeze on
 > > the xemacs-beta repo.
 > 
 > I don't object.

OK.  Once I've ported my scripts to Mercurial I'll make a call for
last-minute APPROVEs to xemacs-beta.

 > > I also wanted to know if I have commit access to it, which question
 > > has sorta been ignored. :-)
 > 
 > Sorry.  You do.

Great!  Thanks!