[PATCH 2 of 4] clone to master bookmark if available

[Victor Suba]

Right, I think if it didn't complain about bookmarked heads that would
cover a lot of ground.  If you want to keep a line of code private,
bookmark it
and don't push the bookmark.  If you want to share a line of code, bookmark
it and push the bookmark.

Question is what to do with a default bookmark?

a) One option is that un-bookmarked heads are all considered "default" heads
and "hg push" will try to push all of them, and complain if they create
multiple
heads on the remote (like today)/

b) Have a single "default" head per-branch that works like a bookmark, and
don't
consider any un-bookmarked heads on push at all.

For bookmarked heads, complain only if the heads you are pushing would
create
forked copies of the bookmarks on the remote (i.e. is not fast-forward).

Cheers,
Vic

On Tue, Nov 15, 2011 at 9:32 AM, David Soria Parra <dsp at php.net> wrote:

> On 2011-11-14, Martin Geisler <mg at lazybytes.net> wrote:
> > --===============2012388182==
> > Content-Type: multipart/signed; boundary="=-=-=";
> >       micalg=pgp-sha1; protocol="application/pgp-signature"
> >
> > --=-=-=
> > Content-Transfer-Encoding: quoted-printable
> >
> > Arne Babenhauserheide <arne_bab at web.de> writes:
> >
> >> Am Sonntag, 13. November 2011, 16:38:52 schrieb Martin Geisler:
> >>> Feature branches are a super powerful concept, but Mercurial fights
> >>> you if you try to use them. It basically tries to make you merge all
> >>> the heads into a single head -- no matter if you have put bookmarks
> >>> on them or not.
> >>
> >> Do you mean non-persistent feature branches?=20
> >
> > I mean branches where you use bookmarks to track them instead of using
> > named branches. Mercurial is happy enough to let you keep several heads
> > around, provided they are on different named branches.
> >
> >> Would it suffice to allow pushing bookmarked heads with an
> >> --new-bookmarked-heads option (which you could also send permanently
> >> and which would also transfer all bookmarks on the pushed heads)?
> >
> > No, that's not the point. Making 'hg push' and 'hg pull' not complain
> > about multiple heads all the time is the important part. Perhaps they
> > should only complain about multiple un-bookmarked heads.
> >
>
> I don't know what the state of this discussion is, but I think hg should
> not complaint about bookmarked heads. THis also includes hg push should
> not complain about newly created heads if a bookmark points to that head
> (and is pushed).
>
> _______________________________________________
> Mercurial-devel mailing list
> Mercurial-devel at selenic.com
> http://selenic.com/mailman/listinfo/mercurial-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://selenic.com/pipermail/mercurial-devel/attachments/20111115/9785747a/attachment.html>