[RFC] Separating last-revision from parents

[Aaron Bentley]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Robert Collins wrote:
> The current behaviour of update and revert is in my mind 'correct' -
> that is, we can probably improve it, buts its doing what I intended.
> 
> I wanted the following to hold true for any tree:
>  'bzr update && bzr revert && bzr info -> no missing revisions, no local
> edits, no new local commits. 

I don't want that.

When I commit locally, I expect that I'm setting a save point that I can
get back to.  If I'm not certain that update is doing to DTRT, I'll
commit locally, and then revert if I get into trouble.  From my
perspective, this behavior is definitely out of the frying pan, into the
fire.

I'm not sure why you want bzr update && revert to throw away local
commits.  Perhaps you could elaborate on that.  But it sounds like
perhaps it could be achieved more conveniently with a flag to update,
anyhow.

Maybe no one else thinks of revert the way I do (wouldn't be the first
time!), but if people like Alexander and I want this behavior, the
separation I've proposed can provide it, as well as the option of your
preferred behavior.

Aaron
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFx+5e0F+nu1YWqI0RAt6eAJwO1TEFQA28n5AL1LFT9/9OlwPOhgCfb886
wnvTgE1Dq9DhKbEiUrGJlQs=
=U3fm
-----END PGP SIGNATURE-----