Announce: bzrtools-44

[Aaron Bentley]

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

Robert Collins wrote:
| It should be possible to represent the partial history - though I don't
| know if the original history horizon stuff we discussed in London is
| still in the bzr design. Even if it isn't listing the last known
| ancestory as the first one should allow automated matchups later.

Tom's post about using human-readable/checksum pairs seemed like a good
idea to me, and I was going to apply it here.  So you'd have

arch-x:lord at emf.net--2004/tla--devo--1.3--patch-5:rwaliuhgzsk

This would make it possible to update baz2bzr conversion algorithm
without breaking everything, because any changes that affected the
contents of the contents of the snapshot (and could therefore lead to
inconsistency) would alter the checksum.

Listing ancestor revisions that we don't have access to would be
impossible if we need a checksum of the snapshot to produce the revision
name.

As for matching up, this requires careful thought.  I don't think you
can have a given revision ID in different places in the revision
histories of different branches and expect sanity.  But maybe you can
have arch-to-bzr matchups still work.

Aaron
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFClfnv0F+nu1YWqI0RAujYAJ9AjSjYvKZ3UV4wNHrB4lD9bZyeBwCcCr1j
gQV31fzqM5w5w1sJqixrx8k=
=FYK/
-----END PGP SIGNATURE-----