[MERGE] Allow external annotation tie breakers

[John Arbash Meinel]

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

Aaron Bentley wrote:
> Aaron Bentley has voted comment.
> Status is now: Semi-approved
> Comment:
> This doesn't look like it provides enough context to the tiebreaker.
> Shouldn't it be providing repositories or branches or something?
> 
> For details, see:
> http://bundlebuggy.aaronbentley.com/project/bzr/request/%3Cm2r60ficu3.fsf%40free.fr%3E
> 
> Project: Bazaar
> 
> 

It provides the same amount of context that "reannotate" has at the
moment. I believe the specific goal of Guilhem was to try to extract a
date string from the revision id and sort on that rather than sorting on
the pure revision_id. (So sort by ~date rather than by username.)

If we wanted to pass in a repository, we would have to pass it through a
lot of intervening layers. I agree that is ultimately better...

John
=:->

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknRGF8ACgkQJdeBCYSNAAOZOwCgsluiYyclFA6DbcoH6NU3sYip
ZGcAoIek8gD9494B3qS7w+I4gKOR+9yc
=C4Td
-----END PGP SIGNATURE-----