MERGE ping

[Aaron Bentley]

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

Martin Pool wrote:
> On 25 Mar 2006, Aaron Bentley <aaron.bentley at utoronto.ca> wrote:
> 

> Rather than treating the conflict list as a set of "lines" perhaps they
> should be conflict objects.  We could have a ConflictSet object and
> change select_conflicts() into a method on that, perhaps called
> remove_all() and remove_paths().  Rather than having conflicts with
> different titles, we can just have subclasses of Conflict.
> The details of how these are packed can then be part of the workingtree
> format.

Okay, I can make these changes.  I think it will make the most sense for
Contlict objects to produce their own stanzas, rather than having
WorkingTree know about that.

> The methods like "conflict_stanzas" and "stanza_conflicts" could be more
> explicit as "conflicts_to_stanzas", etc.

Okay.  I was trying to save space, but maybe trying too hard.

> @@ -48,8 +49,11 @@
>      See also bzr resolve.
>      """
>      def run(self):
> -        for path in
>          WorkingTree.open_containing(u'.')[0].iter_conflicts():
> -            print path
> +        from bzrlib.workingtree import WorkingTree
> +        from transform import conflicts_strings
> +        wt = WorkingTree.open_containing(u'.')[0]
> +        for conflict in conflicts_strings(wt.conflict_lines()):
> +            print conflict
>  
> Should be 'from bzrlib.transform'.

Sorry, I don't follow.

> I don't mind putting this in as a step in the right direction but hope
> you'd think those changes are clearer.

If a ConflictSet is the right way, let's do it that way from the start.
 Otherwise, who knows whether it'll get done?

Aaron
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFEMRsF0F+nu1YWqI0RAkP7AJ9ql7O9bchQ+WUKp7rzE0B0bAXdwACeLdEk
hFz8lwOJFqa0hjos1wJ/WFg=
=efsW
-----END PGP SIGNATURE-----