"0 conflicts encountered."

[Aaron Bentley]

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

John A Meinel wrote:
> I like that too, but I would really like merge to report about what it
> actually did.

I'm not a big fan of doing that, because it's often overwhelming.  With
the rate stuff's been coming into bzr lately, data on successful merges
fills up your terminal buffer.

> Certainly we still have the bug that diff3 conflicts are reported in the
> old locations rather than the new ones. And now it doesn't even tell you
> that it moved anything.

I'm fairly certain we never said when we moved something, unless you
mean when we move files/directories out of the way?

> Also, if we keep --merge-type=diff3 the default, can we include
> --reprocess as the default as well? It is a lot nicer. (Naturally,
> supplying --show-base would turn it off).

That's fine by me.  (you meant merge-type=merge3, right?)

> And when are we thinking to switch to --merge-type=weave?

I'm not sure.  I actually prefer three-way diffs, most of the time
- - they give you a BASE file, so you can see what changes OTHER made.
- - sometimes, weave merges are more confusing because each version
deletes different lines, and you wind up with something that doesn't
make much sense, where Merge3 gives a conflict and shows both sides.

What would be really nice, though, is a "patience"-based
SequenceMatcher.  Did you ever get that working well?  The default
Python one is just too stupid when dealing with files that have had
partially-matching regions inserted.

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

iD8DBQFDkLVG0F+nu1YWqI0RAtjkAJ9sm24twCGkLDIZtGPSf30tcMtmRQCeLJ9i
tYyTupMUMJspsENaatU++FA=
=E0Kt
-----END PGP SIGNATURE-----