[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Gnu-arch-users] Re: <<< conflict markers
From: |
Stefan Monnier |
Subject: |
[Gnu-arch-users] Re: <<< conflict markers |
Date: |
21 Apr 2004 13:32:42 -0400 |
User-agent: |
Gnus/5.09 (Gnus v5.9.0) Emacs/21.3.50 |
> This strikes me as a variant of the "replay --reverse p6 ; commit"
The difference is that it keeps more info about the fact that tla does not
have to apply patch-7 to get to patch-9.
Whether the tla implementation can take advantage of such info to avoid
spurious conflicts when doing things like `replay' from patch-4 to patch-9.
The spurious conflict I'm thinking of is when you have a local change that
conflicts with the undesired patch-6: if `replay' applies patch-6
and then patch-7 (where patch-7 undoes patch-6) you get to resolve the
(same) conflicts twice, whereas if it can skip over the undesired patch,
you avoid the problem.
Relevant in cases where patch-6 erroneously removes a file, or changes the
line-ending from LF to CRLF, ...
Stefan