Re: [PATCH 1/2] Update git to use "git diff-index" instead of "git status"
- From: Peter Tyser <ptyser gmail com>
- To: Kai Willadsen <kai willadsen gmail com>
- Cc: meld-list gnome org, grant b edwards gmail com
- Subject: Re: [PATCH 1/2] Update git to use "git diff-index" instead of "git status"
- Date: Tue, 2 Mar 2010 00:16:57 -0600
<snip>
> Yes, I see the same behaviour. I was worried, since git-diff-index
> *can* return R and C in addition to the statuses that you handle. The
> git-diff-index manpage talks about the option:
> --no-renames
> Turn off rename detection, even when the configuration file gives
> the default to do so.
> which suggested to me that there was a way to turn on rename detection
> through git config. I can't find a definitive answer, but from my
> experiments the git config diff.renames setting doesn't affect
> git-diff-index. So I guess it's okay to not handle R and C statuses,
> as long as we never ask for them...
<snip>
>> Sorry for the long-winded email. Long story short, I think the
>> current patch is an improvement as is. I won't resubmit unless
>> someone sees a flaw in the logic above.
>
> While I'd like a definitive answer on the rename/copy detection, I'm
> inclined to agree. I'll probably poke it a little bit more and then
> push. Thanks.
I finally get it:) I can't affect "git diff-index"'s copy/rename
detection via my .gitconfig either. I added some quick debug to the
git diff-index command and it looks like it never takes the git config
parameter "diff.renames" into consideration. The "diff.renames"
parameter only seems to affect the git commands which use the
git_diff_ui_config() function (eg commit, diff, log, and merge).
So I'm guessing git shouldn't be advertising the "--no-renames" option...
Best,
Peter
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]