git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: Please pull mergetool.git

From
Junio C Hamano <gitster@pobox.com>
Date
Jun 10, 2007, 19:55 UTC
Message-ID
<7v1wgj8tzy.fsf@assigned-by-dhcp.cox.net>
In-Reply-To
<E1HxPt7-0007jV-6V@candygram.thunk.org>
"Theodore Ts'o" <tytso@mit.edu> writes:
Show 13 quoted lines
> Please pull from the "mergetool" branch at:
> 	git://repo.or.cz/git/mergetool.git mergetool
>
> It adds support for vimdiff/gvimdiff as a mergetool program, as well
> Josh's suggestion of making the default merge-tool selection more
> intelligent, although I've rewritten it somewhat take into account the
> comments made by you and others on the git mailing list.
>
> (Note that as of this writing, meld is a pretty sad/dificient tool, and
> even GNOME users may very well prefer kdiff3 over meld --- which is my
> default.  Still, it seems reasonable to default to using KDE tools in an
> KDE login session, and GNOME tools in a GNOME login session, and people
> who care differently should set their own preferences in ~/.gitconfig.)
Thanks for keeping track of the mergetool.

I do not see problems in the mergetool part, other than that I mildly suspect that opendiff -- actually FileMerge -- might want to be in the test -n "$DISPLAY" section, but that is inherited from the previous iteration so in that sense leaving outside is a sane thing to do.

Show 6 quoted lines
> Dan McGee (1):
>       git-mergetool: Allow gvimdiff to be used as a mergetool
>
> Theodore Ts'o (2):
>       git-mergetool: Make default selection of merge-tool more intelligent
>       Add git-applymbox, git-applypatch, and *~ to .gitignore
But I hope you would not be offended if I said I do not want to.

This is not such a strong objection, but I really wish that you did not mix in the .gitignore change; it does not belong to this "series".

If it were an obvious and universally nondisagreeable fix, I would not mind you mixing it in this mergetool updates, but I am of two minds about the .gitignore change, and actually slightly in favor of not adding git-applymbox and git-applypatch back.

About git-applymbox and git-applypatch, it helps people when they switch branches and/or bisect to keep potential build products from older/different revisions listed in .gitignore. That would however imply we would end up carrying old entries forever in it. We do not keep clean rule in Makefile to remove build products from older/different revisions when remove build targets, so why should we keep them in .gitignore?

Also I deliberately have kept *~ out of .gitignore for a reason. I do have that entry in .git/info/exclude, but the choice of Emacs over vi is personal to me.

Previous: Theodore Ts'oNext: Theodore Tso
Message 2 of 7 in “Please pull mergetool.git”
  1. Theodore Ts'oJun 10, 2007
  2. Junio C HamanoJun 10, 2007
  3. Theodore TsoJun 10, 2007
  4. Brian GernhardtJun 10, 2007
  5. Junio C HamanoJun 11, 2007
  6. Martin LanghoffJun 11, 2007
  7. Theodore TsoJun 11, 2007

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.