On Thu, Sep 13, 2007 at 02:56:58PM +0400, Dmitry V. Levin wrote:

quoted text
> The run_status() also defines GIT_INDEX_FILE environment variable which
> may affect other git commands like git-write-tree.

That is already taken care of elsewhere (I think runstatus is
confusingly written -- it should really only set GIT_INDEX_FILE for the
git-runstatus subcommand. This is an artifact from when runstatus was
implemented completely in shell, and I should have cleaned it up during
that transition).

It's probably not worth cleaning up, though, if the shell version is
going away soon.

quoted text
> I need this new option for my work now, I have to backport it to
> previous stable branch, because git-runstatus needs too much memory and
> runs too long in some cases.  Even worse, git-runstatus segfaults in one
> case I'm going to debug yet.
> 
> The --fast option should be easy to port forward to builtin git-commit
> anyway.

Yes, since you have already written the patch and it is fairly trivial,
I think it should be applied. I was trying to subtly hint that you
should make sure the feature makes it into the the new git-commit.

If you can provide a test case for git-runstatus segfaulting, I can help
look at it further.

-Peff
-
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html