On Sat, May 22, 2010 at 08:35:39PM -0500, Jonathan Nieder wrote:
> Without your patch applied, I get
>
> $ cd .git
> $ git --work-tree=/tmp/git symbolic-ref HEAD
> refs/heads/hello
>
> I should have done the following instead:
>
> $ worktree=$(pwd)
> $ cd .git
> $ git --work-tree="$worktree" symbolic-ref HEAD
> fatal: ref HEAD is not a symbolic ref
Thanks
> > +char *get_relative_path(char *cwd, const char *dir)
>
> Looks reasonable. Should be const char *cwd.
That is not possible, because we return a pointer to cwd plus an
offset, which will be subsequently modified by strcat().
> > diff --git a/setup.c b/setup.c
> > index 5716d90..67b5122 100644
> > --- a/setup.c
> > +++ b/setup.c
> > @@ -524,14 +524,18 @@ const char *setup_git_directory(void)
> > const char *retval = setup_git_directory_gently(NULL);
> >
> > /* If the work tree is not the default one, recompute prefix */
> > if (inside_work_tree < 0) {
>
> If I understand correctly, you made two changes here:
>
> - interpret GIT_WORK_TREE and friends relative to .git instead of
> the original cwd (by not calling chdir(retval) before
> get_git_work_tree())
Oops
> - make setup_git_directory stay in the last directory searched for a
> .git directory instead of chdir-ing into the toplevel of the work tree.
>
> Are these safe changes to make?
Indeed, if you put it that way, this change does not look good.
Maybe I'll have a look at Nguyen's patches.
> diff --git a/t/t1501-worktree.sh b/t/t1501-worktree.sh
> index 9df3012..93bf92c 100755
> --- a/t/t1501-worktree.sh
> +++ b/t/t1501-worktree.sh
Thanks
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html