Johannes Schindelin <[EMAIL PROTECTED]> writes:
>> Question: Do you mind if I push a so regressed mingw.git master, or
>> would you like that I wait for a fix (that someone among you would have
>> to provide) that I apply before pushing?
>
> I'd rather have it pushed, and mention it in the ReleaseNotes of (Win-)
> Git, so that anybody who feels inclined can fix it.
I do not mind a "is_absolute_path" helper function in git-sh-setup.sh,
which msysgit folks can override with their own patch.
is_absolute_path () {
case "$1" in
/*) return 0 ;;
*) return 1 ;;
esac
}
I guess no sane person would name their directory in the work tree as C:
on POSIX systems, so we could even do this cross-platform, but it somehow
makes me feel dirty:
is_absolute_path () {
case "$1" in
/*) return 0 ;;
[A-Z]:/*) return 0 ;;
*) return 1 ;;
esac
}
In either case, with such a helper, the problematic part would become
something like this:
git-am.sh | 12 ++++++------
1 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/git-am.sh b/git-am.sh
index 1f6b5e0..0826b09 100755
--- a/git-am.sh
+++ b/git-am.sh
@@ -217,12 +217,12 @@ else
set x
first=
}
- case "$arg" in
- /*)
- set "$@" "$arg" ;;
- *)
- set "$@" "$prefix$arg" ;;
- esac
+ if is_absolute_path "$arg"
+ then
+ set "$@" "$arg"
+ else
+ set "$@" "$prefix$arg"
+ fi
done
shift
fi
