Re: [PATCH] Build in shortlog
- From
Junio C Hamano <junkio@cox.net>
- Date
- Nov 11, 2006, 21:36 UTC
- Message-ID
- <7virhlis3e.fsf@assigned-by-dhcp.cox.net>
- In-Reply-To
- <Pine.LNX.4.63.0610221322370.14200@wbgn013.biozentrum.uni-wuerzburg.de>
Johannes Schindelin <Johannes.Schindelin@gmx.de> writes:
Show 13 quoted lines
> diff --git a/path-list.c b/path-list.c > index 0c332dc..f8800f8 100644 > --- a/path-list.c > +++ b/path-list.c > @@ -57,7 +57,7 @@ struct path_list_item *path_list_insert( > int index = add_entry(list, path); > > if (index < 0) > - index = 1 - index; > + index = -1 - index; > > return list->items + index; > }
This part looks like a genuine bugfix and unrelated to shortlog.
Many callers of path_list_insert() seem to ignore its return value, but merge_recursive.c seems to use it in three places, to find an entry to hang cached rename information to in insert_stage_data().
I haven't come up with an example to demonstrate the breakage of not having this fix in the existing git-merge-recursive, but I think this needs to be applied to even 'maint'.