From: Junio C Hamano Date: Sat, 20 Jan 2007 00:53:07 GMT Subject: Re: [PATCH 3/3] show-branch --reflog: show the reflog message at the top. Message-ID: <7vr6tqbjt8.fsf@assigned-by-dhcp.cox.net> In-Reply-To: Johannes Schindelin writes: > On Fri, 19 Jan 2007, Junio C Hamano wrote: > >> I still am having a hard time trying to explain what these >> commands would do, and why they might be useful: >> >> git-log --walk-reflogs master..next > > I had less problems explaining that when walking fell back to walking the > parents. > > It could even make sense to... (meaning that is not what the code does...) > Of course, you could always say: if you ask for strange things, you will > get strange things. But if you have to say that, doesn't it suggest that accepting the option --walk-reflogs when you have anything other than one single positive commit in revs.pending leads to end user confusion? That's what I was after. The theme is the same about my hesitation against --reverse. On the other hand, >> git-log --walk-reflogs master -- Documentation/ > > Ah, but that is way easier! > > For example: I _know_ that something worked some time ago, and it was > probably in git.c. So ask "git-log --walk-reflogs -- git.c", and you will > soon find out where to start bisecting. there is no reason to forbid pathspec when --walk-reflogs is in effect, which I agree.