[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: git-handler.el
From: |
Eli Zaretskii |
Subject: |
Re: git-handler.el |
Date: |
Sun, 13 Aug 2017 18:26:14 +0300 |
> From: Yuri Khan <address@hidden>
> Date: Sun, 13 Aug 2017 22:08:55 +0700
> Cc: Michael Albinus <address@hidden>, Jonas Bernoulli <address@hidden>,
> Dmitry Gutov <address@hidden>, Emacs developers <address@hidden>
>
> >> Not necessarily. It is today, but two months ago it could have resided
> >> in any of baz.el, quux.el or xyzzy.el.
> >
> > If you don't know its file, how will you find it?
>
> By trial and error. Find the directory at revision, L1: choose the
> next most probable file, look in there. Repeat from L1 until found.
Then going to a directory @VERSION should be fine for your use case,
AFAIU.
> >> In the use case being discussed, I’m not interested in master. Or,
> >> more specifically, not in its current position, and not in the whole
> >> set of its commits that modify baz.el. I am interested in one specific
> >> commit, say ef7a18a071, that does not modify baz.el.
> >
> > Then you should look inside src/emacs/lisp@@/ef7a18a071, I think. Why
> > is that a problem?
>
> I got the impression that, in the proposed scheme, <path>@@<revision>
> only shows files under <path> that are modified by the commit
> <revision>. There is no problem for me if it shows *all* files under
> <path> that existed at that <revision>.
I think it's the latter (for a VCS where a version marks the entire
tree), but maybe I misunderstood what Michael meant.
- Re: git-handler.el, (continued)
- Re: git-handler.el, Michael Albinus, 2017/08/12
- Re: git-handler.el, Yuri Khan, 2017/08/12
- Re: git-handler.el, Michael Albinus, 2017/08/13
- Re: git-handler.el, Yuri Khan, 2017/08/13
- Re: git-handler.el, Eli Zaretskii, 2017/08/13
- Re: git-handler.el, Yuri Khan, 2017/08/13
- Re: git-handler.el,
Eli Zaretskii <=
- Re: git-handler.el, Michael Albinus, 2017/08/14
- Re: git-handler.el, Michael Albinus, 2017/08/14