bug-gnu-emacs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

bug#51897: 29.0.50; keymap-global-unset cannot unset C-x C-z


From: Eli Zaretskii
Subject: bug#51897: 29.0.50; keymap-global-unset cannot unset C-x C-z
Date: Wed, 17 Nov 2021 15:59:38 +0200

> From: Lars Ingebrigtsen <larsi@gnus.org>
> Cc: 51897@debbugs.gnu.org,  manuel.uberti@inventati.org
> Date: Wed, 17 Nov 2021 08:53:32 +0100
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> > Are you using "C-x v v", perhaps?  At some point in the past I
> > proposed to teach "C-x v v" to suggest a push after commit, or even
> > optionally do a push without asking, given some user customization.
> > But that idea was shot down for reasons I still don't understand.
> > I think keeping push out of the "C-x v v" cycle makes that command
> > much less useful nowadays.
> 
> I do see the objection from a standpoint of "well, a commit and a push
> are two totally separate operations" Git standpoint

I didn't suggest to do them both in one go, I suggested that invoking
"C-x v v" _after_ the one that committed a changeset will either push
or offer pushing as one of a small number of possible actions.

> After thinking about it a bit more, I'm not quite sure what I'd prefer
> myself.  It's not uncommon for me that I remember something after I've
> hit `C-x v v' (like adding Copyright-paperwork-exempt: yes to the commit
> message), and having this be a two step operation helps with that,
> because I can just back out the commit and try again, which I can't if
> I've already pushed.

What is reasonable after a commit?  Either (a) show or (b) push or (c)
another commit or (d) pull.  So why not ask the user which one of
these would he/she want to do?  We could even try to be smart about
that: for example, if the clone is ahead of upstream, push might be
the moist reasonable possibility.

What does "C-x v v" do today after push?  If the tree is clean,
nothing useful.  So what are we going to lose?

> But on the other hand, I forget to push so often that it's ridiculous.
> I think one of the contributing factors is that it just takes a while --
> I mean, I have to pull first, wait for it to finish, and then push.

For that reason, I never pull before pushing, unless push tells me I
must.





reply via email to

[Prev in Thread] Current Thread [Next in Thread]