From: Ramkumar Ramachandra <artagnon@gmail.com>
Date: Mon, 24 Jun 2013 07:58:09 GMT
Subject: Re: [PATCH 6/6] push: honor branch.*.push
Message-ID: <CALkWK0=aDkCMB3ROSuH1crTmQKjCFyCmvJfvOTiQnBo4+NckBA@mail.gmail.com>
In-Reply-To: <1372048388-16742-7-git-send-email-gitster@pobox.com>

Junio C Hamano wrote:
> @@ -194,7 +203,14 @@ static int is_workflow_triagular(struct remote *remote)
>  static void setup_default_push_refspecs(struct remote *remote)
>  {
>         struct branch *branch = branch_get(NULL);
> -       int triangular = is_workflow_triagular(remote);
> +       int triangular;
> +
> +       if (branch->push_name) {
> +               setup_per_branch_push(branch);
> +               return;
> +       }

The most obvious question comes first: what result can I expect when
this interacts with remote.<name>.push?

Why did you design this feature like this?  Will the user _not_ want
refspec mapping except when pushing out the current branch with a
plain "git push"?

Also, you managed to throw out all safety out the window.  What
happens when the user does:

  # on branch master, derived from origin
  $ git push ram

And branch.master.push is set to next?  Will you let her shoot herself
in the foot like this?