Re: [PATCH 1/1] builtin/pull.c: use config value of autostash

From: Junio C Hamano <gitster@pobox.com>
Date: Jan 6, 2022, 19:11 UTC
Message-ID: <xmqq35m0el28.fsf@gitster.g>
In-Reply-To: <4fe0e850-1c30-da70-5535-e2a4105ba734@gmail.com>

Philippe Blain <levraiphilippeblain@gmail.com> writes:

Show 22 quoted lines

> We already have a quite useless 'int autostash' local variable in cmd_pull
> a few lines up, so an equivalent fix, I think, would be the following:
>
> diff --git a/builtin/pull.c b/builtin/pull.c
> index 1cfaf9f343..9f8a8dd716 100644
> --- a/builtin/pull.c
> +++ b/builtin/pull.c
> @@ -1036,14 +1036,13 @@ int cmd_pull(int argc, const char **argv, const char *prefix)
>  		oidclr(&orig_head);
>    	if (opt_rebase) {
> -		int autostash = config_autostash;
> -		if (opt_autostash != -1)
> -			autostash = opt_autostash;
> +		if (opt_autostash == -1)
> +			opt_autostash = config_autostash;
>    		if (is_null_oid(&orig_head) && !is_cache_unborn())
>  			die(_("Updating an unborn branch with changes added to the index."));
>  -		if (!autostash)
> +		if (!opt_autostash)
>  			require_clean_work_tree(the_repository,
>  				N_("pull with rebase"),
>  				_("please commit or stash them."), 1, 0);

OK, so the idea is to make opt_autostash the _primary_ thing that matters when deciding to do the auto-stashing. We may make it affected by the value we read from the configuration, if there is no command line option that sets it.

If there is no place that cares about what is in config_autostash (which is rebase.autostash; nobody reads merge.autostash in this command) other than this part, I think that is an even cleaner approach than what I sent. I very much like it.

Here is how I would explain your change.

----- >8 --------- >8 --------- >8 --------- >8 --------- >8 -----
Subject: pull --rebase: honor rebase.autostash even when fast-forwarding

"pull --rebase" internally uses the merge machinery when the other history is a descendant of ours (i.e. perform fast-forward). This came from [1], where the discussion was started from a feature request to do so. It is a bit hard to read the rationale behind it in the discussion, but it seems that it was an established fact for everybody involved that does not even need to be mentioned that fast-forwarding done with "rebase" was much undesirable than done with "merge", and more importantly, the result left by "merge" is as good as (or better than) that by "rebase".

Except for one thing. Because "git merge" does not (and should not) honor rebase.autostash, "git pull" needs to read it and forward it when we use "git merge" as a (hopefully better) substitute for "git rebase" during the fast-forwarding. But we forgot to do so (we only add "--[no-]autostash" to the "git merge" command when "git pull" was invoked with "--[no-]autostash" command line option, so that "git merge" can honor merge.autostash in such a case).

Make sure "git merge" is run with "--autostash" when rebase.autostash is set and used to fast-forward the history on behalf of "git rebase". Incidentally this change also takes care of the case where

 - "git pull --rebase" (without other command line options) is run
 - "rebase.autostash" is not set
 - The history fast-forwards

In such a case, "git merge" is run with an explicit "--no-autostash" to prevent it from honoring merge.autostash configuration, which is what we want. After all, we want the "git merge" to pretend as if it is "git rebase" while being used for this purpose.

[1] https://lore.kernel.org/git/xmqqa8cfbkeq.fsf_-_@gitster.mtv.corp.google.com/