Re: Re* [RFC/PATCH] extend meaning of "--root" option to index comparisons

From: Jeff King <peff@peff.net>
Date: Sep 21, 2008, 13:56 UTC
Message-ID: <20080921135616.GA25238@sigill.intra.peff.net>
In-Reply-To: <7vskrvswxp.fsf@gitster.siamese.dyndns.org>

On Fri, Sep 19, 2008 at 01:27:46PM -0700, Junio C Hamano wrote:

>  (1) A user getting an error message from "git init && git log" may be
>      annoyed, but he very well knows there is no history yet _anyway_.
>      This initial annoyance will pass immediately after creating any
>      commit, so I do not think it is a big issue.

I think there is an additional case of script writers, who want their scripts to fail gracefully or otherwise do the right thing with an initial commit. Right now they have to special-case the initial commit. I don't know if it is possible to have sane enough behavior that the special case can be eliminated, or if it will simply make things worse.

>      "bad default revision 'HEAD'" is a cryptic way to give that indicaion
>      that can be improved but that is a separate issue.  Rewording it so
>      that it explains the situation better in user's terms would be a
>      worthy improvement.

I agree that would be an improvement.

>  (2) "--root" is about "do we show a creation event as a huge diff from
>      emptyness?".  Yes, we turn it on for "git log" but it does not have
>      anything to do with the issue of yet to be born branch, where there
>      isn't even a big creation event yet.

What about index comparisons? What should an index comparison to a branch yet-to-be-born look like? Right now it is an error.

> I am reluctant to agree with the opinion that "git log" should be _silent_
> in a world without any history.

It feels a bit more Unix-y to me. That is, if I am asking for some set of commits, and there are _no_ commits in the set, then I expect no output. That makes sense for text processing.

Show 9 quoted lines

> -	argc = setup_revisions(argc, argv, rev, "HEAD");
> +	argc = setup_revisions(argc, argv, rev, NULL);
> +	if (!rev->pending.nr) {
> +		add_head_to_pending(rev);
> +		if (!rev->pending.nr) {
> +			printf("No commits (yet).\n");
> +			exit(0);
> +		}
> +	}

I like the idea of an improved message, but such a message should definitely not go to stdout; it would feed nonsense to a command like "git log | my_log_filter".

-Peff