Re: [PATCH 3/5] clone: find the current branch more explicitly

From: Jeff King <peff@peff.net>
Date: Nov 30, 2008, 18:10 UTC
Message-ID: <20081130181011.GB10375@coredump.intra.peff.net>
In-Reply-To: <1228039053-31099-3-git-send-email-gitster@pobox.com>

On Sun, Nov 30, 2008 at 01:57:30AM -0800, Junio C Hamano wrote:

Show 8 quoted lines

> +			sym = find_ref_by_name(refs, symref);
> +			if (!sym) {
> +				/*
> +				 * NEEDSWORK: perhaps create the symref ref
> +				 * that is still unborn and queue it?
> +				 */
> +				continue;
> +			}

I don't see any reason not to create the pointer to an unborn branch. I think it would be nice eventually to allow cloning an empty repository (i.e., to facilitate:

  mkdir foo &&
  (cd foo && git --bare init) &&
  git clone foo bar &&
  cd bar &&
  echo content >file && git add file && git commit -m content &&
  git push

) which people have asked for from time to time.

> +			if (sym->symref)
> +				die("symref line says %s points at %s "
> +				    "but earlier it said it points at %s",
> +				    symref, target, sym->symref);

I wonder if there would ever be a use for sending multiple symref packets for the same ref. IOW, should we "be liberal in what we accept" here and just choose some sane behavior (like picking the first or last to be sent), and allow room for expansion there.

I can't actually think of a possible use for that, but nor can I think of any particular reason to die here (save for detecting bugs in upload-pack :) ). But just consider that we will be stuck with clients with the "die" behavior forever.

-Peff