Re: [PATCH 3/6] receive-pack: receive.denyDeleteCurrent

From: Jeff King <peff@peff.net>
Date: Feb 9, 2009, 18:53 UTC
Message-ID: <20090209185310.GB27037@coredump.intra.peff.net>
In-Reply-To: <1234170565-6740-4-git-send-email-gitster@pobox.com>

On Mon, Feb 09, 2009 at 01:09:22AM -0800, Junio C Hamano wrote:

Show 8 quoted lines

> This is a companion patch to the recent 3d95d92 (receive-pack: explain
> what to do when push updates the current branch, 2009-01-31).
> 
> Deleting the current branch from a remote will result in the next clone
> from it not check out anything, among other things.  It also is one of the
> cause that makes remotes/origin/HEAD a dangling symbolic ref.  This patch
> still allows the traditional behaviour but with a big warning, and promises
> that the default will change to 'refuse' in a future release.

This patch looks good to me. One comment:

Show 12 quoted lines

> --- a/t/t5400-send-pack.sh
> +++ b/t/t5400-send-pack.sh
> @@ -193,4 +193,11 @@ test_expect_success 'pushing wildcard refspecs respects forcing' '
>      )
>  '
>  
> +test_expect_success 'deny pushing to delete current branch' '
> +    (
> +	rewound_push_setup &&
> +	git send-pack ../parent/.git :refs/heads/master 2>errs
> +    )
> +'

It was not immediately obvious why the subshell is needed here (for those in the audience, it is because rewound_push_setup changes directory). Perhaps 2/6 should be amended for rewound_push_setup to use a subshell instead, which makes it less likely to be forgotten by new tests.

-Peff