Re: [PATCH 8/8] worktree: make "move" refuse to move atop missing registered worktree

From: Junio C Hamano <gitster@pobox.com>
Date: Jun 8, 2020, 22:06 UTC
Message-ID: <xmqqeeqpjjpe.fsf@gitster.c.googlers.com>
In-Reply-To: <20200608062356.40264-9-sunshine@sunshineco.com>

Eric Sunshine <sunshine@sunshineco.com> writes:

> -	if (file_exists(dst.buf))
> -		die(_("target '%s' already exists"), dst.buf);
> +	check_candidate_path(dst.buf, force, worktrees, "move");

OK. Moving to a location that is already occupied by an existing file or a directory, even if that file or directory is not one of the existing worktree, used to die here, but check_candidate_path() performs that check and dies with almost the same message (it does not say 'target'), so there is no loss of safety here. The check done in the check_candidate_path() helper is even better in that it allows an existing directory as long as it is empty.

Show 10 quoted lines

> diff --git a/t/t2403-worktree-move.sh b/t/t2403-worktree-move.sh
> index 939d18d728..7035c9d72e 100755
> --- a/t/t2403-worktree-move.sh
> +++ b/t/t2403-worktree-move.sh
> @@ -112,6 +112,27 @@ test_expect_success 'move locked worktree (force)' '
>  	git worktree move --force --force flump ploof
>  '
>  
> +test_expect_success 'refuse to move worktree atop existing path' '
> +	> bobble &&

Style?

Show 22 quoted lines

> +	git worktree add --detach beeble &&
> +	test_must_fail git worktree move beeble bobble
> +'
> +
> +test_expect_success 'move atop existing but missing worktree' '
> +	git worktree add --detach gnoo &&
> +	git worktree add --detach pneu &&
> +	rm -fr pneu &&
> +	test_must_fail git worktree move gnoo pneu &&
> +	git worktree move --force gnoo pneu &&
> +
> +	git worktree add --detach nu &&
> +	git worktree lock nu &&
> +	rm -fr nu &&
> +	test_must_fail git worktree move pneu nu &&
> +	test_must_fail git worktree --force move pneu nu &&
> +	git worktree move --force --force pneu nu
> +'
> +
>  test_expect_success 'move a repo with uninitialized submodule' '
>  	git init withsub &&
>  	(

Thanks.