Re: [PATCH v5 1/5] t0060: Add test for manipulating symlinks via absolute paths

From: Junio C Hamano <gitster@pobox.com>
Date: Feb 3, 2014, 18:50 UTC
Message-ID: <xmqq7g9cf2ty.fsf@gitster.dls.corp.google.com>
In-Reply-To: <1391358940-17373-2-git-send-email-martinerikwerner@gmail.com>

Martin Erik Werner <martinerikwerner@gmail.com> writes:

> When symlinks in the working tree are manipulated using the absolute
> path, git dereferences them, and tries to manipulate the link target
> instead.

The above may a very good description of the root cause, but can we have description of a symptom that is visible by end-users that is caused by the bug being fixed with the series here, by ending the above like so:

	... link target instead.  This causes "git foo bar" to
	misbehave in this and that way.

Testing the low-level underlying machinery like this patch does is not wrong per-se, but I suspect that this series was triggered by somebody noticing breakage in a end-user command "git foo $path" with $path being a full pathname to an in-tree symbolic link. It wouldn't be like somebody who was bored and ran "test-path-utils" for fun noticed the root cause without realizing how the fix would affect the behaviour that would be visible to end-users, right?

Can we have that "git foo $path" to the testsuite as well? That is the breakage we do not want to repeat in the future by regressing.

I am guessing "foo" is "add", but I wasn't closely following the progression of the series.

Thanks.