From: Junio C Hamano <gitster@pobox.com>
Date: Tue, 16 Feb 2021 01:37:15 GMT
Subject: Re: [PATCH 1/2] diff: do not display hunk context under -W
Message-ID: <xmqq35xwu71w.fsf@gitster.c.googlers.com>
In-Reply-To: <xmqq7dn8u7dz.fsf@gitster.c.googlers.com>

Junio C Hamano <gitster@pobox.com> writes:

> Ævar Arnfjörð Bjarmason  <avarab@gmail.com> writes:
>
>> But if we viewed it with the -W option we'd show "is_empty_rec()",
>> because we'd first find the "xdl_emit_diff()" context line, extend the
>> diff to that, and then would go look for context to show again.
>
> Makes sense.

Ehh, I did not mean "the current behaviour with -W that shows
is_empty_rec makes sense".  The observation you made (which lead to
the conclusion in your next paragraph that it is not a good idea
to show is_empty_rec on the hunk header line) made sense to me.

But I do not think any change from the current behaviour should be
made if -U<n> wins -W (i.e. the first precontext line shown in the
hunk header is not due to -W).  We should hunt for the name of the
function whose latter half we are seeing at the beginning of the
hunk in that case.

Thanks.