Re: git-diff: --ignore-matching-lines has no effect on the output when --name-only is used

From: Junio C Hamano <gitster@pobox.com>
Date: Jul 25, 2025, 15:20 UTC
Message-ID: <xmqqikjg47qt.fsf@gitster.g>
In-Reply-To: <20250725111139.GB3014187@coredump.intra.peff.net>

Jeff King <peff@peff.net> writes:

Show 7 quoted lines

> I think Junio's response earlier in the thread discusses this, and how
> we already respect "-w" for "--quiet".
>
> I'm not sure I agree with this part that he wrote, though:
>
>> It is just --raw, --name-only, --name-status, and --checkdiff output
>> formats that deliberately ignore content based ignore mechanisms.

Ah, from the end of that non-sentence, something like "[these formats] haven't been adjusted to the new world view". The new world view was described with things like "-w vs --quiet" that changed the semantics over time.

> I can see how it gets weird when you ask for --raw, but ...

The "--raw" output should never change its behaviour (as that designed to be machine readable, and the reader is expected to make good use of the object names by grabbing the contents of the blob---instead of learning a single bit "are they different?"). "--checkdiff" is about detecting malformatted code, and omitting the path from output only because it is a patch that does nothing other than making the indentation worse (hence with -w, diff will say "ah, no non-whitespace chagne") does not make sense, do practically "--name-status" and "--name-only" are the only things that could be updated.

> I do wonder if changing it at this point would somehow break somebody's
> workflow or have unexpected fallout, though.

That is certainly a concern.

I won't be touching the code for this change myself, but I can help reviewing if somebody writes a proposed change.

Thanks.