Re: [PATCH 3/3] t/t4013: add test for --diff-merges=off

From: Sergey Organov <sorganov@gmail.com>
Date: Aug 17, 2020, 20:36 UTC
Message-ID: <87wo1xko0y.fsf@osv.gnss.ru>
In-Reply-To: <xmqqr1s52nv5.fsf@gitster.c.googlers.com>

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

Show 10 quoted lines

> Sergey Organov <sorganov@gmail.com> writes:
>
>> So, do these both effectively resolve to:
>>
>>       git log -m --no-diff-merges --first-parent --patch 
>>
>> where first -m is the one implied by --first-parent, ...
>
> Eh, why does --first-parent even affect the choice of -m/no-m when
> --no-diff-merges is explicitly given?

Well, let's try your question on 3 abstract options:

"Why does --X even affects the choice of --Y/no-Y when --Z is explicitly given?"

And what is the assumed answer?

"--X should only affect the choice of --Y/no-Y when no --Z is explicitly given"

right? Or, in other terms:

"--X implies --Y unless --Z is explicitly given (as opposed to implied?)"

Do you see the problem here? I do see this as a problem. Simple "--X implies --Y" should be sufficient to achieve any of required behaviors. No need for such complications.

I just try to figure if some more or less strict model is hidden behind all this, and it seems there is none.

Thanks, -- Sergey