Re: [GSoC] [PATCH] test: avoid pipes in git related commands for test suite

From: Junio C Hamano <gitster@pobox.com>
Date: Mar 15, 2018, 17:04 UTC
Message-ID: <xmqqsh915kzi.fsf@gitster-ct.c.googlers.com>
In-Reply-To: <CAPig+cTLCswg_=q5ybnyN3As4Au05q5eAcA7Prr643KCgZ0OAw@mail.gmail.com>

Eric Sunshine <sunshine@sunshineco.com> writes:

Show 13 quoted lines

> Thanks for presenting an opposing opinion. While I understand your
> position, the reason for my suggested transformation is that if the
> patch already transformed the code in the way suggested, it would
> increase my confidence, as a reviewer, that the patch author had
> _studied_ and _understood_ the code. Increased confidence is
> especially important for mechanical transformations since -- as seen
> in the unsnipped review comment below -- blindly-applied mechanical
> transformations can be suboptimal or outright incorrect.
>
> It's also the sort of review comment I would make even to very
> seasoned project participants[1].
>
> [1]: https://public-inbox.org/git/CAPig+cQLmYQeRhPxvZHmY7gApnbE25H_KoSWs-ZjuBo4BruimQ@mail.gmail.com/

Yes, it is a good example that mechanical conversions are often mind-numbing and make even seasoned participants miss trivially obvious improvement opportunities ;-)

It however is OK to be more lenient to newer participants and allow deferring such "while at it, make it right" on top of "minimally required for correctness", in order to encourage them by getting something to the tree early ;-)