Re: [PATCH v5 3/3] ref-filter: add support for %(contents:size)

From: Junio C Hamano <gitster@pobox.com>
To: Christian Couder <christian.couder@gmail.com>
Cc: Jeff King <peff@peff.net>, Alban Gruin <alban.gruin@gmail.com>,
	git <git@vger.kernel.org>,
	Christian Couder <chriscool@tuxfamily.org>
Subject: Re: [PATCH v5 3/3] ref-filter: add support for %(contents:size)
Date: Fri, 31 Jul 2020 13:30:19 -0700	[thread overview]
Message-ID: <xmqqa6zf2zs4.fsf@gitster.c.googlers.com> (raw)
In-Reply-To: <CAP8UFD15+p+xKwJ=B9WVsrc+2TvLHKmu78SBCLUFZVSYoTtbbg@mail.gmail.com> (Christian Couder's message of "Fri, 31 Jul 2020 22:12:13 +0200")

Christian Couder <christian.couder@gmail.com> writes:

> Hi Alban and Peff,
>
> On Fri, Jul 31, 2020 at 7:45 PM Jeff King <peff@peff.net> wrote:
>>
>> On Fri, Jul 31, 2020 at 07:37:22PM +0200, Alban Gruin wrote:
>>
>> > > +           # Leave $expect unquoted to lose possible leading whitespaces
>> > > +           echo $expect >expected
>> > > +           test_expect_${4:-sucess} $PREREQ "basic atom: $1 contents:size" '
>> >
>> > There is a typo here, and $expect is written to `expected', but
>> > `test_cmp' wants `expect'.  Fixing those mistakes does not reveal any
>> > broken tests.
>>
>> I thought at first you meant that the typo was s/expected/expect, and
>> wondered how this could possibly have passed. But the typo is
>> s/sucess/success/, so we were in fact not running the test at all (and
>> were generating "test_expect_sucess: not found" messages to stderr, but
>> outside of any test block. Yikes.
>>
>> Thanks for spotting.
>
> Yeah, I copied a suggestion from Junio in the last iteration without
> properly checking it. Sorry about that and thanks for spotting and
> fixing it.

I probably should stop giving "perhaps along the lines of this"
suggestion too lightly and/or when I do not have enough time to
apply and test myself.  Sorry for the gotcha.