Re: [PATCH 05/15] t/Makefile: optimize chainlint self-test

[Eric Sunshine <sunshine@sunshineco.com>]

On Thu, Dec 16, 2021 at 8:22 AM Ævar Arnfjörð Bjarmason
<avarab@gmail.com> wrote:
> > On 13.12.2021 09:27, Eric Sunshine wrote:
> >>It's not seen in the patch context, but earlier in the file we have:
> >>
> >>    CHAINLINTTESTS = $(sort $(...,$(wildcard chainlint/*.test)))
> >>
> >>which provides stability via `sort`, thus ensures that the order of
> >>the ".test" and ".expect" match.
> >>
> >>I think that addresses your concern (unless I misunderstand your observation).
>
> But just FWIW I think both of you are wrong about the potenital for a
> ".test" and ".expect" bug here.
>
> I.e. yes the CHAINLINTTESTS variable is sorted:
>
> But in Eric's patch we just have this relevant to this concern of
> (paraphrased) "would it not be sorted break it?":
>
>         +       sed -e '/^# LINT: /d' $(patsubst %,chainlint/%.test,$(CHAINLINTTESTS)) >'$(CHAINLINTTMP_SQ)'/tests && \
>         +       cat $(patsubst %,chainlint/%.expect,$(CHAINLINTTESTS)) >'$(CHAINLINTTMP_SQ)'/expect && \
>
> So it doesn't matter if it's sorted our not.
>
> I.e. we've got {A,B,C}.{test,expect} files in a directory, and we're
> constructing a "A.test" and "A.expect" via "$(patsubst)".
>
> So if it's "A B C", "C B A", "A C B" etc. won't matter. We'll always get
> ".test" files corresponding to ".expect".

Yes, sorry, I meant to say something along these lines in my reply, in
addition to mentioning `sort`, but forgot. Taking a look at this
again, though, makes me wonder if the CHAINLINTTESTS assignment should
be done with `:=` rather than `=` (unless GNU make is smart enough to
only invoke the `wildcard` operation only once, in which case it
wouldn't particularly matter).