Re: [PATCH] tests: disable fsync everywhere

From: Junio C Hamano <gitster@pobox.com>
Date: Oct 29, 2021, 18:12 UTC
Message-ID: <xmqqpmrnlmyx.fsf@gitster.g>
In-Reply-To: <20211029075640.M183252@dcvr>

Eric Wong <e@80x24.org> writes:

Show 12 quoted lines

> Junio C Hamano <gitster@pobox.com> wrote:
>> Eric Wong <e@80x24.org> writes:
>> > +# n.b. consider using Git.pm
>> > +sub use_fsync {
>> > +    if (!defined($_use_fsync)) {
>> > +        my $x = $ENV{GIT_TEST_FSYNC};
>> > +        if (defined $x) {
>> 
>> I would have expected to see "exists $ENV{GIT_TEST_FSYNC}", but I
>> guess there is no way to place in %ENV anyway, so it would be OK.
>
> Was that meant to say: "no way to place `undef' in %ENV anyway"?

Yes. Nothing the external callers of a Perl script does by futzing the environment with setenv(3) and unsetenv(3) can make undef appear as a value in $ENV{SomeKey}, so defined $ENV{V} and exists $ENV{V} are equivalent.

I still prefer "exists $ENV{V}", which I think conveys the intent of the check better, i.e. "we do this iff the environment variable X is there".

> If so, `undef' can actually be in Perl's %ENV, though it appears
> to get coerced into "" (empty string) when spawning processes.

Yes, but you are talking about the opposite direction, what Perl can do to %ENV to affect processes it spawns, which is not what I meant.

Show 5 quoted lines

> Leaving GIT_CONFIG set was actually causing "git config" to
> exit(1) since git-cvsserver sets GIT_CONFIG and the GIT_CONFIG
> file doesn't have a test.fsync setting.  This is the current
> behavior, I think it's a weird quirk, but intended behavior of
> git-config.

Ah, sorry, I misread the variable. GIT_CONFIG was the thing that says "read from this file and nowhere else"; we do want to disable it locally for "-c var=val" to take effect.

Show 13 quoted lines

> # this assumes you don't have foo.bar set in your ~/.gitconfig :>
> $ GIT_CONFIG=$HOME/.gitconfig git -c foo.bar=0 config --type=bool foo.bar
> $ echo $?
> 1
>
>> > +            my $v = ::safe_pipe_capture('git', '-c', "test.fsync=$x",
>> > +                                        qw(config --type=bool test.fsync));
>> 
>> THis is an interesting idiom.
>
> Heh, I just thought of it before sending my original.  I was
> going to use a regexp originally (in git-svn, too), but didn't
> want to get into corner cases such as hex and +/- prefixes).

And this I think is the best way to do so ;-)