From: Jonathan Tan <jonathantanmy@google.com>
Date: Thu, 07 Feb 2019 00:01:16 GMT
Subject: Re: [PATCH 1/8] tests: define GIT_TEST_PROTOCOL_VERSION
Message-ID: <20190207000116.55126-1-jonathantanmy@google.com>
In-Reply-To: <87d0o4lgtr.fsf@evledraar.gmail.com>

> > @@ -327,6 +327,9 @@ marked strings" in po/README for details.
> >  GIT_TEST_SPLIT_INDEX=<boolean> forces split-index mode on the whole
> >  test suite. Accept any boolean values that are accepted by git-config.
> >
> > +GIT_TEST_PROTOCOL_VERSION=<n>, when set, overrides the
> > +'protocol.version' setting to n if it is less than n.
> > +
> 
> In my version
> (https://public-inbox.org/git/20181213155817.27666-6-avarab@gmail.com/)
> I didn't have this "if it is less than n" caveat. I expect that helped
> with making some tests that were setting e.g. protocol.version=2 Just
> Work, is that the reason for this?

Yes, that's right. I thought that there is not much value in testing
tests that are explicitly protocol v2 as another protocol, since there
was a reason in the first place why the test writer wanted to test it
with v2.

> Mine also had more docs here, but maybe telling people that they can use
> "env" is too much...

With the ability to set =0 to effectively disable the option (because
the minimum of 0 and 0/1/2 is 0/1/2), I thought it was less important.