Re: [PATCH 1/2] tests: use shell negation instead of test_must_fail for test_cmp

From: Jonathan Nieder <jrnieder@gmail.com>
Date: Oct 6, 2017, 21:59 UTC
Message-ID: <20171006215926.GY19555@aiede.mtv.corp.google.com>
In-Reply-To: <20171006192252.3a27enzim5yurhe5@sigill.intra.peff.net>

Hi,

Jeff King wrote:

> On Fri, Oct 06, 2017 at 12:00:05PM -0700, Stefan Beller wrote:

Show 8 quoted lines

>> The `test_must_fail` should only be used to indicate a git command is
>> failing. `test_cmp` is not a git command, such that it doesn't need the
>> special treatment of `test_must_fail` (which e.g. includes checking for
>> segfault)
>
> Hmph. "test_must_fail test_cmp" is a weird thing for somebody to write.
> And your patch is obviously an improvement, but I have to wonder if some
> of these make any sense.

Just for the record: I agree with all the above, and my Reviewed-by still stands.

Thanks for looking closer. I wonder if there's a systematic way to avoid this kind of weak test that can bitrot and still pass too easily --- e.g. should we have a test_must_differ command with an explanation of why it should usually be avoided? Would a lint check that bans this kind of construct completely be going too far?

Jonathan