From: Jeff King <peff@peff.net>
Date: Mon, 07 Nov 2011 16:38:23 GMT
Subject: Re: [PATCH 3/3] grep: get rid of useless x < 0 comparison on an enum member
Message-ID: <20111107163823.GB27055@sigill.intra.peff.net>
In-Reply-To: <m3pqh4krer.fsf@hase.home>

On Mon, Nov 07, 2011 at 02:12:28PM +0100, Andreas Schwab wrote:

> Ævar Arnfjörð Bjarmason <avarab@gmail.com> writes:
> 
> > I.e. we'll always have GREP_HEADER_AUTHOR = 0 and
> > GREP_HEADER_COMMITTER = 1, we'll never have GREP_HEADER_AUTHOR = and
> > GREP_HEADER_COMMITTER = <some int>.
> 
> That is irrelevant.  You can always assign -1 to an object of enumerated
> type and the implicit conversion to the underlying type is fully
> defined, no matter what type the compiler choses.

Yes, but now you are getting into implementation-defined behavior, which
is also something to avoid. IOW, I don't think it's wrong for a static
analyzer to complain about:

  if (enum_with_only_positive_values < 0)

just because you could say:

  enum_with_only_positive_values = -1;

and it would work on _some_ platforms. That same static analyzer should
be complaining about the second line, which is where the real potential
bug is. If you want "-1" as an enumerated value, then add it to your
enum definition and the compiler will do the right thing.

It is perfectly fine to do:

     enum foo { a = -1, b, c };
     ...
     if (foo < 0)
        ...

and the static analyzer should not complain about the conditional there.

-Peff