This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
RE: Warnings about rcs_id strings: let's settle this
- From: "Kean Johnston" <jkj at sco dot com>
- To: "'Zack Weinberg'" <zack at codesourcery dot com>
- Cc: "'Joe Buck'" <jbuck at synopsys dot com>, <gcc at gcc dot gnu dot org>
- Date: Fri, 2 May 2003 11:31:05 -0700
- Subject: RE: Warnings about rcs_id strings: let's settle this
- Organization: The SCO Group
> Right, but to reiterate, GCC has warned about *that* construct since
> forever; anyone using -Wall would have had problems already.
Except people newly using gcc on very large existing legacy source bases
:)
I want to make sure the warning controls whats expected, not what may
already have been worked / hacked / designed around. In in the spirit
of the INTENT, I don't think the const-ness should be a factor. UNLESS
...
If we *really* want to help the user, and give them maximum control,
we could turn this into a two-flag deal, and have
-Wunused-static-variable
and -Wunused-static-const-variable. But I am not really proposing that.
I just don't think const should be a factor, because by FAR the largest
problem area is legacy code, not necessarily code that gets a different
warning between 3.3 version 3.2. That simply exposes the problem, it
doesn't define it.
Kean