This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: backslash-newline, trigraphs, and the C standard
On Wed, Jan 12, 2000 at 09:29:13PM -0800, Mark Mitchell wrote:
> >>>>> "Zack" == Zack Weinberg <zack@wolery.cumb.org> writes:
>
> Zack> Performance is not the main reason why I want to make this
> Zack> change; it's to be able to silence trigraph warnings in
> Zack> comments. There are definitely better places to look for
> Zack> performance improvements.
>
> OK -- understood.
>
> The whole trigraph issue is touchy. Some people really do use them.
>
> I've always thought that they should be on, by default, not off;
> however, I'm not going to try to open that can of worms here.
>
> It would seem to me, though, that a sensible thing would be to not
> warn about trigraphs in comments when not handling trigraphs -- and
> that's the current default situation. I guess I'm missing something.
The current default situation depends which preprocessor you're using.
cccp doesn't warn about trigraphs anywhere even with -Wall, and -Wtrigraphs
just gets you "warning: 5 trigraphs somewhere in this file". cpplib,
on the other hand, issues one warning per trigraph with -Wall.
It is possible that I can rearrange things enough to fix this without
breaking standards conformance - if so I will of course do it that way.
zw