Mail Archives: djgpp-workers/2003/02/05/08:50:20
Hello.
Eli Zaretskii wrote:
>
> On Tue, 4 Feb 2003, cvs-richdawe AT delorie DOT com wrote:
>
> > Update by cvs id: richdawe
> > Update of /cvs/djgpp/djgpp/include
> >
> > Modified Files:
> > string.h strings.h
> > Log Message:
> > Add C99-specific sections for declarations/definitions
> >
> > ===================================================================
> > RCS file: /cvs/djgpp/djgpp/include/string.h,v
> > retrieving revision 1.6
> > retrieving revision 1.7
> > diff -p -2 -r1.6 -r1.7
> > *** djgpp/include/string.h 19 Jan 2003 13:03:55 -0000 1.6
> > --- djgpp/include/string.h 4 Feb 2003 20:24:20 -0000 1.7
> > *************** char * strtok(char *_s1, const char *_s2
> > *** 48,51 ****
> > --- 48,56 ----
> > size_t strxfrm(char *_s1, const char *_s2, size_t _n);
> >
> > + #if (defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L) \
> > + || !defined(__STRICT_ANSI__)
> > +
> > + #endif /* (__STDC_VERSION__ >= 199901L) || !__STRICT_ANSI__ */
> > +
>
> Richard, did you try to see whether an old GCC version swallows this
> successfully? Like compile a program with GCC 2.7.2.x or 2.8?
[snip]
Er, no. But I don't see why I wouldn't work. gcc 3.x currently defaults to C89
mode, which means it doesn't define __STDC_VERSION__. So when it comes to
evaluating this #if/#endif, gcc 2.x and gcc 3.x should behave the same,
which their default settings (e.g.: no -std=c<whatever> or
-std=gnu<whatever>).
Oh. Or do you mean the line-continuation charactre in the multi-line
#if/#endif?
Bye, Rich =]
--
Richard Dawe [ http://www.phekda.freeserve.co.uk/richdawe/ ]
- Raw text -