On 04/04/2012 11:18 AM, Corinna Vinschen wrote:
On Apr 4 16:54, Denis Excoffier wrote:
On Wed, Apr 04, 2012 at 10:02:24AM -0400, Ryan Johnson wrote:
On 04/04/2012 9:32 AM, Denis Excoffier wrote:
Hello,
It could be that snprintf() is not properly declared in<stdio.h>.
According to [1], it's not officially part of c++98 (???). Try
gnu++98 instead. As for why it's not in c++0x, there's a problem
with the macros being defined [2] that AFAIK remains unresolved;
again the workaround is gnu++0x.
[1] http://cygwin.com/ml/cygwin/2010-01/msg00791.html
[2] http://cygwin.com/ml/cygwin/2011-08/msg00311.html
Oh, i should have found these, at least the first one... For c++0x,
couldn't this be addressed now? As a start, i can (humbly) propose:
--- stdio.h 2012-04-04 14:50:32.000000000 +0159
+++ stdio.h++ 2012-04-04 16:38:21.049273700 +0159
@@ -237,7 +237,7 @@
off_t _EXFUN(ftello, ( FILE *));
#endif
#endif
-#if !defined(__STRICT_ANSI__) || (__STDC_VERSION__>= 199901L)
+#if !defined(__STRICT_ANSI__) || (__STDC_VERSION__>= 199901L) ||
(__cplusplus>= 201103L)
How is that supposed to work?
$ gcc -xc++ -std=c++98 -dM -E -< /dev/null | grep cplus
#define __cplusplus 1
$ gcc -xc++ -std=c++0x -dM -E -< /dev/null | grep cplus
#define __cplusplus 1
Actually I'm wondering if that's not a bug in gcc:
$ gcc -xc -std=c89 -dM -E -< /dev/null | grep STDC
#define __STDC_HOSTED__ 1
#define __STDC__ 1
$ gcc -xc -std=c99 -dM -E -< /dev/null | grep STDC
#define __STDC_HOSTED__ 1
#define __STDC_VERSION__ 199901L
#define __GNUC_STDC_INLINE__ 1
#define __STDC__ 1
[~]$ gcc -xc++ -std=c++98 -dM -E -< /dev/null | grep STDC
#define __STDC_HOSTED__ 1
#define __STDC__ 1
[~]$ gcc -xc++ -std=c++0x -dM -E -< /dev/null | grep STDC
#define __STDC_HOSTED__ 1
#define __STDC__ 1
Shouldn't -std=c++0x include the definitions for -std=c99, namely
#define __STDC_VERSION__ 199901L
???
I figured it was a gcc bug from the start, but I don't know enough about
the arcane art of juggling all possible combinations of __STDC__,
__ANSI__, etc. and wouldn't know the first thing about a fix. I just
live with -std=gnu++$WHATEVER for now.
Ryan
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
