[civhr@civ.hw.ac.uk] libc/1178: Multiple declaration of strnlen

[Andreas Jaeger <aj@arthur.rhein-neckar.de>]

This small program:
#include <string.h>

int
main (void)
{
}

shows the problem reported in the appended bug report if you compile
with:

gcc -D_GNU_SOURCE -O2 -Wall -W -Wredundant-decls -c  strlen-test.c 
In file included from /usr/include/string.h:346,
                 from strlen-test.c:1:
/usr/include/bits/string2.h:946: warning: redundant redeclaration of `strnlen' in same scope
/usr/include/string.h:212: warning: previous declaration of `strnlen'

What shall we do?

Andreas

---

• Subject: libc/1178: Multiple declaration of strnlen
• From: civhr@civ.hw.ac.uk
• Date: Fri Jun 25 16:17:55 1999

Topics:
   libc/1178: Multiple declaration of strnlen()


----------------------------------------------------------------------

Date: Fri, 25 Jun 1999 06:25:26 -0400
From: civhr@civ.hw.ac.uk
To: bugs@gnu.org
Subject: libc/1178: Multiple declaration of strnlen()
Message-Id: <199906251025.GAA27125@delysid.gnu.org>


>Number:         1178
>Category:       libc
>Synopsis:       Multiple declaration of strnlen()
>Confidential:   no
>Severity:       non-critical
>Priority:       medium
>Responsible:    libc-gnats
>State:          open
>Class:          sw-bug
>Submitter-Id:   unknown
>Arrival-Date:   Fri Jun 25 06:30:01 EDT 1999
>Last-Modified:
>Originator:     civhr@civ.hw.ac.uk
>Organization:
net
>Release:        2.1.1
>Environment:
Linux 2.2.10, RedHat 6.0 stock
>Description:
With using the right #defines (__USE_GNU etc.), strnlen() gets defined by both
/usr/include/string.h and /usr/include/bits/string2.h 
(if __NO_STRING_INLINES is not defined).
This results in annoying warning messages from gcc.

Wouldn't it make sense to change (/usr/include/string.h line 203)

#ifdef __USE_GNU
/* Find the length of STRING, but scan at most MAXLEN characters.
   If no '\0' terminator is found in that many characters, return MAXLEN.  */
extern size_t strnlen __P ((__const char *__string, size_t __maxlen));
#endif

into

#if defined(__USE_GNU) && defined(__NO_STRING_INLINES)
...
?
That at least keeps gcc happy. Or I'm missing something subtle ?

HTH,

Hannes
>How-To-Repeat:

>Fix:
>Audit-Trail:
>Unformatted:


------------------------------

End of forward__FwPH Digest
***************************

---

-- 
 Andreas Jaeger   aj@arthur.rhein-neckar.de    jaeger@informatik.uni-kl.de
  for pgp-key finger ajaeger@aixd1.rhrk.uni-kl.de