From:	ai26 AT sternwarte DOT uni-erlangen DOT de (Michael Lemke, Sternwarte Bamberg, Phone: +49-951-9522216)
Subject:	RE: strlen on NULL, final test
15 May 1998 15:00:35 -0700 :
Message-ID:	<009C62C2.6ED0BF6E.18.cygnus.gnu-win32@sternwarte.uni-erlangen.de>
To:	GNU-WIN32%CYGNUS DOT COM AT bamp01 DOT sternwarte DOT uni-erlangen DOT de
Cc:	ai26 AT sternwarte DOT uni-erlangen DOT de
X-Unparseable-Date:	Thu, 14 May 1998 13:12:38 MET-1MET DST

In a message of Wed, 13 May 1998 10:37:34 -0400 (EDT)
Received on Thu, 14 May 1998 11:52:22 MET-1MET DST

Ron G. Minnich <rminnich AT sarnoff DOT com> wrote to gnu-win32 AT cygnus DOT com

>to finish this off once and for all, someone raised the question of a
>strlen that might test the argument to see if it's equal to 0, and then
>return 0. 
>
>First, the empty string and the null string are different, and length of 
>a null string has no meaning. 
>Second, if you have null pointers, as others say, fix your code. 
>
>But, even though it's a bad idea, who knows: someone might do it. 
>
>bash$ cat test.c
>main()
>{
>  (void) strlen((char *) 0);
>}
>
>bash$ cc test.c
>bash$ ./a.out
>Segmentation fault (core dumped)
>bash$ 
>
>Again: linux, freebsd, sunos, solaris.
>

Another data point:

LUCKY> cc x

  (void) strlen((char *) 0);
..........^
%CC-I-IMPLICITFUNC, In this statement, the identifier "strlen" is implicitly dec
lared as a function.
at line number 3 in file TMP:[AI26]X.C;1
LUCKY> link x
LUCKY> run x
%SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual address=000000000000
0000, PC=FFFFFFFF8050E8E8, PS=0000001B
%TRACE-F-TRACEBACK, symbolic stack dump follows
  image    module    routine             line      rel PC           abs PC
                                            0 0000000000000000 FFFFFFFF8050E8E8
 X  X  main                                 3 00000000000000C8 00000000000200C8
 X  X  __main                               0 0000000000000070 0000000000020070
                                            0 FFFFFFFF827530F8 FFFFFFFF827530F8
LUCKY> sh sys/noproc/fu
OpenVMS V7.1-1H1  on node LUCKY  14-MAY-1998 13:03:52.43  Uptime  1 01:31:06
        Digital AlphaPC 164 500 MHz
LUCKY> cc/vers
DEC C V5.7-004 on OpenVMS Alpha V7.1-1H1



--
Michael Lemke
Sternwarte Bamberg, University of Erlangen-Nürnberg, Germany
(lemke AT sternwarte DOT uni-erlangen DOT de or michael AT astro DOT as DOT utexas DOT edu)
-
For help on using this list (especially unsubscribing), send a message to
"gnu-win32-request AT cygnus DOT com" with one line of text: "help".

- Raw text -