Xref:	news2.mv.net comp.os.msdos.djgpp:577
Newsgroups:	comp.os.msdos.djgpp
From:	mambuhl AT ripco DOT com (Martin Ambuhl)
Subject:	Re: a bug in sscanf?
Message-ID:	<DLMHs3.Ixr@rci.ripco.com>
Sender:	usenet AT rci DOT ripco DOT com (Net News Admin)
Organization:	Ripco Internet BBS Chicago
Date:	Tue, 23 Jan 1996 07:21:38 GMT
Lines:	61
To:	djgpp AT delorie DOT com
DJ-Gateway:	from newsgroup comp.os.msdos.djgpp

"Alexander V. Lukyanov" <lav AT video DOT yars DOT free DOT net>
in <DLL9Hn DOT 8p7 AT jade DOT mv DOT net> wrote:

>I've found some strangeness in sscanf which I suspect to be a bug.
>I'm using v2beta4.

I think that you are wrong.  The key here is that (7.9.6.6) "Reaching
the end of the string is equivalent to encountering end-of-file for the
fscanf function."

If you change your input string to "abc123 ", you will find the behavior
is what you expect.  Notice that the EOF on "abc123" is encountered
before the attempt to use "%n", and (7.9.6.2) "If end-of-file is
encountered during input, conversion is terminated. If end-of-file
occurs before any characters matching the current directive have been
read (other than leading white space, where permitted), execution of the
current directive terminates with an input failure; otherwise, unless
execution of the current directive is terminated with a matching
failure, execution of the following directive (if any) is terminated
with an input failure."  The fact that %n does not consume any input is
irrelevant here.


>Here is a simple test program:
>---8<---
>#include <stdio.h>

>int main()
>{
>   char name[256];
>   int ch_read;

>   printf("sscanf=%d\n",sscanf("abc123","%255[a-zA-Z0-9]%n",name,&ch_read));
>   printf("ch_read=%d\n",ch_read);
>   printf("name=%s\n",name);

>   return 0;
>}
>---8<---

>It outputs:
>---8<---
>sscanf=1
>ch_read=-1058078715
>name=abc123
>---8<---

>That is ch_read is uninitialized and sscanf succseeded! :-(

>I think it is because a bug in doscan.c. It checks 'fileended' variable
>and immediately exits. But in the format string next token is '%n' which does
>not require more input.

>This test works correctly on Solaris2.4.
>---
>Alexander Lukyanov
>lav AT video DOT yars DOT free DOT net
                                              
--
* Martin Ambuhl       net: mambuhl AT ripco DOT com
* Chicago, IL (USA)    

- Raw text -