mingw-users

[Mingw-users] scanf, fscanf, gets, fgetc, and mostly-every-other-input-function problem

[Felix V. <alp...@ho...>]

Alas, I thought the problem was resolved, but no, it reoccurs:

[test.c file]:
#include <stdio.h>

int main()
{
char a,b,c,d;
int e,f,g,h;
printf("start");
e=fscanf(stdin,"%c",&a);
f=fscanf(stdin,"%c",&b);
g=fscanf(stdin,"%c",&c);
h=fscanf(stdin,"%c",&d);
printf("\n\n\"%c%c%c%c\"=character input\n",a,b,c,d);
printf("%d%d%d%d=fscanf output",e,f,g,h);
}

[entered into command prompt]:
C:\MinGW>gcc test.c -o test.exe

C:\MinGW>test.exe
startu
6

"u
6"=character input
1111=fscanf output
C:\MinGW>test.exe
start


"    "=character input
1111=fscanf output
[the actual characters entered for the first test are 
'u','<return>','<space>', and '6'.  For the second case the characters are 
all spaces.]
[ I have tested this with scanf, gets, fgetc with similar arrangements and 
the results have been the same]
[end]

And that is my problem; I have been told that scanf() stops reading input 
when it encounters a space, but that appears not to be so here.  As I said 
earlier, the scanf and gets functions only work every other time, I believe 
that is because when I hit the <return> key the <return> is automatically 
entered as the input to the next input function.

???

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today - it's FREE! 
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/

Re: [Mingw-users] scanf, fscanf, gets, fgetc, and mostly-every-other-input-function problem

[Jeremy B. <je...@de...>]

>From MSDN:
%c: When used with scanf functions, specifies single-byte character; when 
used with wscanf functions, specifies wide character. White-space characters 
that are ordinarily skipped are read when c is specified. To read next 
non-white-space single-byte character, use %1s; to read next non-white-space 
wide character, use %1ws.

So the program is correctly reading your text.  It reads 'u' '\r' '\n' '6', 
and if you would have used %d in your printf statements you would have seen 
this.

----- Original Message ----- 
From: "Felix VIII" <alp...@ho...>
To: <min...@li...>
Sent: Thursday, July 20, 2006 5:32 PM
Subject: [Mingw-users] scanf, fscanf, gets, fgetc,and 
mostly-every-other-input-function problem


> Alas, I thought the problem was resolved, but no, it reoccurs:
>
> [test.c file]:
> #include <stdio.h>
>
> int main()
> {
> char a,b,c,d;
> int e,f,g,h;
> printf("start");
> e=fscanf(stdin,"%c",&a);
> f=fscanf(stdin,"%c",&b);
> g=fscanf(stdin,"%c",&c);
> h=fscanf(stdin,"%c",&d);
> printf("\n\n\"%c%c%c%c\"=character input\n",a,b,c,d);
> printf("%d%d%d%d=fscanf output",e,f,g,h);
> }
>
> [entered into command prompt]:
> C:\MinGW>gcc test.c -o test.exe
>
> C:\MinGW>test.exe
> startu
> 6
>
> "u
> 6"=character input
> 1111=fscanf output
> C:\MinGW>test.exe
> start
>
>
> "    "=character input
> 1111=fscanf output
> [the actual characters entered for the first test are
> 'u','<return>','<space>', and '6'.  For the second case the characters are
> all spaces.]
> [ I have tested this with scanf, gets, fgetc with similar arrangements and
> the results have been the same]
> [end]
>
> And that is my problem; I have been told that scanf() stops reading input
> when it encounters a space, but that appears not to be so here.  As I said
> earlier, the scanf and gets functions only work every other time, I 
> believe
> that is because when I hit the <return> key the <return> is automatically
> entered as the input to the next input function.
>
> ???
>
> _________________________________________________________________
> Express yourself instantly with MSN Messenger! Download today - it's FREE!
> http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
>
>
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share 
> your
> opinions on IT & business topics through brief surveys -- and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> _______________________________________________
> MinGW-users mailing list
> Min...@li...
>
> You may change your MinGW Account Options or unsubscribe at:
> https://lists.sourceforge.net/lists/listinfo/mingw-users
> 

Re: [Mingw-users] scanf, fscanf, gets, fgetc, and mostly-every-other-input-function problem

[Felix V. <alp...@ho...>]

For some reason when I use "%1s" it only reads the fourth character input 
(d):
[test.c]
#include <stdio.h>

int main()
{
char a,b,c,d;
int e,f,g,h;
printf("start");
e=fscanf(stdin,"%1s",&a);
f=fscanf(stdin,"%1s",&b);
g=fscanf(stdin,"%1s",&c);
h=fscanf(stdin,"%1s",&d);
printf("\n\n\"%c%c%c%c\"=character input\n",a,b,c,d);
printf("%d%d%d%d=fscanf output",e,f,g,h);
}

[command prompt test]:
gcc test.c -o test.exe

test.exe
start abcd [entered 1 space and then 'a','b','c','d','<return>']


"   d"=character input
1111=fscanf output

>From: "Jeremy Bettis" <je...@de...>
>Reply-To: MinGW Users List <min...@li...>
>To: "MinGW Users List" <min...@li...>
>Subject: Re: [Mingw-users] scanf, fscanf, gets, fgetc,and 
>mostly-every-other-input-function problem
>Date: Thu, 20 Jul 2006 17:48:37 -0500
>
> >From MSDN:
>%c: When used with scanf functions, specifies single-byte character; when
>used with wscanf functions, specifies wide character. White-space 
>characters
>that are ordinarily skipped are read when c is specified. To read next
>non-white-space single-byte character, use %1s; to read next 
>non-white-space
>wide character, use %1ws.
>
>So the program is correctly reading your text.  It reads 'u' '\r' '\n' '6',
>and if you would have used %d in your printf statements you would have seen
>this.
>
>----- Original Message -----
>From: "Felix VIII" <alp...@ho...>
>To: <min...@li...>
>Sent: Thursday, July 20, 2006 5:32 PM
>Subject: [Mingw-users] scanf, fscanf, gets, fgetc,and
>mostly-every-other-input-function problem
>
>
> > Alas, I thought the problem was resolved, but no, it reoccurs:
> >
> > [test.c file]:
> > #include <stdio.h>
> >
> > int main()
> > {
> > char a,b,c,d;
> > int e,f,g,h;
> > printf("start");
> > e=fscanf(stdin,"%c",&a);
> > f=fscanf(stdin,"%c",&b);
> > g=fscanf(stdin,"%c",&c);
> > h=fscanf(stdin,"%c",&d);
> > printf("\n\n\"%c%c%c%c\"=character input\n",a,b,c,d);
> > printf("%d%d%d%d=fscanf output",e,f,g,h);
> > }
> >
> > [entered into command prompt]:
> > C:\MinGW>gcc test.c -o test.exe
> >
> > C:\MinGW>test.exe
> > startu
> > 6
> >
> > "u
> > 6"=character input
> > 1111=fscanf output
> > C:\MinGW>test.exe
> > start
> >
> >
> > "    "=character input
> > 1111=fscanf output
> > [the actual characters entered for the first test are
> > 'u','<return>','<space>', and '6'.  For the second case the characters 
>are
> > all spaces.]
> > [ I have tested this with scanf, gets, fgetc with similar arrangements 
>and
> > the results have been the same]
> > [end]
> >
> > And that is my problem; I have been told that scanf() stops reading 
>input
> > when it encounters a space, but that appears not to be so here.  As I 
>said
> > earlier, the scanf and gets functions only work every other time, I
> > believe
> > that is because when I hit the <return> key the <return> is 
>automatically
> > entered as the input to the next input function.
> >
> > ???
> >
> > _________________________________________________________________
> > Express yourself instantly with MSN Messenger! Download today - it's 
>FREE!
> > http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
> >
> >
> > 
>-------------------------------------------------------------------------
> > Take Surveys. Earn Cash. Influence the Future of IT
> > Join SourceForge.net's Techsay panel and you'll get the chance to share
> > your
> > opinions on IT & business topics through brief surveys -- and earn cash
> > 
>http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> > _______________________________________________
> > MinGW-users mailing list
> > Min...@li...
> >
> > You may change your MinGW Account Options or unsubscribe at:
> > https://lists.sourceforge.net/lists/listinfo/mingw-users
> >
>
>
>-------------------------------------------------------------------------
>Take Surveys. Earn Cash. Influence the Future of IT
>Join SourceForge.net's Techsay panel and you'll get the chance to share 
>your
>opinions on IT & business topics through brief surveys -- and earn cash
>http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
>_______________________________________________
>MinGW-users mailing list
>Min...@li...
>
>You may change your MinGW Account Options or unsubscribe at:
>https://lists.sourceforge.net/lists/listinfo/mingw-users

_________________________________________________________________
Don’t just search. Find. Check out the new MSN Search! 
http://search.msn.click-url.com/go/onm00200636ave/direct/01/

Re: [Mingw-users] scanf, fscanf, gets, fgetc, and mostly-every-other-input-function problem

[Gianluca S. <gi...@gm...>]

On 7/21/06, Felix VIII <alp...@ho...> wrote:
> For some reason when I use "%1s" it only reads the fourth character input
> (d):
> [test.c]
> #include <stdio.h>
>
> int main()
> {
> char a,b,c,d;
> int e,f,g,h;
> printf("start");
> e=fscanf(stdin,"%1s",&a);
> f=fscanf(stdin,"%1s",&b);
> g=fscanf(stdin,"%1s",&c);
> h=fscanf(stdin,"%1s",&d);
> printf("\n\n\"%c%c%c%c\"=character input\n",a,b,c,d);
> printf("%d%d%d%d=fscanf output",e,f,g,h);
> }
>

>From the "man printf" page:

s      Matches  a  sequence  of  non-white-space  characters;  the next
        pointer must be a pointer to character array that is long enough
        to  hold  the  input sequence and the terminating null character
        ('\0'), which is added automatically.  The input string stops at
        white  space  or  at  the  maximum field width, whichever occurs
        first.

so I think you are not giving enough room to store the requested 1
character string; try changing
char a,b,c,d;
in
char a[2],b[2],c[2],d[2]; and you will see something different; if you
want to see the correct string from printf, please do not forget to
change the line
printf("\n\n\"%c%c%c%c\"=character input\n",a,b,c,d);
accordingly.

Cheers

Gianluca

Re: [Mingw-users] scanf, fscanf, gets, fgetc, and mostly-every-other-input-function problem

[Keith M. <kei...@to...>]

Gianluca Sforna wrote:
> From the "man printf" page:
>
> s   Matches  a  sequence  of  non-white-space  characters;  the next
>     pointer must be a pointer to character array that is long enough
>     to  hold  the  input sequence and the terminating null character
>     ('\0'), which is added automatically.  The input string stops at
>     white  space  or  at  the  maximum field width, whichever occurs
>     first.

Which, quoted in isolation, is potentially misleading, for it suggests
that if the first char encountered is white space, then the input will
be a null string;  this is *not* the case.

In fact, scanf first skips over any leading white space, *before* it
starts to fetch the string input, and there is no defined mechanism
whereby it can ever return a null (zero length) string.  It *is* true
that parsing stops at the first white-space byte, *after* detecting
the first non-white-space character.  This means that scanf can never
parse a string *containing* white space, with a single `%s' conversion
spec; this is completely inconsistent with the behaviour of `%s' in a
`printf' format spec.

  $ cat scanf-test2.c
  #include <stdio.h>

  int main()
  {
    char c[512];
    scanf( "%511s", c );
    printf( "You entered '%s'\n", c );
    return 0;
  }

  $ gcc -o scanf-test2.exe scanf-test2.c

  $ ./scanf-test2
      this is a test
  You entered 'this'

Incidentally, this little test case can be used to show the importance
of verifying the return value from functions such as scanf, (which this
*doesn't*.  If I run it thus:

  $ ./scanf-test2 < nul
  You entered '<random-garbage-appears-here>'

the unexpected result might lead me to suspect that scanf is broken,
but this is not so, for the standard says:
http://www.opengroup.org/onlinepubs/009695399/functions/scanf.html
| If end-of-file is encountered during input, conversion shall be
| terminated. If end-of-file occurs before any bytes matching the
| current conversion specification (except for %n ) have been read
| (other than leading white-space characters, where permitted),
| execution of the current conversion specification shall terminate
| with an input failure. Otherwise, unless execution of the current
| conversion specification is terminated with a matching failure,
| execution of the following conversion specification (if any) shall
| be terminated with an input failure.

So, this apparent breakage is simply an example of an error, which
*I* should have trapped, but didn't; I should be validating my input,
to avoid such unfortunate surprises:

  $ cat scanf-test3.c
  #include <stdio.h>

  int main()
  {
    char c[512];
    if( scanf( "%511s", c ) == 1 )
      printf( "You entered '%s'\n", c );
    else
      fprintf( stderr, "Input error: no data found\n" );
   return 0;
  }

  $ gcc -o scanf-test3.exe scanf-test3.c

  $ ./scanf-test2
       this is a test
  You entered 'this'

  $ ./scanf-test3 < nul
  Input error: no data found

BTW, that same POSIX Standard reference, cited above, also says:
| Input white-space characters (as specified by isspace()) shall be
| skipped, unless the conversion specification includes a [, c, C, or
| n conversion specifier.

which answers the OP's question about why white-space was not being
skipped, when he used a `%c' conversion spec.

Regards,
Keith.

Re: [Mingw-users] scanf, fscanf, gets, fgetc, and mostly-every-other-input-function problem

[Felix V. <alp...@ho...>]

>BTW, that same POSIX Standard reference, cited above, also says:
>| Input white-space characters (as specified by isspace()) shall be
>| skipped, unless the conversion specification includes a [, c, C, or
>| n conversion specifier.
>
>which answers the OP's question about why white-space was not being
>skipped, when he used a `%c' conversion spec.
>
>Regards,
>Keith.
>

So which format specifier should I use?  '%1s' or ' %c', or the example you 
used:

>#include <stdio.h>
>
>int main()
>{
>  char c[512];
>  if( scanf( "%511s", c ) == 1 )
>    printf( "You entered '%s'\n", c );
>  else
>    fprintf( stderr, "Input error: no data found\n" );
>return 0;
>}

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today - it's FREE! 
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/

Re: [Mingw-users] scanf, fscanf, gets, fgetc, and mostly-every-other-input-function problem

[Keith M. <kei...@to...>]

Felix VIII wrote:
> So which format specifier should I use?

Only *you* can answer that, for it depends on what you are
trying to achieve.  In the simple example you posted, where
you want to skip white space, then fetch one character, I'd
tend to favour:

  int c, status;
  if( (status = scanf( " %c", &c )) == 1 )
    /* do something */;
  else
    /* handle the `no input' error */;

You've been pointed to plenty of references, which should help
you to decide; read them and make your own choice.  If you need
clarification on M$'s implementation, (which Mingw uses), then
you should go here:
http://msdn.microsoft.com/library/default.asp

and search for whatever function you are unsure about.

Regards,
Keith.

Re: [Mingw-users] scanf, fscanf, gets, fgetc, and mostly-every-other-input-function problem

[Felix V. <alp...@ho...>]

>From: Keith MARSHALL <kei...@to...>
>Reply-To: Keith MARSHALL <kei...@to...>,MinGW Users List 
><min...@li...>
>To: MinGW Users List <min...@li...>
>Subject: Re: [Mingw-users] scanf, fscanf, gets, fgetc, and 
>mostly-every-other-input-function problem
>Date: Fri, 21 Jul 2006 16:54:27 +0100
>
>Felix VIII wrote:
> > So which format specifier should I use?
>
>Only *you* can answer that, for it depends on what you are
>trying to achieve.

I was expecting that ;-)

>In the simple example you posted, where
>you want to skip white space, then fetch one character, I'd
>tend to favour:
>
>   int c, status;
>   if( (status = scanf( " %c", &c )) == 1 )
>     /* do something */;
>   else
>     /* handle the `no input' error */;
>
>You've been pointed to plenty of references, which should help
>you to decide; read them and make your own choice.  If you need
>clarification on M$'s implementation, (which Mingw uses), then
>you should go here:
>http://msdn.microsoft.com/library/default.asp
>
>and search for whatever function you are unsure about.
>
>Regards,
>Keith.
>

It seems that " %c" seems to solve my problems, so thank you.
Felix.

_________________________________________________________________
Don’t just search. Find. Check out the new MSN Search! 
http://search.msn.click-url.com/go/onm00200636ave/direct/01/

Re: [Mingw-users] scanf, fscanf, gets, fgetc, and mostly-every-other-input-function problem

[Keith M. <kei...@to...>]

Jeremy Bettis wrote:
> ... To read next non-white-space single-byte character, use %1s;
> to read next non-white-space wide character, use %1ws.

Actually, this probably isn't the best advice.

The OP's code had (my comments appended):

  char c;             /* one byte of storage */
  scanf( "%c", &c );  /* fetch next byte from stdin */

Now, this will fetch the next available byte from stdin, and store
it in `c'.  If that next byte happens to be a space, then `c' will
become a space.

However, if you change that format string to `"%1s"', as you suggest,
then scanf will try to fetch a *string* of one byte in length, but it
will also append a NUL, to terminate the string, so:

  char c;              /* only one byte of storage */
  scanf( "%1s", &c );  /* this is a buffer overrun...
                        * one byte is fetched into `c', then a
                        * terminating NUL is appended, in the
                        * following memory address, where it
                        * DEFINITELY doesn't belong! */

If you want to fetch single character input, but ignore leading white
space, then you could use:

  char c;              /* one byte of storage */
  scanf( " %c", &c );  /* fetch next character from stdin,
                        * ignoring leading white space, (if any).
                        * no buffer overrun, because at most one
                        * byte is stored. */

(note the space in the format string, *before* the `%c' conversion
 spec; this causes scanf to skip *all* leading white space).

Thus:

  $ cat scanf-test.c
  #include <stdio.h>

  int main()
  {
    char c;
    scanf( " %c", &c );
    printf( "You entered a '%c'\n", c );
    return 0;
  }

  $ gcc -o scanf-test scanf-test.c

  $ ./scanf-test
        zzzzzzz
  You entered a 'z'

which seems to be what the OP wanted.

Regards,
Keith.