segfault on strtok

[Fatih Gey]

Hi,

.. following causes a segfault. .. didn't
know why ?!

int main() {
char name[15];
strcpy (name, "ab8bc8cd8e d");

char cur[800];
strcpy (cur, strtok(name, "8"));

while (cur) {
printf ("Output: %s\n", cur);
printf ("Stringleng th %i\n", strlen(cur));
strcpy(cur,strt ok(0, "8"));
printf ("next\n");
}

}

output: -----------------------

Output: ab
Stringlength 2
next
Output: bc
Stringlength 2
next
Output: cd
Stringlength 2
next
Output: ed
Stringlength 2
Segmentationfau lt
------------------------------

Thanks for any help
Regards
Fatih Gey

[Al Bowers]

Fatih Gey wrote:

Hi,

.. following causes a segfault. .. didn't
know why ?!

int main() {
char name[15];
strcpy (name, "ab8bc8cd8e d");

char cur[800];
strcpy (cur, strtok(name, "8"));

while (cur) {
printf ("Output: %s\n", cur);
printf ("Stringleng th %i\n", strlen(cur));
strcpy(cur,strt ok(0, "8"));
printf ("next\n");
}

}

Once all the delimiters, '8' in this case, are exhausted, function
strtok will return NULL. At the point of the seg fault your
statement:
strcpy(cur,strt ok(0,"8"));
becomes
strcpy(cur,NULL );
The second argument NULL is not defined for function strcpy.
The argument must be a string. Therein lies your seg fault.

Actually, if your intent is to only print the tokens, you should
just use a char * instead of using the character array cur.

#include <string.h>
#include <stdio.h>

int main(void)
{
char name[15], *cur;

strcpy (name, "ab8bc8cd8e d");
cur = strtok(name, "8");
while(cur)
{
printf ("Output: %s\n", cur);
printf ("Stringleng th %u\n", strlen(cur));
printf ("next\n");
cur = strtok(NULL,"8" );
}
}

--
Al Bowers
Tampa, Fl USA
mailto: xa******@myrapi dsys.com (remove the x to send email)
http://www.geocities.com/abowers822/

[Andreas Kahari]

In article <3F***********@ mindspring.com> , pete wrote:

Andreas Kahari wrote:

Yes, the integer zero is a null pointer, but only if it's cast
to void * and converted into a pointer type.

A constant expression with a value of zero,
is called a "null pointer constant"
(you can find that phrase in the standard),
whther or not it's cast to (void*).

I was looking at the standard when I wrote it. It says that an
integer which is zero is a null pointer constant if cast to void *.
That null pointer constant is a null pointer if converted to
a pointer type.

--
Andreas Kähäri

[pete]

Andreas Kahari wrote:

In article <3F***********@ mindspring.com> , pete wrote:

Andreas Kahari wrote:

Yes, the integer zero is a null pointer, but only if it's cast
to void * and converted into a pointer type.

A constant expression with a value of zero,
is called a "null pointer constant"
(you can find that phrase in the standard),
whther or not it's cast to (void*).

I was looking at the standard when I wrote it. It says that an
integer which is zero is a null pointer constant if cast to void *.
That null pointer constant is a null pointer if converted to
a pointer type.

I read it differently from the way that you do.

N869
6.3.2.3 Pointers
[#3]
An integer constant expression with the value 0,
or such an expression cast to type void *,
is called a null pointer constant.

--
pete

[Andreas Kahari]

In article <3F**********@m indspring.com>, pete wrote:

Andreas Kahari wrote:

[cut]

I was looking at the standard when I wrote it. It says that an
integer which is zero is a null pointer constant if cast to void *.
That null pointer constant is a null pointer if converted to
a pointer type.

I read it differently from the way that you do.

N869
6.3.2.3 Pointers
[#3]
An integer constant expression with the value 0,
or such an expression cast to type void *,
is called a null pointer constant.

That's the draft standard. I have the book version of the
current standard, but it's at home at the moment. I'll have
to get back to this issue later tonight if noone else has the
current standard at hand and can verify this.
Cheers,
Andreas

--
Andreas Kähäri

[Al Bowers]

Andreas Kahari wrote:

In article <3F***********@ mindspring.com> , pete wrote:

Andreas Kahari wrote:

Yes, the integer zero is a null pointer, but only if it's cast
to void * and converted into a pointer type.

A constant expression with a value of zero,
is called a "null pointer constant"
(you can find that phrase in the standard),
whther or not it's cast to (void*).

I was looking at the standard when I wrote it. It says that an
integer which is zero is a null pointer constant if cast to void *.
That null pointer constant is a null pointer if converted to
a pointer type.

I believe what you are referring to in the standard is the
section 6.3.2.3 Pointers.
Paragraph:
3. An integer constant expression with the value 0, or such an
expression cast to type void *, is called a null pointer constant.

Notice the word "or". It does not say "only if cast".
Among other posibilities, NULL may be defined as:
0, 0L, 0LL, or (void *)0.

---
Al Bowers
Tampa, Fl USA
mailto: xa******@myrapi dsys.com (remove the x to send email)
http://www.geocities.com/abowers822/

[Sheldon Simms]

On Fri, 24 Oct 2003 12:55:34 +0000, Andreas Kahari wrote:

In article <3F**********@m indspring.com>, pete wrote:

Andreas Kahari wrote:

[cut]

I was looking at the standard when I wrote it. It says that an
integer which is zero is a null pointer constant if cast to void *.
That null pointer constant is a null pointer if converted to
a pointer type.

I read it differently from the way that you do.

N869
6.3.2.3 Pointers
[#3]
An integer constant expression with the value 0,
or such an expression cast to type void *,
is called a null pointer constant.

That's the draft standard. I have the book version of the
current standard, but it's at home at the moment. I'll have
to get back to this issue later tonight if noone else has the
current standard at hand and can verify this.

ISO/IEC 9899:1999
6.3.2.3 Pointers
3 An integer constant expression with the value 0, or such an
expressioncast to type void *, is called a null pointer
constant.

[Paul Hsieh]

"Fatih Gey" <fa*******@2way .de> wrote:

...
char cur[800];
strcpy (cur, strtok(name, "8"));

while (cur) {
This is a constant loop -- i.e., the loop will never end.
printf ("Output: %s\n", cur);
printf ("Stringleng th %i\n", strlen(cur));
strcpy(cur,strt ok(0, "8"));
strtok eventually will return NULL. strcpy cannot accept a parameter which is NULL.
printf ("next\n");
}
...

--
Paul Hsieh
http://www.pobox.com/~qed/
http://bstring.sf.net/

[Andreas Kahari]

In article <pa************ *************** *@yahoo.com>, Sheldon Simms wrote:

On Fri, 24 Oct 2003 12:55:34 +0000, Andreas Kahari wrote:

[cut]

That's the draft standard. I have the book version of the
current standard, but it's at home at the moment. I'll have
to get back to this issue later tonight if noone else has the
current standard at hand and can verify this.

ISO/IEC 9899:1999
6.3.2.3 Pointers
3 An integer constant expression with the value 0, or such an
expressioncast to type void *, is called a null pointer
constant.

Yeah yeah, alright.

I wonder what they mean by "or such an expression" though...
--
Andreas Kähäri

[Dan Pop]

In <sl************ **********@vinl and.freeshell.o rg> Andreas Kahari <ak*******@free shell.org> writes:

In article <3F**********@m indspring.com>, pete wrote:

Andreas Kahari wrote:

[cut]

I was looking at the standard when I wrote it. It says that an
integer which is zero is a null pointer constant if cast to void *.
That null pointer constant is a null pointer if converted to
a pointer type.

I read it differently from the way that you do.

N869
6.3.2.3 Pointers
[#3]
An integer constant expression with the value 0,
or such an expression cast to type void *,
is called a null pointer constant.

That's the draft standard.

It doesn't matter: the text hasn't changed at least since the last
public C89 draft, released 15 years ago.

It makes sense to object to N869 quotations *only* when you know that
either C89 or C99 say something different.

Dan
--
Dan Pop
DESY Zeuthen, RZ group
Email: Da*****@ifh.de

[Andreas Kahari]

In article <bn**********@s unnews.cern.ch> , Dan Pop wrote:

In <sl************ **********@vinl and.freeshell.o rg> Andreas
Kahari <ak*******@free shell.org> writes:

[cut]

That's the draft standard.

It doesn't matter: the text hasn't changed at least since the last
public C89 draft, released 15 years ago.

It makes sense to object to N869 quotations *only* when you know that
either C89 or C99 say something different.

I agree, but I thought I remembered the text differently.
Others proved me wrong though.
--
Andreas Kähäri