pointers, structures, and malloc

I switched compilers on a lark. I've gotten rid of my error messages. I have this code:

Code:

---

  struct sdata_2 {
      char d21;
      char d22;
      char d23;
      char d24;
      char d2x[20];

    struct sdata_2 sdata2 = { 'a',  'b', 'c', 'd', 'Mike' }

---

The compiler is very upset over Mike. It does not like multi-byte initial values. Since d2x is defined as a string of bytes, how do I get an initial value?

TIA

try using a string rather than a multibyte character (which possibly would match an integer type, but certainly not a string):

Code:

---

struct sdata_2 sdata2 = { 'a',  'b', 'c', 'd', "Mike" }

---

--
Mats

char is a single character, so 'use these'. char[20] is a bunch of characters, so "use these". The added advantage of the string is that it is properly null-terminated for you.

I assume that the missing '}' in the above is just a copy and paste error, too.

Quote:

---

Originally Posted by lugnut

The compiler is very upset over Mike. It does not like multi-byte initial values. Since d2x is defined as a string of bytes, how do I get an initial value?

TIA

---

The problem is that 'Mike' is an integer, aka int. 'Mike' is 'M' + 'i' + 'k' + 'e' (or some variant, not sure), which certainly does not fit into a char.
Double quotes, on the other hand, is an array of chars, so it fits where single quotes does not.

> The problem is that 'Mike' is an integer, aka int. 'Mike' is 'M' + 'i' + 'k' + 'e' (or some variant, not sure), which certainly does not fit into a char.
Simpler than that. It's syntactically invalid :)

> Double quotes, on the other hand, is an array of chars
is a nul-terminated array of chars
Very, Very, Very, Important :)

Just remember, anything within single quotes must evalute to a single byte. Which of course has an integer value.

Quote:

---

Originally Posted by zacs7

> The problem is that 'Mike' is an integer, aka int. 'Mike' is 'M' + 'i' + 'k' + 'e' (or some variant, not sure), which certainly does not fit into a char.
Simpler than that. It's syntactically invalid :)

---

Are you sure?

Code:

---

#include <iostream>

using namespace std;

int __cdecl main()
{
        unsigned int x = '1234';
        cout << "0x" << hex << x << endl;
}

---

0x31323334

Code:

---

[zac@neux ~]$ cat ln.c
#include <stdio.h>

int main(void)
{
  unsigned int a = '1234';
  unsigned int b = '1' + '2' + '3' + '4';

  printf("&#37;X & %u\n", a, a);
  printf("%X & %u\n", b, b);
  return 0;
}
[zac@neux ~]$ gcc -Wall -pedantic -std=c89 ln.c -o ln
ln.c:5:12: warning: multi-character character constant
[zac@neux ~]$ ./ln
31323334 & 825373492
CA & 202

---

Turns out I was wrong :P. I wonder what it means then? :confused:

Undefined behavior!!!! %X with an int! should be an unsigned int!! woop! *siren sounds*

I'm gonna go do something else for a while.

Quote:

---

Originally Posted by zacs7

Code:

---

[zac@neux ~]$ cat ln.c
#include <stdio.h>

int main(void)
{
  unsigned int a = '1234';
  unsigned int b = '1' + '2' + '3' + '4';

  printf("%X & %u\n", a, a);
  printf("%X & %u\n", b, b);
  return 0;
}
[zac@neux ~]$ gcc -Wall -pedantic -std=c89 ln.c -o ln
ln.c:5:12: warning: multi-character character constant
[zac@neux ~]$ ./ln
31323334 & 825373492
CA & 202

---

Turns out I was wrong :P. I wonder what it means then? :confused:

---

Conveniently, '1' is 0x31, '2' is 0x32, '3' is 0x33, and '4' is 0x34