Thread: Casting character point into structure

Originally Posted by tabstop

If by "cast from string to integer" you mean "using the location in memory of the string as an integer", then fine.

Which AFAICT that is *not* what is happening. In that case, it might as well be:

Code:

Uint16 Type="";

Originally Posted by shaileshk

This is not string but integer but all the problem happening is am unable to print the value of integer structure member and thats why I am trying to Flip the byte and getting it to work.

To be honest, I am not sure what you are really trying to do. What is the bigger picture of the problem that you are trying to solve?

If you believe that "12594678901234567890testnode123456" is an integer, there's not a lot we can do for you.

If you want the integer to have the value of (the first part of) the contents of that string, you need to use strtol to convert the string to a long. The assignment, as you have it, will use the value of "12594678901234567890testnode123456" (note that the value of the string is its address, unrelated to the contents of the string itself) as the value for your variable.

Originally Posted by shaileshk

I have checked the value of Type using gdb as debugger on core file and the value of Type is "12594678901234567890testnode123456"

This may be true in your program, it does not mean that is a good thing, or that it will not cause an error.

You can (by mistake or on purpose) put an incorrect or "dangerous" value into a variable (such as one that is too large for the type), and the debugger will show you this value. It is *your job* as the programmer using the tool to recognize that "12594678901234567890testnode123456" DOES NOT BELONG in an unsigned 16-bit integer. Hopefully you know how that value got there...

Originally Posted by shaileshk

What it is trying to do and as mentioned it dumps the code when changing the value of it.

Yes, very possibly because you put way too big a value into "Type" to start with, which is guaranteed to cause an overflow. The consequences of that may not be immediately apparent, but it probably will not take long for a segfault to occur, during what ever set of operations happens to occur next.

Originally Posted by shaileshk

I havent put this value purposefully but when casting from string to structure member it got inside directly... is there any way to stop this happening...

Yes. Do not cast "from string to structure member". Maybe that is a correct solution, but without the bigger picture of what you are trying to solve we may well be helping you on a wild goose chase.

Originally Posted by shaileshk

I havent put this value purposefully but when casting from string to structure member it got inside directly... is there any way to stop this happening...

Going back to your Original Post:

Code:

typedef struct
{
  char a[10];
  int b;
  char c[10];
}MyStruct;

int main()
{
char *ptr;
MyStruct *pstruct;

/* or C style casting: */
 memset(&ptr, '\0', sizeof(1972));

 ptr = "12394678901234567890testnode123456\0 eventry\0        567\0y8901234567z8";

pstruct = (MyStruct *) ptr;

fprintf(stderr, "value of b %u\n", pstruct->b);

return 0;
}

Understand that the value of pstruct->b WILL NOT be 1234[...], although you may think that.

On my system, an int is four bytes. A C string is a series of one byte characters. So you have cast a string into MyStruct, the first ten bytes of which is a char array. Then you have an int, four bytes, which those four bytes will be the characters '1','2','3','4'. According to the ASCII table, the *integer* value of those four characters are

1 = 49
2 = 50
3 = 51
4 = 52

These are signed values, so the bits set for each byte are
0011 0001
0011 0010
0011 0011
0011 0100
On a little endian machine, the first byte is the least significant, so it has the same value, 49. The next byte's bits represent

12686 64384 32192 16096 4048 1024 512 256

so 0011 0010 will add 32192+16096+512 to the total, so we now have 48,849 and the most significant bytes are still to come.

Do you see what happened?