CODEWITHSUNDEEP
cstringcomparison
Bogdan M.
Bogdan M.

Reputation: 2181

String comparison C - strcmp()

I'm trying to compare two strings, but I fail achieving that. Why?

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

int main(){
    float a = 1231.23123;
    char b[32];
    sprintf(b, "%f", a);
    printf("%s\n", b);
    char c[32] = "1231.23123";
    if (strcmp(c, b) == 0){
        printf("SUCCES\n");
    }
    else{
        printf("DIFFER\n");
    }
    return 0;
}

Result:

1231.231201
DIFFER

Upvotes: 1

Views: 13203

Answers (4)

LihO
LihO

Reputation: 42083

You are comparing these 2 strings here:

1231.23123
1231.231201

which are different indeed, thus strcmp returns non-zero value.

The actual problem here is that when you do float a = 1231.23123;, the number you want to store in a can't be represented as a float, the nearest number that can be represented as a float is 1231.231201171875 in this case. Have a look at OMG Ponies!!! (Aka Humanity: Epic Fail) ;)

To solve your problem I would start with using double instead of float to get more precise accuracy. Then you could specify the precision (%.5lf) while printing this number into the string to make sure that the number is rounded just like you need it:

double d = 1231.23123;
char str[32];
sprintf(str, "%.5lf", d);
// strcmp(str, "1231.23123") would return 0 here

Upvotes: 2

teppic
teppic

Reputation: 8195

If you want a set number of digits to compare against in a string, use the precision specifier in sprintf - %.5f, and as others have pointed out, the number you've picked cannot be represented by a float, but can be represented by a double. i.e.

double a = 1231.23123;
char b[32];
sprintf(b, "%.5f",a);

Upvotes: 4

NPE
NPE

Reputation: 500357

The two strings are clearly different, so strcmp() is working as it should.

The issue is that 1231.23123 cannot be represented as a float. In fact, the nearest number that can be represented as a float is 1231.231201171875, which is what you're seeing (rounded by sprintf() to six decimal places).

Upvotes: 10

Wangfan Fu
Wangfan Fu

Reputation: 139

It's because the precision of float cannot support so many digits. So b is not "1231.23123". In my test, it's "1231.231201".

Upvotes: 3

Related Questions