C - Determining alphabetical order of characters/strings - c

Question

I'm trying to write a function that compares two strings (s1 and s2) and works out whether s1 comes before, after or is equal to the s2 string, alphabetically (in the same way as a dictionary is read). If s1 comes before s2 it should return -1. If it's equal to s2 it should return 0. If it comes after s2 it should return 1.
I'm having difficulty getting the function to work - I can only seem to get returns for the first chars in each string and only using the same case. Grateful for any help you can give.
Here's the code so far:
#include <stdio.h>
#include <stdlib.h>
int cmpstr(const char *, const char *);
int main()
{
printf("Test 1: %d\n", cmpstr( "Hello", "World"));
printf("Test 2: %d\n", cmpstr( "Hello", "Hello"));
printf("Test 3: %d\n", cmpstr( "World", "Hello"));
return 0;
}
int cmpstr(const char *s1, const char *s2)
{
/*compare corresponding string characters until null is reached*/
while(*s1 != '\0' && *s2 != '\0' )
{
if (*s1 < *s2)
{
return -1;
}
else if (*s1 > *s2)
{
return 1;
}
else
{
return 0;
s1++;
s2++;
}
}
return 0;
}

Answer 1

just remove the last else part and put return 0 out of loop because both string are only equal if if part and else-if part will not be true, when it will come out from loop it will return 0.
int cmpstr(const char *s1, const char *s2)
{
/*compare corresponding string characters until null is reached*/
while(*s1 != '\0' && *s2 != '\0' )
{
if (*s1 < *s2)
{
return -1;
}
else if (*s1 > *s2)
{
return 1;
}
s1++;
s2++;
}
return 0;
}

Answer 2

Your code has a very obvious mistake, which is the return 0-statement making the s1++;s2++ to unreachable code (your compiler should have warned you about that).
But it has also a conceptual mistake, as it ignores situations where s1 is longer than s2 or vice versa. So in your approach (once corrected the return 0-thing, "Hello" and "Hello there" would compare equal.
See the following code with works in a different manner. It skips equal characters until one (or both) strings has (have) ended. Then, according to this state, result is determined:
int cmpstr(const char *s1, const char *s2)
{
while (*s1 && *s2 && *s1 == *s2) { // move forward until either one of the strings ends or the first difference is detected.
s1++;
s2++;
}
int result = (*s1 - *s2);
// if both strings are equal, s1 and s2 have reached their ends and result is 0
// if *s1 > *s2, s1 is lexographically greater than s2 and result is positive
// if *s1 < *s2, s1 is lexographically lower than s2 and result is negative
// normalize "positive" and "negative" to 1 and -1, respectively
if (result < 0)
result = -1;
else if (result > 0)
result = 1;
return result;
}

Answer 3

Removing 'return 0' in else statement will work. If the chars are equal in same level, you need to look next ones until the equality breaks.
Edit: Also, you need to think about when lengths of strings are not equal.
int cmpstrMY(const char *s1, const char *s2)
{
char sc1, sc2;
/*compare corresponding string characters until null is reached*/
while (1)
{
sc1 = towlower(*s1);
sc2 = towlower(*s2);
if (sc1 == '\0' && sc2 == '\0') {
break;
}
else if (sc1 == '\0' && sc2 != '\0') {
return -1;
}
else if (sc1 != '\0' && sc2 == '\0') {
return 1;
}
else if (sc1 < sc2)
{
return -1;
}
else if (sc1 > sc2)
{
return 1;
}
else
{
s1++;
s2++;
}
}
return 0;
}
Your cmpstr must be something like the code above.