Strcmp function.

[Mr.Lnx]

Hello again. I have some questions here about strcmp function.

For instance... let's go in some code.

Assuming that we have

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

int main(void)
{
	
	if(strcmp("22" , "17 December Testing") < 0)
	puts("Ok1");	
	
	printf("\n%d" , strcmp("25" , "23 December Testing"));	
	
	return 0;
}

I have read from the theory when we have a string that is shorter from one other is also less from this. For example if we have "abc" and "abcd" the first literal is less than the second literal.

However I have the need to discuss what can I do when I have literals as previous. For example "25" and "23 December Testing" on this we take the value 1 or some value greater than 0 hence first literal is not less than second.

Anyway the question is .... what it occurs when we have two strings that do not match their characters (the opposite case from "abc" , "abcd") and the second or the first is shorter.... as I wrote before .

So far I believe that strcmp will compare the first two 25 and 23 ... 25 > 23 ... so it returns with the > 0 value... I am right ?

Thank you in advance for your time.

Mr. Lnx

[laserlight]

You can think of string length as a "tie breaker", e.g., strcmp("ab", "a") > 0 but strcmp("ab", "b") < 0 since 'a' < 'b', assuming ASCII or the like.

[iMalc]

The rule about length that your referring to only applies when one string is a prefix of the other.

[Mr.Lnx]

The rule about length that your referring to only applies when one string is a prefix of the other.

Give one example please....

According to laserlight I checked some examples on myown.

 	printf("\n%d\n" , strcmp("f" , "ejh "));

Here strcmp compares f and e at first f has greater ASCII value than e .... then we have the '\0' character of the first literal so the work of strcmpy ends here?
Either or not the second literal "ejh" has more than n-1 character from the first .

what about when we have

 printf("\n%d\n" , strcmp("  a" , "a  "));

strcmp compares space with a then space with space then a with space ???

[c99tutorial]

     printf("\n%d\n" , strcmp("f" , "ejh "));

Here strcmp compares f and e at first f has greater ASCII value than e .... then we have the '\0' character of the first literal so the work of strcmpy ends here?

The comparison ends at the first mismatch -- its that simple. What's all this about the length of the string and such...

Compare "f" and "ejh ":

Index 0: 'f' > 'e' therefore strcmp ends with result: >0 (positive integer for result means the left is greater than the right)

By the way, you can get the result of strcmp by just doing a simple calculation on the first letters that mismatch: 'f'-'e' obviously results in a positive number, so the result from strcmp will be positive. If they are equal... the comparison continues.

Compare "aaa" and "aa"

Index 0: 'a' == 'a'
Index 1: 'a' == 'a'
Index 2: 'a' > '\0', so again the comparison ends with a positive number because 'a'-'\0' is positive

Compare " a" and "a ":

Index 0: ' ' < 'a', therefore the comparison ends with a negative number because ' '-'a' is negative

[Mr.Lnx]

Very good explanation thank you but I have some last questions....

Consider "22" and "a" will give negative result from strcmp ..... I think function compares 'a' with '2' not 'a' with '22' ... and then ends

that is the same with "22" and "33" ? I mean again will compare with '2' and '3' and '2' and '3' ???

[nonoob]

Yes, character by character. strcmp does not interpret decimal numbers (multiple digits) in any more intelligent form. It just goes by comparing individual corresponding characters according to their magnitudes in the ASCII chart ( Ascii Table - ASCII character codes and html, octal, hex and decimal chart conversion ).

Compare '2' with 'a': '2' is 50, and 'a' is 97. 50 - 97 is negative.

In your second example, no it will not "compare with '2' and '3' and '2' and '3' ". It will do the first comparison and stop because the characters are different. There is no need to check further.

[Mr.Lnx]

Very well. Thank you ... it is clear now

EDIT: Sorry me guys.... really something last :P

"abc" , "abcd"

The comparisons are:

'a'=='a'
'b' == 'b'
'c' == 'c'
'\0' < 'd'

so negative result ?! ok

forget it..... is the same example with "aaa" and "aa" as c99tutorial wrote before... thanks again.

[iMalc]

The ordering strcmp uses is called Lexicographical Ordering.
Please have a read of the wikipedia article on it: Lexicographical order - Wikipedia, the free encyclopedia