strcmp returning 1...

why is strcmp returning 1 in this case?? doesn't the contents of "somestring" match "test".
I've tried removing the trailing \n's that fgets puts in but still the same. I input test when prompted.

Code:

---


#include <stdio.h>
#include <string.h>
#include <ctype.h>
int main()
{
        char somestring[100];
       
        printf("Please enter a string\n");

    fgets(somestring, 100, stdin);
   
    //the following will rebuild the string to get rid of any new lines so we can compare
    //it with "test".
       
        int i;
        for (i=0; i < strlen(somestring)-1; i++)
        {
                if (somestring[i] != '\n'){
                        somestring[i] = somestring[i];
                }
        }
       
        printf("\n%d", strcmp(somestring, "test"));
       
        return 0;
}

---

Code:

---

for (i=0; i < strlen(somestring)-1; i++)

---

Because of the -1, this loop will never loop over the '\n' and you won't have a chance to remove it. Get rid of the -1, and perhaps save the result of strlen() because ot's a somewhat expensive operation.

[edit] eg

Code:

---

#include <stdio.h>
#include <string.h>
#include <ctype.h>
int main()
{
        char somestring[100];
       
        printf("Please enter a string\n");

    fgets(somestring, 100, stdin);
   
    //the following will rebuild the string to get rid of any new lines so we can compare
    //it with "test".
       
        int i;  // C99-style declarations in the middle of a block
        size_t len = strlen(somestring);
        for (i=0; i < len/*-1*/; i++)
        {
                if (somestring[i] != '\n'){
                        somestring[i] = somestring[i];
                }
        }
       
        printf("\n%d", strcmp(somestring, "test"));
       
        return 0;
}

---

Declaring variables in the middle of a block is C99; you might want to declare them at the start of the block.
[/edit]

hmm weird same thing:

Code:

---


#include <stdio.h>
#include <string.h>
#include <ctype.h>
int main()
{
        char somestring[100];
       
        printf("Please enter a string\n");

    fgets(somestring, 100, stdin);
   
    //the following will rebuild the string to get rid of any new lines so we can compare
    //it with "test".
       
        int i;
        int stringLength = strlen(somestring);
        for (i=0; i < stringLength; i++)
        {
                if (somestring[i] != '\n'){
                        somestring[i] = somestring[i];
                }
        }
       
        printf("\n%d", strcmp(somestring, "test"));
       
        return 0;
}

---

Change

Code:

---

                if (somestring[i] != '\n'){
                        somestring[i] = somestring[i];
                }

---

to

Code:

---

                if (somestring[i] == '\n'){
                        somestring[i] = 0;
                        // and maybe break;
                }

---

Otherwise it just copies the existing string, skipping over the '\n' which is already there.

Quote:

---

Originally Posted by dwks

Change

Code:

---

                if (somestring[i] != '\n'){
                        somestring[i] = somestring[i];
                }

---

to

Code:

---

                if (somestring[i] == '\n'){
                        somestring[i] = 0;
                        // and maybe break;
                }

---

Otherwise it just copies the existing string, skipping over the '\n' which is already there.

---

It works however i'm interested as to why the new lines are being replaced with 0?

All strings should have a null at the end. I think '\0' is better than 0.

[edit]

dwks's algorithm loops until it reaches a '\n'. Then it change that char to null, so it will be end of string.
Imagine we have this in memory: trust your self\n\n223anything\0. It will be like this after changing first '\n' to null:
self\0

Sticking a null where the newline was just terminates the string at that particular point, at the same time obliterating the newline. [edit] The whole line isn't being replaced with 0, only the character that was the newline. [/edit] [edit=2] 0 is the same as '\0': http://faq.cprogramming.com/cgi-bin/...&id=1043284376 [/edit]

so it's basically the same as '\0' ? (tried it and gives the same results suppose it is, just checking if its something dangerous.)

See second edit. [edit] Yes. [/edit]

There's a function that's particularly great for this job:

Code:

---

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

int main(void) {
  char somestring[100];

  puts("Please enter a string:");
  fgets(somestring, sizeof somestring, stdin);
  strtok(somestring, "\n"); /* remove a newline if present */

  puts( strcmp(somestring, "test") ? "no match" : "match" );

  return 0;
}

---

Owner@pavilion ~
$ ./strcmp
Please enter a string:
test
match

Have a nice day everyone.

There are many ways to do it, here are some of my favorites, listed in order of efficiency:

• Code:

  ---

  size_t len = strlen(buffer);  /* relys on '\n' being the last char in the string */
if(buffer[len-1] == '\n') buffer[--len] = 0;

  ---

• Code:

  ---

  char *p = buffer;
while(*p && *p != '\n') p++;
*p = 0;

  ---

• Code:

  ---

  char *p;
if((p = strchr(buffer, '\n'))) *p = 0;

  ---

• Code:

  ---

  buffer[strcspn(buffer, "\n")] = 0;  /* not sure about this one, it might be strspn() */

  ---

• etc.

I actually prefer strchr().

Code:

---

{
  char str[] = "foo\n", *p;

  if((p = strchr(str, '\n')))
    *p = '\0';
}

---

EDIT: dwks beat me.

Quote:

---

Originally Posted by dwks

There are many ways to do it, here are some of my favorites, listed in order of efficiency

---

The similarity to this thread (joined already in progress) made me post this.