squeeze() function

[xion]

void squeeze(char s1[], char s2[])
{
   int i,j,k;
   
   for(i = k = 0; s1[i] != '\0'; i++)
   {
      for(j = 0; s2[j] != '\0' && s2[j] != s1[i]; j++)
         ;
      if (s1[j] == '\0')
         s1[k++] = s1[i];
   }

   s1[k] = '\0';
}

im havin trouble understanding this piece of code. i know what it does. it deletes, in s1[], whatever occurs in s2[]. ive been at this k&r exercise for a while and cheated by looking at the answer book. yet i still dont understand the concept of this function. the for loop, i know, steps through each array element and tests. and while s2[j] != s1[i], j keeps incrementing. but once it reaches a match, what happens?

if (s2[j] == '\0')
   s1[k++] = s1[i];

if it reaches a match before the actual '\0' what happens? thanks in advance for response. and sorry if this post seemed a little long-winded. just trying to understand everything. =)

[Hammer]

That solution doesn't appear to work (try compiling and running it yourself).

A quick search found this solution.

[vasu15]

i guess this piece of code works if s2 > s1 and the similar chars are found within s1 length

let s1=hello and s2=bull$$$$
As the for loop checks for s1[i] j will end up with a number more than length of string s1
so untill there is a match k keeps on increasing
and assigns the chars to s1[k++]

now when the match l is found j=2 as s1[2]!='\0'
nothing happens and loop iterates..like this the chars ll are skipped and when o is checked s1[2]is
assigned o;
uot of the loop string s1 is terminated after k chars

i think u got it dude......

[Hammer]

>>>let s1=hello and s2=bull$$$$ <<<
(I'm presuming you're talking about the OPs code)
Let's run that example:

int main(void)
{
    char    a1[] = "hello";
    char    a2[] = "bull$$$$";

    squeeze(a1, a2);

    printf (">a1:%s<\n", a1);
    printf (">a2:%s<\n", a2);

    return 0;
}

/*
 Output:
 
 >a1:h<
 >a2:bull$$$$<
 
 */

The output is broken, and so is the code.

Note: The "bull$$$$" was the real word in my code. If you actually use $ signs, the output is:
>a1:<
>a2:bull$$$$<

[xion]

the two for loops loop on each array element until '\0' is encountered or a match is found between s2[] and s1[]. when i match is found it breaks out of the loop right?

the if statement:

if (s2[j] == '\0')
   s1[k++] = s1[i];

will only execute if s2[j] equals the null character. so if it finds a match before the actual '\0' the if statement wont execute. im lost. =\

[vasu15]

if the match is not found before the null whatever chars r there in s1 get assigned as s1[k++]
when there is a match found before null then that char is missed
in assigning to s1[k++]
so after the loops r over s1[k] is set to null so the string teminates there itself

for better understanding u do the itewrations manually taking s1 hello s2 as nullchar

[Hammer]

>>still puzzled.
You do understand what I'm saying, right? The original code you posted doesn't appear to work properly, so trying to learn from it ain't a good idea, unless you're looking to find how it's broken.

[vasu15]

the piece of code works for my system what do mean by OPs code
u try taking s1 s2 as fixed array of length 10

[Hammer]

OP = Original Poster (thread starter)

It's broken in that it requires both arrays to be the same length. I was assuming that you wouldn't want that, as you'd most likely want to do something like:

squeeze(myarray, "abc"); /* remove abc from myarray */

[xion]

thanks for the responses everyone.

for(i = k = 0; s1[i] != '\0'; i++)
   {
      for(j = 0; s2[j] != '\0' && s2[j] != s1[i]; j++)
         ;
      if (s1[j] == '\0')
         s1[k++] = s1[i];
   }

it compares s1[i] to s2[j] when the 2nd for loop reaches either '\0' or a match it goes to the if statement. the if statement then checks to see if the current j array element is '\0' if it is a '\0' then no match was found and it is copied to s1. now i understand! =)

[P.Phant]

if (s1[j] == '\0')
         s1[k++] = s1[i];

I think the problem is there no? (didn't try to run it)
it should be:

if (s2[j] == '\0')
         s1[k++] = s1[i];

It's s2 you are parsing in the nested for loop. If s2[j] is NUL then it means there was no match, hence you should keep s1[i], else you should discard it.

s1[j] on the other hand could be out of bound or whatever...

[xion]

Originally posted by P.Phant
it should be:

if (s2[j] == '\0')
         s1[k++] = s1[i];

It's s2 you are parsing in the nested for loop. If s2[j] is NUL then it means there was no match, hence you should keep s1[i], else you should discard it.

s1[j] on the other hand could be out of bound or whatever...

yeah. youre right. my code is a typo. had to type it out because the actual file is on my linux box. thanks for pointing that out. =)