• Use code tags: Writeup Formatting Tips
• Post code that does not have syntax errors (the code you actually ran).
• Tell us what you expect the count to be.

UPDATE: OP has been edited without notice since this reply and the others to date. Original code (with code tags added by /me) was:

$cnt =0
$str="hi hi harshmane hi hi";
 if ($str=~/([harsh]+)/g)        # cf current content
{
   $cnt++;
}
print $cnt;
[download]

OP has been msgd that such changes without notice are unhelpful.

I think he did the same thing in to find string. And this question is virtually the same as that one. That might also explain why he only matches once in this thread -- he probably used different data in the other one.

Elda Taluta; Sarks Sark; Ark Arks
My deviantART gallery

perlfaq4 - How can I count the number of occurrences of a substring within a string?

Code tags are your (and our) friends.

Your problem is that you don't store the number of matches. The following code will fix your issue.

use warnings;
use strict;
my $cnt = 0;
my $str = "hi hi harshmane hi hi";


$cnt = ($str =~ s/(harsh)/$1/g);
print "'harsh' appears $cnt times.\n";

$cnt = ($str =~ s/(hi)/$1/g);
print "'hi' appears $cnt times.\n";

#OUTPUT
#'harsh' appears 1 times.
#'hi' appears 4 times.
[download]

No need for the capture and substitution by itself!

knoppix@Microknoppix:~$ perl -E '
> $str = q{hi hi harshmane hi hi};
> $cnt = () = $str =~ m{hi}g;
> say $cnt;'
4
knoppix@Microknoppix:~$
[download]

I hope this is of interest.

Cheers,

JohnGG

Your code would increment $cnt once if the expression is true (matches one or more times). You could turn it around like this:

$cnt++ while $str=~/harsh/g;
[download]

Note also that you don't need to store the matches (with parentheses) for what you are doing.

And the '+' modifier in your example would match 'harsh' and 'harshhhhhhhhhh'. Is that what you meant?