How have you tried? It would make more sense to tell you where you took the wrong road instead of just giving you a solution.

I think a better solution than 'split' is using regex. Something like this (untested):

$line =~ m/^\S+\.(.*)=\S+$/;

If you are matching many instances of this (say reading a log file that is 700MB or more and checking each line for this string) then you are likely to find that the performance improves if you tie the match down to the bare minimum of wildcard characters, specify everything that you know is going to be in the match, and also anchor the match where possible to the beginning and end of the string (or line) so it doesn't have to try iterating over the string:

use strict;
use warnings;

my $line = "SNMPv2-MI::enterprises.343.2.10.3.5.100.1.0=INT";

my $match = $1 if $line =~ /^SNMPv2-MI::enterprises([0-9\.]+)=INT$/;

perl -e '$line="SNMPv2-MI::enterprises.343.2.10.3.5.100.1.0=INT";
print $1 if $line=~/^[\w|\d|-]*::\w*.(((\d{1,3})\.?){8})/'

Thanks everybody. I tried split function and got the results. I know this is a long procedure and bad programming but i learnt a lot in this process. Here is what i have done First step input is SNMPv2-SMI::enterprises.343.2.10.3.5.100.1.0 = INTEGER: 5 output is SNMPv2-SMI::enterprises.343.2.10.3.5.100.1.0=INTEGER:5 removed all the intermediate spaces using $line=~s/\s//g; Then replaced all the . with : so that i get result as SNMPv2-SMI::enterprises:343:2:10:3:5:100:1:0:INTEGER:5 Then put this string in a array. Read through the array. concatinate only numbers until the word INTEGER is reached. finally i will get .343.2.10.3.5.100.1.0