cipher has asked for the wisdom of the Perl Monks concerning the following question:

Dear Perlmonks, I have an array which contains lots of IP addresses. 
@ip = (1.1.1.1 2.2.2.2 1.1.1.1 4.4.4.4 1.1.1.1 4.4.4.4)

[download]
I want to find the top unique ip's with their count from this array. Example: Output should be top n, In this case n = 2. 
IP   -- Count
1.1.1.1  3
4.4.4.4  2

[download]
I have read many forums and posts and I see most perl experts recommend using hash for this purpose. I already have a running code but here I have my IP address in a scalar, My new code is different and has IP address in an array. Here is my running code: 
use strict;
use warnings;

my $fields = ();
my @logs = ();
my $reg_sip = ();
my $si = ();
my $sIp = ();
my $key = ();
my %sIp = ();

open FILE, "c:/perl/fw.log" or die $!;
        while ($fields = <FILE>)

        {
        @logs = split (/ /,$fields);
             foreach $reg_sip (@logs)
                {if ($reg_sip =~ m/src=\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1
+,3}/)
                    {$si = substr $reg_sip, 4;     $sIp{$si} += 1; }}
        }
my $top = 10;
print "\nTop Ten Sources are\n";
my $i =0; 
foreach $key (sort {$sIp{$b} <=> $sIp{$a}} keys %sIp) 
{print "\n$key\t - $sIp{$key}"; $i++; if ($i == $top){last;}}


## Result
Top Ten Sources are
10.71.74.100     - 2706
192.168.237.103  - 1896
192.168.237.119  - 1473
10.71.74.73      - 1302
192.168.237.110  - 650
10.71.74.139     - 325
10.71.74.74      - 238
192.168.33.32    - 213
192.168.237.247  - 124
192.168.237.250  - 18

[download]
I want to acheive similar output but from array. Example: 
use strict;
use warnings;
my @srcs = ();

open FILE, "c:/perl/fw.log" or die $!;
    while (<FILE>)
    {
    if (my ($src) =/\bsrc=(\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3})/){push 
+@srcs, $src;}
    }

####
Code to print Top n unique ip's in descending order from @srcs.

[download]

Replies are listed 'Best First'.
Re: Top n Unique count from Array
by davido (Cardinal) on Mar 25, 2011 at 08:20 UTC

    How about this?

    use Modern::Perl;

my %seen;

while( <DATA> ) {
    chomp;
    $seen{$_}++;
}

my @top = ( sort { $seen{$b} <=> $seen{$a} } keys %seen )[0,1];

foreach ( @top ) {
    say "$_ $seen{$_}";
}

__DATA__
1.1.1.1
2.2.2.2
1.1.1.1
4.4.4.4
1.1.1.1
4.4.4.4

    [download]

    Dave

[reply]
[d/l]
      UPDATE: Looks like I found the solution. This one is working for me: 
      my $top = 10;
my %cnt_hash; 
$cnt_hash{$_}++ foreach @srcs; 
foreach $key (sort {$cnt_hash{$b} <=> $cnt_hash{$a}} keys %cnt_hash)
{ print "$key\t$cnt_hash{$key}\n";
   $i++;
   if ($i == $top){last;}

}

      [download]
[reply]
[d/l]
Re: Top n Unique count from Array
by JavaFan (Canon) on Mar 25, 2011 at 10:13 UTC
    I want to acheive similar output but from array.
    But the first thing your initial program does when reading in a line is to turn it into an array!
    if (my ($src) =/\bsrc=(\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3})/){push @srcs, $src;}
    So, your question is, I replaced @logs = split (/ /,$fields); with a similar line that uses @srcs instead of @logs, how do I continue?. Maybe the easiest is to add @logs = @srcs;?
[reply]
[d/l]
[select]
Re: Top n Unique count from Array
by Anonymous Monk on Mar 25, 2011 at 07:36 UTC
    I want to acheive similar output but from array.

    Why?

    Where is your attempt?

[reply]
      Why?
      I want to use the IP address multiple time for various purpose in my script so I am pushing them in an array.

      Where is your attempt?
      Tried this, following an example but output is not as I want it.
      Output is not sorted and where do I specify that I only want top n unique values. 
      %seen = (); foreach $item (@srcs) {     push(@uniq, $item) unless $see
+n{$item}++; }
print "@uniq\n";

      [download]

      This one only gives the unique ip addresses and not count. 
      my %count; 
foreach $1 ( @srcs ) { 
  $count{$1}++; 
} 

my @srcs = sort keys %count;

      [download]
[reply]
[d/l]
[select]
        I want to acheive similar output but from array.

        Why?

        I want to use the IP address multiple time for various purpose in my script so I am pushing them in an array.

        The question was why do you want an array?

        Tried this, following an example but output is not as I want it. Output is not sorted and where do I specify that I only want top n unique values.

        But why did you try that?

        Why should that code produce something sorted?

        What series of steps does the code perform to produce a sorted array of IPs and hits?

        Also, what do you mean where?

        Please think about it, and try to answer these questions out loud ; speak the answer to your monitor or an object on your desk.

        What I would do is redirect the output of the first program to a file, then execute head --lines=N file to get the top N results

        Read more... (966 Bytes)
[reply]
[d/l]
[select]

Back to Seekers of Perl Wisdom