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

I need to parse my links and get server name info between "http://" and the first "/"
I need to then get a count of each one. Here an example of what I have: 
http://riverserver/dir1/dir2/index.html
http://riverserver/dir1/dir2/index.html
http://perlmonks.org/index.pl
http://webserver.company.serv/Adir/Bdir/thePage.cfm
http://riverserver/dir1/dir2/index.html
http://webserver.company.serv/Adir/Bdir/thePage.cfm

[download]
I want to parse the info and get this: 
riverserver = 3
perlmonks.org = 1
webserver.company.serv = 2

[download]
I think I have a problem with how to parse the links and add a hash but not sure if I am doing this right?? 
use strict;
my $link1 = 'http://perlmonks.org/index.pl';
my $link2 = 'http://riverserver/dir1/dir2/index.html';
#more links etc...
$link1 =~ s/http\:\/\///;
$link2 =~ s/http\:\/\///;

my @link1 = split /\//, $link1;
my @link2 = split /\//, $link2;

print "$link1[0]\n";
print "$link2[0]\n";

[download]

Replies are listed 'Best First'.
Re: Parsing to get server info
by ctilmes (Vicar) on Jul 15, 2003 at 13:18 UTC
    Use URI. 
    use URI;

my %hostcount;

while (<DATA>)
{
    my $u = URI->new($_);
    $hostcount{$u->host}++;
}

foreach my $host (keys %hostcount)
{
    print "$host = $hostcount{$host}\n";
}

__DATA__
http://riverserver/dir1/dir2/index.html
http://riverserver/dir1/dir2/index.html
http://perlmonks.org/index.pl
http://webserver.company.serv/Adir/Bdir/thePage.cfm
http://riverserver/dir1/dir2/index.html
http://webserver.company.serv/Adir/Bdir/thePage.cfm

    [download]

    Output: 

    webserver.company.serv = 2
riverserver = 3
perlmonks.org = 1

    [download]
[reply]
[d/l]
[select]
      I tried exactly as you had and got this message: 
      Can't locate object method "host" via package "URI::_generic" (perhaps
+ you forgo
t to load "URI::_generic"?) at C:\Perl\bin\url1.pl line 11, <DATA> lin
+e 7.

      [download]
      I do have "URI" module on my Windows NT. Please advise.
[reply]
[d/l]
Re: Parsing to get server info
by gjb (Vicar) on Jul 15, 2003 at 13:14 UTC

    If you're sure that the data you process contains only valid URLs, you can do it a bit more conveniently with: 

      $link =~ m{http://([^/]+)};
  $server = $1;

    [download]
    Hope this helps, -gjb-

[reply]
[d/l]
      As an add on here. maybe 
      $line =~ m{^https?://([^/]+)};
$server{$1}++;

for ( keys %server ) {
   print "$_ : $server{$_}\n";
}

      [download]
      This way it will also catch https URLs. Though using a module as recommended below is probably the best way.


      MMMMM... Chocolaty Perl Goodness.....
[reply]
[d/l]

Back to Seekers of Perl Wisdom