CODEWITHSUNDEEP
perlurihtml-treebuilder
user3425810
user3425810

Reputation: 51

Perl extract pattern from html file

I have a .html file full of links, I would like to extract the domains without the http:// (so just the hostname portion of the link, e.g blah.com) list them and remove duplicates.

This is what I have come up with so far - i think the issue is the way I am trying to pass the $tree data 

#!/usr/local/bin/perl -w

use HTML::TreeBuilder 5 -weak; # Ensure weak references in use
use URI;
  foreach my $file_name (@ARGV) {
    my $tree = HTML::TreeBuilder->new; # empty tree
    $tree->parse_file($file_name);
    my $u1 = URI->new($tree);
    print "host: ", $u1->host, "\n";
    print "Hey, here's a dump of the parse tree of $file_name:\n";

    # Now that we're done with it, we must destroy it.
    # $tree = $tree->delete; # Not required with weak references
  }

Upvotes: 5

Views: 291

Answers (2)

Kenosis
Kenosis

Reputation: 6204

Here's another option:

use strict;
use warnings;
use Regexp::Common qw/URI/;
use URI;

my %hosts;

while (<>) {
    $hosts{ URI->new($1)->host }++ while /$RE{URI}{-keep}/g;
}

print "$_\n" for keys %hosts;

Command-line usage: perl script.pl htmlFile1 [htmlFile2 ...] [>outFile]

You can send the script multiple html files. The last, optional parameter directs output to a file.

Partial output using cnn.com home page as html source:

www.huffingtonpost.com
a.visualrevenue.com
earlystart.blogs.cnn.com
reliablesources.blogs.cnn.com
insideman.blogs.cnn.com
cnnphotos.blogs.cnn.com
cnnpresents.blogs.cnn.com
i.cdn.turner.com
www.stylelist.com
js.revsci.net
z.cdn.turner.com
www.cnn.com
...

Hope this helps!

Upvotes: 2

user2410502
user2410502

Reputation:

Personally, I'd go with Mojo::DOM for this, and using URI module to extract domains: `

  use Mojo::DOM;
  use URI;
  use List::AllUtils qw/uniq/;

  my @domains = sort +uniq
    map eval { URI->new( $_->{href} )->authority } // (),
        Mojo::DOM->new( $html_code )->find("a[href]")->each;

(P.S. the exception handing on ->authority is because some URI's will croak here; like mailto:s)

Upvotes: 4

Related Questions