Question

I have an html file “statistics.htm”. Got the data in the html file to a variable.

suppose in the variable i have below data:

<ul class="chart-legend">
    <li class="label-1">
      <div><em></em>FTP<br>
      0 B</div>
    </li>
    <li class="label-2">
      <div><em></em>HTTP<br>
      589 KB</div>
    </li>
    <li class="label-3">
      <div><em></em>POP3/IMAP<br>
      0 B</div>
    </li>
    <li class="label-4">
      <div><em></em>SMTP<br>
      0 B</div>
    </li>
</ul>

suppose if customer gives FTP as an argument i want the FTP value which is 0 B in above case.

How can i achieve this?

Answer 1

You can do this very simply with HTML::TreeBuilder::XPath (OK, very simply until you get to the fun XPath query!):

#!/usr/bin/perl

use strict;
use warnings;

use HTML::TreeBuilder::XPath;

my $html= HTML::TreeBuilder::XPath->new->parse_file( \*DATA);
my $ftp= $html->findnodes( 'normalize-space( //div/br[./preceding-sibling::text()="FTP"]/following-sibling::text())');
print "$ftp\n";


__DATA__
<ul class="chart-legend">
    <li class="label-1">
      <div><em></em>FTP<br>
      0 Ba</div>
    </li>
    <li class="label-2">
      <div><em></em>HTTP<br>
      589 KB</div>
    </li>
    <li class="label-3">
      <div><em></em>POP3/IMAP<br>
      0 Bb</div>
    </li>
    <li class="label-4">
      <div><em></em>SMTP<br>
      0 Bc</div>
    </li>
</ul>

The XPath expression: look for a br in a div, one which previous sibling text is 'FTB' (you may want to normalize spaces there). Then take the next sibling text. Wrap this in a normalize-space to clean-up the result, Voilà!