How to extract a column of a table from html page using perl modules?

Question

I have the following html code of a part of a webpage.

<h2 id="failed_process">Failed Process</h2>
<table border="1">
  <thead>
    <tr>
      <th>
        <b>pid</b>
      </th>
      <th>
        <b>Priority</b>
      </th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <td id="90"><a href="details.jsp?pid=p_201211162334&refresh=0">p_201211162334</a></td>
      <td id="priority_90">NORMAL</td>
    </tr>
    <tr>
      <td id="91"><a href="details.jsp?pid=p_201211163423&refresh=0">p_201211163423</a></td>
      <td id="priority_91">NORMAL</td>
    </tr>
    <tr>
      <td id="98"><a href="details.jsp?pid=p_201211166543&refresh=0">p_201211166543</a></td>
      <td id="priority_98">NORMAL</td>
    </tr>
  </tbody>
</table>
<hr>

I need to extract the pid column . The output should look like

pid
p_201211162334
p_201211163423
p_201211166543

The table count for "Failed Process" table is 4. But the problem is if I mention the table count as 4 and if there are no failed tasks in the webpage, it'll go to the next table and fetch the pid's of next table resulting in wrong pid's.

I am using the below code to get the result.

#!/usr/bin/perl
 use strict; 
 use warnings;
 use lib qw(..);
 use HTML::TableExtract;

 my $content = get("URL");
 my $te = HTML::TableExtract->new(
 headers => [qw(pid)], attribs => { id => 'failed_process' },
 );

 $te->parse($content);

 foreach my $col ($te->rows) {
 print ("\t", @$col), "\n";
 }

But I am getting the following error:

Can't call method "rows" on an undefined value 

Answer 1

With my favourite DOM parser Mojo::DOM from the Mojolicious suite it would look like that:

#!/usr/bin/env perl

use strict;
use warnings;
use feature 'say';
use Mojo::DOM;

# instantiate with all DATA lines
my $dom = Mojo::DOM->new(do { local $/; <DATA> });

# extract all first column cells
$dom->find('table tr')->each(sub {
    my $cell = shift->children->[0];
    say $cell->all_text;
});

__DATA__
<h2 id="failed_process">Failed Process</h2>
<table border="1">
    ...

Output:

pid
p_201211162334
p_201211163423
p_201211166543

Answer 2

After $te->parse($html) you may add some like foreach my $table ($te->tables) .. then you can get rows $table->rows. You may also use Data::Dumper to analyze $te.