I've been trying to use HTML::TagParser to pull reviews from Yelp and modifying formatting slightly to post on our website, though I'm having difficulty parsing through child nodes of the HTML. So I started a basic html file to play with, and still having some difficulty getting the beginning and ending tags to encapsulate properly with inner-text. Here's the basic version of my HTML

<table>
  <tr id="row1">
    <td><div id="r1d1"><b>Test11</b></div></td>
    <td><div id="r1d2"><i>Test12</i></div></td>
    <td><div id="r1d3">test13</div></td>
    <td><b>test14</b></td>
  </tr>
  <tr id="row2">
    <td><div id="r2d1"><b><em>Test21</em></b></div></td>
    <td>2</td>
    <td>3</td>
    <td>4</td>
  </tr>
</table>
[download]

and my code, which pulls things out of place since I can't seem to get my mind to grasp the structuring correctly. Certainly I'm missing something about the "parentNode" after a "childNode" is found... and perhaps "nextSibling" can be useful somewhere.

#!/usr/bin/perl
use strict; use CGI ':standard';
use HTML::TagParser; my @Body;

my $html=HTML::TagParser->new('table.txt');
my @elem=$html->getElementsByTagName("table");

foreach my $parent(@elem){
my $text=$parent->innerText();
my $tag=$parent->tagName;

my $get=$parent->childNodes();
if (@$get){push @Body,digChild(@$get);}
push @Body,"<$tag>$text</$tag>";
} #foreach parent

print header,start_html('table');
print join("\n",@Body)."\n";
print end_html;


sub digChild { my @ar=@_;
foreach my $row(@ar){
my $tag=$row->tagName;
my $txt=$row->innerText;
my $c=$row->childNodes();
if (@$c){push @Body,digChild(@$c);}
return "<$tag>$txt</$tag>";
} # foreach row
} # sub
[download]

Thank you.