http://qs321.pair.com?node_id=814093

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

hello all

I'm trying to extract table rows from following HTML. It is showing the desired output but is there any other way that I can map my table rows with key-value pair or array of arrays of td elements under a tr

<html><head><title>Person Profile</title></head>
<center>
<font size=5><b>Profile</b></font>

<table cellspacing="1" cellpadding="1">
<tr>
    <td class="rlab">Short Name:</td>
    <td class="l">John</td>
</tr>
<tr>
    <td class="rlab">Long Name:</td>
    <td class="l">John Abraham</td>
</tr>
<tr>
    <td class="rlab">Company:</td>
    <td class="l">Idea</a></td>
</tr>
</tr>
<tr>
    <td class="rlab">Currency:</td>
    <td class="l">EUR</td>
</tr>
</table>
</body></html>

[download]

I'm trying the following code

use strict;
use warnings;
use HTML::TreeBuilder;

#Parse html content using html-treebuilder:
my $root = HTML::TreeBuilder->new();
$root->parse($html);
$root->eof();

my @tables = $root->look_down(_tag => 'table');
while (@tables) {
    my $node = shift @tables;
    if (ref $node) {
        unshift @tables, $node->content_list;
    }
    else {
        print $node,"\n";
    }
}
$root = $root->delete;

[download]

OUTPUT is 

---------- Perl ----------
Short Name:
John
Long Name:
John Abraham
Company:
Idea
Currency:
EUR

Output completed (0 sec consumed) - Normal Termination

[download]

Replies are listed 'Best First'.
Re: Extract HTML Table rows
by bobf (Monsignor) on Dec 23, 2009 at 14:44 UTC

    I have found HTML::TableExtract to be easy to use in simple cases: 

    use strict;
use warnings;

use HTML::TableExtract;

my $content;
{
    local $/ = undef; # slurp mode
    $content = <DATA>;
}

my $te = HTML::TableExtract->new();
$te->parse( $content );

foreach my $ts ( $te->tables() )
{
    foreach my $row ( $ts->rows() )
    {
        print join ( "\t", @$row ), "\n";
    }
}

__DATA__
<html><head><title>Person Profile</title></head>
<center>
<font size=5><b>Profile</b></font>

<table cellspacing="1" cellpadding="1">
<tr>
    <td class="rlab">Short Name:</td>
    <td class="l">John</td>
</tr>
<tr>
    <td class="rlab">Long Name:</td>
    <td class="l">John Abraham</td>
</tr>
<tr>
    <td class="rlab">Company:</td>
    <td class="l">Idea</a></td>
</tr>
</tr>
<tr>
    <td class="rlab">Currency:</td>
    <td class="l">EUR</td>
</tr>
</table>
</body></html>

    [download]

[reply]
[d/l]
Re: Extract HTML Table rows
by suaveant (Parson) on Dec 23, 2009 at 14:40 UTC
    There are modules specifically to handle html tables...

    HTML::TableExtract
    HTML::TableParser

                    - Ant
                    - Some of my best work - (1 2 3)

[reply]
Re: Extract HTML Table rows
by wfsp (Abbot) on Dec 23, 2009 at 15:16 UTC
    The modules recommended by suaveant and bobf are a good bet. If you wanted to use HTML::TreeBuilder the following would be one way to do it. 
    #! /usr/bin/perl

use strict;
use warnings;

use Data::Dumper;
$Data::Dumper::Indent=1;

use HTML::TreeBuilder;

my $t = HTML::TreeBuilder->new_from_file(*DATA);

my ($table) = $t->look_down(_tag => q{table});
my @rows = $table->look_down(_tag => q{tr});

my %db;

for my $row (@rows){
  my $key   = $row->look_down(class => q{rlab})->as_text;
  my $value = $row->look_down(class => q{l})->as_text;
  $db{$key} = $value;
}

for my $key (keys %db){
  printf qq{%s -> %s\n}, $key, $db{$key};
}

__DATA__
<html><head><title>Person Profile</title></head>
<center>
<font size=5><b>Profile</b></font>

<table cellspacing="1" cellpadding="1">
<tr>
  <td class="rlab">Short Name:</td>
  <td class="l">John</td>
</tr>
<tr>
  <td class="rlab">Long Name:</td>
  <td class="l">John Abraham</td>
</tr>
<tr>
  <td class="rlab">Company:</td>
  <td class="l">Idea</a></td>
</tr>
</tr>
<tr>
  <td class="rlab">Currency:</td>
  <td class="l">EUR</td>
</tr>
</table>

</body></html>

    [download]
    Company: -> Idea
Long Name: -> John Abraham
Currency: -> EUR
Short Name: -> John

    [download]
    I've assumed that
    • there is one table,
    • each row has two columns each with a class as in your sample data
    You would probably want to include some error checking to confirm those assumptions though.
[reply]
[d/l]
[select]

Back to Seekers of Perl Wisdom