Why don't you use HTML::Table::Extract to extract the rows and columns of that table? After the matching problem becomes trivial.

while (<DATA>) {
    if (my ($a, $b, $c, $d) = (m/TD class=i><A href="([^"]*)"\s*>([^<]
+*)<.*TD class=j><A href="([^"]*)"\s*>([^<]*)</)) {
        print "($a, $b, $c, $d)\n";
    }
}
__END__
<TR class=g><TD class=i><A href="/page1.html">Page name here</A></TD><
+TD class=j><A href="/page2.html">Other page name here</A></TD></TR>
<TR class=h><TD class=i><A href="/blahblah.html">lah blah</A></TD><TD 
+class=j><A href="/ribbit.html">Ribbit</A></TD></TR>
[download]

yes!!! That is exactly it! That's what I need to do.

I can get your demo working but do I need to save the html into an array in order to do this while loop?

Thanks.

My solution saves the data into an array of hashes.

---




($_='kkvvttuu bbooppuuiiffss qqffssmm iibbddllffss')
=~y~b-v~a-z~s; print

I don't understand what you want. There are a number of things in each page which match. Within the match there are more than two URLs, but you only want the first two?

Give us an example like "here's a page, (some code). In this page I need to extract the links (list of links you need to extract)".

Let me retry to explain what I need to do.

I am scraping a web page and I have $source which is the html dump. From the entire source code there is a table that has a ton of rows but always 2 columns (as you can see from the two samples above.

There is 1 URL in each table cell which means 2 URLS per table row. I need to collect the URL and the text of the URL for EACH table cell (which means 2 urls per row).

Those 2 samples displayed in the OP is what the table looks like. There's dozens upon dozens of <TR>s and I need to extract each url and the text for it and at the same time keep each row of links connected. This is to say I need to keep the two URLS per row together somehow because they are relative to each other. This code is the beginning of the table and 3 rows. You see it's just the same type of code I pasted earlier. I need the urls and the text but keep each row of data together magically.

<TABLE width=468 style="border-collapse:collapse"><TR><TD><TABLE width
+=468 style="border-collapse:collapse" cellpadding=2> <TR><TD class=he
+ad>Artist</TD><TD class=head>Song</TD></TR> 

<TR class=g><TD class=i><A href="/my-bloody-valentine-rxzg3.html">My B
+loody Valentine</A></TD><TD class=j><A href="/my-bloody-valentine-lov
+ely-sweet-darlene-9lgrnkb.html">Lovely Sweet Darlene *</A></TD></TR>

<TR class=h><TD class=i><A href="/jennifer-love-hewitt-d193f.html">Jen
+nifer Love Hewitt</A></TD><TD class=j><A href="/jennifer-love-hewitt-
+i-want-a-love-i-can-see-jv2mj67.html">I Want A Love I Can See *</A></
+TD></TR>

<TR class=g><TD class=i><A href="/jennifer-love-hewitt-d193f.html">Jen
+nifer Love Hewitt</A></TD><TD class=j><A href="/jennifer-love-hewitt-
+no-ordinary-love-4rftntr.html">No Ordinary Love *</A></TD></TR>
[download]

From that code I want to get /my-bloody-valentine-rxzg3.html and My Bloody Valentine from the first cell from the first table and ="/my-bloody-valentine-rxzg3.html and Lovely Sweet Darlene . I need this data collected for each row.

Does this make any more sense?

Edited by Chady -- escaped <TR>

Well, this does it, but it's not the optimal solution. There's lots of repetition and one side is clearly the artist and the other a song, so it looks kind of off to me. You know better than me what you want though...

I never thought I'd see MBF and JLH mentioned in the same breath...

#!/usr/bin/perl
use strict;
use warnings;
use diagnostics;
use Data::Dumper;

my $html = '<TABLE width=468 style="border-collapse:collapse">
<TR><TD><TABLE width=468 style="border-collapse:collapse" cellpadding=
+2> <TR><TD class=head>Artist</TD><TD class=head>Song</TD></TR>
<TR class=g><TD class=i><A href="/my-bloody-valentine-rxzg3.html">My B
+loody Valentine</A></TD><TD class=j><A href="/my-bloody-valentine-lov
+ely-sweet-darlene-9lgrnkb.html">Lovely Sweet Darlene *</A></TD></TR>
<TR class=h><TD class=i><A href="/jennifer-love-hewitt-d193f.html">Jen
+nifer Love Hewitt</A></TD><TD class=j><A href="/jennifer-love-hewitt-
+i-want-a-love-i-can-see-jv2mj67.html">I Want A Love I Can See *</A></
+TD></TR>
<TR class=g><TD class=i><A href="/jennifer-love-hewitt-d193f.html">Jen
+nifer Love Hewitt</A></TD><TD class=j><A href="/jennifer-love-hewitt-
+no-ordinary-love-4rftntr.html">No Ordinary Love *</A></TD></TR>';


my @data = ();
while ( $html =~
    m|<TD class=\w><A href="([^"]+)">([^<]+)</A></TD>|g )
{
    push( @data, { link => $1, name => $2 } );
}




print Dumper(\@data);

#### result ####
# $VAR1 = [
#    {
#      'link' => '/my-bloody-valentine-rxzg3.html',
#      'name' => 'My Bloody Valentine'
#    },
#    {
#      'link' => '/my-bloody-valentine-lovely-sweet-darlene-9lgrnkb.ht
+ml',
#      'name' => 'Lovely Sweet Darlene *'
#    },
#    {
#      'link' => '/jennifer-love-hewitt-d193f.html',
#      'name' => 'Jennifer Love Hewitt'
#    },
#    {
#      'link' => '/jennifer-love-hewitt-i-want-a-love-i-can-see-jv2mj6
+7.html',
#      'name' => 'I Want A Love I Can See *'
#    },
#    {
#      'link' => '/jennifer-love-hewitt-d193f.html',
#      'name' => 'Jennifer Love Hewitt'
#    },
#    {
#      'link' => '/jennifer-love-hewitt-no-ordinary-love-4rftntr.html'
+,
#      'name' => 'No Ordinary Love *'
#    }
#  ];
#
[download]

---




($_='kkvvttuu bbooppuuiiffss qqffssmm iibbddllffss')
=~y~b-v~a-z~s; print