#!/usr/bin/perl

use strict;
use warnings;

# load in 'tree' mode for working with
# HTML::Element structures. note that in
# this case, subtables are *not* decoupled
# from one another.

use HTML::TableExtract 'tree';

my $te = HTML::TableExtract->new(
  # extraction parameters here...note that
  # in tree mode, keep_html is irrelevant
);
$te->parse_file("./myfile.html");

my $t = $te->first_table_found or die "oops, no tables.\n";

# at this point we can work with $t->rows and the
# cells within, but rather than text or html, the
# content is now individual element objects/trees

# for html...
print "H::TE as html:\n";
foreach my $row ($t->rows) {
  print join(':', map { $_->as_HTML } @$row), "\n";
}

# for text...
print "H::TE as text:\n";
foreach my $row ($t->rows) {
  print join(':', map { $_->as_text } @$row), "\n";
}

# Alternatively, you could switch entirely over
# to the HTML::ElementTable structure

my $et = $t->tree;

# as html
print "H::ET as html:\n";
print $et->as_HTML, "\n";

# as text
print "H::ET as text:\n";
print $et->as_text, "\n";