#!/usr/bin/perl -w
use strict;
use HTML::TokeParser;

my $string = qq,

clear text <tag var1=".." var2="..>.." > clear text

,;

printf "\" appears %s times in this string\n", ($string)=~ y/"/"/;

my $p = HTML::TokeParser->new(\$string);

die $! unless $p;

while (my $token = $p->get_token)
{
    # ["S", $tag, %$attr, @$attrseq, $origtext]
    if( ($token->[0] eq 'S') ) # is it a starting link tag
    {
        my ($typeotag, $tag, $attr, $attrseq, $origtext)=@$token;
        print "this start tag is ($tag)\n";
        print "its attribues are (in original sequence):\n";
        printf "(%s)=(%s)\n",$_,$attr->{$_} for(@$attrseq);
        print "\n\n-- a start tag no more --\n\n";
    }
    else
    {
        printf "something else  (%s)\n\t\t(%s)\n",
               @{$token};
    }

}

__END__
F:\dev\HTML_Tokeparser_Tutorial>perl liar.pl
" appears 4 times in this string
something else  (T)
                (

clear text )
this start tag is (tag)
its attribues are (in original sequence):
(var1)=(..)
(var2)=(..>..)


-- a start tag no more --

something else  (T)
                ( clear text)
something else  (T)
                (

)

F:\dev\HTML_Tokeparser_Tutorial>