1package HTML::Filter; 2 3use strict; 4 5require HTML::Parser; 6our @ISA = qw(HTML::Parser); 7our $VERSION = '3.76'; 8 9sub declaration { $_[0]->output("<!$_[1]>") } 10sub process { $_[0]->output($_[2]) } 11sub comment { $_[0]->output("<!--$_[1]-->") } 12sub start { $_[0]->output($_[4]) } 13sub end { $_[0]->output($_[2]) } 14sub text { $_[0]->output($_[1]) } 15 16sub output { print $_[1] } 17 181; 19 20__END__ 21 22=head1 NAME 23 24HTML::Filter - Filter HTML text through the parser 25 26=head1 NOTE 27 28B<This module is deprecated.> The C<HTML::Parser> now provides the 29functionally of C<HTML::Filter> much more efficiently with the 30C<default> handler. 31 32=head1 SYNOPSIS 33 34 require HTML::Filter; 35 $p = HTML::Filter->new->parse_file("index.html"); 36 37=head1 DESCRIPTION 38 39C<HTML::Filter> is an HTML parser that by default prints the 40original text of each HTML element (a slow version of cat(1) basically). 41The callback methods may be overridden to modify the filtering for some 42HTML elements and you can override output() method which is called to 43print the HTML text. 44 45C<HTML::Filter> is a subclass of C<HTML::Parser>. This means that 46the document should be given to the parser by calling the $p->parse() 47or $p->parse_file() methods. 48 49=head1 EXAMPLES 50 51The first example is a filter that will remove all comments from an 52HTML file. This is achieved by simply overriding the comment method 53to do nothing. 54 55 package CommentStripper; 56 require HTML::Filter; 57 @ISA=qw(HTML::Filter); 58 sub comment { } # ignore comments 59 60The second example shows a filter that will remove any E<lt>TABLE>s 61found in the HTML file. We specialize the start() and end() methods 62to count table tags and then make output not happen when inside a 63table. 64 65 package TableStripper; 66 require HTML::Filter; 67 @ISA=qw(HTML::Filter); 68 sub start 69 { 70 my $self = shift; 71 $self->{table_seen}++ if $_[0] eq "table"; 72 $self->SUPER::start(@_); 73 } 74 75 sub end 76 { 77 my $self = shift; 78 $self->SUPER::end(@_); 79 $self->{table_seen}-- if $_[0] eq "table"; 80 } 81 82 sub output 83 { 84 my $self = shift; 85 unless ($self->{table_seen}) { 86 $self->SUPER::output(@_); 87 } 88 } 89 90If you want to collect the parsed text internally you might want to do 91something like this: 92 93 package FilterIntoString; 94 require HTML::Filter; 95 @ISA=qw(HTML::Filter); 96 sub output { push(@{$_[0]->{fhtml}}, $_[1]) } 97 sub filtered_html { join("", @{$_[0]->{fhtml}}) } 98 99=head1 SEE ALSO 100 101L<HTML::Parser> 102 103=head1 COPYRIGHT 104 105Copyright 1997-1999 Gisle Aas. 106 107This library is free software; you can redistribute it and/or 108modify it under the same terms as Perl itself. 109 110=cut 111