my $parser = HTML::Parser->new( start_h => [ \&_starttag, 'self, tagname, text, attr' ], end_h => [ \&_endtag, 'self, tagname' ], text_h => [ \&_text, 'self, text, dtext' ] ); $parser->parse($response->content()); sub _starttag { my ($self, $tag, $attr) = @_; $self->{'_body'} = 1 if($tag eq 'body'); $self->{'_body'} = 0 if($tag eq 'script'); $self->{'_body'} = 0 if($tag eq 'a'); $self->{'_body'} = 0 if($tag eq 'label'); $self->{'_body'} = 0 if($tag eq 'option'); $self->{'_body'} = 0 if($tag eq 'form'); } sub _endtag { my ($self, $tag) = @_; $self->{'_body' } = 0 if($tag eq 'body' ); $self->{'_body' } = 1 if($tag eq 'a' ); $self->{'_body'} = 1 if($tag eq 'script'); $self->{'_body'} = 1 if($tag eq 'label'); $self->{'_body'} = 1 if($tag eq 'option'); $self->{'_body'} = 1 if($tag eq 'form'); } sub _text { my ($self, $dtext) = @_; $dtext =~ s/\A\s+//; $dtext =~ s/\s+\z//; return() unless ( length($dtext) > 0 and $dtext =~ /[^\s]/ ); if ($self->{'_body'} == 1) { $content=$content.$dtext." "; } }