#!/usr/bin/perl
use strict;
use warnings;
use diagnostics;
use HTML::TokeParser;

# my $file = 'school.html';

my @html_files = File::Find::Rule->file->name( '*.html' )->in( $html_dir );
my $p = HTML::TokeParser->new($file) or die "Can't open: $!";

my %school;
while (my $tag = $p->get_tag('div', '/html')) {
        # first move to the right div that contains the information
        last if $tag->[0] eq '/html';
        next unless exists $tag->[1]{'id'} and $tag->[1]{'id'} eq 'inhalt_large';
        
        $p->get_tag('h1');
        $school{'location'} = $p->get_text('/h1');
        
        while (my $tag = $p->get_tag('div')) {
                last if exists $tag->[1]{'id'} and $tag->[1]{'id'} eq 'fusszeile';
                
                # get the school name from the heading
                next unless exists $tag->[1]{'class'} and $tag->[1]{'class'} eq 'fm_linkeSpalte';
                $p->get_tag('h2');
                $school{'name'} = $p->get_text('/h2');
                
                # verify format for school type
                $tag = $p->get_tag('span');
                unless (exists $tag->[1]{'class'} and $tag->[1]{'class'} eq 'schulart_text') {
                        warn "unexpected format: parsing stopped";
                        last;
                }
                $school{'type'} = $p->get_text('/span');
                
                # verify format for address
                $tag = $p->get_tag('p');
                unless (exists $tag->[1]{'class'} and $tag->[1]{'class'} eq 'einzel_text') {
                        warn "unexpected format: parsing stopped";
                        last;
                }
                $school{'address'} = clean_address($p->get_text('/p'));
                
                # find the description
                $tag = $p->get_tag('p');
                $school{'description'} = $p->get_text('/p');
        }
}

print qq/$school{'name'}\n/;
print qq/$school{'location'}\n/;
print qq/$school{'type'}\n/;

foreach (@{$school{'address'}}) {
        print "$_\n";
}

print qq/\nDescription: $school{'description'}\n/;

sub clean_address {
        my $text = shift;
        my @lines = split "\n", $text;
        foreach (@lines) {
                s/^\s+//;
                s/\s+$//;
        }
        return \@lines;
}