#!/usr/local/bin/perl -w
#
# MySpaceMusic v1e (c) 2008 Martin Hassel, code(cat)lacasahassel(dog)net
# cat is at and dog is dot ;-)
#
# Developed with help and encouragement from Barbara Lublinn
#
# This program takes a list of artists/bands, together with the url to their
# MySpace page, and checks them all against a list of towns/cities.
# If a given city is found in a bands schedule the date, time and place is
# extracted and presented.
#
# This program is free software; you can redistribute it and/or modify it under
# the terms of either the Artistic License (which comes with Perl 5) or the
# GNU General Public License as published by the Free Software Foundation;
# either version 2 of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# A full copy of the GNU General Public License can be retrieved from
# http://www.gnu.org/copyleft/gpl.html

use LWP;
$browser = LWP::UserAgent->new;

use POSIX;
%month = (
        jan => 0,
        feb => 1,
        mar => 2,
        apr => 3,
        may => 4,
        jun => 5,
        jul => 6,
        aug => 7,
        sep => 8,
        oct => 9,
        nov => 10,
        dec => 11
    );

# Load the list of bands to look for: myspacemusicbands.txt
print "Reading: myspacemusicbands.txt\n";
open(BANDS, 'myspacemusicbands.txt') or die $!;
@bands = <BANDS>;
close(BANDS);

# Load the list of cities to look for: myspacemusiccities.txt
print "Reading: myspacemusiccities.txt\n";
open(CITIES, 'myspacemusiccities.txt') or die $!;
@cities = <CITIES>;
close(CITIES);

$retries = 1;
$maxretries = 100;
print "Processing band list...\n";
foreach $band (@bands) {
	@band = split(/\t/,$band);
	$band = $band[0];
	$url = $band[1];
  $response = $browser->get($url);
  while((!$response->is_success || $response->content_type ne 'text/html') && $retries<$maxretries) {
    $response = $browser->get($url);
    print "Retry: ".($retries++)." ($band)\n";
    sleep $retries;
  }
  if($retries>=$maxretries) {
    die "Can't get $url -- ", $response->status_line;
    die "Expected HTML; found ", $response->content_type unless $response->content_type eq 'text/html';
  }
  $retries = 1;
  $content = $response->content;
  $content =~ s/^.*?<div id="profile_bandschedule">(.*?)<\/div>.*$/$1/gsi;
  @schedule = split(/<\/tr>\s*?<tr>/i,$content);
  foreach $city (@cities) {
    chomp $city;
    foreach $scheduleitem (@schedule) {
      $scheduleitem = fixsechars($scheduleitem);
      if($scheduleitem =~ m/$city/i) {
        $scheduleitem  =~ m/<td width="85".*?sans-serif\">(.*?)</gsi;
        $date = $1;
        $scheduleitem  =~ m/<td width="35".*?sans-serif\">(.*?)</gsi;
        $time = $1;
        $scheduleitem =~ m/<td width="191".*?Band_Show_ID.*?\">(.*?)</gsi;
        $place = $1;
        $scheduleitem =~ m/<td width="115".*?sans-serif\">(.*?)</gsi;
        $town = $1;
        if($date) {
          @date = split(" ", $date);
          $unixtime = mktime(0, 0, 0, $date[0], $month{$date[1]}, $date[2]-1900, 0, 0);
          if($unixtime > time) {
            print "$band:\t";
            print "$date\t$time\t$place\t$town\n";
          }
        }
      }
    }
  }
  sleep 3;
}

sub fixsechars {
  $string = shift;
  $string =~ s/Ã¥/å/gsi;
  $string =~ s/ä/ä/gsi;
  $string =~ s/ö/ö/gsi;
  $string =~ s/Ã…/Å/gsi;
  $string =~ s/Ö/Ö/gsi;
  return $string;
}