#!/usr/bin/perl
# spider.plx

use warnings;
use strict;
#parse Videos.html source code for subpages
open LYNX, "lynx -source http://www.regenmag.com/Videos.html/ |" or die "Can't open lynx: $!";
my @subpages = ();
my %seen = ();
while (<LYNX>) {
  	my($line) = $_;
  	chomp($line);
  	if($line =~ /<a href="(\/Videos-.{1,50}\.html)/) {
		push @subpages, "http://www.regenmag.com$1";
		if ($1 =~ /Videos-\d+-(.+)\.html/){
			my $filename = $1;
			$filename =~ s/-/\./g;
			push @subpages, $filename;
		}			
	}
}
#remove duplicate links
my %subpages = @subpages;
#search each subpage for youtube links
for my $i (keys %subpages) {
	open LYNX, "lynx -source $i |" or die "Can't open lynx: $!";
	while (<LYNX>) {
	  	my($line) = $_;
	  	chomp($line);
	  	if($line =~ /youtube\.com\/v\/(\w{6,20})&autoplay/) {
			my $file = "http://www.youtube.com/watch?v=$1";
#download each one
			system ("youtube-dl -o /media/Primary/Videos/Music/$subpages{$i} $file") or print STDERR "couldn't exec youtube-dl: $!";
			sleep 2;
#transcode to mpeg
			system ("ffmpeg -i /media/Primary/Videos/Music/$subpages{$i} -r 25 -target vcd /media/Primary/Videos/Music/$subpages{$i}.mpg") or print STDERR "couldn't exec ffmpeg: $!";
			sleep 2;
			system ("rm /media/Primary/Videos/Music/$subpages{$i}") or print STDERR "couldn't exec rm: $!";
		}
	}
}