# download strategy revised using
# http://www.kde-apps.org/content/show.php?content=41456
-package YouTubeGrabber;
+package videosite::YouTubeGrabber;
-use GrabberBase;
-@ISA = qw(GrabberBase);
+use videosite::GrabberBase;
+@ISA = qw(videosite::GrabberBase);
use LWP::UserAgent;
use HTTP::Cookies;
use HTML::TokeParser;
+use HTML::Entities qw(decode_entities);
+use Encode;
use Data::Dumper;
+use videosite::JSArrayParser;
use strict;
my $self = $class->SUPER::new();
$self->{'NAME'} = 'youtube';
- $self->{'PATTERNS'} = ['(http://(?:[-a-zA-Z0-9_.]+\.)*youtube.(?:com|de|co.uk)/watch\?(?:.+=.+&)*v=([-a-zA-Z0-9_]+))',
- '(http://(?:[-a-zA-Z0-9_.]+\.)*youtube.(?:com|de|co.uk)/v/([-a-zA-Z0-9_]+))'];
- $self->{'_PARAMS'} = {'QUALITY' => ['normal', 'Quality of the video to download. normal = standard resolution flash video, high = higher resolution flash video, h264 = high resolution MPEG4 video'], 'USERNAME' => ['', 'Username to use for YouTube login'], 'PASSWORD' => ['', 'Password to use for YouTube login']};
+ $self->{'PATTERNS'} = ['(http://(?:[-a-zA-Z0-9_.]+\.)*youtube\.(?:com|de|co.uk)/watch\?(?:.+=.+&)*v=([-a-zA-Z0-9_]+))',
+ '(http://(?:[-a-zA-Z0-9_.]+\.)*youtube\.(?:com|de|co.uk)/watch\#\!v=([-a-zA-Z0-9_]+))',
+ '(http://(?:[-a-zA-Z0-9_.]+\.)*youtube\.(?:com|de|co.uk)/v/([-a-zA-Z0-9_]+))',
+ '(http://(?:[-a-zA-Z0-9_.]+\.)*youtu\.be/([-a-zA-Z0-9_]+))',
+ '(http://(?:[-a-zA-Z0-9_.]+\.)*youtube\.(?:com|de|co.uk)/user/\w+\?.*/([-a-zA-Z0-9_]+))'];
+ $self->{'_PARAMS'} = {
+ 'QUALITY' => ['normal', 'Quality of the video to download.', {
+ 'normal' => 'standard resolution flash video',
+ 'high' => 'higher resolution flash video',
+ 'h264' => 'high resolution MPEG4 video',
+ 'hd' => 'HD720 resolution'}],
+ 'USERNAME' => ['', 'Username to use for YouTube login'],
+ 'PASSWORD' => ['', 'Password to use for YouTube login']};
bless($self, $class);
$self->_prepare_parameters();
my $r;
my $videourl;
my $quality = $self->_getval('QUALITY');
- my $append = '';
+ my %preflist = (
+ 'insane' => [38, 37, 22, 35, 18, 34, 6, 5],
+ 'hd' => [37, 22, 35, 18, 34, 6, 5, 38],
+ 'h264' => [18, 34, 37, 22, 35, 6, 5, 38],
+ 'high' => [34, 35, 18, 37, 22, 6, 5, 38],
+ 'normal' => [6, 5, 34, 35, 18, 22, 37, 38]);
+ my %videoformats = (
+ 38 => 'mp4,h264,4k',
+ 37 => 'mp4,h264,1080p',
+ 35 => 'flv,h264,large',
+ 34 => 'flv,h264',
+ 22 => 'mp4,h264,720p',
+ 18 => 'mp4,h264',
+ 5 => 'flv,flv',
+ );
+ my $preflist;
+ my $jsp;
$url =~ m|$pattern|;
$url = $1;
$metadata->{'SOURCE'} = $self->{'NAME'};
$metadata->{'TITLE'} = undef;
$metadata->{'DLURL'} = undef;
+ $metadata->{'COOKIE'} = undef;
- if ($quality eq 'high') {
- $append = '&fmt=6';
- } elsif ($quality eq 'h264') {
- $append = '&fmt=18';
- }
+ $self->debug("Matched id %s from pattern %s", $2, $pattern);
+
+ $preflist = $preflist{$quality};
+ $self->debug("Quality: %s, preflist: [%s]", $quality, join(", ", @{$preflist}));
- $videourl = sprintf('http://www.youtube.com/watch?v=%s%s', $2, $append);
+ $videourl = sprintf('http://www.youtube.com/watch?v=%s', $2);
unless(defined($r = $ua->get($videourl))) {
$self->error('Could not download %s', $url);
if ($r->base->as_string() =~ m,/verify_age,) {
$self->debug('Video requires age verification');
$ua->cookie_jar($jar);
- $r = $self->__login($videourl, $ua);
+ my @logindata = $self->__login($videourl, $ua);
+ $r = $logindata[0];
+ $metadata->{'COOKIE'} = $logindata[1];
unless(defined($r)) {
$self->error('Could not log into YouTube');
return undef;
$p = HTML::TokeParser->new(\$content);
- while ($tag = $p->get_tag('div', 'meta', 'script')) {
+ SWF_ARGS: while ($tag = $p->get_tag('div', 'meta', 'script')) {
if ('meta' eq $tag->[0]) {
if ('title' eq $tag->[1]->{'name'}) {
$metadata->{'TITLE'} = $tag->[1]->{'content'};
+ # Convert HTML entities in the title. This is a bit convoluted.
+ $metadata->{'TITLE'} = encode("utf8",
+ decode_entities(
+ decode("utf8", $metadata->{'TITLE'})));
+
$self->debug('Title found: %s', $metadata->{'TITLE'});
}
} elsif ('script' eq $tag->[0]) {
+ my %urls;
+
$e = $p->get_text();
- if ($e =~ m|/watch_fullscreen\?(.+)\x27|) {
- my %args = map { split(/=/, $_, 2); } split(/&/, $1);
- $metadata->{'DLURL'} = sprintf('http://www.youtube.com/get_video.php?video_id=%s&t=%s%s',
- $metadata->{'ID'}, $args{'t'}, $append);
+ $self->debug("Found script: %s", $e);
+
+ if ($e =~ m|\x27SWF_ARGS\x27:\s+(.+),|) {
+ my $args = $1;
+
+ $self->debug("Found SWF_ARGS: %s", $args);
+ $jsp = videosite::JSArrayParser->new();
+ $self->debug("Using %s to parse", ref($jsp));
+ $r = $jsp->parse($args);
+
+ unless(defined($r)) {
+ $self->error("Found information hash, but could not parse");
+ return undef;
+ }
+
+ if (exists($r->{'fmt_url_map'}) and ($r->{'fmt_url_map'} ne '')) {
+ my $urls = $r->{'fmt_url_map'};
+
+ $self->debug("Video has fmt_url_map: %s", $urls);
+
+ $urls =~ s/%([[:xdigit:]]{2})/chr(hex($1))/ge;
+ %urls = split(/[\|,]/, $urls);
+ $self->debug("Pagetype: old (SWF_ARGS), fmt_url_map");
+
+ } elsif (exists($r->{'t'}) and ($r->{'t'} ne '')) {
+ my $thash = $r->{'t'};
+
+ if (exists($r->{'fmt_map'}) && ($r->{'fmt_map'} ne '')) {
+ my $fmt = $r->{'fmt_map'};
+ my @fmt;
+
+ $self->debug('Video has fmt_map');
+ $fmt =~ s/%([[:xdigit:]]{2})/chr(hex($1))/ge;
+ @fmt = split(/,/, $fmt);
+ foreach (@fmt) {
+ @_=split(/\//);
+ $urls{$_[0]} = sprintf('http://www.youtube.com/get_video?video_id=%s&fmt=%d&t=%s',
+ $metadata->{'ID'},
+ $_[0],
+ $thash);
+ }
+ $self->debug("Pagetype: 2009 (SWF_ARGS), t with fmt_map");
+
+ } else {
+ $urls{5} = sprintf('http://www.youtube.com/get_video?video_id=%s&t=%s',
+ $metadata->{'ID'},
+ $thash);
+ $self->debug("Pagetype: 2009 (SWF_ARGS), t without fmt_map");
+ }
+ } else {
+ $self->error('Neither fmt_url_map nor t found in video information hash');
+ return undef;
+ }
+ } elsif ($e =~ m|var swfHTML = .*fmt_url_map=([^\&]+)\&|) {
+ my $urls = $1;
+ $self->debug("Video has fmt_url_map: %s", $urls);
+
+ $urls =~ s/%([[:xdigit:]]{2})/chr(hex($1))/ge;
+ %urls = split(/[\|,]/, $urls);
+ $self->debug("Pagetype: 2010 (swfHTML), fmt_url_map");
+ }
+
+ if (%urls) {
+ foreach (keys(%urls)) {
+ if (exists($videoformats{$_})) {
+ $self->debug('Found URL for format %s (%s): %s', $_, $videoformats{$_}, $urls{$_});
+ } else {
+ $self->error('Unknown format %s: %s', $_, $urls{$_});
+ }
+ }
+
+ foreach (@{$preflist}) {
+ if (exists($urls{$_})) {
+ $self->debug("Selected URL with quality level %s", $_);
+ $metadata->{'DLURL'} = $urls{$_};
+ last;
+ }
+ }
+
$self->debug('URL found: %s', $metadata->{'DLURL'});
+ last SWF_ARGS;
}
} elsif ('div' eq $tag->[0]) {
- if (exists($tag->[1]->{'class'}) and ('errorBox' eq $tag->[1]->{'class'})) {
+ if (exists($tag->[1]->{'class'}) and ('yt-alert-content' eq $tag->[1]->{'class'})) {
$self->error("Could not get video data for youtube %s: %s",
$metadata->{'ID'}, $p->get_trimmed_text());
return undef;
return $found;
}
+ sub get_all_cookies {
+
+ my $jar = shift;
+ my $key = shift;
+ my $val = "";
+ $jar->scan(sub { $val .= "; " if !( $val eq "" ); $val .= "$_[1]=$_[2]" });
+
+ return $val;
+ }
+
if (($user eq '') or ($pass eq '')) {
$self->error('No username or password defined for YouTube');
return undef;
}
$self->debug('Logging in');
- $r = $ua->post('http://www.youtube.com/signup', { 'next' => '/', 'current_form' => 'login', 'action_login' => '1', 'username' => $user, 'password' => $pass });
-
+ $r = $ua->get('https://www.google.com/accounts/ServiceLoginAuth?service=youtube');
+ $c = $r->decoded_content();
+ $p = HTML::TokeParser->new(\$c);
+ while (my $tag = $p->get_tag('input')) {
+ if ($tag->[1]{name} eq 'GALX') {
+ $token = $tag->[1]{value};
+ last;
+ }
+ }
+ $self->debug("GALX = .$token");
+ $r = $ua->post('https://www.google.com/accounts/ServiceLoginAuth?service=youtube', { 'service' => 'youtube', 'Email' => $user, 'Passwd' => $pass, 'GALX' => $token });
+ $c = $r -> decoded_content();
+ $p = HTML::TokeParser->new(\$c);
+ while (my $tag = $p->get_tag('script')) {
+ if($p->get_text() =~ /location\.replace\("(.+)"\)/) {
+ $token = $1;
+ $token =~ s/\\x([A-Fa-f0-9]{2})/pack('C', hex($1))/seg;
+ last;
+ }
+ }
+ $r = $ua->get($token);
unless(check_cookie($ua->cookie_jar, 'LOGIN_INFO')) {
$self->error('Could not log into YouTube');
return undef;
}
+ $self->debug("Got a cookie");
+
$r = $ua->get($videourl);
if ($r->base->as_string() =~ m,/verify_age,) {
$self->debug("Looking for session token...");
$c = $r->decoded_content();
$p = HTML::TokeParser->new(\$c);
while (my $tag = $p->get_tag('script')) {
- if ($p->get_text() =~ /gXSRF_token = '(.+)'/) {
+ if ($p->get_text() =~ /'XSRF_TOKEN': '(.+)'/) {
$token = $1;
last;
}
$r = $ua->post($r->base->as_string, { 'next_url' => $r->base->path, 'action_confirm' => 'Confirm Birth Date', 'session_token' => $token });
}
- unless(check_cookie($ua->cookie_jar, 'is_adult')) {
- $self->error('Could not authenticate session');
- return undef;
- }
+# Apparently there is no longer a specific "is_adult" cookie
+# or, by the looks of it, anything similar
+#
+# unless(check_cookie($ua->cookie_jar, 'is_adult')) {
+# $self->error('Could not authenticate session');
+# return undef;
+# }
- return $ua->get($videourl);
+ my $cookie = get_all_cookies($ua->cookie_jar);
+ return ($ua->get($videourl), $cookie);
}
1;