X-Git-Url: https://git.camperquake.de/gitweb.cgi?a=blobdiff_plain;f=videosite%2FYouTubeGrabber.pm;h=ad03f6a6366f9860de6377502a5b5abd98808fd1;hb=58d2597498d4162e2a3811978f59be7bedf2f9b6;hp=a70993e481dce9284055fa661a2e65567c4ad064;hpb=0f81249e945336a480456645a710f04589af9437;p=videosite.git diff --git a/videosite/YouTubeGrabber.pm b/videosite/YouTubeGrabber.pm index a70993e..ad03f6a 100644 --- a/videosite/YouTubeGrabber.pm +++ b/videosite/YouTubeGrabber.pm @@ -20,47 +20,84 @@ use videosite::JSArrayParser; use strict; 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]); + 'insane' => [38, 37, 137, 22, 136, 35, 135, 18, 134, 34, 6, 5, 43], + 'hd' => [37, 137, 22, 136, 35, 135, 18, 134, 34, 6, 5, 38, 43], + 'h264' => [18, 134, 34, 37, 137, 22, 136, 35, 135, 6, 5, 38, 43], + 'high' => [34, 35, 135, 18, 134, 37, 137, 22, 136, 6, 5, 38, 43], + 'normal' => [6, 5, 34, 35, 135, 18, 134, 22, 136, 37, 137, 38, 43]); +my %audiopreflist = ( + 'insane' => [172,141,171,140,139], + 'hd' => [172,141,171,140,139], + 'h264' => [172,141,171,140,139], + 'high' => [171,140,172,141,139], + 'normal' => [171,140,172,141,139]); 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', + # Container/Video codec/Audio codec/Resolution + 5 => 'FLV/Sorenson/MP3/240p', + 6 => 'FLV/Sorenson/MP3/270p', + 13 => '3GP/MPEG4-Visual/144p', # 0.5MBit + 17 => '3GP/MPEG4-Visual/144p', # 2MBit + 18 => 'MP4/H264/AAC/360p', # isommp42, Baseline + 22 => 'MP4/H264/AAC/720p', # isommp42, High + 34 => 'FLV/H264/AAC/360p', # Main + 35 => 'FLV/H264/AAC/480p', # Main + 36 => '3GP/MPEG4-Visual/240p', + 37 => 'MP4/H264/AAC/1080p', # High + 38 => 'MP4/H264/AAC/3072p', # High + 43 => 'WebM/VP8/Vorbis/360p', + 44 => 'WebM/VP8/Vorbis/480p', + 45 => 'WebM/VP8/Vorbis/720p', + 46 => 'WebM/VP8/Vorbis/1080p/3D', # effective 540p + 82 => 'MP4/H264/AAC/360p/3D', # isomavc1mp42 + 83 => 'MP4/H264/AAC/240p/3D', # isomavc1mp42 + 84 => 'MP4/H264/AAC/720p/3D', # isomavc1mp42 + 85 => 'MP4/H264/AAC/1080p/3D', # isomavc1mp42, effective 540p + 100 => 'WebM/VP8/Vorbis/360p/3D', + 101 => 'WebM/VP8/Vorbis/480p/3D', + 102 => 'WebM/VP8/Vorbis/720p/3D', + 133 => 'MP4/H264/240p/AdaptiveVideo', + 134 => 'MP4/H264/360p/AdaptiveVideo', + 135 => 'MP4/H264/480p/AdaptiveVideo', + 136 => 'MP4/H264/720p/AdaptiveVideo', + 137 => 'MP4/H264/1080p/AdaptiveVideo', + 139 => 'MP4/AAC/48kbit/AdaptiveAudio', + 140 => 'MP4/AAC/128kbit/AdaptiveAudio', + 141 => 'MP4/AAC/256kbit/AdaptiveAudio', + 160 => 'MP4/H264/144p/AdaptiveVideo', + 171 => 'WebM/Vorbis/128kbit/AdaptiveAudio', + 172 => 'WebM/Vorbis/160kbit/AdaptiveAudio', ); sub new { my $class = shift; - my $self = $class->SUPER::new(); - - $self->{'NAME'} = 'youtube'; - $self->{'PATTERNS'} = ['(https?://(?:[-a-zA-Z0-9_.]+\.)*youtube\.(?:com|de|co.uk)/watch(?:_popup)?\?(?:.+=.+&)*v=([-a-zA-Z0-9_]+))', - '(https?://(?:[-a-zA-Z0-9_.]+\.)*youtube\.(?:com|de|co.uk)/watch\#\!v=([-a-zA-Z0-9_]+))', - '(https?://(?:[-a-zA-Z0-9_.]+\.)*youtube\.(?:com|de|co.uk)/v/([-a-zA-Z0-9_]+))', - '(https?://(?:[-a-zA-Z0-9_.]+\.)*youtube\.(?:com|de|co.uk)/user/[[:alnum:]]+\?v=([-a-zA-Z0-9_]+))', - '(https?://(?:[-a-zA-Z0-9_.]+\.)*youtube\.(?:com|de|co.uk)/(?:user/)?[[:alnum:]]+#p/(?:\w+/)+\d+/([-a-zA-Z0-9_]+))', - '(https?://(?:[-a-zA-Z0-9_.]+\.)*youtu\.be/watch\?v=([-a-zA-Z0-9_]+))', - '(https?://(?:[-a-zA-Z0-9_.]+\.)*youtu\.be/([-a-zA-Z0-9_]+))', - '(https?://(?:[-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(); - - return $self; + my $self = $class->SUPER::new( + NAME => 'youtube', + _SELFTESTURL => 'http://www.youtube.com/watch?v=dMH0bHeiRNg', + _SELFTESTTITLE => 'Evolution of Dance - By Judson Laipply', + PATTERNS => ['(https?://(?:[-a-zA-Z0-9_.]+\.)*youtube\.(?:com|de|co.uk)/watch(?:_popup)?/?\?.*?v=([-a-zA-Z0-9_]+))', + '(https?://(?:[-a-zA-Z0-9_.]+\.)*youtube\.(?:com|de|co.uk)/watch\#\!v=([-a-zA-Z0-9_]+))', + '(https?://(?:[-a-zA-Z0-9_.]+\.)*youtube\.(?:com|de|co.uk)/v/([-a-zA-Z0-9_]+))', + '(https?://(?:[-a-zA-Z0-9_.]+\.)*youtube\.(?:com|de|co.uk)/embed/([-a-zA-Z0-9_]+))', + '(https?://(?:[-a-zA-Z0-9_.]+\.)*youtube\.(?:com|de|co.uk)/user/[[:alnum:]]+\?v=([-a-zA-Z0-9_]+))', + '(https?://(?:[-a-zA-Z0-9_.]+\.)*youtube\.(?:com|de|co.uk)/(?:user/)?[[:alnum:]]+#p/(?:\w+/)+\d+/([-a-zA-Z0-9_]+))', + '(https?://(?:[-a-zA-Z0-9_.]+\.)*youtu\.be/watch\?v=([-a-zA-Z0-9_]+))', + '(https?://(?:[-a-zA-Z0-9_.]+\.)*youtu\.be/([-a-zA-Z0-9_]+))', + '(https?://(?:[-a-zA-Z0-9_.]+\.)*youtube\.(?:com|de|co.uk)/user/\w+\?.*/([-a-zA-Z0-9_]+))'], + _PARAMS => { + QUALITY => ['normal', 'Quality of the video to download.', { + normal => 'standard resolution flash video', + high => 'higher resolution flash video', + h264 => 'high resolution MPEG4 video', + hd => 'HD720 resolution'}], + USERNAME => ['', 'Username to use for YouTube login'], + PASSWORD => ['', 'Password to use for YouTube login'], + HTTPS => [1, 'Whether to use HTTPS (if available) to connect to YouTube'], + ADAPTIVE => [0, 'Whether to use adaptive rate URLs (if available)'], + }, + @_, + ); + + return bless($self, $class); } sub _parse { @@ -70,22 +107,35 @@ sub _parse { my $id; my $res; - my $url =~ m|$pattern|; - my $url = $1; - my $id = $2; + $url =~ m|$pattern|; + $url = $1; + $id = $2; $self->debug("Matched id %s from pattern %s", $id, $pattern); $res = $self->_parse_by_video_info($url, $id); - if (defined($res) && ref($res)) { - return $res; + if (defined($res)) { + if (ref($res)) { + return $res; + } else { + $self->debug("_parse_by_video_info failed with status noretry"); + return undef; + } } else { + $self->debug("_parse_by_video_info failed with status retry"); $res = $self->_parse_by_scrape($url, $id); } return $res; } +# +# Try to get video information by using the API call. +# +# Returns hashref on success +# Returns undef on retryable error (try to scrape the website) +# Returns 0 on non-retryable error +# sub _parse_by_video_info { my $self = shift; my $url = shift; @@ -95,6 +145,7 @@ sub _parse_by_video_info { my $videourl; my $ua = $self->ua(); my $preflist; + my $audiopreflist; my $r; my $content; my $urls; @@ -105,13 +156,16 @@ sub _parse_by_video_info { $metadata->{'SOURCE'} = $self->{'NAME'}; $metadata->{'TITLE'} = undef; $metadata->{'DLURL'} = undef; - $metadata->{'COOKIE'} = undef; $preflist = $preflist{$quality}; - $self->debug("Quality: %s, preflist: [%s]", $quality, join(", ", @{$preflist})); - - $videourl = sprintf('https://www.youtube.com/get_video_info?video_id=%s&eurl=%s', - $id, 'http%3A%2F%2Fwww%2Eyoutube%2Ecom%2F'); + $audiopreflist = $audiopreflist{$quality}; + $self->debug("Quality: %s, preflist: [%s], audiopreflist: [%s]", + $quality, + join(", ", @{$preflist}), + join(", ", @{$audiopreflist})); + + $videourl = sprintf('%s://www.youtube.com/get_video_info?video_id=%s&eurl=%s', + $self->_getval('HTTPS')?'https':'http', $id, 'http%3A%2F%2Fwww%2Eyoutube%2Ecom%2F'); $self->debug("Video info URL: %s", $videourl); $r = $ua->get($videourl); @@ -124,28 +178,54 @@ sub _parse_by_video_info { $self->debug('Content from get_video_info: %s', $content); # Decode content - $content = { split /[&=]/, $content }; + $content = $self->decode_querystring($content); + + $self->debug("Decoded get_video_info: %s", Dumper($content)); if ($content->{'status'} ne 'ok') { $self->debug("Non OK status code found: %s", $content->{'status'}); return undef; } - unless(exists($content->{'fmt_url_map'}) and exists($content->{'title'})) { - $self->debug("No fmt_url_map or no title found"); + # Check if this is live content + if (exists($content->{ps}) and ($content->{ps} eq 'live')) { + $self->error("Video URL seems to point to a live stream, cannot save this"); + return 0; + } + + if (exists($content->{'fmt_url_map'})) { + # Decode fmt_url_map + $urls = $self->decode_hexurl($content->{'fmt_url_map'}); + $urls = { split /[\|,]/, $urls }; + } elsif (exists($content->{'url_encoded_fmt_stream_map'})) { + $urls = $self->_decode_url_encoded_fmt_stream_map($content->{'url_encoded_fmt_stream_map'}, 1); + } else { + $self->debug("No URL data found"); return undef; } - # Decode fmt_url_map - $urls = $content->{'fmt_url_map'}; - $urls =~ s/%(..)/chr(hex($1))/ge; - $urls = { split /[\|,]/, $urls }; + if (exists($content->{'adaptive_fmts'})) { + $self->debug('Video has adaptive_fmts'); + if ($self->_getval('ADAPTIVE')) { + my %af = %{$self->_decode_url_encoded_fmt_stream_map($content->{'adaptive_fmts'}, 1)}; + while (my ($k, $v) = each %af) { + $urls->{$k} = $v; + } + } else { + $self->debug('Use of adaptive URLs not enabled'); + } + } - $self->__pick_url($urls, $preflist, $metadata); + unless(exists($content->{'title'})) { + $self->debug("No title found"); + return undef; + } + + $self->__pick_url($urls, $preflist, $audiopreflist, $metadata); $metadata->{'TITLE'} = $content->{'title'}; $metadata->{'TITLE'} =~ s/\+/ /g; - $metadata->{'TITLE'} =~ s/%(..)/chr(hex($1))/ge; + $metadata->{'TITLE'} = $self->decode_hexurl($metadata->{'TITLE'}); $metadata->{'TITLE'} = decode("utf8", $metadata->{'TITLE'}); $self->debug('Title found: %s', $metadata->{'TITLE'}); @@ -180,13 +260,12 @@ sub _parse_by_scrape { $metadata->{'SOURCE'} = $self->{'NAME'}; $metadata->{'TITLE'} = undef; $metadata->{'DLURL'} = undef; - $metadata->{'COOKIE'} = undef; $preflist = $preflist{$quality}; $self->debug("Quality: %s, preflist: [%s]", $quality, join(", ", @{$preflist})); - $videourl = sprintf('https://www.youtube.com/watch?v=%s', $id); + $videourl = sprintf('%s://www.youtube.com/watch?v=%s', $self->_getval('HTTPS')?'https':'http', $id); unless(defined($r = $ua->get($videourl))) { $self->error('Could not download %s', $url); @@ -197,7 +276,6 @@ sub _parse_by_scrape { $self->debug('Video requires age verification'); 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; @@ -223,10 +301,10 @@ sub _parse_by_scrape { $e = $p->get_text(); $self->debug("Found script: %s", $e); - if ($e =~ m|\x27SWF_ARGS\x27:\s+(.+),|) { + if ($e =~ m|ytplayer\.config\s*=\s*(.+);$|) { my $args = $1; + $self->debug("Found PLAYER_CONFIG: %s", $args); - $self->debug("Found SWF_ARGS: %s", $args); $jsp = videosite::JSArrayParser->new(); $self->debug("Using %s to parse", ref($jsp)); $r = $jsp->parse($args); @@ -236,54 +314,32 @@ sub _parse_by_scrape { return undef; } - if (exists($r->{'fmt_url_map'}) and ($r->{'fmt_url_map'} ne '')) { - my $urls = $r->{'fmt_url_map'}; + if (exists($r->{'args'}) and exists($r->{'args'}->{'ps'}) and ($r->{'args'}->{'ps'} eq 'live')) { + $self->error("Video URL seems to point to a live stream, cannot save this"); + return undef; + } + + if (exists($r->{'args'}) and exists($r->{'args'}->{'fmt_url_map'}) and ($r->{'args'}->{'fmt_url_map'} ne '')) { + my $urls = $r->{'args'}->{'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"); + foreach (keys(%urls)) { + $urls{$_} = $self->decode_hexurl($urls{$_}); } + $self->debug("Pagetype: 2011 (PLAYER_CONFIG), fmt_url_map"); + } elsif (exists($r->{'args'}) and exists($r->{'args'}->{'url_encoded_fmt_stream_map'}) and ($r->{'args'}->{'url_encoded_fmt_stream_map'} ne '')) { + %urls = %{$self->_decode_url_encoded_fmt_stream_map($r->{'args'}->{'url_encoded_fmt_stream_map'}, 0)}; + + $self->debug("Pagetype: 2011 (PLAYER_CONFIG), url_encoded_fmt_stream_map"); } else { - $self->error('Neither fmt_url_map nor t found in video information hash'); + $self->error('fmt_url_map not found in PLAYER_CONFIG'); return undef; } - } elsif ($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); - $self->debug("Pagetype: 2010 (swfHTML), fmt_url_map"); - } elsif ($e =~ m|\x27PLAYER_CONFIG\x27:\s+(.+)\}\);|) { + } elsif ($e =~ m|yt\.playerConfig\s*=\s*(.+);\n|) { my $args = $1; - $self->debug("Found PLAYER_CONFIG: %s", $args); + $self->debug("Found yt.playerConfig: %s", $args); $jsp = videosite::JSArrayParser->new(); $self->debug("Using %s to parse", ref($jsp)); @@ -299,26 +355,19 @@ sub _parse_by_scrape { return undef; } - if (exists($r->{'args'}) and exists($r->{'args'}->{'fmt_url_map'}) and ($r->{'args'}->{'fmt_url_map'} ne '')) { - my $urls = $r->{'args'}->{'fmt_url_map'}; - - $self->debug("Video has fmt_url_map: %s", $urls); + if (exists($r->{'args'}) and exists($r->{'args'}->{'url_encoded_fmt_stream_map'}) and ($r->{'args'}->{'url_encoded_fmt_stream_map'} ne '')) { + %urls = %{$self->_decode_url_encoded_fmt_stream_map($r->{'args'}->{'url_encoded_fmt_stream_map'}, 0)}; - %urls = split(/[\|,]/, $urls); - foreach (keys(%urls)) { - my $u = $urls{$_}; - $u =~ s/%([[:xdigit:]]{2})/chr(hex($1))/ge; - $urls{$_} = $u; - } - $self->debug("Pagetype: 2011 (PLAYER_CONFIG), fmt_url_map"); + $self->debug("Pagetype: 2012 (yt.playerConfig), url_encoded_fmt_stream_map"); } else { - $self->error('fmt_url_map not found in PLAYER_CONFIG'); + $self->error('url_map not found in yt.playerConfig'); return undef; } } + if (%urls) { - $self->__pick_url(\%urls, $preflist, $metadata); + $self->__pick_url(\%urls, $preflist, undef, $metadata); last SWF_ARGS; } } elsif ('div' eq $tag->[0]) { @@ -448,11 +497,50 @@ sub __login { return ($ua->get($videourl), $cookie); } +# Take an encoded url_encoded_fmt_stream_map and return a hash +# matching video IDs to download URLs +sub _decode_url_encoded_fmt_stream_map { + my $self = shift; + my $data = shift; + my $dataencoded = shift; + my @data; + my $h = {}; + + $data = $self->decode_hexurl($data) if $dataencoded; + # This will + # - Split the decoded string into segments (along ,) + # - Interpret each segment as a concatenated key-value list (key and value separated by =, pairs separated by & + # - URL-decode each key and value _again_ + # + # @data will be an array of hash references + + @data = map { { map { $self->decode_hexurl($_) } split /[&=]/ } } split /,/, $data; + $self->debug("_decode_url_encoded_fmt_stream_map() decoded %s", Dumper(\@data)); + + # From each array entry, pick the itag and the url values and return that + # as a hash reference + + + foreach (@data) { + if (exists($_->{'sig'})) { + $h->{$_->{'itag'}} = sprintf('%s&signature=%s', $_->{'url'}, $_->{'sig'}); + } else { + $h->{$_->{'itag'}} = $_->{'url'}; + } + } + + return $h; +} + + + sub __pick_url { my $self = shift; my $urls = shift; my $preflist = shift; + my $audiopreflist = shift; my $metadata = shift; + my $picked_format; foreach (keys(%{$urls})) { if (exists($videoformats{$_})) { @@ -466,11 +554,40 @@ sub __pick_url { if (exists($urls->{$_})) { $self->debug("Selected URL with quality level %s", $_); $metadata->{'DLURL'} = $urls->{$_}; + $picked_format = $_; + if (exists($videoformats{$_})) { + $metadata->{'FORMAT'} = $videoformats{$_}; + } else { + $metadata->{'FORMAT'} = 'unknown'; + } last; } } + if (defined($picked_format)) { + if ($metadata->{'FORMAT'} =~ m/Adaptive/) { + if (defined($audiopreflist)) { + # Pick an audio format as well + foreach (@{$audiopreflist}) { + if (exists($urls->{$_})) { + $self->debug("Selected audio URL with quality level %s", $_); + $metadata->{'DLURL_AUDIO'} = $urls->{$_}; + if (exists($videoformats{$_})) { + $metadata->{'FORMAT_AUDIO'} = $videoformats{$_}; + } else { + $metadata->{'FORMAT_AUDIO'} = 'unknown'; + } + last; + } + } + } else { + $self->error('Need an audio preflist, but none provided'); + } + } + } + $self->debug('URL found: %s', $metadata->{'DLURL'}); } 1; +