| 1 | package Anime::ANN; |
| 2 | |
| 3 | use LWP::UserAgent; |
| 4 | use HTML::Entities; |
| 5 | |
| 6 | $ver = "0.1"; |
| 7 | |
| 8 | sub _get |
| 9 | { |
| 10 | my($uri, $ua, $res); |
| 11 | ($uri) = @_; |
| 12 | |
| 13 | $ua = LWP::UserAgent->new; |
| 14 | $ua->agent("ANNData/$ver "); |
| 15 | |
| 16 | $res = $ua->request(HTTP::Request->new("GET", "$uri")); |
| 17 | |
| 18 | die "could not fetch $uri\n" unless $res->is_success; |
| 19 | return $res->content; |
| 20 | } |
| 21 | |
| 22 | sub getlist |
| 23 | { |
| 24 | my($name, $il, $html, @ret); |
| 25 | ($name) = @_; |
| 26 | |
| 27 | $il = uc(($name =~ /^(.)/)[0]); |
| 28 | $il = "9" if (!($il =~ /[A-Z]/)); |
| 29 | $html = _get "http://www.animenewsnetwork.com/encyclopedia/anime.php?list=$il"; |
| 30 | |
| 31 | # The only way to recognize entries that seems sure is to look |
| 32 | # after the "HOVERLINE" class. |
| 33 | |
| 34 | push @ret, $1 while $html =~ /<A\s.*CLASS=HOVERLINE\s.*>.*<FONT.*>([^<>]*$name[^<>]*)<\/FONT/ig; |
| 35 | |
| 36 | return(@ret); |
| 37 | } |
| 38 | |
| 39 | sub getid |
| 40 | { |
| 41 | my($name, $il, $html, $url); |
| 42 | ($name) = @_; |
| 43 | |
| 44 | $il = uc(($name =~ /^(.)/)[0]); |
| 45 | $il = "9" if (!($il =~ /[A-Z]/)); |
| 46 | $html = _get "http://www.animenewsnetwork.com/encyclopedia/anime.php?list=$il"; |
| 47 | |
| 48 | # The only way to recognize entries that seems sure is to look |
| 49 | # after the "HOVERLINE" class. |
| 50 | |
| 51 | (($url) = ($html =~ /<A\s.*CLASS=HOVERLINE\s.*HREF=\"([^\"]+)\".*$name/i)) || return; |
| 52 | |
| 53 | return ($url =~ /\?id=(\d+)$/)[0]; |
| 54 | } |
| 55 | |
| 56 | sub getthemes |
| 57 | { |
| 58 | my($html, $kind, @ret); |
| 59 | ($html, $kind) = @_; |
| 60 | |
| 61 | if($html =~ /$kind theme:<\/b>\n/igc) { |
| 62 | my(@parts, $ct, $buf); |
| 63 | while($html =~ /\G\<br\> (([^<>]|\<i\>|<\/i>)+)/igc) { |
| 64 | $buf = $1; |
| 65 | # 0 1 2 3 4 5 6 7 8 9 10 11 |
| 66 | if(@parts = ($buf =~ /(\#(\d+):)?\s*\"([^\"\(]+)(\s+\((\<i\>(.*)<\/i>(;\s*)?)?([^<>]+)?\))?\"\s+by\s+([^\(]*[^\(\s])(\s*\(eps (\d+)-(\d+)?\))?/i)) { |
| 67 | $ct = {}; |
| 68 | $ct->{"num"} = $parts[1] if defined $parts[1]; |
| 69 | if(defined $parts[5]) { |
| 70 | $ct->{"tit"} = $parts[5]; |
| 71 | $ct->{"jat"} = decode_entities($parts[2]) if defined $parts[2]; |
| 72 | } else { |
| 73 | $ct->{"tit"} = $parts[2] if defined $parts[2]; |
| 74 | } |
| 75 | $ct->{"ent"} = $parts[7] if defined $parts[7]; |
| 76 | $ct->{"prf"} = $parts[8] if defined $parts[8]; |
| 77 | $ct->{"fep"} = $parts[10] if defined $parts[10]; |
| 78 | $ct->{"lep"} = $parts[11] if defined $parts[11]; |
| 79 | push @ret, $ct; |
| 80 | } |
| 81 | } |
| 82 | } |
| 83 | |
| 84 | return \@ret; |
| 85 | } |
| 86 | |
| 87 | sub getseries |
| 88 | { |
| 89 | my($id, $buf, $html, %ret); |
| 90 | ($id) = @_; |
| 91 | |
| 92 | $html = _get "http://www.animenewsnetwork.com/encyclopedia/anime.php?id=$id"; |
| 93 | |
| 94 | $ret{"url"} = "http://www.animenewsnetwork.com/encyclopedia/anime.php?id=$id"; |
| 95 | ($ret{"name"}) = ($html =~ /\<TITLE\>Anime News Network - ([^<]*)<\/TITLE>/); |
| 96 | if(($buf) = ($html =~ /vintage:<\/b>\n([^<]+)</is)) { |
| 97 | $ret{"vtg"} = $buf; |
| 98 | } |
| 99 | if(($buf) = ($html =~ /number of episodes:<\/b>\n([^<]+)</is)) { |
| 100 | $ret{"eps"} = $buf; |
| 101 | } |
| 102 | $buf = getthemes $html, "opening"; |
| 103 | $ret{"op"} = $buf if(@{$buf}); |
| 104 | $buf = getthemes $html, "ending"; |
| 105 | $ret{"ed"} = $buf if(@{$buf}); |
| 106 | |
| 107 | return \%ret; |
| 108 | } |