_ Git - youtube-dl/blob - youtube_dl/extractor/animeondemand.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_urlparse,
   8     compat_str,
   9 )
  10 from ..utils import (
  11     determine_ext,
  12     extract_attributes,
  13     ExtractorError,
  14     sanitized_Request,
  15     urlencode_postdata,
  16 )
  17
  18
  19 class AnimeOnDemandIE(InfoExtractor):
  20     _VALID_URL = r'https?://(?:www\.)?anime-on-demand\.de/anime/(?P<id>\d+)'
  21     _LOGIN_URL = 'https://www.anime-on-demand.de/users/sign_in'
  22     _APPLY_HTML5_URL = 'https://www.anime-on-demand.de/html5apply'
  23     _NETRC_MACHINE = 'animeondemand'
  24     _TESTS = [{
  25         # jap, OmU
  26         'url': 'https://www.anime-on-demand.de/anime/161',
  27         'info_dict': {
  28             'id': '161',
  29             'title': 'Grimgar, Ashes and Illusions (OmU)',
  30             'description': 'md5:6681ce3c07c7189d255ac6ab23812d31',
  31         },
  32         'playlist_mincount': 4,
  33     }, {
  34         # Film wording is used instead of Episode, ger/jap, Dub/OmU
  35         'url': 'https://www.anime-on-demand.de/anime/39',
  36         'only_matching': True,
  37     }, {
  38         # Episodes without titles, jap, OmU
  39         'url': 'https://www.anime-on-demand.de/anime/162',
  40         'only_matching': True,
  41     }, {
  42         # ger/jap, Dub/OmU, account required
  43         'url': 'https://www.anime-on-demand.de/anime/169',
  44         'only_matching': True,
  45     }, {
  46         # Full length film, non-series, ger/jap, Dub/OmU, account required
  47         'url': 'https://www.anime-on-demand.de/anime/185',
  48         'only_matching': True,
  49     }, {
  50         # Flash videos
  51         'url': 'https://www.anime-on-demand.de/anime/12',
  52         'only_matching': True,
  53     }]
  54
  55     def _login(self):
  56         (username, password) = self._get_login_info()
  57         if username is None:
  58             return
  59
  60         login_page = self._download_webpage(
  61             self._LOGIN_URL, None, 'Downloading login page')
  62
  63         if '>Our licensing terms allow the distribution of animes only to German-speaking countries of Europe' in login_page:
  64             self.raise_geo_restricted(
  65                 '%s is only available in German-speaking countries of Europe' % self.IE_NAME)
  66
  67         login_form = self._form_hidden_inputs('new_user', login_page)
  68
  69         login_form.update({
  70             'user[login]': username,
  71             'user[password]': password,
  72         })
  73
  74         post_url = self._search_regex(
  75             r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
  76             'post url', default=self._LOGIN_URL, group='url')
  77
  78         if not post_url.startswith('http'):
  79             post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
  80
  81         request = sanitized_Request(
  82             post_url, urlencode_postdata(login_form))
  83         request.add_header('Referer', self._LOGIN_URL)
  84
  85         response = self._download_webpage(
  86             request, None, 'Logging in as %s' % username)
  87
  88         if all(p not in response for p in ('>Logout<', 'href="/users/sign_out"')):
  89             error = self._search_regex(
  90                 r'<p class="alert alert-danger">(.+?)</p>',
  91                 response, 'error', default=None)
  92             if error:
  93                 raise ExtractorError('Unable to login: %s' % error, expected=True)
  94             raise ExtractorError('Unable to log in')
  95
  96     def _real_initialize(self):
  97         self._login()
  98
  99     def _real_extract(self, url):
 100         anime_id = self._match_id(url)
 101
 102         webpage = self._download_webpage(url, anime_id)
 103
 104         if 'data-playlist=' not in webpage:
 105             self._download_webpage(
 106                 self._APPLY_HTML5_URL, anime_id,
 107                 'Activating HTML5 beta', 'Unable to apply HTML5 beta')
 108             webpage = self._download_webpage(url, anime_id)
 109
 110         csrf_token = self._html_search_meta(
 111             'csrf-token', webpage, 'csrf token', fatal=True)
 112
 113         anime_title = self._html_search_regex(
 114             r'(?s)<h1[^>]+itemprop="name"[^>]*>(.+?)</h1>',
 115             webpage, 'anime name')
 116         anime_description = self._html_search_regex(
 117             r'(?s)<div[^>]+itemprop="description"[^>]*>(.+?)</div>',
 118             webpage, 'anime description', default=None)
 119
 120         entries = []
 121
 122         def extract_info(html, video_id, num=None):
 123             title, description = [None] * 2
 124             formats = []
 125
 126             for input_ in re.findall(
 127                     r'<input[^>]+class=["\'].*?streamstarter[^>]+>', html):
 128                 attributes = extract_attributes(input_)
 129                 title = attributes.get('data-dialog-header')
 130                 playlist_urls = []
 131                 for playlist_key in ('data-playlist', 'data-otherplaylist', 'data-stream'):
 132                     playlist_url = attributes.get(playlist_key)
 133                     if isinstance(playlist_url, compat_str) and re.match(
 134                             r'/?[\da-zA-Z]+', playlist_url):
 135                         playlist_urls.append(attributes[playlist_key])
 136                 if not playlist_urls:
 137                     continue
 138
 139                 lang = attributes.get('data-lang')
 140                 lang_note = attributes.get('value')
 141
 142                 for playlist_url in playlist_urls:
 143                     kind = self._search_regex(
 144                         r'videomaterialurl/\d+/([^/]+)/',
 145                         playlist_url, 'media kind', default=None)
 146                     format_id_list = []
 147                     if lang:
 148                         format_id_list.append(lang)
 149                     if kind:
 150                         format_id_list.append(kind)
 151                     if not format_id_list and num is not None:
 152                         format_id_list.append(compat_str(num))
 153                     format_id = '-'.join(format_id_list)
 154                     format_note = ', '.join(filter(None, (kind, lang_note)))
 155                     request = sanitized_Request(
 156                         compat_urlparse.urljoin(url, playlist_url),
 157                         headers={
 158                             'X-Requested-With': 'XMLHttpRequest',
 159                             'X-CSRF-Token': csrf_token,
 160                             'Referer': url,
 161                             'Accept': 'application/json, text/javascript, */*; q=0.01',
 162                         })
 163                     playlist = self._download_json(
 164                         request, video_id, 'Downloading %s playlist JSON' % format_id,
 165                         fatal=False)
 166                     if not playlist:
 167                         continue
 168                     stream_url = playlist.get('streamurl')
 169                     if stream_url:
 170                         rtmp = re.search(
 171                             r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+/))(?P<playpath>mp[34]:.+)',
 172                             stream_url)
 173                         if rtmp:
 174                             formats.append({
 175                                 'url': rtmp.group('url'),
 176                                 'app': rtmp.group('app'),
 177                                 'play_path': rtmp.group('playpath'),
 178                                 'page_url': url,
 179                                 'player_url': 'https://www.anime-on-demand.de/assets/jwplayer.flash-55abfb34080700304d49125ce9ffb4a6.swf',
 180                                 'rtmp_real_time': True,
 181                                 'format_id': 'rtmp',
 182                                 'ext': 'flv',
 183                             })
 184                             continue
 185                     start_video = playlist.get('startvideo', 0)
 186                     playlist = playlist.get('playlist')
 187                     if not playlist or not isinstance(playlist, list):
 188                         continue
 189                     playlist = playlist[start_video]
 190                     title = playlist.get('title')
 191                     if not title:
 192                         continue
 193                     description = playlist.get('description')
 194                     for source in playlist.get('sources', []):
 195                         file_ = source.get('file')
 196                         if not file_:
 197                             continue
 198                         ext = determine_ext(file_)
 199                         format_id_list = [lang, kind]
 200                         if ext == 'm3u8':
 201                             format_id_list.append('hls')
 202                         elif source.get('type') == 'video/dash' or ext == 'mpd':
 203                             format_id_list.append('dash')
 204                         format_id = '-'.join(filter(None, format_id_list))
 205                         if ext == 'm3u8':
 206                             file_formats = self._extract_m3u8_formats(
 207                                 file_, video_id, 'mp4',
 208                                 entry_protocol='m3u8_native', m3u8_id=format_id, fatal=False)
 209                         elif source.get('type') == 'video/dash' or ext == 'mpd':
 210                             continue
 211                             file_formats = self._extract_mpd_formats(
 212                                 file_, video_id, mpd_id=format_id, fatal=False)
 213                         else:
 214                             continue
 215                         for f in file_formats:
 216                             f.update({
 217                                 'language': lang,
 218                                 'format_note': format_note,
 219                             })
 220                         formats.extend(file_formats)
 221
 222             return {
 223                 'title': title,
 224                 'description': description,
 225                 'formats': formats,
 226             }
 227
 228         def extract_entries(html, video_id, common_info, num=None):
 229             info = extract_info(html, video_id, num)
 230
 231             if info['formats']:
 232                 self._sort_formats(info['formats'])
 233                 f = common_info.copy()
 234                 f.update(info)
 235                 entries.append(f)
 236
 237             # Extract teaser/trailer only when full episode is not available
 238             if not info['formats']:
 239                 m = re.search(
 240                     r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>(?P<kind>Teaser|Trailer)<',
 241                     html)
 242                 if m:
 243                     f = common_info.copy()
 244                     f.update({
 245                         'id': '%s-%s' % (f['id'], m.group('kind').lower()),
 246                         'title': m.group('title'),
 247                         'url': compat_urlparse.urljoin(url, m.group('href')),
 248                     })
 249                     entries.append(f)
 250
 251         def extract_episodes(html):
 252             for num, episode_html in enumerate(re.findall(
 253                     r'(?s)<h3[^>]+class="episodebox-title".+?>Episodeninhalt<', html), 1):
 254                 episodebox_title = self._search_regex(
 255                     (r'class="episodebox-title"[^>]+title=(["\'])(?P<title>.+?)\1',
 256                      r'class="episodebox-title"[^>]+>(?P<title>.+?)<'),
 257                     episode_html, 'episodebox title', default=None, group='title')
 258                 if not episodebox_title:
 259                     continue
 260
 261                 episode_number = int(self._search_regex(
 262                     r'(?:Episode|Film)\s*(\d+)',
 263                     episodebox_title, 'episode number', default=num))
 264                 episode_title = self._search_regex(
 265                     r'(?:Episode|Film)\s*\d+\s*-\s*(.+)',
 266                     episodebox_title, 'episode title', default=None)
 267
 268                 video_id = 'episode-%d' % episode_number
 269
 270                 common_info = {
 271                     'id': video_id,
 272                     'series': anime_title,
 273                     'episode': episode_title,
 274                     'episode_number': episode_number,
 275                 }
 276
 277                 extract_entries(episode_html, video_id, common_info)
 278
 279         def extract_film(html, video_id):
 280             common_info = {
 281                 'id': anime_id,
 282                 'title': anime_title,
 283                 'description': anime_description,
 284             }
 285             extract_entries(html, video_id, common_info)
 286
 287         extract_episodes(webpage)
 288
 289         if not entries:
 290             extract_film(webpage, anime_id)
 291
 292         return self.playlist_result(entries, anime_id, anime_title, anime_description)