Raphaël G. Git Repositories - youtubedl/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 compat_str
   7 from ..utils import (
   8     determine_ext,
   9     extract_attributes,
  10     ExtractorError,
  11     url_or_none,
  12     urlencode_postdata,
  13     urljoin,
  14 )
  15
  16
  17 class AnimeOnDemandIE(InfoExtractor):
  18     _VALID_URL = r'https?://(?:www\.)?anime-on-demand\.de/anime/(?P<id>\d+)'
  19     _LOGIN_URL = 'https://www.anime-on-demand.de/users/sign_in'
  20     _APPLY_HTML5_URL = 'https://www.anime-on-demand.de/html5apply'
  21     _NETRC_MACHINE = 'animeondemand'
  22     # German-speaking countries of Europe
  23     _GEO_COUNTRIES = ['AT', 'CH', 'DE', 'LI', 'LU']
  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 = urljoin(self._LOGIN_URL, post_url)
  80
  81         response = self._download_webpage(
  82             post_url, None, 'Logging in',
  83             data=urlencode_postdata(login_form), headers={
  84                 'Referer': self._LOGIN_URL,
  85             })
  86
  87         if all(p not in response for p in ('>Logout<', 'href="/users/sign_out"')):
  88             error = self._search_regex(
  89                 r'<p[^>]+\bclass=(["\'])(?:(?!\1).)*\balert\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</p>',
  90                 response, 'error', default=None, group='error')
  91             if error:
  92                 raise ExtractorError('Unable to login: %s' % error, expected=True)
  93             raise ExtractorError('Unable to log in')
  94
  95     def _real_initialize(self):
  96         self._login()
  97
  98     def _real_extract(self, url):
  99         anime_id = self._match_id(url)
 100
 101         webpage = self._download_webpage(url, anime_id)
 102
 103         if 'data-playlist=' not in webpage:
 104             self._download_webpage(
 105                 self._APPLY_HTML5_URL, anime_id,
 106                 'Activating HTML5 beta', 'Unable to apply HTML5 beta')
 107             webpage = self._download_webpage(url, anime_id)
 108
 109         csrf_token = self._html_search_meta(
 110             'csrf-token', webpage, 'csrf token', fatal=True)
 111
 112         anime_title = self._html_search_regex(
 113             r'(?s)<h1[^>]+itemprop="name"[^>]*>(.+?)</h1>',
 114             webpage, 'anime name')
 115         anime_description = self._html_search_regex(
 116             r'(?s)<div[^>]+itemprop="description"[^>]*>(.+?)</div>',
 117             webpage, 'anime description', default=None)
 118
 119         entries = []
 120
 121         def extract_info(html, video_id, num=None):
 122             title, description = [None] * 2
 123             formats = []
 124
 125             for input_ in re.findall(
 126                     r'<input[^>]+class=["\'].*?streamstarter[^>]+>', html):
 127                 attributes = extract_attributes(input_)
 128                 title = attributes.get('data-dialog-header')
 129                 playlist_urls = []
 130                 for playlist_key in ('data-playlist', 'data-otherplaylist', 'data-stream'):
 131                     playlist_url = attributes.get(playlist_key)
 132                     if isinstance(playlist_url, compat_str) and re.match(
 133                             r'/?[\da-zA-Z]+', playlist_url):
 134                         playlist_urls.append(attributes[playlist_key])
 135                 if not playlist_urls:
 136                     continue
 137
 138                 lang = attributes.get('data-lang')
 139                 lang_note = attributes.get('value')
 140
 141                 for playlist_url in playlist_urls:
 142                     kind = self._search_regex(
 143                         r'videomaterialurl/\d+/([^/]+)/',
 144                         playlist_url, 'media kind', default=None)
 145                     format_id_list = []
 146                     if lang:
 147                         format_id_list.append(lang)
 148                     if kind:
 149                         format_id_list.append(kind)
 150                     if not format_id_list and num is not None:
 151                         format_id_list.append(compat_str(num))
 152                     format_id = '-'.join(format_id_list)
 153                     format_note = ', '.join(filter(None, (kind, lang_note)))
 154                     item_id_list = []
 155                     if format_id:
 156                         item_id_list.append(format_id)
 157                     item_id_list.append('videomaterial')
 158                     playlist = self._download_json(
 159                         urljoin(url, playlist_url), video_id,
 160                         'Downloading %s JSON' % ' '.join(item_id_list),
 161                         headers={
 162                             'X-Requested-With': 'XMLHttpRequest',
 163                             'X-CSRF-Token': csrf_token,
 164                             'Referer': url,
 165                             'Accept': 'application/json, text/javascript, */*; q=0.01',
 166                         }, fatal=False)
 167                     if not playlist:
 168                         continue
 169                     stream_url = url_or_none(playlist.get('streamurl'))
 170                     if stream_url:
 171                         rtmp = re.search(
 172                             r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+/))(?P<playpath>mp[34]:.+)',
 173                             stream_url)
 174                         if rtmp:
 175                             formats.append({
 176                                 'url': rtmp.group('url'),
 177                                 'app': rtmp.group('app'),
 178                                 'play_path': rtmp.group('playpath'),
 179                                 'page_url': url,
 180                                 'player_url': 'https://www.anime-on-demand.de/assets/jwplayer.flash-55abfb34080700304d49125ce9ffb4a6.swf',
 181                                 'rtmp_real_time': True,
 182                                 'format_id': 'rtmp',
 183                                 'ext': 'flv',
 184                             })
 185                             continue
 186                     start_video = playlist.get('startvideo', 0)
 187                     playlist = playlist.get('playlist')
 188                     if not playlist or not isinstance(playlist, list):
 189                         continue
 190                     playlist = playlist[start_video]
 191                     title = playlist.get('title')
 192                     if not title:
 193                         continue
 194                     description = playlist.get('description')
 195                     for source in playlist.get('sources', []):
 196                         file_ = source.get('file')
 197                         if not file_:
 198                             continue
 199                         ext = determine_ext(file_)
 200                         format_id_list = [lang, kind]
 201                         if ext == 'm3u8':
 202                             format_id_list.append('hls')
 203                         elif source.get('type') == 'video/dash' or ext == 'mpd':
 204                             format_id_list.append('dash')
 205                         format_id = '-'.join(filter(None, format_id_list))
 206                         if ext == 'm3u8':
 207                             file_formats = self._extract_m3u8_formats(
 208                                 file_, video_id, 'mp4',
 209                                 entry_protocol='m3u8_native', m3u8_id=format_id, fatal=False)
 210                         elif source.get('type') == 'video/dash' or ext == 'mpd':
 211                             continue
 212                             file_formats = self._extract_mpd_formats(
 213                                 file_, video_id, mpd_id=format_id, fatal=False)
 214                         else:
 215                             continue
 216                         for f in file_formats:
 217                             f.update({
 218                                 'language': lang,
 219                                 'format_note': format_note,
 220                             })
 221                         formats.extend(file_formats)
 222
 223             return {
 224                 'title': title,
 225                 'description': description,
 226                 'formats': formats,
 227             }
 228
 229         def extract_entries(html, video_id, common_info, num=None):
 230             info = extract_info(html, video_id, num)
 231
 232             if info['formats']:
 233                 self._sort_formats(info['formats'])
 234                 f = common_info.copy()
 235                 f.update(info)
 236                 entries.append(f)
 237
 238             # Extract teaser/trailer only when full episode is not available
 239             if not info['formats']:
 240                 m = re.search(
 241                     r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>(?P<kind>Teaser|Trailer)<',
 242                     html)
 243                 if m:
 244                     f = common_info.copy()
 245                     f.update({
 246                         'id': '%s-%s' % (f['id'], m.group('kind').lower()),
 247                         'title': m.group('title'),
 248                         'url': urljoin(url, m.group('href')),
 249                     })
 250                     entries.append(f)
 251
 252         def extract_episodes(html):
 253             for num, episode_html in enumerate(re.findall(
 254                     r'(?s)<h3[^>]+class="episodebox-title".+?>Episodeninhalt<', html), 1):
 255                 episodebox_title = self._search_regex(
 256                     (r'class="episodebox-title"[^>]+title=(["\'])(?P<title>.+?)\1',
 257                      r'class="episodebox-title"[^>]+>(?P<title>.+?)<'),
 258                     episode_html, 'episodebox title', default=None, group='title')
 259                 if not episodebox_title:
 260                     continue
 261
 262                 episode_number = int(self._search_regex(
 263                     r'(?:Episode|Film)\s*(\d+)',
 264                     episodebox_title, 'episode number', default=num))
 265                 episode_title = self._search_regex(
 266                     r'(?:Episode|Film)\s*\d+\s*-\s*(.+)',
 267                     episodebox_title, 'episode title', default=None)
 268
 269                 video_id = 'episode-%d' % episode_number
 270
 271                 common_info = {
 272                     'id': video_id,
 273                     'series': anime_title,
 274                     'episode': episode_title,
 275                     'episode_number': episode_number,
 276                 }
 277
 278                 extract_entries(episode_html, video_id, common_info)
 279
 280         def extract_film(html, video_id):
 281             common_info = {
 282                 'id': anime_id,
 283                 'title': anime_title,
 284                 'description': anime_description,
 285             }
 286             extract_entries(html, video_id, common_info)
 287
 288         extract_episodes(webpage)
 289
 290         if not entries:
 291             extract_film(webpage, anime_id)
 292
 293         return self.playlist_result(entries, anime_id, anime_title, anime_description)