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     urlencode_postdata,
  12     urljoin,
  13 )
  14
  15
  16 class AnimeOnDemandIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?anime-on-demand\.de/anime/(?P<id>\d+)'
  18     _LOGIN_URL = 'https://www.anime-on-demand.de/users/sign_in'
  19     _APPLY_HTML5_URL = 'https://www.anime-on-demand.de/html5apply'
  20     _NETRC_MACHINE = 'animeondemand'
  21     # German-speaking countries of Europe
  22     _GEO_COUNTRIES = ['AT', 'CH', 'DE', 'LI', 'LU']
  23     _TESTS = [{
  24         # jap, OmU
  25         'url': 'https://www.anime-on-demand.de/anime/161',
  26         'info_dict': {
  27             'id': '161',
  28             'title': 'Grimgar, Ashes and Illusions (OmU)',
  29             'description': 'md5:6681ce3c07c7189d255ac6ab23812d31',
  30         },
  31         'playlist_mincount': 4,
  32     }, {
  33         # Film wording is used instead of Episode, ger/jap, Dub/OmU
  34         'url': 'https://www.anime-on-demand.de/anime/39',
  35         'only_matching': True,
  36     }, {
  37         # Episodes without titles, jap, OmU
  38         'url': 'https://www.anime-on-demand.de/anime/162',
  39         'only_matching': True,
  40     }, {
  41         # ger/jap, Dub/OmU, account required
  42         'url': 'https://www.anime-on-demand.de/anime/169',
  43         'only_matching': True,
  44     }, {
  45         # Full length film, non-series, ger/jap, Dub/OmU, account required
  46         'url': 'https://www.anime-on-demand.de/anime/185',
  47         'only_matching': True,
  48     }, {
  49         # Flash videos
  50         'url': 'https://www.anime-on-demand.de/anime/12',
  51         'only_matching': True,
  52     }]
  53
  54     def _login(self):
  55         (username, password) = self._get_login_info()
  56         if username is None:
  57             return
  58
  59         login_page = self._download_webpage(
  60             self._LOGIN_URL, None, 'Downloading login page')
  61
  62         if '>Our licensing terms allow the distribution of animes only to German-speaking countries of Europe' in login_page:
  63             self.raise_geo_restricted(
  64                 '%s is only available in German-speaking countries of Europe' % self.IE_NAME)
  65
  66         login_form = self._form_hidden_inputs('new_user', login_page)
  67
  68         login_form.update({
  69             'user[login]': username,
  70             'user[password]': password,
  71         })
  72
  73         post_url = self._search_regex(
  74             r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
  75             'post url', default=self._LOGIN_URL, group='url')
  76
  77         if not post_url.startswith('http'):
  78             post_url = urljoin(self._LOGIN_URL, post_url)
  79
  80         response = self._download_webpage(
  81             post_url, None, 'Logging in as %s' % username,
  82             data=urlencode_postdata(login_form), headers={
  83                 'Referer': self._LOGIN_URL,
  84             })
  85
  86         if all(p not in response for p in ('>Logout<', 'href="/users/sign_out"')):
  87             error = self._search_regex(
  88                 r'<p class="alert alert-danger">(.+?)</p>',
  89                 response, 'error', default=None)
  90             if error:
  91                 raise ExtractorError('Unable to login: %s' % error, expected=True)
  92             raise ExtractorError('Unable to log in')
  93
  94     def _real_initialize(self):
  95         self._login()
  96
  97     def _real_extract(self, url):
  98         anime_id = self._match_id(url)
  99
 100         webpage = self._download_webpage(url, anime_id)
 101
 102         if 'data-playlist=' not in webpage:
 103             self._download_webpage(
 104                 self._APPLY_HTML5_URL, anime_id,
 105                 'Activating HTML5 beta', 'Unable to apply HTML5 beta')
 106             webpage = self._download_webpage(url, anime_id)
 107
 108         csrf_token = self._html_search_meta(
 109             'csrf-token', webpage, 'csrf token', fatal=True)
 110
 111         anime_title = self._html_search_regex(
 112             r'(?s)<h1[^>]+itemprop="name"[^>]*>(.+?)</h1>',
 113             webpage, 'anime name')
 114         anime_description = self._html_search_regex(
 115             r'(?s)<div[^>]+itemprop="description"[^>]*>(.+?)</div>',
 116             webpage, 'anime description', default=None)
 117
 118         entries = []
 119
 120         def extract_info(html, video_id, num=None):
 121             title, description = [None] * 2
 122             formats = []
 123
 124             for input_ in re.findall(
 125                     r'<input[^>]+class=["\'].*?streamstarter[^>]+>', html):
 126                 attributes = extract_attributes(input_)
 127                 title = attributes.get('data-dialog-header')
 128                 playlist_urls = []
 129                 for playlist_key in ('data-playlist', 'data-otherplaylist', 'data-stream'):
 130                     playlist_url = attributes.get(playlist_key)
 131                     if isinstance(playlist_url, compat_str) and re.match(
 132                             r'/?[\da-zA-Z]+', playlist_url):
 133                         playlist_urls.append(attributes[playlist_key])
 134                 if not playlist_urls:
 135                     continue
 136
 137                 lang = attributes.get('data-lang')
 138                 lang_note = attributes.get('value')
 139
 140                 for playlist_url in playlist_urls:
 141                     kind = self._search_regex(
 142                         r'videomaterialurl/\d+/([^/]+)/',
 143                         playlist_url, 'media kind', default=None)
 144                     format_id_list = []
 145                     if lang:
 146                         format_id_list.append(lang)
 147                     if kind:
 148                         format_id_list.append(kind)
 149                     if not format_id_list and num is not None:
 150                         format_id_list.append(compat_str(num))
 151                     format_id = '-'.join(format_id_list)
 152                     format_note = ', '.join(filter(None, (kind, lang_note)))
 153                     item_id_list = []
 154                     if format_id:
 155                         item_id_list.append(format_id)
 156                     item_id_list.append('videomaterial')
 157                     playlist = self._download_json(
 158                         urljoin(url, playlist_url), video_id,
 159                         'Downloading %s JSON' % ' '.join(item_id_list),
 160                         headers={
 161                             'X-Requested-With': 'XMLHttpRequest',
 162                             'X-CSRF-Token': csrf_token,
 163                             'Referer': url,
 164                             'Accept': 'application/json, text/javascript, */*; q=0.01',
 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': 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)