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
  51     def _login(self):
  52         (username, password) = self._get_login_info()
  53         if username is None:
  54             return
  55
  56         login_page = self._download_webpage(
  57             self._LOGIN_URL, None, 'Downloading login page')
  58
  59         if '>Our licensing terms allow the distribution of animes only to German-speaking countries of Europe' in login_page:
  60             self.raise_geo_restricted(
  61                 '%s is only available in German-speaking countries of Europe' % self.IE_NAME)
  62
  63         login_form = self._form_hidden_inputs('new_user', login_page)
  64
  65         login_form.update({
  66             'user[login]': username,
  67             'user[password]': password,
  68         })
  69
  70         post_url = self._search_regex(
  71             r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
  72             'post url', default=self._LOGIN_URL, group='url')
  73
  74         if not post_url.startswith('http'):
  75             post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
  76
  77         request = sanitized_Request(
  78             post_url, urlencode_postdata(login_form))
  79         request.add_header('Referer', self._LOGIN_URL)
  80
  81         response = self._download_webpage(
  82             request, None, 'Logging in as %s' % username)
  83
  84         if all(p not in response for p in ('>Logout<', 'href="/users/sign_out"')):
  85             error = self._search_regex(
  86                 r'<p class="alert alert-danger">(.+?)</p>',
  87                 response, 'error', default=None)
  88             if error:
  89                 raise ExtractorError('Unable to login: %s' % error, expected=True)
  90             raise ExtractorError('Unable to log in')
  91
  92     def _real_initialize(self):
  93         self._login()
  94
  95     def _real_extract(self, url):
  96         anime_id = self._match_id(url)
  97
  98         webpage = self._download_webpage(url, anime_id)
  99
 100         if 'data-playlist=' not in webpage:
 101             self._download_webpage(
 102                 self._APPLY_HTML5_URL, anime_id,
 103                 'Activating HTML5 beta', 'Unable to apply HTML5 beta')
 104             webpage = self._download_webpage(url, anime_id)
 105
 106         csrf_token = self._html_search_meta(
 107             'csrf-token', webpage, 'csrf token', fatal=True)
 108
 109         anime_title = self._html_search_regex(
 110             r'(?s)<h1[^>]+itemprop="name"[^>]*>(.+?)</h1>',
 111             webpage, 'anime name')
 112         anime_description = self._html_search_regex(
 113             r'(?s)<div[^>]+itemprop="description"[^>]*>(.+?)</div>',
 114             webpage, 'anime description', default=None)
 115
 116         entries = []
 117
 118         def extract_info(html, video_id, num=None):
 119             title, description = [None] * 2
 120             formats = []
 121
 122             for input_ in re.findall(
 123                     r'<input[^>]+class=["\'].*?streamstarter_html5[^>]+>', html):
 124                 attributes = extract_attributes(input_)
 125                 playlist_urls = []
 126                 for playlist_key in ('data-playlist', 'data-otherplaylist'):
 127                     playlist_url = attributes.get(playlist_key)
 128                     if isinstance(playlist_url, compat_str) and re.match(
 129                             r'/?[\da-zA-Z]+', playlist_url):
 130                         playlist_urls.append(attributes[playlist_key])
 131                 if not playlist_urls:
 132                     continue
 133
 134                 lang = attributes.get('data-lang')
 135                 lang_note = attributes.get('value')
 136
 137                 for playlist_url in playlist_urls:
 138                     kind = self._search_regex(
 139                         r'videomaterialurl/\d+/([^/]+)/',
 140                         playlist_url, 'media kind', default=None)
 141                     format_id_list = []
 142                     if lang:
 143                         format_id_list.append(lang)
 144                     if kind:
 145                         format_id_list.append(kind)
 146                     if not format_id_list and num is not None:
 147                         format_id_list.append(compat_str(num))
 148                     format_id = '-'.join(format_id_list)
 149                     format_note = ', '.join(filter(None, (kind, lang_note)))
 150                     request = sanitized_Request(
 151                         compat_urlparse.urljoin(url, playlist_url),
 152                         headers={
 153                             'X-Requested-With': 'XMLHttpRequest',
 154                             'X-CSRF-Token': csrf_token,
 155                             'Referer': url,
 156                             'Accept': 'application/json, text/javascript, */*; q=0.01',
 157                         })
 158                     playlist = self._download_json(
 159                         request, video_id, 'Downloading %s playlist JSON' % format_id,
 160                         fatal=False)
 161                     if not playlist:
 162                         continue
 163                     start_video = playlist.get('startvideo', 0)
 164                     playlist = playlist.get('playlist')
 165                     if not playlist or not isinstance(playlist, list):
 166                         continue
 167                     playlist = playlist[start_video]
 168                     title = playlist.get('title')
 169                     if not title:
 170                         continue
 171                     description = playlist.get('description')
 172                     for source in playlist.get('sources', []):
 173                         file_ = source.get('file')
 174                         if not file_:
 175                             continue
 176                         ext = determine_ext(file_)
 177                         format_id_list = [lang, kind]
 178                         if ext == 'm3u8':
 179                             format_id_list.append('hls')
 180                         elif source.get('type') == 'video/dash' or ext == 'mpd':
 181                             format_id_list.append('dash')
 182                         format_id = '-'.join(filter(None, format_id_list))
 183                         if ext == 'm3u8':
 184                             file_formats = self._extract_m3u8_formats(
 185                                 file_, video_id, 'mp4',
 186                                 entry_protocol='m3u8_native', m3u8_id=format_id, fatal=False)
 187                         elif source.get('type') == 'video/dash' or ext == 'mpd':
 188                             continue
 189                             file_formats = self._extract_mpd_formats(
 190                                 file_, video_id, mpd_id=format_id, fatal=False)
 191                         else:
 192                             continue
 193                         for f in file_formats:
 194                             f.update({
 195                                 'language': lang,
 196                                 'format_note': format_note,
 197                             })
 198                         formats.extend(file_formats)
 199
 200             return {
 201                 'title': title,
 202                 'description': description,
 203                 'formats': formats,
 204             }
 205
 206         def extract_entries(html, video_id, common_info, num=None):
 207             info = extract_info(html, video_id, num)
 208
 209             if info['formats']:
 210                 self._sort_formats(info['formats'])
 211                 f = common_info.copy()
 212                 f.update(info)
 213                 entries.append(f)
 214
 215             # Extract teaser/trailer only when full episode is not available
 216             if not info['formats']:
 217                 m = re.search(
 218                     r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>(?P<kind>Teaser|Trailer)<',
 219                     html)
 220                 if m:
 221                     f = common_info.copy()
 222                     f.update({
 223                         'id': '%s-%s' % (f['id'], m.group('kind').lower()),
 224                         'title': m.group('title'),
 225                         'url': compat_urlparse.urljoin(url, m.group('href')),
 226                     })
 227                     entries.append(f)
 228
 229         def extract_episodes(html):
 230             for num, episode_html in enumerate(re.findall(
 231                     r'(?s)<h3[^>]+class="episodebox-title".+?>Episodeninhalt<', html), 1):
 232                 episodebox_title = self._search_regex(
 233                     (r'class="episodebox-title"[^>]+title=(["\'])(?P<title>.+?)\1',
 234                      r'class="episodebox-title"[^>]+>(?P<title>.+?)<'),
 235                     episode_html, 'episodebox title', default=None, group='title')
 236                 if not episodebox_title:
 237                     continue
 238
 239                 episode_number = int(self._search_regex(
 240                     r'(?:Episode|Film)\s*(\d+)',
 241                     episodebox_title, 'episode number', default=num))
 242                 episode_title = self._search_regex(
 243                     r'(?:Episode|Film)\s*\d+\s*-\s*(.+)',
 244                     episodebox_title, 'episode title', default=None)
 245
 246                 video_id = 'episode-%d' % episode_number
 247
 248                 common_info = {
 249                     'id': video_id,
 250                     'series': anime_title,
 251                     'episode': episode_title,
 252                     'episode_number': episode_number,
 253                 }
 254
 255                 extract_entries(episode_html, video_id, common_info)
 256
 257         def extract_film(html, video_id):
 258             common_info = {
 259                 'id': anime_id,
 260                 'title': anime_title,
 261                 'description': anime_description,
 262             }
 263             extract_entries(html, video_id, common_info)
 264
 265         extract_episodes(webpage)
 266
 267         if not entries:
 268             extract_film(webpage, anime_id)
 269
 270         return self.playlist_result(entries, anime_id, anime_title, anime_description)