_ 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     encode_dict,
  13     extract_attributes,
  14     ExtractorError,
  15     sanitized_Request,
  16     urlencode_postdata,
  17 )
  18
  19
  20 class AnimeOnDemandIE(InfoExtractor):
  21     _VALID_URL = r'https?://(?:www\.)?anime-on-demand\.de/anime/(?P<id>\d+)'
  22     _LOGIN_URL = 'https://www.anime-on-demand.de/users/sign_in'
  23     _APPLY_HTML5_URL = 'https://www.anime-on-demand.de/html5apply'
  24     _NETRC_MACHINE = 'animeondemand'
  25     _TESTS = [{
  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
  35         'url': 'https://www.anime-on-demand.de/anime/39',
  36         'only_matching': True,
  37     }, {
  38         # Episodes without titles
  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
  47     def _login(self):
  48         (username, password) = self._get_login_info()
  49         if username is None:
  50             return
  51
  52         login_page = self._download_webpage(
  53             self._LOGIN_URL, None, 'Downloading login page')
  54
  55         if '>Our licensing terms allow the distribution of animes only to German-speaking countries of Europe' in login_page:
  56             self.raise_geo_restricted(
  57                 '%s is only available in German-speaking countries of Europe' % self.IE_NAME)
  58
  59         login_form = self._form_hidden_inputs('new_user', login_page)
  60
  61         login_form.update({
  62             'user[login]': username,
  63             'user[password]': password,
  64         })
  65
  66         post_url = self._search_regex(
  67             r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
  68             'post url', default=self._LOGIN_URL, group='url')
  69
  70         if not post_url.startswith('http'):
  71             post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
  72
  73         request = sanitized_Request(
  74             post_url, urlencode_postdata(encode_dict(login_form)))
  75         request.add_header('Referer', self._LOGIN_URL)
  76
  77         response = self._download_webpage(
  78             request, None, 'Logging in as %s' % username)
  79
  80         if all(p not in response for p in ('>Logout<', 'href="/users/sign_out"')):
  81             error = self._search_regex(
  82                 r'<p class="alert alert-danger">(.+?)</p>',
  83                 response, 'error', default=None)
  84             if error:
  85                 raise ExtractorError('Unable to login: %s' % error, expected=True)
  86             raise ExtractorError('Unable to log in')
  87
  88     def _real_initialize(self):
  89         self._login()
  90
  91     def _real_extract(self, url):
  92         anime_id = self._match_id(url)
  93
  94         webpage = self._download_webpage(url, anime_id)
  95
  96         if 'data-playlist=' not in webpage:
  97             self._download_webpage(
  98                 self._APPLY_HTML5_URL, anime_id,
  99                 'Activating HTML5 beta', 'Unable to apply HTML5 beta')
 100             webpage = self._download_webpage(url, anime_id)
 101
 102         csrf_token = self._html_search_meta(
 103             'csrf-token', webpage, 'csrf token', fatal=True)
 104
 105         anime_title = self._html_search_regex(
 106             r'(?s)<h1[^>]+itemprop="name"[^>]*>(.+?)</h1>',
 107             webpage, 'anime name')
 108         anime_description = self._html_search_regex(
 109             r'(?s)<div[^>]+itemprop="description"[^>]*>(.+?)</div>',
 110             webpage, 'anime description', default=None)
 111
 112         entries = []
 113
 114         for num, episode_html in enumerate(re.findall(
 115                 r'(?s)<h3[^>]+class="episodebox-title".+?>Episodeninhalt<', webpage), 1):
 116             episodebox_title = self._search_regex(
 117                 (r'class="episodebox-title"[^>]+title=(["\'])(?P<title>.+?)\1',
 118                  r'class="episodebox-title"[^>]+>(?P<title>.+?)<'),
 119                 episode_html, 'episodebox title', default=None, group='title')
 120             if not episodebox_title:
 121                 continue
 122
 123             episode_number = int(self._search_regex(
 124                 r'(?:Episode|Film)\s*(\d+)',
 125                 episodebox_title, 'episode number', default=num))
 126             episode_title = self._search_regex(
 127                 r'(?:Episode|Film)\s*\d+\s*-\s*(.+)',
 128                 episodebox_title, 'episode title', default=None)
 129
 130             video_id = 'episode-%d' % episode_number
 131
 132             common_info = {
 133                 'id': video_id,
 134                 'series': anime_title,
 135                 'episode': episode_title,
 136                 'episode_number': episode_number,
 137             }
 138
 139             formats = []
 140
 141             for input_ in re.findall(
 142                     r'<input[^>]+class=["\'].*?streamstarter_html5[^>]+>', episode_html):
 143                 attributes = extract_attributes(input_)
 144                 playlist_urls = []
 145                 for playlist_key in ('data-playlist', 'data-otherplaylist'):
 146                     playlist_url = attributes.get(playlist_key)
 147                     if isinstance(playlist_url, compat_str) and re.match(
 148                             r'/?[\da-zA-Z]+', playlist_url):
 149                         playlist_urls.append(attributes[playlist_key])
 150                 if not playlist_urls:
 151                     continue
 152
 153                 lang = attributes.get('data-lang')
 154                 lang_note = attributes.get('value')
 155
 156                 for playlist_url in playlist_urls:
 157                     kind = self._search_regex(
 158                         r'videomaterialurl/\d+/([^/]+)/',
 159                         playlist_url, 'media kind', default=None)
 160                     format_id_list = []
 161                     if lang:
 162                         format_id_list.append(lang)
 163                     if kind:
 164                         format_id_list.append(kind)
 165                     if not format_id_list:
 166                         format_id_list.append(compat_str(num))
 167                     format_id = '-'.join(format_id_list)
 168                     format_note = ', '.join(filter(None, (kind, lang_note)))
 169                     request = sanitized_Request(
 170                         compat_urlparse.urljoin(url, playlist_url),
 171                         headers={
 172                             'X-Requested-With': 'XMLHttpRequest',
 173                             'X-CSRF-Token': csrf_token,
 174                             'Referer': url,
 175                             'Accept': 'application/json, text/javascript, */*; q=0.01',
 176                         })
 177                     playlist = self._download_json(
 178                         request, video_id, 'Downloading %s playlist JSON' % format_id,
 179                         fatal=False)
 180                     if not playlist:
 181                         continue
 182                     start_video = playlist.get('startvideo', 0)
 183                     playlist = playlist.get('playlist')
 184                     if not playlist or not isinstance(playlist, list):
 185                         continue
 186                     playlist = playlist[start_video]
 187                     title = playlist.get('title')
 188                     if not title:
 189                         continue
 190                     description = playlist.get('description')
 191                     for source in playlist.get('sources', []):
 192                         file_ = source.get('file')
 193                         if not file_:
 194                             continue
 195                         ext = determine_ext(file_)
 196                         format_id_list = [lang, kind]
 197                         if ext == 'm3u8':
 198                             format_id_list.append('hls')
 199                         elif source.get('type') == 'video/dash' or ext == 'mpd':
 200                             format_id_list.append('dash')
 201                         format_id = '-'.join(filter(None, format_id_list))
 202                         if ext == 'm3u8':
 203                             file_formats = self._extract_m3u8_formats(
 204                                 file_, video_id, 'mp4',
 205                                 entry_protocol='m3u8_native', m3u8_id=format_id, fatal=False)
 206                         elif source.get('type') == 'video/dash' or ext == 'mpd':
 207                             file_formats = self._extract_mpd_formats(
 208                                 file_, video_id, mpd_id=format_id, fatal=False)
 209                         else:
 210                             continue
 211                         for f in file_formats:
 212                             f.update({
 213                                 'language': lang,
 214                                 'format_note': format_note,
 215                             })
 216                         formats.extend(file_formats)
 217
 218             if formats:
 219                 self._sort_formats(formats)
 220                 f = common_info.copy()
 221                 f.update({
 222                     'title': title,
 223                     'description': description,
 224                     'formats': formats,
 225                 })
 226                 entries.append(f)
 227
 228             # Extract teaser only when full episode is not available
 229             if not formats:
 230                 m = re.search(
 231                     r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>Teaser<',
 232                     episode_html)
 233                 if m:
 234                     f = common_info.copy()
 235                     f.update({
 236                         'id': '%s-teaser' % f['id'],
 237                         'title': m.group('title'),
 238                         'url': compat_urlparse.urljoin(url, m.group('href')),
 239                     })
 240                     entries.append(f)
 241
 242         return self.playlist_result(entries, anime_id, anime_title, anime_description)