yt_dlp/extractor/animelab.py

   1 from .common import InfoExtractor
   2
   3 from ..utils import (
   4     ExtractorError,
   5     urlencode_postdata,
   6     int_or_none,
   7     str_or_none,
   8     determine_ext,
   9 )
  10
  11 from ..compat import compat_HTTPError
  12
  13
  14 class AnimeLabBaseIE(InfoExtractor):
  15     _LOGIN_URL = 'https://www.animelab.com/login'
  16     _NETRC_MACHINE = 'animelab'
  17     _LOGGED_IN = False
  18
  19     def _is_logged_in(self, login_page=None):
  20         if not self._LOGGED_IN:
  21             if not login_page:
  22                 login_page = self._download_webpage(self._LOGIN_URL, None, 'Downloading login page')
  23             AnimeLabBaseIE._LOGGED_IN = 'Sign In' not in login_page
  24         return self._LOGGED_IN
  25
  26     def _perform_login(self, username, password):
  27         if self._is_logged_in():
  28             return
  29
  30         login_form = {
  31             'email': username,
  32             'password': password,
  33         }
  34
  35         try:
  36             response = self._download_webpage(
  37                 self._LOGIN_URL, None, 'Logging in', 'Wrong login info',
  38                 data=urlencode_postdata(login_form),
  39                 headers={'Content-Type': 'application/x-www-form-urlencoded'})
  40         except ExtractorError as e:
  41             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
  42                 raise ExtractorError('Unable to log in (wrong credentials?)', expected=True)
  43             raise
  44
  45         if not self._is_logged_in(response):
  46             raise ExtractorError('Unable to login (cannot verify if logged in)')
  47
  48     def _real_initialize(self):
  49         if not self._is_logged_in():
  50             self.raise_login_required('Login is required to access any AnimeLab content')
  51
  52
  53 class AnimeLabIE(AnimeLabBaseIE):
  54     _VALID_URL = r'https?://(?:www\.)?animelab\.com/player/(?P<id>[^/]+)'
  55
  56     # the following tests require authentication, but a free account will suffice
  57     # just set 'usenetrc' to true in test/local_parameters.json if you use a .netrc file
  58     # or you can set 'username' and 'password' there
  59     # the tests also select a specific format so that the same video is downloaded
  60     # regardless of whether the user is premium or not (needs testing on a premium account)
  61     _TEST = {
  62         'url': 'https://www.animelab.com/player/fullmetal-alchemist-brotherhood-episode-42',
  63         'md5': '05bde4b91a5d1ff46ef5b94df05b0f7f',
  64         'info_dict': {
  65             'id': '383',
  66             'ext': 'mp4',
  67             'display_id': 'fullmetal-alchemist-brotherhood-episode-42',
  68             'title': 'Fullmetal Alchemist: Brotherhood - Episode 42 - Signs of a Counteroffensive',
  69             'description': 'md5:103eb61dd0a56d3dfc5dbf748e5e83f4',
  70             'series': 'Fullmetal Alchemist: Brotherhood',
  71             'episode': 'Signs of a Counteroffensive',
  72             'episode_number': 42,
  73             'duration': 1469,
  74             'season': 'Season 1',
  75             'season_number': 1,
  76             'season_id': '38',
  77         },
  78         'params': {
  79             'format': '[format_id=21711_yeshardsubbed_ja-JP][height=480]',
  80         },
  81         'skip': 'All AnimeLab content requires authentication',
  82     }
  83
  84     def _real_extract(self, url):
  85         display_id = self._match_id(url)
  86
  87         # unfortunately we can get different URLs for the same formats
  88         # e.g. if we are using a "free" account so no dubs available
  89         # (so _remove_duplicate_formats is not effective)
  90         # so we use a dictionary as a workaround
  91         formats = {}
  92         for language_option_url in ('https://www.animelab.com/player/%s/subtitles',
  93                                     'https://www.animelab.com/player/%s/dubbed'):
  94             actual_url = language_option_url % display_id
  95             webpage = self._download_webpage(actual_url, display_id, 'Downloading URL ' + actual_url)
  96
  97             video_collection = self._parse_json(self._search_regex(r'new\s+?AnimeLabApp\.VideoCollection\s*?\((.*?)\);', webpage, 'AnimeLab VideoCollection'), display_id)
  98             position = int_or_none(self._search_regex(r'playlistPosition\s*?=\s*?(\d+)', webpage, 'Playlist Position'))
  99
 100             raw_data = video_collection[position]['videoEntry']
 101
 102             video_id = str_or_none(raw_data['id'])
 103
 104             # create a title from many sources (while grabbing other info)
 105             # TODO use more fallback sources to get some of these
 106             series = raw_data.get('showTitle')
 107             video_type = raw_data.get('videoEntryType', {}).get('name')
 108             episode_number = raw_data.get('episodeNumber')
 109             episode_name = raw_data.get('name')
 110
 111             title_parts = (series, video_type, episode_number, episode_name)
 112             if None not in title_parts:
 113                 title = '%s - %s %s - %s' % title_parts
 114             else:
 115                 title = episode_name
 116
 117             description = raw_data.get('synopsis') or self._og_search_description(webpage, default=None)
 118
 119             duration = int_or_none(raw_data.get('duration'))
 120
 121             thumbnail_data = raw_data.get('images', [])
 122             thumbnails = []
 123             for thumbnail in thumbnail_data:
 124                 for instance in thumbnail['imageInstances']:
 125                     image_data = instance.get('imageInfo', {})
 126                     thumbnails.append({
 127                         'id': str_or_none(image_data.get('id')),
 128                         'url': image_data.get('fullPath'),
 129                         'width': image_data.get('width'),
 130                         'height': image_data.get('height'),
 131                     })
 132
 133             season_data = raw_data.get('season', {}) or {}
 134             season = str_or_none(season_data.get('name'))
 135             season_number = int_or_none(season_data.get('seasonNumber'))
 136             season_id = str_or_none(season_data.get('id'))
 137
 138             for video_data in raw_data['videoList']:
 139                 current_video_list = {}
 140                 current_video_list['language'] = video_data.get('language', {}).get('languageCode')
 141
 142                 is_hardsubbed = video_data.get('hardSubbed')
 143
 144                 for video_instance in video_data['videoInstances']:
 145                     httpurl = video_instance.get('httpUrl')
 146                     url = httpurl if httpurl else video_instance.get('rtmpUrl')
 147                     if url is None:
 148                         # this video format is unavailable to the user (not premium etc.)
 149                         continue
 150
 151                     current_format = current_video_list.copy()
 152
 153                     format_id_parts = []
 154
 155                     format_id_parts.append(str_or_none(video_instance.get('id')))
 156
 157                     if is_hardsubbed is not None:
 158                         if is_hardsubbed:
 159                             format_id_parts.append('yeshardsubbed')
 160                         else:
 161                             format_id_parts.append('nothardsubbed')
 162
 163                     format_id_parts.append(current_format['language'])
 164
 165                     format_id = '_'.join([x for x in format_id_parts if x is not None])
 166
 167                     ext = determine_ext(url)
 168                     if ext == 'm3u8':
 169                         for format_ in self._extract_m3u8_formats(
 170                                 url, video_id, m3u8_id=format_id, fatal=False):
 171                             formats[format_['format_id']] = format_
 172                         continue
 173                     elif ext == 'mpd':
 174                         for format_ in self._extract_mpd_formats(
 175                                 url, video_id, mpd_id=format_id, fatal=False):
 176                             formats[format_['format_id']] = format_
 177                         continue
 178
 179                     current_format['url'] = url
 180                     quality_data = video_instance.get('videoQuality')
 181                     if quality_data:
 182                         quality = quality_data.get('name') or quality_data.get('description')
 183                     else:
 184                         quality = None
 185
 186                     height = None
 187                     if quality:
 188                         height = int_or_none(self._search_regex(r'(\d+)p?$', quality, 'Video format height', default=None))
 189
 190                     if height is None:
 191                         self.report_warning('Could not get height of video')
 192                     else:
 193                         current_format['height'] = height
 194                     current_format['format_id'] = format_id
 195
 196                     formats[current_format['format_id']] = current_format
 197
 198         formats = list(formats.values())
 199         self._sort_formats(formats)
 200
 201         return {
 202             'id': video_id,
 203             'display_id': display_id,
 204             'title': title,
 205             'description': description,
 206             'series': series,
 207             'episode': episode_name,
 208             'episode_number': int_or_none(episode_number),
 209             'thumbnails': thumbnails,
 210             'duration': duration,
 211             'formats': formats,
 212             'season': season,
 213             'season_number': season_number,
 214             'season_id': season_id,
 215         }
 216
 217
 218 class AnimeLabShowsIE(AnimeLabBaseIE):
 219     _VALID_URL = r'https?://(?:www\.)?animelab\.com/shows/(?P<id>[^/]+)'
 220
 221     _TEST = {
 222         'url': 'https://www.animelab.com/shows/attack-on-titan',
 223         'info_dict': {
 224             'id': '45',
 225             'title': 'Attack on Titan',
 226             'description': 'md5:989d95a2677e9309368d5cf39ba91469',
 227         },
 228         'playlist_count': 59,
 229         'skip': 'All AnimeLab content requires authentication',
 230     }
 231
 232     def _real_extract(self, url):
 233         _BASE_URL = 'http://www.animelab.com'
 234         _SHOWS_API_URL = '/api/videoentries/show/videos/'
 235         display_id = self._match_id(url)
 236
 237         webpage = self._download_webpage(url, display_id, 'Downloading requested URL')
 238
 239         show_data_str = self._search_regex(r'({"id":.*}),\svideoEntry', webpage, 'AnimeLab show data')
 240         show_data = self._parse_json(show_data_str, display_id)
 241
 242         show_id = str_or_none(show_data.get('id'))
 243         title = show_data.get('name')
 244         description = show_data.get('shortSynopsis') or show_data.get('longSynopsis')
 245
 246         entries = []
 247         for season in show_data['seasons']:
 248             season_id = season['id']
 249             get_data = urlencode_postdata({
 250                 'seasonId': season_id,
 251                 'limit': 1000,
 252             })
 253             # despite using urlencode_postdata, we are sending a GET request
 254             target_url = _BASE_URL + _SHOWS_API_URL + show_id + "?" + get_data.decode('utf-8')
 255             response = self._download_webpage(
 256                 target_url,
 257                 None, 'Season id %s' % season_id)
 258
 259             season_data = self._parse_json(response, display_id)
 260
 261             for video_data in season_data['list']:
 262                 entries.append(self.url_result(
 263                     _BASE_URL + '/player/' + video_data['slug'], 'AnimeLab',
 264                     str_or_none(video_data.get('id')), video_data.get('name')
 265                 ))
 266
 267         return {
 268             '_type': 'playlist',
 269             'id': show_id,
 270             'title': title,
 271             'description': description,
 272             'entries': entries,
 273         }
 274
 275 # TODO implement myqueue