yt_dlp/extractor/adn.py

   1 import base64
   2 import binascii
   3 import json
   4 import os
   5 import random
   6
   7 from .common import InfoExtractor
   8 from ..aes import aes_cbc_decrypt_bytes, unpad_pkcs7
   9 from ..compat import (
  10     compat_HTTPError,
  11     compat_b64decode,
  12 )
  13 from ..utils import (
  14     ass_subtitles_timecode,
  15     bytes_to_intlist,
  16     bytes_to_long,
  17     ExtractorError,
  18     float_or_none,
  19     int_or_none,
  20     intlist_to_bytes,
  21     long_to_bytes,
  22     pkcs1pad,
  23     strip_or_none,
  24     try_get,
  25     unified_strdate,
  26     urlencode_postdata,
  27 )
  28
  29
  30 class ADNIE(InfoExtractor):
  31     IE_DESC = 'Animation Digital Network'
  32     _VALID_URL = r'https?://(?:www\.)?(?:animation|anime)digitalnetwork\.fr/video/[^/]+/(?P<id>\d+)'
  33     _TESTS = [{
  34         'url': 'https://animationdigitalnetwork.fr/video/fruits-basket/9841-episode-1-a-ce-soir',
  35         'md5': '1c9ef066ceb302c86f80c2b371615261',
  36         'info_dict': {
  37             'id': '9841',
  38             'ext': 'mp4',
  39             'title': 'Fruits Basket - Episode 1',
  40             'description': 'md5:14be2f72c3c96809b0ca424b0097d336',
  41             'series': 'Fruits Basket',
  42             'duration': 1437,
  43             'release_date': '20190405',
  44             'comment_count': int,
  45             'average_rating': float,
  46             'season_number': 1,
  47             'episode': 'À ce soir !',
  48             'episode_number': 1,
  49         },
  50         'skip': 'Only available in region (FR, ...)',
  51     }, {
  52         'url': 'http://animedigitalnetwork.fr/video/blue-exorcist-kyoto-saga/7778-episode-1-debut-des-hostilites',
  53         'only_matching': True,
  54     }]
  55
  56     _NETRC_MACHINE = 'animationdigitalnetwork'
  57     _BASE = 'animationdigitalnetwork.fr'
  58     _API_BASE_URL = 'https://gw.api.' + _BASE + '/'
  59     _PLAYER_BASE_URL = _API_BASE_URL + 'player/'
  60     _HEADERS = {}
  61     _LOGIN_ERR_MESSAGE = 'Unable to log in'
  62     _RSA_KEY = (0x9B42B08905199A5CCE2026274399CA560ECB209EE9878A708B1C0812E1BB8CB5D1FB7441861147C1A1F2F3A0476DD63A9CAC20D3E983613346850AA6CB38F16DC7D720FD7D86FC6E5B3D5BBC72E14CD0BF9E869F2CEA2CCAD648F1DCE38F1FF916CEFB2D339B64AA0264372344BC775E265E8A852F88144AB0BD9AA06C1A4ABB, 65537)
  63     _POS_ALIGN_MAP = {
  64         'start': 1,
  65         'end': 3,
  66     }
  67     _LINE_ALIGN_MAP = {
  68         'middle': 8,
  69         'end': 4,
  70     }
  71
  72     def _get_subtitles(self, sub_url, video_id):
  73         if not sub_url:
  74             return None
  75
  76         enc_subtitles = self._download_webpage(
  77             sub_url, video_id, 'Downloading subtitles location', fatal=False) or '{}'
  78         subtitle_location = (self._parse_json(enc_subtitles, video_id, fatal=False) or {}).get('location')
  79         if subtitle_location:
  80             enc_subtitles = self._download_webpage(
  81                 subtitle_location, video_id, 'Downloading subtitles data',
  82                 fatal=False, headers={'Origin': 'https://' + self._BASE})
  83         if not enc_subtitles:
  84             return None
  85
  86         # http://animationdigitalnetwork.fr/components/com_vodvideo/videojs/adn-vjs.min.js
  87         dec_subtitles = unpad_pkcs7(aes_cbc_decrypt_bytes(
  88             compat_b64decode(enc_subtitles[24:]),
  89             binascii.unhexlify(self._K + '7fac1178830cfe0c'),
  90             compat_b64decode(enc_subtitles[:24])))
  91         subtitles_json = self._parse_json(dec_subtitles.decode(), None, fatal=False)
  92         if not subtitles_json:
  93             return None
  94
  95         subtitles = {}
  96         for sub_lang, sub in subtitles_json.items():
  97             ssa = '''[Script Info]
  98 ScriptType:V4.00
  99 [V4 Styles]
 100 Format: Name,Fontname,Fontsize,PrimaryColour,SecondaryColour,TertiaryColour,BackColour,Bold,Italic,BorderStyle,Outline,Shadow,Alignment,MarginL,MarginR,MarginV,AlphaLevel,Encoding
 101 Style: Default,Arial,18,16777215,16777215,16777215,0,-1,0,1,1,0,2,20,20,20,0,0
 102 [Events]
 103 Format: Marked,Start,End,Style,Name,MarginL,MarginR,MarginV,Effect,Text'''
 104             for current in sub:
 105                 start, end, text, line_align, position_align = (
 106                     float_or_none(current.get('startTime')),
 107                     float_or_none(current.get('endTime')),
 108                     current.get('text'), current.get('lineAlign'),
 109                     current.get('positionAlign'))
 110                 if start is None or end is None or text is None:
 111                     continue
 112                 alignment = self._POS_ALIGN_MAP.get(position_align, 2) + self._LINE_ALIGN_MAP.get(line_align, 0)
 113                 ssa += os.linesep + 'Dialogue: Marked=0,%s,%s,Default,,0,0,0,,%s%s' % (
 114                     ass_subtitles_timecode(start),
 115                     ass_subtitles_timecode(end),
 116                     '{\\a%d}' % alignment if alignment != 2 else '',
 117                     text.replace('\n', '\\N').replace('<i>', '{\\i1}').replace('</i>', '{\\i0}'))
 118
 119             if sub_lang == 'vostf':
 120                 sub_lang = 'fr'
 121             subtitles.setdefault(sub_lang, []).extend([{
 122                 'ext': 'json',
 123                 'data': json.dumps(sub),
 124             }, {
 125                 'ext': 'ssa',
 126                 'data': ssa,
 127             }])
 128         return subtitles
 129
 130     def _perform_login(self, username, password):
 131         try:
 132             access_token = (self._download_json(
 133                 self._API_BASE_URL + 'authentication/login', None,
 134                 'Logging in', self._LOGIN_ERR_MESSAGE, fatal=False,
 135                 data=urlencode_postdata({
 136                     'password': password,
 137                     'rememberMe': False,
 138                     'source': 'Web',
 139                     'username': username,
 140                 })) or {}).get('accessToken')
 141             if access_token:
 142                 self._HEADERS = {'authorization': 'Bearer ' + access_token}
 143         except ExtractorError as e:
 144             message = None
 145             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
 146                 resp = self._parse_json(
 147                     e.cause.read().decode(), None, fatal=False) or {}
 148                 message = resp.get('message') or resp.get('code')
 149             self.report_warning(message or self._LOGIN_ERR_MESSAGE)
 150
 151     def _real_extract(self, url):
 152         video_id = self._match_id(url)
 153         video_base_url = self._PLAYER_BASE_URL + 'video/%s/' % video_id
 154         player = self._download_json(
 155             video_base_url + 'configuration', video_id,
 156             'Downloading player config JSON metadata',
 157             headers=self._HEADERS)['player']
 158         options = player['options']
 159
 160         user = options['user']
 161         if not user.get('hasAccess'):
 162             self.raise_login_required()
 163
 164         token = self._download_json(
 165             user.get('refreshTokenUrl') or (self._PLAYER_BASE_URL + 'refresh/token'),
 166             video_id, 'Downloading access token', headers={
 167                 'x-player-refresh-token': user['refreshToken']
 168             }, data=b'')['token']
 169
 170         links_url = try_get(options, lambda x: x['video']['url']) or (video_base_url + 'link')
 171         self._K = ''.join([random.choice('0123456789abcdef') for _ in range(16)])
 172         message = bytes_to_intlist(json.dumps({
 173             'k': self._K,
 174             't': token,
 175         }))
 176
 177         # Sometimes authentication fails for no good reason, retry with
 178         # a different random padding
 179         links_data = None
 180         for _ in range(3):
 181             padded_message = intlist_to_bytes(pkcs1pad(message, 128))
 182             n, e = self._RSA_KEY
 183             encrypted_message = long_to_bytes(pow(bytes_to_long(padded_message), e, n))
 184             authorization = base64.b64encode(encrypted_message).decode()
 185
 186             try:
 187                 links_data = self._download_json(
 188                     links_url, video_id, 'Downloading links JSON metadata', headers={
 189                         'X-Player-Token': authorization
 190                     }, query={
 191                         'freeWithAds': 'true',
 192                         'adaptive': 'false',
 193                         'withMetadata': 'true',
 194                         'source': 'Web'
 195                     })
 196                 break
 197             except ExtractorError as e:
 198                 if not isinstance(e.cause, compat_HTTPError):
 199                     raise e
 200
 201                 if e.cause.code == 401:
 202                     # This usually goes away with a different random pkcs1pad, so retry
 203                     continue
 204
 205                 error = self._parse_json(e.cause.read(), video_id)
 206                 message = error.get('message')
 207                 if e.cause.code == 403 and error.get('code') == 'player-bad-geolocation-country':
 208                     self.raise_geo_restricted(msg=message)
 209                 raise ExtractorError(message)
 210         else:
 211             raise ExtractorError('Giving up retrying')
 212
 213         links = links_data.get('links') or {}
 214         metas = links_data.get('metadata') or {}
 215         sub_url = (links.get('subtitles') or {}).get('all')
 216         video_info = links_data.get('video') or {}
 217         title = metas['title']
 218
 219         formats = []
 220         for format_id, qualities in (links.get('streaming') or {}).items():
 221             if not isinstance(qualities, dict):
 222                 continue
 223             for quality, load_balancer_url in qualities.items():
 224                 load_balancer_data = self._download_json(
 225                     load_balancer_url, video_id,
 226                     'Downloading %s %s JSON metadata' % (format_id, quality),
 227                     fatal=False) or {}
 228                 m3u8_url = load_balancer_data.get('location')
 229                 if not m3u8_url:
 230                     continue
 231                 m3u8_formats = self._extract_m3u8_formats(
 232                     m3u8_url, video_id, 'mp4', 'm3u8_native',
 233                     m3u8_id=format_id, fatal=False)
 234                 if format_id == 'vf':
 235                     for f in m3u8_formats:
 236                         f['language'] = 'fr'
 237                 formats.extend(m3u8_formats)
 238
 239         video = (self._download_json(
 240             self._API_BASE_URL + 'video/%s' % video_id, video_id,
 241             'Downloading additional video metadata', fatal=False) or {}).get('video') or {}
 242         show = video.get('show') or {}
 243
 244         return {
 245             'id': video_id,
 246             'title': title,
 247             'description': strip_or_none(metas.get('summary') or video.get('summary')),
 248             'thumbnail': video_info.get('image') or player.get('image'),
 249             'formats': formats,
 250             'subtitles': self.extract_subtitles(sub_url, video_id),
 251             'episode': metas.get('subtitle') or video.get('name'),
 252             'episode_number': int_or_none(video.get('shortNumber')),
 253             'series': show.get('title'),
 254             'season_number': int_or_none(video.get('season')),
 255             'duration': int_or_none(video_info.get('duration') or video.get('duration')),
 256             'release_date': unified_strdate(video.get('releaseDate')),
 257             'average_rating': float_or_none(video.get('rating') or metas.get('rating')),
 258             'comment_count': int_or_none(video.get('commentsCount')),
 259         }