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 = 'Anime Digital Network'
  32     _VALID_URL = r'https?://(?:www\.)?animedigitalnetwork\.fr/video/[^/]+/(?P<id>\d+)'
  33     _TEST = {
  34         'url': 'http://animedigitalnetwork.fr/video/blue-exorcist-kyoto-saga/7778-episode-1-debut-des-hostilites',
  35         'md5': '0319c99885ff5547565cacb4f3f9348d',
  36         'info_dict': {
  37             'id': '7778',
  38             'ext': 'mp4',
  39             'title': 'Blue Exorcist - Kyôto Saga - Episode 1',
  40             'description': 'md5:2f7b5aa76edbc1a7a92cedcda8a528d5',
  41             'series': 'Blue Exorcist - Kyôto Saga',
  42             'duration': 1467,
  43             'release_date': '20170106',
  44             'comment_count': int,
  45             'average_rating': float,
  46             'season_number': 2,
  47             'episode': 'Début des hostilités',
  48             'episode_number': 1,
  49         }
  50     }
  51
  52     _NETRC_MACHINE = 'animedigitalnetwork'
  53     _BASE_URL = 'http://animedigitalnetwork.fr'
  54     _API_BASE_URL = 'https://gw.api.animedigitalnetwork.fr/'
  55     _PLAYER_BASE_URL = _API_BASE_URL + 'player/'
  56     _HEADERS = {}
  57     _LOGIN_ERR_MESSAGE = 'Unable to log in'
  58     _RSA_KEY = (0x9B42B08905199A5CCE2026274399CA560ECB209EE9878A708B1C0812E1BB8CB5D1FB7441861147C1A1F2F3A0476DD63A9CAC20D3E983613346850AA6CB38F16DC7D720FD7D86FC6E5B3D5BBC72E14CD0BF9E869F2CEA2CCAD648F1DCE38F1FF916CEFB2D339B64AA0264372344BC775E265E8A852F88144AB0BD9AA06C1A4ABB, 65537)
  59     _POS_ALIGN_MAP = {
  60         'start': 1,
  61         'end': 3,
  62     }
  63     _LINE_ALIGN_MAP = {
  64         'middle': 8,
  65         'end': 4,
  66     }
  67
  68     def _get_subtitles(self, sub_url, video_id):
  69         if not sub_url:
  70             return None
  71
  72         enc_subtitles = self._download_webpage(
  73             sub_url, video_id, 'Downloading subtitles location', fatal=False) or '{}'
  74         subtitle_location = (self._parse_json(enc_subtitles, video_id, fatal=False) or {}).get('location')
  75         if subtitle_location:
  76             enc_subtitles = self._download_webpage(
  77                 subtitle_location, video_id, 'Downloading subtitles data',
  78                 fatal=False, headers={'Origin': 'https://animedigitalnetwork.fr'})
  79         if not enc_subtitles:
  80             return None
  81
  82         # http://animedigitalnetwork.fr/components/com_vodvideo/videojs/adn-vjs.min.js
  83         dec_subtitles = unpad_pkcs7(aes_cbc_decrypt_bytes(
  84             compat_b64decode(enc_subtitles[24:]),
  85             binascii.unhexlify(self._K + '7fac1178830cfe0c'),
  86             compat_b64decode(enc_subtitles[:24])))
  87         subtitles_json = self._parse_json(dec_subtitles.decode(), None, fatal=False)
  88         if not subtitles_json:
  89             return None
  90
  91         subtitles = {}
  92         for sub_lang, sub in subtitles_json.items():
  93             ssa = '''[Script Info]
  94 ScriptType:V4.00
  95 [V4 Styles]
  96 Format: Name,Fontname,Fontsize,PrimaryColour,SecondaryColour,TertiaryColour,BackColour,Bold,Italic,BorderStyle,Outline,Shadow,Alignment,MarginL,MarginR,MarginV,AlphaLevel,Encoding
  97 Style: Default,Arial,18,16777215,16777215,16777215,0,-1,0,1,1,0,2,20,20,20,0,0
  98 [Events]
  99 Format: Marked,Start,End,Style,Name,MarginL,MarginR,MarginV,Effect,Text'''
 100             for current in sub:
 101                 start, end, text, line_align, position_align = (
 102                     float_or_none(current.get('startTime')),
 103                     float_or_none(current.get('endTime')),
 104                     current.get('text'), current.get('lineAlign'),
 105                     current.get('positionAlign'))
 106                 if start is None or end is None or text is None:
 107                     continue
 108                 alignment = self._POS_ALIGN_MAP.get(position_align, 2) + self._LINE_ALIGN_MAP.get(line_align, 0)
 109                 ssa += os.linesep + 'Dialogue: Marked=0,%s,%s,Default,,0,0,0,,%s%s' % (
 110                     ass_subtitles_timecode(start),
 111                     ass_subtitles_timecode(end),
 112                     '{\\a%d}' % alignment if alignment != 2 else '',
 113                     text.replace('\n', '\\N').replace('<i>', '{\\i1}').replace('</i>', '{\\i0}'))
 114
 115             if sub_lang == 'vostf':
 116                 sub_lang = 'fr'
 117             subtitles.setdefault(sub_lang, []).extend([{
 118                 'ext': 'json',
 119                 'data': json.dumps(sub),
 120             }, {
 121                 'ext': 'ssa',
 122                 'data': ssa,
 123             }])
 124         return subtitles
 125
 126     def _perform_login(self, username, password):
 127         try:
 128             access_token = (self._download_json(
 129                 self._API_BASE_URL + 'authentication/login', None,
 130                 'Logging in', self._LOGIN_ERR_MESSAGE, fatal=False,
 131                 data=urlencode_postdata({
 132                     'password': password,
 133                     'rememberMe': False,
 134                     'source': 'Web',
 135                     'username': username,
 136                 })) or {}).get('accessToken')
 137             if access_token:
 138                 self._HEADERS = {'authorization': 'Bearer ' + access_token}
 139         except ExtractorError as e:
 140             message = None
 141             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
 142                 resp = self._parse_json(
 143                     e.cause.read().decode(), None, fatal=False) or {}
 144                 message = resp.get('message') or resp.get('code')
 145             self.report_warning(message or self._LOGIN_ERR_MESSAGE)
 146
 147     def _real_extract(self, url):
 148         video_id = self._match_id(url)
 149         video_base_url = self._PLAYER_BASE_URL + 'video/%s/' % video_id
 150         player = self._download_json(
 151             video_base_url + 'configuration', video_id,
 152             'Downloading player config JSON metadata',
 153             headers=self._HEADERS)['player']
 154         options = player['options']
 155
 156         user = options['user']
 157         if not user.get('hasAccess'):
 158             self.raise_login_required()
 159
 160         token = self._download_json(
 161             user.get('refreshTokenUrl') or (self._PLAYER_BASE_URL + 'refresh/token'),
 162             video_id, 'Downloading access token', headers={
 163                 'x-player-refresh-token': user['refreshToken']
 164             }, data=b'')['token']
 165
 166         links_url = try_get(options, lambda x: x['video']['url']) or (video_base_url + 'link')
 167         self._K = ''.join([random.choice('0123456789abcdef') for _ in range(16)])
 168         message = bytes_to_intlist(json.dumps({
 169             'k': self._K,
 170             't': token,
 171         }))
 172
 173         # Sometimes authentication fails for no good reason, retry with
 174         # a different random padding
 175         links_data = None
 176         for _ in range(3):
 177             padded_message = intlist_to_bytes(pkcs1pad(message, 128))
 178             n, e = self._RSA_KEY
 179             encrypted_message = long_to_bytes(pow(bytes_to_long(padded_message), e, n))
 180             authorization = base64.b64encode(encrypted_message).decode()
 181
 182             try:
 183                 links_data = self._download_json(
 184                     links_url, video_id, 'Downloading links JSON metadata', headers={
 185                         'X-Player-Token': authorization
 186                     }, query={
 187                         'freeWithAds': 'true',
 188                         'adaptive': 'false',
 189                         'withMetadata': 'true',
 190                         'source': 'Web'
 191                     })
 192                 break
 193             except ExtractorError as e:
 194                 if not isinstance(e.cause, compat_HTTPError):
 195                     raise e
 196
 197                 if e.cause.code == 401:
 198                     # This usually goes away with a different random pkcs1pad, so retry
 199                     continue
 200
 201                 error = self._parse_json(e.cause.read(), video_id)
 202                 message = error.get('message')
 203                 if e.cause.code == 403 and error.get('code') == 'player-bad-geolocation-country':
 204                     self.raise_geo_restricted(msg=message)
 205                 raise ExtractorError(message)
 206         else:
 207             raise ExtractorError('Giving up retrying')
 208
 209         links = links_data.get('links') or {}
 210         metas = links_data.get('metadata') or {}
 211         sub_url = (links.get('subtitles') or {}).get('all')
 212         video_info = links_data.get('video') or {}
 213         title = metas['title']
 214
 215         formats = []
 216         for format_id, qualities in (links.get('streaming') or {}).items():
 217             if not isinstance(qualities, dict):
 218                 continue
 219             for quality, load_balancer_url in qualities.items():
 220                 load_balancer_data = self._download_json(
 221                     load_balancer_url, video_id,
 222                     'Downloading %s %s JSON metadata' % (format_id, quality),
 223                     fatal=False) or {}
 224                 m3u8_url = load_balancer_data.get('location')
 225                 if not m3u8_url:
 226                     continue
 227                 m3u8_formats = self._extract_m3u8_formats(
 228                     m3u8_url, video_id, 'mp4', 'm3u8_native',
 229                     m3u8_id=format_id, fatal=False)
 230                 if format_id == 'vf':
 231                     for f in m3u8_formats:
 232                         f['language'] = 'fr'
 233                 formats.extend(m3u8_formats)
 234         self._sort_formats(formats)
 235
 236         video = (self._download_json(
 237             self._API_BASE_URL + 'video/%s' % video_id, video_id,
 238             'Downloading additional video metadata', fatal=False) or {}).get('video') or {}
 239         show = video.get('show') or {}
 240
 241         return {
 242             'id': video_id,
 243             'title': title,
 244             'description': strip_or_none(metas.get('summary') or video.get('summary')),
 245             'thumbnail': video_info.get('image') or player.get('image'),
 246             'formats': formats,
 247             'subtitles': self.extract_subtitles(sub_url, video_id),
 248             'episode': metas.get('subtitle') or video.get('name'),
 249             'episode_number': int_or_none(video.get('shortNumber')),
 250             'series': show.get('title'),
 251             'season_number': int_or_none(video.get('season')),
 252             'duration': int_or_none(video_info.get('duration') or video.get('duration')),
 253             'release_date': unified_strdate(video.get('releaseDate')),
 254             'average_rating': float_or_none(video.get('rating') or metas.get('rating')),
 255             'comment_count': int_or_none(video.get('commentsCount')),
 256         }