yt_dlp/extractor/adn.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import base64
   5 import binascii
   6 import json
   7 import os
   8 import random
   9
  10 from .common import InfoExtractor
  11 from ..aes import aes_cbc_decrypt_bytes, unpad_pkcs7
  12 from ..compat import (
  13     compat_HTTPError,
  14     compat_b64decode,
  15 )
  16 from ..utils import (
  17     ass_subtitles_timecode,
  18     bytes_to_intlist,
  19     bytes_to_long,
  20     ExtractorError,
  21     float_or_none,
  22     int_or_none,
  23     intlist_to_bytes,
  24     long_to_bytes,
  25     pkcs1pad,
  26     strip_or_none,
  27     try_get,
  28     unified_strdate,
  29     urlencode_postdata,
  30 )
  31
  32
  33 class ADNIE(InfoExtractor):
  34     IE_DESC = 'Anime Digital Network'
  35     _VALID_URL = r'https?://(?:www\.)?animedigitalnetwork\.fr/video/[^/]+/(?P<id>\d+)'
  36     _TEST = {
  37         'url': 'http://animedigitalnetwork.fr/video/blue-exorcist-kyoto-saga/7778-episode-1-debut-des-hostilites',
  38         'md5': '0319c99885ff5547565cacb4f3f9348d',
  39         'info_dict': {
  40             'id': '7778',
  41             'ext': 'mp4',
  42             'title': 'Blue Exorcist - Kyôto Saga - Episode 1',
  43             'description': 'md5:2f7b5aa76edbc1a7a92cedcda8a528d5',
  44             'series': 'Blue Exorcist - Kyôto Saga',
  45             'duration': 1467,
  46             'release_date': '20170106',
  47             'comment_count': int,
  48             'average_rating': float,
  49             'season_number': 2,
  50             'episode': 'Début des hostilités',
  51             'episode_number': 1,
  52         }
  53     }
  54
  55     _NETRC_MACHINE = 'animedigitalnetwork'
  56     _BASE_URL = 'http://animedigitalnetwork.fr'
  57     _API_BASE_URL = 'https://gw.api.animedigitalnetwork.fr/'
  58     _PLAYER_BASE_URL = _API_BASE_URL + 'player/'
  59     _HEADERS = {}
  60     _LOGIN_ERR_MESSAGE = 'Unable to log in'
  61     _RSA_KEY = (0x9B42B08905199A5CCE2026274399CA560ECB209EE9878A708B1C0812E1BB8CB5D1FB7441861147C1A1F2F3A0476DD63A9CAC20D3E983613346850AA6CB38F16DC7D720FD7D86FC6E5B3D5BBC72E14CD0BF9E869F2CEA2CCAD648F1DCE38F1FF916CEFB2D339B64AA0264372344BC775E265E8A852F88144AB0BD9AA06C1A4ABB, 65537)
  62     _POS_ALIGN_MAP = {
  63         'start': 1,
  64         'end': 3,
  65     }
  66     _LINE_ALIGN_MAP = {
  67         'middle': 8,
  68         'end': 4,
  69     }
  70
  71     def _get_subtitles(self, sub_url, video_id):
  72         if not sub_url:
  73             return None
  74
  75         enc_subtitles = self._download_webpage(
  76             sub_url, video_id, 'Downloading subtitles location', fatal=False) or '{}'
  77         subtitle_location = (self._parse_json(enc_subtitles, video_id, fatal=False) or {}).get('location')
  78         if subtitle_location:
  79             enc_subtitles = self._download_webpage(
  80                 subtitle_location, video_id, 'Downloading subtitles data',
  81                 fatal=False, headers={'Origin': 'https://animedigitalnetwork.fr'})
  82         if not enc_subtitles:
  83             return None
  84
  85         # http://animedigitalnetwork.fr/components/com_vodvideo/videojs/adn-vjs.min.js
  86         dec_subtitles = unpad_pkcs7(aes_cbc_decrypt_bytes(
  87             compat_b64decode(enc_subtitles[24:]),
  88             binascii.unhexlify(self._K + 'ab9f52f5baae7c72'),
  89             compat_b64decode(enc_subtitles[:24])))
  90         subtitles_json = self._parse_json(dec_subtitles.decode(), None, fatal=False)
  91         if not subtitles_json:
  92             return None
  93
  94         subtitles = {}
  95         for sub_lang, sub in subtitles_json.items():
  96             ssa = '''[Script Info]
  97 ScriptType:V4.00
  98 [V4 Styles]
  99 Format: Name,Fontname,Fontsize,PrimaryColour,SecondaryColour,TertiaryColour,BackColour,Bold,Italic,BorderStyle,Outline,Shadow,Alignment,MarginL,MarginR,MarginV,AlphaLevel,Encoding
 100 Style: Default,Arial,18,16777215,16777215,16777215,0,-1,0,1,1,0,2,20,20,20,0,0
 101 [Events]
 102 Format: Marked,Start,End,Style,Name,MarginL,MarginR,MarginV,Effect,Text'''
 103             for current in sub:
 104                 start, end, text, line_align, position_align = (
 105                     float_or_none(current.get('startTime')),
 106                     float_or_none(current.get('endTime')),
 107                     current.get('text'), current.get('lineAlign'),
 108                     current.get('positionAlign'))
 109                 if start is None or end is None or text is None:
 110                     continue
 111                 alignment = self._POS_ALIGN_MAP.get(position_align, 2) + self._LINE_ALIGN_MAP.get(line_align, 0)
 112                 ssa += os.linesep + 'Dialogue: Marked=0,%s,%s,Default,,0,0,0,,%s%s' % (
 113                     ass_subtitles_timecode(start),
 114                     ass_subtitles_timecode(end),
 115                     '{\\a%d}' % alignment if alignment != 2 else '',
 116                     text.replace('\n', '\\N').replace('<i>', '{\\i1}').replace('</i>', '{\\i0}'))
 117
 118             if sub_lang == 'vostf':
 119                 sub_lang = 'fr'
 120             subtitles.setdefault(sub_lang, []).extend([{
 121                 'ext': 'json',
 122                 'data': json.dumps(sub),
 123             }, {
 124                 'ext': 'ssa',
 125                 'data': ssa,
 126             }])
 127         return subtitles
 128
 129     def _real_initialize(self):
 130         username, password = self._get_login_info()
 131         if not username:
 132             return
 133         try:
 134             access_token = (self._download_json(
 135                 self._API_BASE_URL + 'authentication/login', None,
 136                 'Logging in', self._LOGIN_ERR_MESSAGE, fatal=False,
 137                 data=urlencode_postdata({
 138                     'password': password,
 139                     'rememberMe': False,
 140                     'source': 'Web',
 141                     'username': username,
 142                 })) or {}).get('accessToken')
 143             if access_token:
 144                 self._HEADERS = {'authorization': 'Bearer ' + access_token}
 145         except ExtractorError as e:
 146             message = None
 147             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
 148                 resp = self._parse_json(
 149                     e.cause.read().decode(), None, fatal=False) or {}
 150                 message = resp.get('message') or resp.get('code')
 151             self.report_warning(message or self._LOGIN_ERR_MESSAGE)
 152
 153     def _real_extract(self, url):
 154         video_id = self._match_id(url)
 155         video_base_url = self._PLAYER_BASE_URL + 'video/%s/' % video_id
 156         player = self._download_json(
 157             video_base_url + 'configuration', video_id,
 158             'Downloading player config JSON metadata',
 159             headers=self._HEADERS)['player']
 160         options = player['options']
 161
 162         user = options['user']
 163         if not user.get('hasAccess'):
 164             self.raise_login_required()
 165
 166         token = self._download_json(
 167             user.get('refreshTokenUrl') or (self._PLAYER_BASE_URL + 'refresh/token'),
 168             video_id, 'Downloading access token', headers={
 169                 'x-player-refresh-token': user['refreshToken']
 170             }, data=b'')['token']
 171
 172         links_url = try_get(options, lambda x: x['video']['url']) or (video_base_url + 'link')
 173         self._K = ''.join([random.choice('0123456789abcdef') for _ in range(16)])
 174         message = bytes_to_intlist(json.dumps({
 175             'k': self._K,
 176             't': token,
 177         }))
 178
 179         # Sometimes authentication fails for no good reason, retry with
 180         # a different random padding
 181         links_data = None
 182         for _ in range(3):
 183             padded_message = intlist_to_bytes(pkcs1pad(message, 128))
 184             n, e = self._RSA_KEY
 185             encrypted_message = long_to_bytes(pow(bytes_to_long(padded_message), e, n))
 186             authorization = base64.b64encode(encrypted_message).decode()
 187
 188             try:
 189                 links_data = self._download_json(
 190                     links_url, video_id, 'Downloading links JSON metadata', headers={
 191                         'X-Player-Token': authorization
 192                     }, query={
 193                         'freeWithAds': 'true',
 194                         'adaptive': 'false',
 195                         'withMetadata': 'true',
 196                         'source': 'Web'
 197                     })
 198                 break
 199             except ExtractorError as e:
 200                 if not isinstance(e.cause, compat_HTTPError):
 201                     raise e
 202
 203                 if e.cause.code == 401:
 204                     # This usually goes away with a different random pkcs1pad, so retry
 205                     continue
 206
 207                 error = self._parse_json(e.cause.read(), video_id)
 208                 message = error.get('message')
 209                 if e.cause.code == 403 and error.get('code') == 'player-bad-geolocation-country':
 210                     self.raise_geo_restricted(msg=message)
 211                 raise ExtractorError(message)
 212         else:
 213             raise ExtractorError('Giving up retrying')
 214
 215         links = links_data.get('links') or {}
 216         metas = links_data.get('metadata') or {}
 217         sub_url = (links.get('subtitles') or {}).get('all')
 218         video_info = links_data.get('video') or {}
 219         title = metas['title']
 220
 221         formats = []
 222         for format_id, qualities in (links.get('streaming') or {}).items():
 223             if not isinstance(qualities, dict):
 224                 continue
 225             for quality, load_balancer_url in qualities.items():
 226                 load_balancer_data = self._download_json(
 227                     load_balancer_url, video_id,
 228                     'Downloading %s %s JSON metadata' % (format_id, quality),
 229                     fatal=False) or {}
 230                 m3u8_url = load_balancer_data.get('location')
 231                 if not m3u8_url:
 232                     continue
 233                 m3u8_formats = self._extract_m3u8_formats(
 234                     m3u8_url, video_id, 'mp4', 'm3u8_native',
 235                     m3u8_id=format_id, fatal=False)
 236                 if format_id == 'vf':
 237                     for f in m3u8_formats:
 238                         f['language'] = 'fr'
 239                 formats.extend(m3u8_formats)
 240         self._sort_formats(formats)
 241
 242         video = (self._download_json(
 243             self._API_BASE_URL + 'video/%s' % video_id, video_id,
 244             'Downloading additional video metadata', fatal=False) or {}).get('video') or {}
 245         show = video.get('show') or {}
 246
 247         return {
 248             'id': video_id,
 249             'title': title,
 250             'description': strip_or_none(metas.get('summary') or video.get('summary')),
 251             'thumbnail': video_info.get('image') or player.get('image'),
 252             'formats': formats,
 253             'subtitles': self.extract_subtitles(sub_url, video_id),
 254             'episode': metas.get('subtitle') or video.get('name'),
 255             'episode_number': int_or_none(video.get('shortNumber')),
 256             'series': show.get('title'),
 257             'season_number': int_or_none(video.get('season')),
 258             'duration': int_or_none(video_info.get('duration') or video.get('duration')),
 259             'release_date': unified_strdate(video.get('releaseDate')),
 260             'average_rating': float_or_none(video.get('rating') or metas.get('rating')),
 261             'comment_count': int_or_none(video.get('commentsCount')),
 262         }