summaryrefslogtreecommitdiff
path: root/youtube_dl/extractor/adn.py
blob: a55ebbcbd6898121cf19e96c69e050bbea609b3d (plain)
    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
   12 from ..compat import (
   13     compat_HTTPError,
   14     compat_b64decode,
   15     compat_ord,
   16 )
   17 from ..utils import (
   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     @staticmethod
   72     def _ass_subtitles_timecode(seconds):
   73         return '%01d:%02d:%02d.%02d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 100)
   74 
   75     def _get_subtitles(self, sub_url, video_id):
   76         if not sub_url:
   77             return None
   78 
   79         enc_subtitles = self._download_webpage(
   80             sub_url, video_id, 'Downloading subtitles location', fatal=False) or '{}'
   81         subtitle_location = (self._parse_json(enc_subtitles, video_id, fatal=False) or {}).get('location')
   82         if subtitle_location:
   83             enc_subtitles = self._download_webpage(
   84                 subtitle_location, video_id, 'Downloading subtitles data',
   85                 fatal=False, headers={'Origin': 'https://animedigitalnetwork.fr'})
   86         if not enc_subtitles:
   87             return None
   88 
   89         # http://animedigitalnetwork.fr/components/com_vodvideo/videojs/adn-vjs.min.js
   90         dec_subtitles = intlist_to_bytes(aes_cbc_decrypt(
   91             bytes_to_intlist(compat_b64decode(enc_subtitles[24:])),
   92             bytes_to_intlist(binascii.unhexlify(self._K + 'ab9f52f5baae7c72')),
   93             bytes_to_intlist(compat_b64decode(enc_subtitles[:24]))
   94         ))
   95         subtitles_json = self._parse_json(
   96             dec_subtitles[:-compat_ord(dec_subtitles[-1])].decode(),
   97             None, fatal=False)
   98         if not subtitles_json:
   99             return None
  100 
  101         subtitles = {}
  102         for sub_lang, sub in subtitles_json.items():
  103             ssa = '''[Script Info]
  104 ScriptType:V4.00
  105 [V4 Styles]
  106 Format: Name,Fontname,Fontsize,PrimaryColour,SecondaryColour,TertiaryColour,BackColour,Bold,Italic,BorderStyle,Outline,Shadow,Alignment,MarginL,MarginR,MarginV,AlphaLevel,Encoding
  107 Style: Default,Arial,18,16777215,16777215,16777215,0,-1,0,1,1,0,2,20,20,20,0,0
  108 [Events]
  109 Format: Marked,Start,End,Style,Name,MarginL,MarginR,MarginV,Effect,Text'''
  110             for current in sub:
  111                 start, end, text, line_align, position_align = (
  112                     float_or_none(current.get('startTime')),
  113                     float_or_none(current.get('endTime')),
  114                     current.get('text'), current.get('lineAlign'),
  115                     current.get('positionAlign'))
  116                 if start is None or end is None or text is None:
  117                     continue
  118                 alignment = self._POS_ALIGN_MAP.get(position_align, 2) + self._LINE_ALIGN_MAP.get(line_align, 0)
  119                 ssa += os.linesep + 'Dialogue: Marked=0,%s,%s,Default,,0,0,0,,%s%s' % (
  120                     self._ass_subtitles_timecode(start),
  121                     self._ass_subtitles_timecode(end),
  122                     '{\\a%d}' % alignment if alignment != 2 else '',
  123                     text.replace('\n', '\\N').replace('<i>', '{\\i1}').replace('</i>', '{\\i0}'))
  124 
  125             if sub_lang == 'vostf':
  126                 sub_lang = 'fr'
  127             subtitles.setdefault(sub_lang, []).extend([{
  128                 'ext': 'json',
  129                 'data': json.dumps(sub),
  130             }, {
  131                 'ext': 'ssa',
  132                 'data': ssa,
  133             }])
  134         return subtitles
  135 
  136     def _real_initialize(self):
  137         username, password = self._get_login_info()
  138         if not username:
  139             return
  140         try:
  141             access_token = (self._download_json(
  142                 self._API_BASE_URL + 'authentication/login', None,
  143                 'Logging in', self._LOGIN_ERR_MESSAGE, fatal=False,
  144                 data=urlencode_postdata({
  145                     'password': password,
  146                     'rememberMe': False,
  147                     'source': 'Web',
  148                     'username': username,
  149                 })) or {}).get('accessToken')
  150             if access_token:
  151                 self._HEADERS = {'authorization': 'Bearer ' + access_token}
  152         except ExtractorError as e:
  153             message = None
  154             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
  155                 resp = self._parse_json(
  156                     e.cause.read().decode(), None, fatal=False) or {}
  157                 message = resp.get('message') or resp.get('code')
  158             self.report_warning(message or self._LOGIN_ERR_MESSAGE)
  159 
  160     def _real_extract(self, url):
  161         video_id = self._match_id(url)
  162         video_base_url = self._PLAYER_BASE_URL + 'video/%s/' % video_id
  163         player = self._download_json(
  164             video_base_url + 'configuration', video_id,
  165             'Downloading player config JSON metadata',
  166             headers=self._HEADERS)['player']
  167         options = player['options']
  168 
  169         user = options['user']
  170         if not user.get('hasAccess'):
  171             self.raise_login_required()
  172 
  173         token = self._download_json(
  174             user.get('refreshTokenUrl') or (self._PLAYER_BASE_URL + 'refresh/token'),
  175             video_id, 'Downloading access token', headers={
  176                 'x-player-refresh-token': user['refreshToken']
  177             }, data=b'')['token']
  178 
  179         links_url = try_get(options, lambda x: x['video']['url']) or (video_base_url + 'link')
  180         self._K = ''.join([random.choice('0123456789abcdef') for _ in range(16)])
  181         message = bytes_to_intlist(json.dumps({
  182             'k': self._K,
  183             't': token,
  184         }))
  185 
  186         # Sometimes authentication fails for no good reason, retry with
  187         # a different random padding
  188         links_data = None
  189         for _ in range(3):
  190             padded_message = intlist_to_bytes(pkcs1pad(message, 128))
  191             n, e = self._RSA_KEY
  192             encrypted_message = long_to_bytes(pow(bytes_to_long(padded_message), e, n))
  193             authorization = base64.b64encode(encrypted_message).decode()
  194 
  195             try:
  196                 links_data = self._download_json(
  197                     links_url, video_id, 'Downloading links JSON metadata', headers={
  198                         'X-Player-Token': authorization
  199                     }, query={
  200                         'freeWithAds': 'true',
  201                         'adaptive': 'false',
  202                         'withMetadata': 'true',
  203                         'source': 'Web'
  204                     })
  205                 break
  206             except ExtractorError as e:
  207                 if not isinstance(e.cause, compat_HTTPError):
  208                     raise e
  209 
  210                 if e.cause.code == 401:
  211                     # This usually goes away with a different random pkcs1pad, so retry
  212                     continue
  213 
  214                 error = self._parse_json(e.cause.read(), video_id)
  215                 message = error.get('message')
  216                 if e.cause.code == 403 and error.get('code') == 'player-bad-geolocation-country':
  217                     self.raise_geo_restricted(msg=message)
  218                 raise ExtractorError(message)
  219         else:
  220             raise ExtractorError('Giving up retrying')
  221 
  222         links = links_data.get('links') or {}
  223         metas = links_data.get('metadata') or {}
  224         sub_url = (links.get('subtitles') or {}).get('all')
  225         video_info = links_data.get('video') or {}
  226         title = metas['title']
  227 
  228         formats = []
  229         for format_id, qualities in (links.get('streaming') or {}).items():
  230             if not isinstance(qualities, dict):
  231                 continue
  232             for quality, load_balancer_url in qualities.items():
  233                 load_balancer_data = self._download_json(
  234                     load_balancer_url, video_id,
  235                     'Downloading %s %s JSON metadata' % (format_id, quality),
  236                     fatal=False) or {}
  237                 m3u8_url = load_balancer_data.get('location')
  238                 if not m3u8_url:
  239                     continue
  240                 m3u8_formats = self._extract_m3u8_formats(
  241                     m3u8_url, video_id, 'mp4', 'm3u8_native',
  242                     m3u8_id=format_id, fatal=False)
  243                 if format_id == 'vf':
  244                     for f in m3u8_formats:
  245                         f['language'] = 'fr'
  246                 formats.extend(m3u8_formats)
  247         self._sort_formats(formats)
  248 
  249         video = (self._download_json(
  250             self._API_BASE_URL + 'video/%s' % video_id, video_id,
  251             'Downloading additional video metadata', fatal=False) or {}).get('video') or {}
  252         show = video.get('show') or {}
  253 
  254         return {
  255             'id': video_id,
  256             'title': title,
  257             'description': strip_or_none(metas.get('summary') or video.get('summary')),
  258             'thumbnail': video_info.get('image') or player.get('image'),
  259             'formats': formats,
  260             'subtitles': self.extract_subtitles(sub_url, video_id),
  261             'episode': metas.get('subtitle') or video.get('name'),
  262             'episode_number': int_or_none(video.get('shortNumber')),
  263             'series': show.get('title'),
  264             'season_number': int_or_none(video.get('season')),
  265             'duration': int_or_none(video_info.get('duration') or video.get('duration')),
  266             'release_date': unified_strdate(video.get('releaseDate')),
  267             'average_rating': float_or_none(video.get('rating') or metas.get('rating')),
  268             'comment_count': int_or_none(video.get('commentsCount')),
  269         }

Generated by cgit