_ Git - youtube-dl/blob - youtube_dl/extractor/adn.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import base64
   5 import json
   6 import os
   7
   8 from .common import InfoExtractor
   9 from ..aes import aes_cbc_decrypt
  10 from ..compat import compat_ord
  11 from ..utils import (
  12     bytes_to_intlist,
  13     ExtractorError,
  14     float_or_none,
  15     intlist_to_bytes,
  16     srt_subtitles_timecode,
  17     strip_or_none,
  18     urljoin,
  19 )
  20
  21
  22 class ADNIE(InfoExtractor):
  23     IE_DESC = 'Anime Digital Network'
  24     _VALID_URL = r'https?://(?:www\.)?animedigitalnetwork\.fr/video/[^/]+/(?P<id>\d+)'
  25     _TEST = {
  26         'url': 'http://animedigitalnetwork.fr/video/blue-exorcist-kyoto-saga/7778-episode-1-debut-des-hostilites',
  27         'md5': 'e497370d847fd79d9d4c74be55575c7a',
  28         'info_dict': {
  29             'id': '7778',
  30             'ext': 'mp4',
  31             'title': 'Blue Exorcist - Kyôto Saga - Épisode 1',
  32             'description': 'md5:2f7b5aa76edbc1a7a92cedcda8a528d5',
  33         }
  34     }
  35     _BASE_URL = 'http://animedigitalnetwork.fr'
  36
  37     def _get_subtitles(self, sub_path, video_id):
  38         if not sub_path:
  39             return None
  40
  41         enc_subtitles = self._download_webpage(
  42             urljoin(self._BASE_URL, sub_path),
  43             video_id, fatal=False, headers={
  44                 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:53.0) Gecko/20100101 Firefox/53.0',
  45             })
  46         if not enc_subtitles:
  47             return None
  48
  49         # http://animedigitalnetwork.fr/components/com_vodvideo/videojs/adn-vjs.min.js
  50         dec_subtitles = intlist_to_bytes(aes_cbc_decrypt(
  51             bytes_to_intlist(base64.b64decode(enc_subtitles[24:])),
  52             bytes_to_intlist(b'\x1b\xe0\x29\x61\x38\x94\x24\x00\x12\xbd\xc5\x80\xac\xce\xbe\xb0'),
  53             bytes_to_intlist(base64.b64decode(enc_subtitles[:24]))
  54         ))
  55         subtitles_json = self._parse_json(
  56             dec_subtitles[:-compat_ord(dec_subtitles[-1])].decode(),
  57             None, fatal=False)
  58         if not subtitles_json:
  59             return None
  60
  61         subtitles = {}
  62         for sub_lang, sub in subtitles_json.items():
  63             srt = ''
  64             for num, current in enumerate(sub):
  65                 start, end, text = (
  66                     float_or_none(current.get('startTime')),
  67                     float_or_none(current.get('endTime')),
  68                     current.get('text'))
  69                 if start is None or end is None or text is None:
  70                     continue
  71                 srt += os.linesep.join(
  72                     (
  73                         '%d' % num,
  74                         '%s --> %s' % (
  75                             srt_subtitles_timecode(start),
  76                             srt_subtitles_timecode(end)),
  77                         text,
  78                         os.linesep,
  79                     ))
  80
  81             if sub_lang == 'vostf':
  82                 sub_lang = 'fr'
  83             subtitles.setdefault(sub_lang, []).extend([{
  84                 'ext': 'json',
  85                 'data': json.dumps(sub),
  86             }, {
  87                 'ext': 'srt',
  88                 'data': srt,
  89             }])
  90         return subtitles
  91
  92     def _real_extract(self, url):
  93         video_id = self._match_id(url)
  94         webpage = self._download_webpage(url, video_id)
  95         player_config = self._parse_json(self._search_regex(
  96             r'playerConfig\s*=\s*({.+});', webpage, 'player config'), video_id)
  97
  98         video_info = {}
  99         video_info_str = self._search_regex(
 100             r'videoInfo\s*=\s*({.+});', webpage,
 101             'video info', fatal=False)
 102         if video_info_str:
 103             video_info = self._parse_json(
 104                 video_info_str, video_id, fatal=False) or {}
 105
 106         options = player_config.get('options') or {}
 107         metas = options.get('metas') or {}
 108         title = metas.get('title') or video_info['title']
 109         links = player_config.get('links') or {}
 110         if not links:
 111             links_url = player_config['linksurl']
 112             links_data = self._download_json(urljoin(
 113                 self._BASE_URL, links_url), video_id)
 114             links = links_data.get('links') or {}
 115
 116         formats = []
 117         for format_id, qualities in links.items():
 118             if not isinstance(qualities, dict):
 119                 continue
 120             for load_balancer_url in qualities.values():
 121                 load_balancer_data = self._download_json(
 122                     load_balancer_url, video_id, fatal=False) or {}
 123                 m3u8_url = load_balancer_data.get('location')
 124                 if not m3u8_url:
 125                     continue
 126                 m3u8_formats = self._extract_m3u8_formats(
 127                     m3u8_url, video_id, 'mp4', 'm3u8_native',
 128                     m3u8_id=format_id, fatal=False)
 129                 if format_id == 'vf':
 130                     for f in m3u8_formats:
 131                         f['language'] = 'fr'
 132                 formats.extend(m3u8_formats)
 133         error = options.get('error')
 134         if not formats and error:
 135             raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
 136         self._sort_formats(formats)
 137
 138         return {
 139             'id': video_id,
 140             'title': title,
 141             'description': strip_or_none(metas.get('summary') or video_info.get('resume')),
 142             'thumbnail': video_info.get('image'),
 143             'formats': formats,
 144             'subtitles': self.extract_subtitles(player_config.get('subtitles'), video_id),
 145             'episode': metas.get('subtitle') or video_info.get('videoTitle'),
 146             'series': video_info.get('playlistTitle'),
 147         }