Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/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
  12 from ..compat import (
  13     compat_b64decode,
  14     compat_ord,
  15 )
  16 from ..utils import (
  17     bytes_to_intlist,
  18     bytes_to_long,
  19     ExtractorError,
  20     float_or_none,
  21     intlist_to_bytes,
  22     long_to_bytes,
  23     pkcs1pad,
  24     strip_or_none,
  25     urljoin,
  26 )
  27
  28
  29 class ADNIE(InfoExtractor):
  30     IE_DESC = 'Anime Digital Network'
  31     _VALID_URL = r'https?://(?:www\.)?animedigitalnetwork\.fr/video/[^/]+/(?P<id>\d+)'
  32     _TEST = {
  33         'url': 'http://animedigitalnetwork.fr/video/blue-exorcist-kyoto-saga/7778-episode-1-debut-des-hostilites',
  34         'md5': 'e497370d847fd79d9d4c74be55575c7a',
  35         'info_dict': {
  36             'id': '7778',
  37             'ext': 'mp4',
  38             'title': 'Blue Exorcist - Kyôto Saga - Épisode 1',
  39             'description': 'md5:2f7b5aa76edbc1a7a92cedcda8a528d5',
  40         }
  41     }
  42     _BASE_URL = 'http://animedigitalnetwork.fr'
  43     _RSA_KEY = (0xc35ae1e4356b65a73b551493da94b8cb443491c0aa092a357a5aee57ffc14dda85326f42d716e539a34542a0d3f363adf16c5ec222d713d5997194030ee2e4f0d1fb328c01a81cf6868c090d50de8e169c6b13d1675b9eeed1cbc51e1fffca9b38af07f37abd790924cd3bee59d0257cfda4fe5f3f0534877e21ce5821447d1b, 65537)
  44     _POS_ALIGN_MAP = {
  45         'start': 1,
  46         'end': 3,
  47     }
  48     _LINE_ALIGN_MAP = {
  49         'middle': 8,
  50         'end': 4,
  51     }
  52
  53     @staticmethod
  54     def _ass_subtitles_timecode(seconds):
  55         return '%01d:%02d:%02d.%02d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 100)
  56
  57     def _get_subtitles(self, sub_path, video_id):
  58         if not sub_path:
  59             return None
  60
  61         enc_subtitles = self._download_webpage(
  62             urljoin(self._BASE_URL, sub_path),
  63             video_id, 'Downloading subtitles location', fatal=False) or '{}'
  64         subtitle_location = (self._parse_json(enc_subtitles, video_id, fatal=False) or {}).get('location')
  65         if subtitle_location:
  66             enc_subtitles = self._download_webpage(
  67                 urljoin(self._BASE_URL, subtitle_location),
  68                 video_id, 'Downloading subtitles data', fatal=False,
  69                 headers={'Origin': 'https://animedigitalnetwork.fr'})
  70         if not enc_subtitles:
  71             return None
  72
  73         # http://animedigitalnetwork.fr/components/com_vodvideo/videojs/adn-vjs.min.js
  74         dec_subtitles = intlist_to_bytes(aes_cbc_decrypt(
  75             bytes_to_intlist(compat_b64decode(enc_subtitles[24:])),
  76             bytes_to_intlist(binascii.unhexlify(self._K + '4b8ef13ec1872730')),
  77             bytes_to_intlist(compat_b64decode(enc_subtitles[:24]))
  78         ))
  79         subtitles_json = self._parse_json(
  80             dec_subtitles[:-compat_ord(dec_subtitles[-1])].decode(),
  81             None, fatal=False)
  82         if not subtitles_json:
  83             return None
  84
  85         subtitles = {}
  86         for sub_lang, sub in subtitles_json.items():
  87             ssa = '''[Script Info]
  88 ScriptType:V4.00
  89 [V4 Styles]
  90 Format: Name,Fontname,Fontsize,PrimaryColour,SecondaryColour,TertiaryColour,BackColour,Bold,Italic,BorderStyle,Outline,Shadow,Alignment,MarginL,MarginR,MarginV,AlphaLevel,Encoding
  91 Style: Default,Arial,18,16777215,16777215,16777215,0,-1,0,1,1,0,2,20,20,20,0,0
  92 [Events]
  93 Format: Marked,Start,End,Style,Name,MarginL,MarginR,MarginV,Effect,Text'''
  94             for current in sub:
  95                 start, end, text, line_align, position_align = (
  96                     float_or_none(current.get('startTime')),
  97                     float_or_none(current.get('endTime')),
  98                     current.get('text'), current.get('lineAlign'),
  99                     current.get('positionAlign'))
 100                 if start is None or end is None or text is None:
 101                     continue
 102                 alignment = self._POS_ALIGN_MAP.get(position_align, 2) + self._LINE_ALIGN_MAP.get(line_align, 0)
 103                 ssa += os.linesep + 'Dialogue: Marked=0,%s,%s,Default,,0,0,0,,%s%s' % (
 104                     self._ass_subtitles_timecode(start),
 105                     self._ass_subtitles_timecode(end),
 106                     '{\\a%d}' % alignment if alignment != 2 else '',
 107                     text.replace('\n', '\\N').replace('<i>', '{\\i1}').replace('</i>', '{\\i0}'))
 108
 109             if sub_lang == 'vostf':
 110                 sub_lang = 'fr'
 111             subtitles.setdefault(sub_lang, []).extend([{
 112                 'ext': 'json',
 113                 'data': json.dumps(sub),
 114             }, {
 115                 'ext': 'ssa',
 116                 'data': ssa,
 117             }])
 118         return subtitles
 119
 120     def _real_extract(self, url):
 121         video_id = self._match_id(url)
 122         webpage = self._download_webpage(url, video_id)
 123         player_config = self._parse_json(self._search_regex(
 124             r'playerConfig\s*=\s*({.+});', webpage,
 125             'player config', default='{}'), video_id, fatal=False)
 126         if not player_config:
 127             config_url = urljoin(self._BASE_URL, self._search_regex(
 128                 r'(?:id="player"|class="[^"]*adn-player-container[^"]*")[^>]+data-url="([^"]+)"',
 129                 webpage, 'config url'))
 130             player_config = self._download_json(
 131                 config_url, video_id,
 132                 'Downloading player config JSON metadata')['player']
 133
 134         video_info = {}
 135         video_info_str = self._search_regex(
 136             r'videoInfo\s*=\s*({.+});', webpage,
 137             'video info', fatal=False)
 138         if video_info_str:
 139             video_info = self._parse_json(
 140                 video_info_str, video_id, fatal=False) or {}
 141
 142         options = player_config.get('options') or {}
 143         metas = options.get('metas') or {}
 144         links = player_config.get('links') or {}
 145         sub_path = player_config.get('subtitles')
 146         error = None
 147         if not links:
 148             links_url = player_config.get('linksurl') or options['videoUrl']
 149             token = options['token']
 150             self._K = ''.join([random.choice('0123456789abcdef') for _ in range(16)])
 151             message = bytes_to_intlist(json.dumps({
 152                 'k': self._K,
 153                 'e': 60,
 154                 't': token,
 155             }))
 156             padded_message = intlist_to_bytes(pkcs1pad(message, 128))
 157             n, e = self._RSA_KEY
 158             encrypted_message = long_to_bytes(pow(bytes_to_long(padded_message), e, n))
 159             authorization = base64.b64encode(encrypted_message).decode()
 160             links_data = self._download_json(
 161                 urljoin(self._BASE_URL, links_url), video_id,
 162                 'Downloading links JSON metadata', headers={
 163                     'Authorization': 'Bearer ' + authorization,
 164                 })
 165             links = links_data.get('links') or {}
 166             metas = metas or links_data.get('meta') or {}
 167             sub_path = sub_path or links_data.get('subtitles') or \
 168                 'index.php?option=com_vodapi&task=subtitles.getJSON&format=json&id=' + video_id
 169             sub_path += '&token=' + token
 170             error = links_data.get('error')
 171         title = metas.get('title') or video_info['title']
 172
 173         formats = []
 174         for format_id, qualities in links.items():
 175             if not isinstance(qualities, dict):
 176                 continue
 177             for quality, load_balancer_url in qualities.items():
 178                 load_balancer_data = self._download_json(
 179                     load_balancer_url, video_id,
 180                     'Downloading %s %s JSON metadata' % (format_id, quality),
 181                     fatal=False) or {}
 182                 m3u8_url = load_balancer_data.get('location')
 183                 if not m3u8_url:
 184                     continue
 185                 m3u8_formats = self._extract_m3u8_formats(
 186                     m3u8_url, video_id, 'mp4', 'm3u8_native',
 187                     m3u8_id=format_id, fatal=False)
 188                 if format_id == 'vf':
 189                     for f in m3u8_formats:
 190                         f['language'] = 'fr'
 191                 formats.extend(m3u8_formats)
 192         if not error:
 193             error = options.get('error')
 194         if not formats and error:
 195             raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
 196         self._sort_formats(formats)
 197
 198         return {
 199             'id': video_id,
 200             'title': title,
 201             'description': strip_or_none(metas.get('summary') or video_info.get('resume')),
 202             'thumbnail': video_info.get('image'),
 203             'formats': formats,
 204             'subtitles': self.extract_subtitles(sub_path, video_id),
 205             'episode': metas.get('subtitle') or video_info.get('videoTitle'),
 206             'series': video_info.get('playlistTitle'),
 207         }