Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/letv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import datetime
   5 import re
   6 import time
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_urllib_parse,
  11     compat_ord,
  12 )
  13 from ..utils import (
  14     determine_ext,
  15     ExtractorError,
  16     parse_iso8601,
  17     sanitized_Request,
  18     int_or_none,
  19     encode_data_uri,
  20 )
  21
  22
  23 class LetvIE(InfoExtractor):
  24     IE_DESC = '乐视网'
  25     _VALID_URL = r'http://www\.letv\.com/ptv/vplay/(?P<id>\d+).html'
  26
  27     _TESTS = [{
  28         'url': 'http://www.letv.com/ptv/vplay/22005890.html',
  29         'md5': 'edadcfe5406976f42f9f266057ee5e40',
  30         'info_dict': {
  31             'id': '22005890',
  32             'ext': 'mp4',
  33             'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家',
  34             'description': 'md5:a9cb175fd753e2962176b7beca21a47c',
  35         },
  36         'params': {
  37             'hls_prefer_native': True,
  38         },
  39     }, {
  40         'url': 'http://www.letv.com/ptv/vplay/1415246.html',
  41         'info_dict': {
  42             'id': '1415246',
  43             'ext': 'mp4',
  44             'title': '美人天下01',
  45             'description': 'md5:f88573d9d7225ada1359eaf0dbf8bcda',
  46         },
  47         'params': {
  48             'hls_prefer_native': True,
  49         },
  50     }, {
  51         'note': 'This video is available only in Mainland China, thus a proxy is needed',
  52         'url': 'http://www.letv.com/ptv/vplay/1118082.html',
  53         'md5': '2424c74948a62e5f31988438979c5ad1',
  54         'info_dict': {
  55             'id': '1118082',
  56             'ext': 'mp4',
  57             'title': '与龙共舞 完整版',
  58             'description': 'md5:7506a5eeb1722bb9d4068f85024e3986',
  59         },
  60         'params': {
  61             'hls_prefer_native': True,
  62         },
  63         'skip': 'Only available in China',
  64     }]
  65
  66     @staticmethod
  67     def urshift(val, n):
  68         return val >> n if val >= 0 else (val + 0x100000000) >> n
  69
  70     # ror() and calc_time_key() are reversed from a embedded swf file in KLetvPlayer.swf
  71     def ror(self, param1, param2):
  72         _loc3_ = 0
  73         while _loc3_ < param2:
  74             param1 = self.urshift(param1, 1) + ((param1 & 1) << 31)
  75             _loc3_ += 1
  76         return param1
  77
  78     def calc_time_key(self, param1):
  79         _loc2_ = 773625421
  80         _loc3_ = self.ror(param1, _loc2_ % 13)
  81         _loc3_ = _loc3_ ^ _loc2_
  82         _loc3_ = self.ror(_loc3_, _loc2_ % 17)
  83         return _loc3_
  84
  85     # see M3U8Encryption class in KLetvPlayer.swf
  86     @staticmethod
  87     def decrypt_m3u8(encrypted_data):
  88         if encrypted_data[:5].decode('utf-8').lower() != 'vc_01':
  89             return encrypted_data
  90         encrypted_data = encrypted_data[5:]
  91
  92         _loc4_ = bytearray()
  93         while encrypted_data:
  94             b = compat_ord(encrypted_data[0])
  95             _loc4_.extend([b // 16, b & 0x0f])
  96             encrypted_data = encrypted_data[1:]
  97         idx = len(_loc4_) - 11
  98         _loc4_ = _loc4_[idx:] + _loc4_[:idx]
  99         _loc7_ = bytearray()
 100         while _loc4_:
 101             _loc7_.append(_loc4_[0] * 16 + _loc4_[1])
 102             _loc4_ = _loc4_[2:]
 103
 104         return bytes(_loc7_)
 105
 106     def _real_extract(self, url):
 107         media_id = self._match_id(url)
 108         page = self._download_webpage(url, media_id)
 109         params = {
 110             'id': media_id,
 111             'platid': 1,
 112             'splatid': 101,
 113             'format': 1,
 114             'tkey': self.calc_time_key(int(time.time())),
 115             'domain': 'www.letv.com'
 116         }
 117         play_json_req = sanitized_Request(
 118             'http://api.letv.com/mms/out/video/playJson?' + compat_urllib_parse.urlencode(params)
 119         )
 120         cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
 121         if cn_verification_proxy:
 122             play_json_req.add_header('Ytdl-request-proxy', cn_verification_proxy)
 123
 124         play_json = self._download_json(
 125             play_json_req,
 126             media_id, 'Downloading playJson data')
 127
 128         # Check for errors
 129         playstatus = play_json['playstatus']
 130         if playstatus['status'] == 0:
 131             flag = playstatus['flag']
 132             if flag == 1:
 133                 msg = 'Country %s auth error' % playstatus['country']
 134             else:
 135                 msg = 'Generic error. flag = %d' % flag
 136             raise ExtractorError(msg, expected=True)
 137
 138         playurl = play_json['playurl']
 139
 140         formats = ['350', '1000', '1300', '720p', '1080p']
 141         dispatch = playurl['dispatch']
 142
 143         urls = []
 144         for format_id in formats:
 145             if format_id in dispatch:
 146                 media_url = playurl['domain'][0] + dispatch[format_id][0]
 147                 media_url += '&' + compat_urllib_parse.urlencode({
 148                     'm3v': 1,
 149                     'format': 1,
 150                     'expect': 3,
 151                     'rateid': format_id,
 152                 })
 153
 154                 nodes_data = self._download_json(
 155                     media_url, media_id,
 156                     'Download JSON metadata for format %s' % format_id)
 157
 158                 req = self._request_webpage(
 159                     nodes_data['nodelist'][0]['location'], media_id,
 160                     note='Downloading m3u8 information for format %s' % format_id)
 161
 162                 m3u8_data = self.decrypt_m3u8(req.read())
 163
 164                 url_info_dict = {
 165                     'url': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
 166                     'ext': determine_ext(dispatch[format_id][1]),
 167                     'format_id': format_id,
 168                     'protocol': 'm3u8',
 169                 }
 170
 171                 if format_id[-1:] == 'p':
 172                     url_info_dict['height'] = int_or_none(format_id[:-1])
 173
 174                 urls.append(url_info_dict)
 175
 176         publish_time = parse_iso8601(self._html_search_regex(
 177             r'发布时间&nbsp;([^<>]+) ', page, 'publish time', default=None),
 178             delimiter=' ', timezone=datetime.timedelta(hours=8))
 179         description = self._html_search_meta('description', page, fatal=False)
 180
 181         return {
 182             'id': media_id,
 183             'formats': urls,
 184             'title': playurl['title'],
 185             'thumbnail': playurl['pic'],
 186             'description': description,
 187             'timestamp': publish_time,
 188         }
 189
 190
 191 class LetvTvIE(InfoExtractor):
 192     _VALID_URL = r'http://www.letv.com/tv/(?P<id>\d+).html'
 193     _TESTS = [{
 194         'url': 'http://www.letv.com/tv/46177.html',
 195         'info_dict': {
 196             'id': '46177',
 197             'title': '美人天下',
 198             'description': 'md5:395666ff41b44080396e59570dbac01c'
 199         },
 200         'playlist_count': 35
 201     }]
 202
 203     def _real_extract(self, url):
 204         playlist_id = self._match_id(url)
 205         page = self._download_webpage(url, playlist_id)
 206
 207         media_urls = list(set(re.findall(
 208             r'http://www.letv.com/ptv/vplay/\d+.html', page)))
 209         entries = [self.url_result(media_url, ie='Letv')
 210                    for media_url in media_urls]
 211
 212         title = self._html_search_meta('keywords', page,
 213                                        fatal=False).split('，')[0]
 214         description = self._html_search_meta('description', page, fatal=False)
 215
 216         return self.playlist_result(entries, playlist_id, playlist_title=title,
 217                                     playlist_description=description)
 218
 219
 220 class LetvPlaylistIE(LetvTvIE):
 221     _VALID_URL = r'http://tv.letv.com/[a-z]+/(?P<id>[a-z]+)/index.s?html'
 222     _TESTS = [{
 223         'url': 'http://tv.letv.com/izt/wuzetian/index.html',
 224         'info_dict': {
 225             'id': 'wuzetian',
 226             'title': '武媚娘传奇',
 227             'description': 'md5:e12499475ab3d50219e5bba00b3cb248'
 228         },
 229         # This playlist contains some extra videos other than the drama itself
 230         'playlist_mincount': 96
 231     }, {
 232         'url': 'http://tv.letv.com/pzt/lswjzzjc/index.shtml',
 233         'info_dict': {
 234             'id': 'lswjzzjc',
 235             # The title should be "劲舞青春", but I can't find a simple way to
 236             # determine the playlist title
 237             'title': '乐视午间自制剧场',
 238             'description': 'md5:b1eef244f45589a7b5b1af9ff25a4489'
 239         },
 240         'playlist_mincount': 7
 241     }]