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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import base64
   5 import datetime
   6 import hashlib
   7 import re
   8 import time
   9
  10 from .common import InfoExtractor
  11 from ..compat import (
  12     compat_ord,
  13     compat_str,
  14     compat_urllib_parse_urlencode,
  15 )
  16 from ..utils import (
  17     determine_ext,
  18     encode_data_uri,
  19     ExtractorError,
  20     int_or_none,
  21     orderedSet,
  22     parse_iso8601,
  23     sanitized_Request,
  24     str_or_none,
  25     url_basename,
  26 )
  27
  28
  29 class LeIE(InfoExtractor):
  30     IE_DESC = '乐视网'
  31     _VALID_URL = r'https?://(?:www\.le\.com/ptv/vplay|sports\.le\.com/video)/(?P<id>\d+)\.html'
  32
  33     _URL_TEMPLATE = 'http://www.le.com/ptv/vplay/%s.html'
  34
  35     _TESTS = [{
  36         'url': 'http://www.le.com/ptv/vplay/22005890.html',
  37         'md5': 'edadcfe5406976f42f9f266057ee5e40',
  38         'info_dict': {
  39             'id': '22005890',
  40             'ext': 'mp4',
  41             'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家',
  42             'description': 'md5:a9cb175fd753e2962176b7beca21a47c',
  43         },
  44         'params': {
  45             'hls_prefer_native': True,
  46         },
  47     }, {
  48         'url': 'http://www.le.com/ptv/vplay/1415246.html',
  49         'info_dict': {
  50             'id': '1415246',
  51             'ext': 'mp4',
  52             'title': '美人天下01',
  53             'description': 'md5:f88573d9d7225ada1359eaf0dbf8bcda',
  54         },
  55         'params': {
  56             'hls_prefer_native': True,
  57         },
  58     }, {
  59         'note': 'This video is available only in Mainland China, thus a proxy is needed',
  60         'url': 'http://www.le.com/ptv/vplay/1118082.html',
  61         'md5': '2424c74948a62e5f31988438979c5ad1',
  62         'info_dict': {
  63             'id': '1118082',
  64             'ext': 'mp4',
  65             'title': '与龙共舞 完整版',
  66             'description': 'md5:7506a5eeb1722bb9d4068f85024e3986',
  67         },
  68         'params': {
  69             'hls_prefer_native': True,
  70         },
  71         'skip': 'Only available in China',
  72     }, {
  73         'url': 'http://sports.le.com/video/25737697.html',
  74         'only_matching': True,
  75     }]
  76
  77     @staticmethod
  78     def urshift(val, n):
  79         return val >> n if val >= 0 else (val + 0x100000000) >> n
  80
  81     # ror() and calc_time_key() are reversed from a embedded swf file in KLetvPlayer.swf
  82     def ror(self, param1, param2):
  83         _loc3_ = 0
  84         while _loc3_ < param2:
  85             param1 = self.urshift(param1, 1) + ((param1 & 1) << 31)
  86             _loc3_ += 1
  87         return param1
  88
  89     def calc_time_key(self, param1):
  90         _loc2_ = 773625421
  91         _loc3_ = self.ror(param1, _loc2_ % 13)
  92         _loc3_ = _loc3_ ^ _loc2_
  93         _loc3_ = self.ror(_loc3_, _loc2_ % 17)
  94         return _loc3_
  95
  96     # see M3U8Encryption class in KLetvPlayer.swf
  97     @staticmethod
  98     def decrypt_m3u8(encrypted_data):
  99         if encrypted_data[:5].decode('utf-8').lower() != 'vc_01':
 100             return encrypted_data
 101         encrypted_data = encrypted_data[5:]
 102
 103         _loc4_ = bytearray(2 * len(encrypted_data))
 104         for idx, val in enumerate(encrypted_data):
 105             b = compat_ord(val)
 106             _loc4_[2 * idx] = b // 16
 107             _loc4_[2 * idx + 1] = b % 16
 108         idx = len(_loc4_) - 11
 109         _loc4_ = _loc4_[idx:] + _loc4_[:idx]
 110         _loc7_ = bytearray(len(encrypted_data))
 111         for i in range(len(encrypted_data)):
 112             _loc7_[i] = _loc4_[2 * i] * 16 + _loc4_[2 * i + 1]
 113
 114         return bytes(_loc7_)
 115
 116     def _real_extract(self, url):
 117         media_id = self._match_id(url)
 118         page = self._download_webpage(url, media_id)
 119         params = {
 120             'id': media_id,
 121             'platid': 1,
 122             'splatid': 101,
 123             'format': 1,
 124             'tkey': self.calc_time_key(int(time.time())),
 125             'domain': 'www.le.com'
 126         }
 127         play_json_req = sanitized_Request(
 128             'http://api.le.com/mms/out/video/playJson?' + compat_urllib_parse_urlencode(params)
 129         )
 130         cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
 131         if cn_verification_proxy:
 132             play_json_req.add_header('Ytdl-request-proxy', cn_verification_proxy)
 133
 134         play_json = self._download_json(
 135             play_json_req,
 136             media_id, 'Downloading playJson data')
 137
 138         # Check for errors
 139         playstatus = play_json['playstatus']
 140         if playstatus['status'] == 0:
 141             flag = playstatus['flag']
 142             if flag == 1:
 143                 msg = 'Country %s auth error' % playstatus['country']
 144             else:
 145                 msg = 'Generic error. flag = %d' % flag
 146             raise ExtractorError(msg, expected=True)
 147
 148         playurl = play_json['playurl']
 149
 150         formats = ['350', '1000', '1300', '720p', '1080p']
 151         dispatch = playurl['dispatch']
 152
 153         urls = []
 154         for format_id in formats:
 155             if format_id in dispatch:
 156                 media_url = playurl['domain'][0] + dispatch[format_id][0]
 157                 media_url += '&' + compat_urllib_parse_urlencode({
 158                     'm3v': 1,
 159                     'format': 1,
 160                     'expect': 3,
 161                     'rateid': format_id,
 162                 })
 163
 164                 nodes_data = self._download_json(
 165                     media_url, media_id,
 166                     'Download JSON metadata for format %s' % format_id)
 167
 168                 req = self._request_webpage(
 169                     nodes_data['nodelist'][0]['location'], media_id,
 170                     note='Downloading m3u8 information for format %s' % format_id)
 171
 172                 m3u8_data = self.decrypt_m3u8(req.read())
 173
 174                 url_info_dict = {
 175                     'url': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
 176                     'ext': determine_ext(dispatch[format_id][1]),
 177                     'format_id': format_id,
 178                     'protocol': 'm3u8',
 179                 }
 180
 181                 if format_id[-1:] == 'p':
 182                     url_info_dict['height'] = int_or_none(format_id[:-1])
 183
 184                 urls.append(url_info_dict)
 185
 186         publish_time = parse_iso8601(self._html_search_regex(
 187             r'发布时间&nbsp;([^<>]+) ', page, 'publish time', default=None),
 188             delimiter=' ', timezone=datetime.timedelta(hours=8))
 189         description = self._html_search_meta('description', page, fatal=False)
 190
 191         return {
 192             'id': media_id,
 193             'formats': urls,
 194             'title': playurl['title'],
 195             'thumbnail': playurl['pic'],
 196             'description': description,
 197             'timestamp': publish_time,
 198         }
 199
 200
 201 class LePlaylistIE(InfoExtractor):
 202     _VALID_URL = r'https?://[a-z]+\.le\.com/(?!video)[a-z]+/(?P<id>[a-z0-9_]+)'
 203
 204     _TESTS = [{
 205         'url': 'http://www.le.com/tv/46177.html',
 206         'info_dict': {
 207             'id': '46177',
 208             'title': '美人天下',
 209             'description': 'md5:395666ff41b44080396e59570dbac01c'
 210         },
 211         'playlist_count': 35
 212     }, {
 213         'url': 'http://tv.le.com/izt/wuzetian/index.html',
 214         'info_dict': {
 215             'id': 'wuzetian',
 216             'title': '武媚娘传奇',
 217             'description': 'md5:e12499475ab3d50219e5bba00b3cb248'
 218         },
 219         # This playlist contains some extra videos other than the drama itself
 220         'playlist_mincount': 96
 221     }, {
 222         'url': 'http://tv.le.com/pzt/lswjzzjc/index.shtml',
 223         # This series is moved to http://www.le.com/tv/10005297.html
 224         'only_matching': True,
 225     }, {
 226         'url': 'http://www.le.com/comic/92063.html',
 227         'only_matching': True,
 228     }, {
 229         'url': 'http://list.le.com/listn/c1009_sc532002_d2_p1_o1.html',
 230         'only_matching': True,
 231     }]
 232
 233     @classmethod
 234     def suitable(cls, url):
 235         return False if LeIE.suitable(url) else super(LePlaylistIE, cls).suitable(url)
 236
 237     def _real_extract(self, url):
 238         playlist_id = self._match_id(url)
 239         page = self._download_webpage(url, playlist_id)
 240
 241         # Currently old domain names are still used in playlists
 242         media_ids = orderedSet(re.findall(
 243             r'<a[^>]+href="http://www\.letv\.com/ptv/vplay/(\d+)\.html', page))
 244         entries = [self.url_result(LeIE._URL_TEMPLATE % media_id, ie='Le')
 245                    for media_id in media_ids]
 246
 247         title = self._html_search_meta('keywords', page,
 248                                        fatal=False).split('，')[0]
 249         description = self._html_search_meta('description', page, fatal=False)
 250
 251         return self.playlist_result(entries, playlist_id, playlist_title=title,
 252                                     playlist_description=description)
 253
 254
 255 class LetvCloudIE(InfoExtractor):
 256     # Most of *.letv.com is changed to *.le.com on 2016/01/02
 257     # but yuntv.letv.com is kept, so also keep the extractor name
 258     IE_DESC = '乐视云'
 259     _VALID_URL = r'https?://yuntv\.letv\.com/bcloud.html\?.+'
 260
 261     _TESTS = [{
 262         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=467623dedf',
 263         'md5': '26450599afd64c513bc77030ad15db44',
 264         'info_dict': {
 265             'id': 'p7jnfw5hw9_467623dedf',
 266             'ext': 'mp4',
 267             'title': 'Video p7jnfw5hw9_467623dedf',
 268         },
 269     }, {
 270         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=ec93197892&pu=2c7cd40209&auto_play=1&gpcflag=1&width=640&height=360',
 271         'md5': 'e03d9cc8d9c13191e1caf277e42dbd31',
 272         'info_dict': {
 273             'id': 'p7jnfw5hw9_ec93197892',
 274             'ext': 'mp4',
 275             'title': 'Video p7jnfw5hw9_ec93197892',
 276         },
 277     }, {
 278         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=187060b6fd',
 279         'md5': 'cb988699a776b22d4a41b9d43acfb3ac',
 280         'info_dict': {
 281             'id': 'p7jnfw5hw9_187060b6fd',
 282             'ext': 'mp4',
 283             'title': 'Video p7jnfw5hw9_187060b6fd',
 284         },
 285     }]
 286
 287     @staticmethod
 288     def sign_data(obj):
 289         if obj['cf'] == 'flash':
 290             salt = '2f9d6924b33a165a6d8b5d3d42f4f987'
 291             items = ['cf', 'format', 'ran', 'uu', 'ver', 'vu']
 292         elif obj['cf'] == 'html5':
 293             salt = 'fbeh5player12c43eccf2bec3300344'
 294             items = ['cf', 'ran', 'uu', 'bver', 'vu']
 295         input_data = ''.join([item + obj[item] for item in items]) + salt
 296         obj['sign'] = hashlib.md5(input_data.encode('utf-8')).hexdigest()
 297
 298     def _get_formats(self, cf, uu, vu, media_id):
 299         def get_play_json(cf, timestamp):
 300             data = {
 301                 'cf': cf,
 302                 'ver': '2.2',
 303                 'bver': 'firefox44.0',
 304                 'format': 'json',
 305                 'uu': uu,
 306                 'vu': vu,
 307                 'ran': compat_str(timestamp),
 308             }
 309             self.sign_data(data)
 310             return self._download_json(
 311                 'http://api.letvcloud.com/gpc.php?' + compat_urllib_parse_urlencode(data),
 312                 media_id, 'Downloading playJson data for type %s' % cf)
 313
 314         play_json = get_play_json(cf, time.time())
 315         # The server time may be different from local time
 316         if play_json.get('code') == 10071:
 317             play_json = get_play_json(cf, play_json['timestamp'])
 318
 319         if not play_json.get('data'):
 320             if play_json.get('message'):
 321                 raise ExtractorError('Letv cloud said: %s' % play_json['message'], expected=True)
 322             elif play_json.get('code'):
 323                 raise ExtractorError('Letv cloud returned error %d' % play_json['code'], expected=True)
 324             else:
 325                 raise ExtractorError('Letv cloud returned an unknwon error')
 326
 327         def b64decode(s):
 328             return base64.b64decode(s.encode('utf-8')).decode('utf-8')
 329
 330         formats = []
 331         for media in play_json['data']['video_info']['media'].values():
 332             play_url = media['play_url']
 333             url = b64decode(play_url['main_url'])
 334             decoded_url = b64decode(url_basename(url))
 335             formats.append({
 336                 'url': url,
 337                 'ext': determine_ext(decoded_url),
 338                 'format_id': str_or_none(play_url.get('vtype')),
 339                 'format_note': str_or_none(play_url.get('definition')),
 340                 'width': int_or_none(play_url.get('vwidth')),
 341                 'height': int_or_none(play_url.get('vheight')),
 342             })
 343
 344         return formats
 345
 346     def _real_extract(self, url):
 347         uu_mobj = re.search('uu=([\w]+)', url)
 348         vu_mobj = re.search('vu=([\w]+)', url)
 349
 350         if not uu_mobj or not vu_mobj:
 351             raise ExtractorError('Invalid URL: %s' % url, expected=True)
 352
 353         uu = uu_mobj.group(1)
 354         vu = vu_mobj.group(1)
 355         media_id = uu + '_' + vu
 356
 357         formats = self._get_formats('flash', uu, vu, media_id) + self._get_formats('html5', uu, vu, media_id)
 358         self._sort_formats(formats)
 359
 360         return {
 361             'id': media_id,
 362             'title': 'Video %s' % media_id,
 363             'formats': formats,
 364         }