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     str_or_none,
  24     url_basename,
  25     urshift,
  26     update_url_query,
  27 )
  28
  29
  30 class LeIE(InfoExtractor):
  31     IE_DESC = '乐视网'
  32     _VALID_URL = r'https?://(?:www\.le\.com/ptv/vplay|sports\.le\.com/video)/(?P<id>\d+)\.html'
  33
  34     _URL_TEMPLATE = 'http://www.le.com/ptv/vplay/%s.html'
  35
  36     _TESTS = [{
  37         'url': 'http://www.le.com/ptv/vplay/22005890.html',
  38         'md5': 'edadcfe5406976f42f9f266057ee5e40',
  39         'info_dict': {
  40             'id': '22005890',
  41             'ext': 'mp4',
  42             'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家',
  43             'description': 'md5:a9cb175fd753e2962176b7beca21a47c',
  44         },
  45         'params': {
  46             'hls_prefer_native': True,
  47         },
  48     }, {
  49         'url': 'http://www.le.com/ptv/vplay/1415246.html',
  50         'info_dict': {
  51             'id': '1415246',
  52             'ext': 'mp4',
  53             'title': '美人天下01',
  54             'description': 'md5:f88573d9d7225ada1359eaf0dbf8bcda',
  55         },
  56         'params': {
  57             'hls_prefer_native': True,
  58         },
  59     }, {
  60         'note': 'This video is available only in Mainland China, thus a proxy is needed',
  61         'url': 'http://www.le.com/ptv/vplay/1118082.html',
  62         'md5': '2424c74948a62e5f31988438979c5ad1',
  63         'info_dict': {
  64             'id': '1118082',
  65             'ext': 'mp4',
  66             'title': '与龙共舞 完整版',
  67             'description': 'md5:7506a5eeb1722bb9d4068f85024e3986',
  68         },
  69         'params': {
  70             'hls_prefer_native': True,
  71         },
  72         'skip': 'Only available in China',
  73     }, {
  74         'url': 'http://sports.le.com/video/25737697.html',
  75         'only_matching': True,
  76     }]
  77
  78     # ror() and calc_time_key() are reversed from a embedded swf file in KLetvPlayer.swf
  79     def ror(self, param1, param2):
  80         _loc3_ = 0
  81         while _loc3_ < param2:
  82             param1 = urshift(param1, 1) + ((param1 & 1) << 31)
  83             _loc3_ += 1
  84         return param1
  85
  86     def calc_time_key(self, param1):
  87         _loc2_ = 773625421
  88         _loc3_ = self.ror(param1, _loc2_ % 13)
  89         _loc3_ = _loc3_ ^ _loc2_
  90         _loc3_ = self.ror(_loc3_, _loc2_ % 17)
  91         return _loc3_
  92
  93     # reversed from http://jstatic.letvcdn.com/sdk/player.js
  94     def get_mms_key(self, time):
  95         return self.ror(time, 8) ^ 185025305
  96
  97     # see M3U8Encryption class in KLetvPlayer.swf
  98     @staticmethod
  99     def decrypt_m3u8(encrypted_data):
 100         if encrypted_data[:5].decode('utf-8').lower() != 'vc_01':
 101             return encrypted_data
 102         encrypted_data = encrypted_data[5:]
 103
 104         _loc4_ = bytearray(2 * len(encrypted_data))
 105         for idx, val in enumerate(encrypted_data):
 106             b = compat_ord(val)
 107             _loc4_[2 * idx] = b // 16
 108             _loc4_[2 * idx + 1] = b % 16
 109         idx = len(_loc4_) - 11
 110         _loc4_ = _loc4_[idx:] + _loc4_[:idx]
 111         _loc7_ = bytearray(len(encrypted_data))
 112         for i in range(len(encrypted_data)):
 113             _loc7_[i] = _loc4_[2 * i] * 16 + _loc4_[2 * i + 1]
 114
 115         return bytes(_loc7_)
 116
 117     def _check_errors(self, play_json):
 118         # Check for errors
 119         playstatus = play_json['playstatus']
 120         if playstatus['status'] == 0:
 121             flag = playstatus['flag']
 122             if flag == 1:
 123                 msg = 'Country %s auth error' % playstatus['country']
 124             else:
 125                 msg = 'Generic error. flag = %d' % flag
 126             raise ExtractorError(msg, expected=True)
 127
 128     def _real_extract(self, url):
 129         media_id = self._match_id(url)
 130         page = self._download_webpage(url, media_id)
 131
 132         play_json_h5 = self._download_json(
 133             'http://api.le.com/mms/out/video/playJsonH5',
 134             media_id, 'Downloading html5 playJson data', query={
 135                 'id': media_id,
 136                 'platid': 3,
 137                 'splatid': 304,
 138                 'format': 1,
 139                 'tkey': self.get_mms_key(int(time.time())),
 140                 'domain': 'www.le.com',
 141                 'tss': 'no',
 142             },
 143             headers=self.geo_verification_headers())
 144         self._check_errors(play_json_h5)
 145
 146         play_json_flash = self._download_json(
 147             'http://api.le.com/mms/out/video/playJson',
 148             media_id, 'Downloading flash playJson data', query={
 149                 'id': media_id,
 150                 'platid': 1,
 151                 'splatid': 101,
 152                 'format': 1,
 153                 'tkey': self.calc_time_key(int(time.time())),
 154                 'domain': 'www.le.com',
 155             },
 156             headers=self.geo_verification_headers())
 157         self._check_errors(play_json_flash)
 158
 159         def get_h5_urls(media_url, format_id):
 160             location = self._download_json(
 161                 media_url, media_id,
 162                 'Download JSON metadata for format %s' % format_id, query={
 163                     'format': 1,
 164                     'expect': 3,
 165                     'tss': 'no',
 166                 })['location']
 167
 168             return {
 169                 'http': update_url_query(location, {'tss': 'no'}),
 170                 'hls': update_url_query(location, {'tss': 'ios'}),
 171             }
 172
 173         def get_flash_urls(media_url, format_id):
 174             media_url += '&' + compat_urllib_parse_urlencode({
 175                 'm3v': 1,
 176                 'format': 1,
 177                 'expect': 3,
 178                 'rateid': format_id,
 179             })
 180
 181             nodes_data = self._download_json(
 182                 media_url, media_id,
 183                 'Download JSON metadata for format %s' % format_id)
 184
 185             req = self._request_webpage(
 186                 nodes_data['nodelist'][0]['location'], media_id,
 187                 note='Downloading m3u8 information for format %s' % format_id)
 188
 189             m3u8_data = self.decrypt_m3u8(req.read())
 190
 191             return {
 192                 'hls': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
 193             }
 194
 195         extracted_formats = []
 196         formats = []
 197         for play_json, get_urls in ((play_json_h5, get_h5_urls), (play_json_flash, get_flash_urls)):
 198             playurl = play_json['playurl']
 199             play_domain = playurl['domain'][0]
 200
 201             for format_id, format_data in playurl.get('dispatch', []).items():
 202                 if format_id in extracted_formats:
 203                     continue
 204                 extracted_formats.append(format_id)
 205
 206                 media_url = play_domain + format_data[0]
 207                 for protocol, format_url in get_urls(media_url, format_id).items():
 208                     f = {
 209                         'url': format_url,
 210                         'ext': determine_ext(format_data[1]),
 211                         'format_id': '%s-%s' % (protocol, format_id),
 212                         'protocol': 'm3u8_native' if protocol == 'hls' else 'http',
 213                         'quality': int_or_none(format_id),
 214                     }
 215
 216                     if format_id[-1:] == 'p':
 217                         f['height'] = int_or_none(format_id[:-1])
 218
 219                     formats.append(f)
 220         self._sort_formats(formats, ('height', 'quality', 'format_id'))
 221
 222         publish_time = parse_iso8601(self._html_search_regex(
 223             r'发布时间&nbsp;([^<>]+) ', page, 'publish time', default=None),
 224             delimiter=' ', timezone=datetime.timedelta(hours=8))
 225         description = self._html_search_meta('description', page, fatal=False)
 226
 227         return {
 228             'id': media_id,
 229             'formats': formats,
 230             'title': playurl['title'],
 231             'thumbnail': playurl['pic'],
 232             'description': description,
 233             'timestamp': publish_time,
 234         }
 235
 236
 237 class LePlaylistIE(InfoExtractor):
 238     _VALID_URL = r'https?://[a-z]+\.le\.com/(?!video)[a-z]+/(?P<id>[a-z0-9_]+)'
 239
 240     _TESTS = [{
 241         'url': 'http://www.le.com/tv/46177.html',
 242         'info_dict': {
 243             'id': '46177',
 244             'title': '美人天下',
 245             'description': 'md5:395666ff41b44080396e59570dbac01c'
 246         },
 247         'playlist_count': 35
 248     }, {
 249         'url': 'http://tv.le.com/izt/wuzetian/index.html',
 250         'info_dict': {
 251             'id': 'wuzetian',
 252             'title': '武媚娘传奇',
 253             'description': 'md5:e12499475ab3d50219e5bba00b3cb248'
 254         },
 255         # This playlist contains some extra videos other than the drama itself
 256         'playlist_mincount': 96
 257     }, {
 258         'url': 'http://tv.le.com/pzt/lswjzzjc/index.shtml',
 259         # This series is moved to http://www.le.com/tv/10005297.html
 260         'only_matching': True,
 261     }, {
 262         'url': 'http://www.le.com/comic/92063.html',
 263         'only_matching': True,
 264     }, {
 265         'url': 'http://list.le.com/listn/c1009_sc532002_d2_p1_o1.html',
 266         'only_matching': True,
 267     }]
 268
 269     @classmethod
 270     def suitable(cls, url):
 271         return False if LeIE.suitable(url) else super(LePlaylistIE, cls).suitable(url)
 272
 273     def _real_extract(self, url):
 274         playlist_id = self._match_id(url)
 275         page = self._download_webpage(url, playlist_id)
 276
 277         # Currently old domain names are still used in playlists
 278         media_ids = orderedSet(re.findall(
 279             r'<a[^>]+href="http://www\.letv\.com/ptv/vplay/(\d+)\.html', page))
 280         entries = [self.url_result(LeIE._URL_TEMPLATE % media_id, ie='Le')
 281                    for media_id in media_ids]
 282
 283         title = self._html_search_meta('keywords', page,
 284                                        fatal=False).split('，')[0]
 285         description = self._html_search_meta('description', page, fatal=False)
 286
 287         return self.playlist_result(entries, playlist_id, playlist_title=title,
 288                                     playlist_description=description)
 289
 290
 291 class LetvCloudIE(InfoExtractor):
 292     # Most of *.letv.com is changed to *.le.com on 2016/01/02
 293     # but yuntv.letv.com is kept, so also keep the extractor name
 294     IE_DESC = '乐视云'
 295     _VALID_URL = r'https?://yuntv\.letv\.com/bcloud.html\?.+'
 296
 297     _TESTS = [{
 298         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=467623dedf',
 299         'md5': '26450599afd64c513bc77030ad15db44',
 300         'info_dict': {
 301             'id': 'p7jnfw5hw9_467623dedf',
 302             'ext': 'mp4',
 303             'title': 'Video p7jnfw5hw9_467623dedf',
 304         },
 305     }, {
 306         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=ec93197892&pu=2c7cd40209&auto_play=1&gpcflag=1&width=640&height=360',
 307         'md5': 'e03d9cc8d9c13191e1caf277e42dbd31',
 308         'info_dict': {
 309             'id': 'p7jnfw5hw9_ec93197892',
 310             'ext': 'mp4',
 311             'title': 'Video p7jnfw5hw9_ec93197892',
 312         },
 313     }, {
 314         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=187060b6fd',
 315         'md5': 'cb988699a776b22d4a41b9d43acfb3ac',
 316         'info_dict': {
 317             'id': 'p7jnfw5hw9_187060b6fd',
 318             'ext': 'mp4',
 319             'title': 'Video p7jnfw5hw9_187060b6fd',
 320         },
 321     }]
 322
 323     @staticmethod
 324     def sign_data(obj):
 325         if obj['cf'] == 'flash':
 326             salt = '2f9d6924b33a165a6d8b5d3d42f4f987'
 327             items = ['cf', 'format', 'ran', 'uu', 'ver', 'vu']
 328         elif obj['cf'] == 'html5':
 329             salt = 'fbeh5player12c43eccf2bec3300344'
 330             items = ['cf', 'ran', 'uu', 'bver', 'vu']
 331         input_data = ''.join([item + obj[item] for item in items]) + salt
 332         obj['sign'] = hashlib.md5(input_data.encode('utf-8')).hexdigest()
 333
 334     def _get_formats(self, cf, uu, vu, media_id):
 335         def get_play_json(cf, timestamp):
 336             data = {
 337                 'cf': cf,
 338                 'ver': '2.2',
 339                 'bver': 'firefox44.0',
 340                 'format': 'json',
 341                 'uu': uu,
 342                 'vu': vu,
 343                 'ran': compat_str(timestamp),
 344             }
 345             self.sign_data(data)
 346             return self._download_json(
 347                 'http://api.letvcloud.com/gpc.php?' + compat_urllib_parse_urlencode(data),
 348                 media_id, 'Downloading playJson data for type %s' % cf)
 349
 350         play_json = get_play_json(cf, time.time())
 351         # The server time may be different from local time
 352         if play_json.get('code') == 10071:
 353             play_json = get_play_json(cf, play_json['timestamp'])
 354
 355         if not play_json.get('data'):
 356             if play_json.get('message'):
 357                 raise ExtractorError('Letv cloud said: %s' % play_json['message'], expected=True)
 358             elif play_json.get('code'):
 359                 raise ExtractorError('Letv cloud returned error %d' % play_json['code'], expected=True)
 360             else:
 361                 raise ExtractorError('Letv cloud returned an unknwon error')
 362
 363         def b64decode(s):
 364             return base64.b64decode(s.encode('utf-8')).decode('utf-8')
 365
 366         formats = []
 367         for media in play_json['data']['video_info']['media'].values():
 368             play_url = media['play_url']
 369             url = b64decode(play_url['main_url'])
 370             decoded_url = b64decode(url_basename(url))
 371             formats.append({
 372                 'url': url,
 373                 'ext': determine_ext(decoded_url),
 374                 'format_id': str_or_none(play_url.get('vtype')),
 375                 'format_note': str_or_none(play_url.get('definition')),
 376                 'width': int_or_none(play_url.get('vwidth')),
 377                 'height': int_or_none(play_url.get('vheight')),
 378             })
 379
 380         return formats
 381
 382     def _real_extract(self, url):
 383         uu_mobj = re.search('uu=([\w]+)', url)
 384         vu_mobj = re.search('vu=([\w]+)', url)
 385
 386         if not uu_mobj or not vu_mobj:
 387             raise ExtractorError('Invalid URL: %s' % url, expected=True)
 388
 389         uu = uu_mobj.group(1)
 390         vu = vu_mobj.group(1)
 391         media_id = uu + '_' + vu
 392
 393         formats = self._get_formats('flash', uu, vu, media_id) + self._get_formats('html5', uu, vu, media_id)
 394         self._sort_formats(formats)
 395
 396         return {
 397             'id': media_id,
 398             'title': 'Video %s' % media_id,
 399             'formats': formats,
 400         }