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|(?:www\.)?lesports)\.com/(?:match|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         'url': 'http://www.lesports.com/match/1023203003.html',
  78         'only_matching': True,
  79     }, {
  80         'url': 'http://sports.le.com/match/1023203003.html',
  81         'only_matching': True,
  82     }]
  83
  84     # ror() and calc_time_key() are reversed from a embedded swf file in KLetvPlayer.swf
  85     def ror(self, param1, param2):
  86         _loc3_ = 0
  87         while _loc3_ < param2:
  88             param1 = urshift(param1, 1) + ((param1 & 1) << 31)
  89             _loc3_ += 1
  90         return param1
  91
  92     def calc_time_key(self, param1):
  93         _loc2_ = 773625421
  94         _loc3_ = self.ror(param1, _loc2_ % 13)
  95         _loc3_ = _loc3_ ^ _loc2_
  96         _loc3_ = self.ror(_loc3_, _loc2_ % 17)
  97         return _loc3_
  98
  99     # reversed from http://jstatic.letvcdn.com/sdk/player.js
 100     def get_mms_key(self, time):
 101         return self.ror(time, 8) ^ 185025305
 102
 103     # see M3U8Encryption class in KLetvPlayer.swf
 104     @staticmethod
 105     def decrypt_m3u8(encrypted_data):
 106         if encrypted_data[:5].decode('utf-8').lower() != 'vc_01':
 107             return encrypted_data
 108         encrypted_data = encrypted_data[5:]
 109
 110         _loc4_ = bytearray(2 * len(encrypted_data))
 111         for idx, val in enumerate(encrypted_data):
 112             b = compat_ord(val)
 113             _loc4_[2 * idx] = b // 16
 114             _loc4_[2 * idx + 1] = b % 16
 115         idx = len(_loc4_) - 11
 116         _loc4_ = _loc4_[idx:] + _loc4_[:idx]
 117         _loc7_ = bytearray(len(encrypted_data))
 118         for i in range(len(encrypted_data)):
 119             _loc7_[i] = _loc4_[2 * i] * 16 + _loc4_[2 * i + 1]
 120
 121         return bytes(_loc7_)
 122
 123     def _check_errors(self, play_json):
 124         # Check for errors
 125         playstatus = play_json['playstatus']
 126         if playstatus['status'] == 0:
 127             flag = playstatus['flag']
 128             if flag == 1:
 129                 msg = 'Country %s auth error' % playstatus['country']
 130             else:
 131                 msg = 'Generic error. flag = %d' % flag
 132             raise ExtractorError(msg, expected=True)
 133
 134     def _real_extract(self, url):
 135         media_id = self._match_id(url)
 136         page = self._download_webpage(url, media_id)
 137
 138         play_json_h5 = self._download_json(
 139             'http://api.le.com/mms/out/video/playJsonH5',
 140             media_id, 'Downloading html5 playJson data', query={
 141                 'id': media_id,
 142                 'platid': 3,
 143                 'splatid': 304,
 144                 'format': 1,
 145                 'tkey': self.get_mms_key(int(time.time())),
 146                 'domain': 'www.le.com',
 147                 'tss': 'no',
 148             },
 149             headers=self.geo_verification_headers())
 150         self._check_errors(play_json_h5)
 151
 152         play_json_flash = self._download_json(
 153             'http://api.le.com/mms/out/video/playJson',
 154             media_id, 'Downloading flash playJson data', query={
 155                 'id': media_id,
 156                 'platid': 1,
 157                 'splatid': 101,
 158                 'format': 1,
 159                 'tkey': self.calc_time_key(int(time.time())),
 160                 'domain': 'www.le.com',
 161             },
 162             headers=self.geo_verification_headers())
 163         self._check_errors(play_json_flash)
 164
 165         def get_h5_urls(media_url, format_id):
 166             location = self._download_json(
 167                 media_url, media_id,
 168                 'Download JSON metadata for format %s' % format_id, query={
 169                     'format': 1,
 170                     'expect': 3,
 171                     'tss': 'no',
 172                 })['location']
 173
 174             return {
 175                 'http': update_url_query(location, {'tss': 'no'}),
 176                 'hls': update_url_query(location, {'tss': 'ios'}),
 177             }
 178
 179         def get_flash_urls(media_url, format_id):
 180             media_url += '&' + compat_urllib_parse_urlencode({
 181                 'm3v': 1,
 182                 'format': 1,
 183                 'expect': 3,
 184                 'rateid': format_id,
 185             })
 186
 187             nodes_data = self._download_json(
 188                 media_url, media_id,
 189                 'Download JSON metadata for format %s' % format_id)
 190
 191             req = self._request_webpage(
 192                 nodes_data['nodelist'][0]['location'], media_id,
 193                 note='Downloading m3u8 information for format %s' % format_id)
 194
 195             m3u8_data = self.decrypt_m3u8(req.read())
 196
 197             return {
 198                 'hls': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
 199             }
 200
 201         extracted_formats = []
 202         formats = []
 203         for play_json, get_urls in ((play_json_h5, get_h5_urls), (play_json_flash, get_flash_urls)):
 204             playurl = play_json['playurl']
 205             play_domain = playurl['domain'][0]
 206
 207             for format_id, format_data in playurl.get('dispatch', []).items():
 208                 if format_id in extracted_formats:
 209                     continue
 210                 extracted_formats.append(format_id)
 211
 212                 media_url = play_domain + format_data[0]
 213                 for protocol, format_url in get_urls(media_url, format_id).items():
 214                     f = {
 215                         'url': format_url,
 216                         'ext': determine_ext(format_data[1]),
 217                         'format_id': '%s-%s' % (protocol, format_id),
 218                         'protocol': 'm3u8_native' if protocol == 'hls' else 'http',
 219                         'quality': int_or_none(format_id),
 220                     }
 221
 222                     if format_id[-1:] == 'p':
 223                         f['height'] = int_or_none(format_id[:-1])
 224
 225                     formats.append(f)
 226         self._sort_formats(formats, ('height', 'quality', 'format_id'))
 227
 228         publish_time = parse_iso8601(self._html_search_regex(
 229             r'发布时间&nbsp;([^<>]+) ', page, 'publish time', default=None),
 230             delimiter=' ', timezone=datetime.timedelta(hours=8))
 231         description = self._html_search_meta('description', page, fatal=False)
 232
 233         return {
 234             'id': media_id,
 235             'formats': formats,
 236             'title': playurl['title'],
 237             'thumbnail': playurl['pic'],
 238             'description': description,
 239             'timestamp': publish_time,
 240         }
 241
 242
 243 class LePlaylistIE(InfoExtractor):
 244     _VALID_URL = r'https?://[a-z]+\.le\.com/(?!video)[a-z]+/(?P<id>[a-z0-9_]+)'
 245
 246     _TESTS = [{
 247         'url': 'http://www.le.com/tv/46177.html',
 248         'info_dict': {
 249             'id': '46177',
 250             'title': '美人天下',
 251             'description': 'md5:395666ff41b44080396e59570dbac01c'
 252         },
 253         'playlist_count': 35
 254     }, {
 255         'url': 'http://tv.le.com/izt/wuzetian/index.html',
 256         'info_dict': {
 257             'id': 'wuzetian',
 258             'title': '武媚娘传奇',
 259             'description': 'md5:e12499475ab3d50219e5bba00b3cb248'
 260         },
 261         # This playlist contains some extra videos other than the drama itself
 262         'playlist_mincount': 96
 263     }, {
 264         'url': 'http://tv.le.com/pzt/lswjzzjc/index.shtml',
 265         # This series is moved to http://www.le.com/tv/10005297.html
 266         'only_matching': True,
 267     }, {
 268         'url': 'http://www.le.com/comic/92063.html',
 269         'only_matching': True,
 270     }, {
 271         'url': 'http://list.le.com/listn/c1009_sc532002_d2_p1_o1.html',
 272         'only_matching': True,
 273     }]
 274
 275     @classmethod
 276     def suitable(cls, url):
 277         return False if LeIE.suitable(url) else super(LePlaylistIE, cls).suitable(url)
 278
 279     def _real_extract(self, url):
 280         playlist_id = self._match_id(url)
 281         page = self._download_webpage(url, playlist_id)
 282
 283         # Currently old domain names are still used in playlists
 284         media_ids = orderedSet(re.findall(
 285             r'<a[^>]+href="http://www\.letv\.com/ptv/vplay/(\d+)\.html', page))
 286         entries = [self.url_result(LeIE._URL_TEMPLATE % media_id, ie='Le')
 287                    for media_id in media_ids]
 288
 289         title = self._html_search_meta('keywords', page,
 290                                        fatal=False).split('，')[0]
 291         description = self._html_search_meta('description', page, fatal=False)
 292
 293         return self.playlist_result(entries, playlist_id, playlist_title=title,
 294                                     playlist_description=description)
 295
 296
 297 class LetvCloudIE(InfoExtractor):
 298     # Most of *.letv.com is changed to *.le.com on 2016/01/02
 299     # but yuntv.letv.com is kept, so also keep the extractor name
 300     IE_DESC = '乐视云'
 301     _VALID_URL = r'https?://yuntv\.letv\.com/bcloud.html\?.+'
 302
 303     _TESTS = [{
 304         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=467623dedf',
 305         'md5': '26450599afd64c513bc77030ad15db44',
 306         'info_dict': {
 307             'id': 'p7jnfw5hw9_467623dedf',
 308             'ext': 'mp4',
 309             'title': 'Video p7jnfw5hw9_467623dedf',
 310         },
 311     }, {
 312         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=ec93197892&pu=2c7cd40209&auto_play=1&gpcflag=1&width=640&height=360',
 313         'md5': 'e03d9cc8d9c13191e1caf277e42dbd31',
 314         'info_dict': {
 315             'id': 'p7jnfw5hw9_ec93197892',
 316             'ext': 'mp4',
 317             'title': 'Video p7jnfw5hw9_ec93197892',
 318         },
 319     }, {
 320         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=187060b6fd',
 321         'md5': 'cb988699a776b22d4a41b9d43acfb3ac',
 322         'info_dict': {
 323             'id': 'p7jnfw5hw9_187060b6fd',
 324             'ext': 'mp4',
 325             'title': 'Video p7jnfw5hw9_187060b6fd',
 326         },
 327     }]
 328
 329     @staticmethod
 330     def sign_data(obj):
 331         if obj['cf'] == 'flash':
 332             salt = '2f9d6924b33a165a6d8b5d3d42f4f987'
 333             items = ['cf', 'format', 'ran', 'uu', 'ver', 'vu']
 334         elif obj['cf'] == 'html5':
 335             salt = 'fbeh5player12c43eccf2bec3300344'
 336             items = ['cf', 'ran', 'uu', 'bver', 'vu']
 337         input_data = ''.join([item + obj[item] for item in items]) + salt
 338         obj['sign'] = hashlib.md5(input_data.encode('utf-8')).hexdigest()
 339
 340     def _get_formats(self, cf, uu, vu, media_id):
 341         def get_play_json(cf, timestamp):
 342             data = {
 343                 'cf': cf,
 344                 'ver': '2.2',
 345                 'bver': 'firefox44.0',
 346                 'format': 'json',
 347                 'uu': uu,
 348                 'vu': vu,
 349                 'ran': compat_str(timestamp),
 350             }
 351             self.sign_data(data)
 352             return self._download_json(
 353                 'http://api.letvcloud.com/gpc.php?' + compat_urllib_parse_urlencode(data),
 354                 media_id, 'Downloading playJson data for type %s' % cf)
 355
 356         play_json = get_play_json(cf, time.time())
 357         # The server time may be different from local time
 358         if play_json.get('code') == 10071:
 359             play_json = get_play_json(cf, play_json['timestamp'])
 360
 361         if not play_json.get('data'):
 362             if play_json.get('message'):
 363                 raise ExtractorError('Letv cloud said: %s' % play_json['message'], expected=True)
 364             elif play_json.get('code'):
 365                 raise ExtractorError('Letv cloud returned error %d' % play_json['code'], expected=True)
 366             else:
 367                 raise ExtractorError('Letv cloud returned an unknwon error')
 368
 369         def b64decode(s):
 370             return base64.b64decode(s.encode('utf-8')).decode('utf-8')
 371
 372         formats = []
 373         for media in play_json['data']['video_info']['media'].values():
 374             play_url = media['play_url']
 375             url = b64decode(play_url['main_url'])
 376             decoded_url = b64decode(url_basename(url))
 377             formats.append({
 378                 'url': url,
 379                 'ext': determine_ext(decoded_url),
 380                 'format_id': str_or_none(play_url.get('vtype')),
 381                 'format_note': str_or_none(play_url.get('definition')),
 382                 'width': int_or_none(play_url.get('vwidth')),
 383                 'height': int_or_none(play_url.get('vheight')),
 384             })
 385
 386         return formats
 387
 388     def _real_extract(self, url):
 389         uu_mobj = re.search(r'uu=([\w]+)', url)
 390         vu_mobj = re.search(r'vu=([\w]+)', url)
 391
 392         if not uu_mobj or not vu_mobj:
 393             raise ExtractorError('Invalid URL: %s' % url, expected=True)
 394
 395         uu = uu_mobj.group(1)
 396         vu = vu_mobj.group(1)
 397         media_id = uu + '_' + vu
 398
 399         formats = self._get_formats('flash', uu, vu, media_id) + self._get_formats('html5', uu, vu, media_id)
 400         self._sort_formats(formats)
 401
 402         return {
 403             'id': media_id,
 404             'title': 'Video %s' % media_id,
 405             'formats': formats,
 406         }