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