yt_dlp/extractor/leeco.py

   1 import base64
   2 import datetime as dt
   3 import hashlib
   4 import re
   5 import time
   6 import urllib.parse
   7
   8 from .common import InfoExtractor
   9 from ..compat import compat_ord
  10 from ..utils import (
  11     ExtractorError,
  12     determine_ext,
  13     encode_data_uri,
  14     int_or_none,
  15     orderedSet,
  16     parse_iso8601,
  17     str_or_none,
  18     url_basename,
  19     urshift,
  20 )
  21
  22
  23 class LeIE(InfoExtractor):
  24     IE_DESC = '乐视网'
  25     _VALID_URL = r'https?://(?:www\.le\.com/ptv/vplay|(?:sports\.le|(?:www\.)?lesports)\.com/(?:match|video))/(?P<id>\d+)\.html'
  26     _GEO_COUNTRIES = ['CN']
  27     _URL_TEMPLATE = 'http://www.le.com/ptv/vplay/%s.html'
  28
  29     _TESTS = [{
  30         'url': 'http://www.le.com/ptv/vplay/22005890.html',
  31         'md5': 'edadcfe5406976f42f9f266057ee5e40',
  32         'info_dict': {
  33             'id': '22005890',
  34             'ext': 'mp4',
  35             'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家',
  36             'description': 'md5:a9cb175fd753e2962176b7beca21a47c',
  37         },
  38         'params': {
  39             'hls_prefer_native': True,
  40         },
  41     }, {
  42         'url': 'http://www.le.com/ptv/vplay/1415246.html',
  43         'info_dict': {
  44             'id': '1415246',
  45             'ext': 'mp4',
  46             'title': '美人天下01',
  47             'description': 'md5:28942e650e82ed4fcc8e4de919ee854d',
  48         },
  49         'params': {
  50             'hls_prefer_native': True,
  51         },
  52     }, {
  53         'note': 'This video is available only in Mainland China, thus a proxy is needed',
  54         'url': 'http://www.le.com/ptv/vplay/1118082.html',
  55         'md5': '2424c74948a62e5f31988438979c5ad1',
  56         'info_dict': {
  57             'id': '1118082',
  58             'ext': 'mp4',
  59             'title': '与龙共舞 完整版',
  60             'description': 'md5:7506a5eeb1722bb9d4068f85024e3986',
  61         },
  62         'params': {
  63             'hls_prefer_native': True,
  64         },
  65     }, {
  66         'url': 'http://sports.le.com/video/25737697.html',
  67         'only_matching': True,
  68     }, {
  69         'url': 'http://www.lesports.com/match/1023203003.html',
  70         'only_matching': True,
  71     }, {
  72         'url': 'http://sports.le.com/match/1023203003.html',
  73         'only_matching': True,
  74     }]
  75
  76     # ror() and calc_time_key() are reversed from a embedded swf file in LetvPlayer.swf
  77     def ror(self, param1, param2):
  78         _loc3_ = 0
  79         while _loc3_ < param2:
  80             param1 = urshift(param1, 1) + ((param1 & 1) << 31)
  81             _loc3_ += 1
  82         return param1
  83
  84     def calc_time_key(self, param1):
  85         _loc2_ = 185025305
  86         return self.ror(param1, _loc2_ % 17) ^ _loc2_
  87
  88     # see M3U8Encryption class in KLetvPlayer.swf
  89     @staticmethod
  90     def decrypt_m3u8(encrypted_data):
  91         if encrypted_data[:5].decode('utf-8').lower() != 'vc_01':
  92             return encrypted_data
  93         encrypted_data = encrypted_data[5:]
  94
  95         _loc4_ = bytearray(2 * len(encrypted_data))
  96         for idx, val in enumerate(encrypted_data):
  97             b = compat_ord(val)
  98             _loc4_[2 * idx] = b // 16
  99             _loc4_[2 * idx + 1] = b % 16
 100         idx = len(_loc4_) - 11
 101         _loc4_ = _loc4_[idx:] + _loc4_[:idx]
 102         _loc7_ = bytearray(len(encrypted_data))
 103         for i in range(len(encrypted_data)):
 104             _loc7_[i] = _loc4_[2 * i] * 16 + _loc4_[2 * i + 1]
 105
 106         return bytes(_loc7_)
 107
 108     def _check_errors(self, play_json):
 109         # Check for errors
 110         playstatus = play_json['msgs']['playstatus']
 111         if playstatus['status'] == 0:
 112             flag = playstatus['flag']
 113             if flag == 1:
 114                 self.raise_geo_restricted()
 115             else:
 116                 raise ExtractorError('Generic error. flag = %d' % flag, expected=True)
 117
 118     def _real_extract(self, url):
 119         media_id = self._match_id(url)
 120         page = self._download_webpage(url, media_id)
 121
 122         play_json_flash = self._download_json(
 123             'http://player-pc.le.com/mms/out/video/playJson',
 124             media_id, 'Downloading flash playJson data', query={
 125                 'id': media_id,
 126                 'platid': 1,
 127                 'splatid': 105,
 128                 'format': 1,
 129                 'source': 1000,
 130                 'tkey': self.calc_time_key(int(time.time())),
 131                 'domain': 'www.le.com',
 132                 'region': 'cn',
 133             },
 134             headers=self.geo_verification_headers())
 135         self._check_errors(play_json_flash)
 136
 137         def get_flash_urls(media_url, format_id):
 138             nodes_data = self._download_json(
 139                 media_url, media_id,
 140                 f'Download JSON metadata for format {format_id}',
 141                 query={
 142                     'm3v': 1,
 143                     'format': 1,
 144                     'expect': 3,
 145                     'tss': 'ios',
 146                 })
 147
 148             req = self._request_webpage(
 149                 nodes_data['nodelist'][0]['location'], media_id,
 150                 note=f'Downloading m3u8 information for format {format_id}')
 151
 152             m3u8_data = self.decrypt_m3u8(req.read())
 153
 154             return {
 155                 'hls': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
 156             }
 157
 158         extracted_formats = []
 159         formats = []
 160         playurl = play_json_flash['msgs']['playurl']
 161         play_domain = playurl['domain'][0]
 162
 163         for format_id, format_data in playurl.get('dispatch', []).items():
 164             if format_id in extracted_formats:
 165                 continue
 166             extracted_formats.append(format_id)
 167
 168             media_url = play_domain + format_data[0]
 169             for protocol, format_url in get_flash_urls(media_url, format_id).items():
 170                 f = {
 171                     'url': format_url,
 172                     'ext': determine_ext(format_data[1]),
 173                     'format_id': f'{protocol}-{format_id}',
 174                     'protocol': 'm3u8_native' if protocol == 'hls' else 'http',
 175                     'quality': int_or_none(format_id),
 176                 }
 177
 178                 if format_id[-1:] == 'p':
 179                     f['height'] = int_or_none(format_id[:-1])
 180
 181                 formats.append(f)
 182
 183         publish_time = parse_iso8601(self._html_search_regex(
 184             r'发布时间&nbsp;([^<>]+) ', page, 'publish time', default=None),
 185             delimiter=' ', timezone=dt.timedelta(hours=8))
 186         description = self._html_search_meta('description', page, fatal=False)
 187
 188         return {
 189             'id': media_id,
 190             'formats': formats,
 191             'title': playurl['title'],
 192             'thumbnail': playurl['pic'],
 193             'description': description,
 194             'timestamp': publish_time,
 195             '_format_sort_fields': ('res', 'quality'),
 196         }
 197
 198
 199 class LePlaylistIE(InfoExtractor):
 200     _VALID_URL = r'https?://[a-z]+\.le\.com/(?!video)[a-z]+/(?P<id>[a-z0-9_]+)'
 201
 202     _TESTS = [{
 203         'url': 'http://www.le.com/tv/46177.html',
 204         'info_dict': {
 205             'id': '46177',
 206             'title': '美人天下',
 207             'description': 'md5:395666ff41b44080396e59570dbac01c',
 208         },
 209         'playlist_count': 35,
 210     }, {
 211         'url': 'http://tv.le.com/izt/wuzetian/index.html',
 212         'info_dict': {
 213             'id': 'wuzetian',
 214             'title': '武媚娘传奇',
 215             'description': 'md5:e12499475ab3d50219e5bba00b3cb248',
 216         },
 217         # This playlist contains some extra videos other than the drama itself
 218         'playlist_mincount': 96,
 219     }, {
 220         'url': 'http://tv.le.com/pzt/lswjzzjc/index.shtml',
 221         # This series is moved to http://www.le.com/tv/10005297.html
 222         'only_matching': True,
 223     }, {
 224         'url': 'http://www.le.com/comic/92063.html',
 225         'only_matching': True,
 226     }, {
 227         'url': 'http://list.le.com/listn/c1009_sc532002_d2_p1_o1.html',
 228         'only_matching': True,
 229     }]
 230
 231     @classmethod
 232     def suitable(cls, url):
 233         return False if LeIE.suitable(url) else super().suitable(url)
 234
 235     def _real_extract(self, url):
 236         playlist_id = self._match_id(url)
 237         page = self._download_webpage(url, playlist_id)
 238
 239         # Currently old domain names are still used in playlists
 240         media_ids = orderedSet(re.findall(
 241             r'<a[^>]+href="http://www\.letv\.com/ptv/vplay/(\d+)\.html', page))
 242         entries = [self.url_result(LeIE._URL_TEMPLATE % media_id, ie='Le')
 243                    for media_id in media_ids]
 244
 245         title = self._html_search_meta('keywords', page,
 246                                        fatal=False).split('，')[0]
 247         description = self._html_search_meta('description', page, fatal=False)
 248
 249         return self.playlist_result(entries, playlist_id, playlist_title=title,
 250                                     playlist_description=description)
 251
 252
 253 class LetvCloudIE(InfoExtractor):
 254     # Most of *.letv.com is changed to *.le.com on 2016/01/02
 255     # but yuntv.letv.com is kept, so also keep the extractor name
 256     IE_DESC = '乐视云'
 257     _VALID_URL = r'https?://yuntv\.letv\.com/bcloud.html\?.+'
 258
 259     _TESTS = [{
 260         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=467623dedf',
 261         'md5': '26450599afd64c513bc77030ad15db44',
 262         'info_dict': {
 263             'id': 'p7jnfw5hw9_467623dedf',
 264             'ext': 'mp4',
 265             'title': 'Video p7jnfw5hw9_467623dedf',
 266         },
 267     }, {
 268         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=ec93197892&pu=2c7cd40209&auto_play=1&gpcflag=1&width=640&height=360',
 269         'md5': 'e03d9cc8d9c13191e1caf277e42dbd31',
 270         'info_dict': {
 271             'id': 'p7jnfw5hw9_ec93197892',
 272             'ext': 'mp4',
 273             'title': 'Video p7jnfw5hw9_ec93197892',
 274         },
 275     }, {
 276         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=187060b6fd',
 277         'md5': 'cb988699a776b22d4a41b9d43acfb3ac',
 278         'info_dict': {
 279             'id': 'p7jnfw5hw9_187060b6fd',
 280             'ext': 'mp4',
 281             'title': 'Video p7jnfw5hw9_187060b6fd',
 282         },
 283     }]
 284
 285     @staticmethod
 286     def sign_data(obj):
 287         if obj['cf'] == 'flash':
 288             salt = '2f9d6924b33a165a6d8b5d3d42f4f987'
 289             items = ['cf', 'format', 'ran', 'uu', 'ver', 'vu']
 290         elif obj['cf'] == 'html5':
 291             salt = 'fbeh5player12c43eccf2bec3300344'
 292             items = ['cf', 'ran', 'uu', 'bver', 'vu']
 293         input_data = ''.join([item + obj[item] for item in items]) + salt
 294         obj['sign'] = hashlib.md5(input_data.encode()).hexdigest()
 295
 296     def _get_formats(self, cf, uu, vu, media_id):
 297         def get_play_json(cf, timestamp):
 298             data = {
 299                 'cf': cf,
 300                 'ver': '2.2',
 301                 'bver': 'firefox44.0',
 302                 'format': 'json',
 303                 'uu': uu,
 304                 'vu': vu,
 305                 'ran': str(timestamp),
 306             }
 307             self.sign_data(data)
 308             return self._download_json(
 309                 'http://api.letvcloud.com/gpc.php?' + urllib.parse.urlencode(data),
 310                 media_id, f'Downloading playJson data for type {cf}')
 311
 312         play_json = get_play_json(cf, time.time())
 313         # The server time may be different from local time
 314         if play_json.get('code') == 10071:
 315             play_json = get_play_json(cf, play_json['timestamp'])
 316
 317         if not play_json.get('data'):
 318             if play_json.get('message'):
 319                 raise ExtractorError('Letv cloud said: {}'.format(play_json['message']), expected=True)
 320             elif play_json.get('code'):
 321                 raise ExtractorError('Letv cloud returned error %d' % play_json['code'], expected=True)
 322             else:
 323                 raise ExtractorError('Letv cloud returned an unknown error')
 324
 325         def b64decode(s):
 326             return base64.b64decode(s).decode('utf-8')
 327
 328         formats = []
 329         for media in play_json['data']['video_info']['media'].values():
 330             play_url = media['play_url']
 331             url = b64decode(play_url['main_url'])
 332             decoded_url = b64decode(url_basename(url))
 333             formats.append({
 334                 'url': url,
 335                 'ext': determine_ext(decoded_url),
 336                 'format_id': str_or_none(play_url.get('vtype')),
 337                 'format_note': str_or_none(play_url.get('definition')),
 338                 'width': int_or_none(play_url.get('vwidth')),
 339                 'height': int_or_none(play_url.get('vheight')),
 340             })
 341
 342         return formats
 343
 344     def _real_extract(self, url):
 345         uu_mobj = re.search(r'uu=([\w]+)', url)
 346         vu_mobj = re.search(r'vu=([\w]+)', url)
 347
 348         if not uu_mobj or not vu_mobj:
 349             raise ExtractorError(f'Invalid URL: {url}', expected=True)
 350
 351         uu = uu_mobj.group(1)
 352         vu = vu_mobj.group(1)
 353         media_id = uu + '_' + vu
 354
 355         formats = self._get_formats('flash', uu, vu, media_id) + self._get_formats('html5', uu, vu, media_id)
 356
 357         return {
 358             'id': media_id,
 359             'title': f'Video {media_id}',
 360             'formats': formats,
 361         }