yt_dlp/extractor/leeco.py

   1 import datetime
   2 import hashlib
   3 import re
   4 import time
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_b64decode,
   9     compat_ord,
  10     compat_str,
  11     compat_urllib_parse_urlencode,
  12 )
  13 from ..utils import (
  14     determine_ext,
  15     encode_data_uri,
  16     ExtractorError,
  17     int_or_none,
  18     orderedSet,
  19     parse_iso8601,
  20     str_or_none,
  21     url_basename,
  22     urshift,
  23 )
  24
  25
  26 class LeIE(InfoExtractor):
  27     IE_DESC = '乐视网'
  28     _VALID_URL = r'https?://(?:www\.le\.com/ptv/vplay|(?:sports\.le|(?:www\.)?lesports)\.com/(?:match|video))/(?P<id>\d+)\.html'
  29     _GEO_COUNTRIES = ['CN']
  30     _URL_TEMPLATE = 'http://www.le.com/ptv/vplay/%s.html'
  31
  32     _TESTS = [{
  33         'url': 'http://www.le.com/ptv/vplay/22005890.html',
  34         'md5': 'edadcfe5406976f42f9f266057ee5e40',
  35         'info_dict': {
  36             'id': '22005890',
  37             'ext': 'mp4',
  38             'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家',
  39             'description': 'md5:a9cb175fd753e2962176b7beca21a47c',
  40         },
  41         'params': {
  42             'hls_prefer_native': True,
  43         },
  44     }, {
  45         'url': 'http://www.le.com/ptv/vplay/1415246.html',
  46         'info_dict': {
  47             'id': '1415246',
  48             'ext': 'mp4',
  49             'title': '美人天下01',
  50             'description': 'md5:28942e650e82ed4fcc8e4de919ee854d',
  51         },
  52         'params': {
  53             'hls_prefer_native': True,
  54         },
  55     }, {
  56         'note': 'This video is available only in Mainland China, thus a proxy is needed',
  57         'url': 'http://www.le.com/ptv/vplay/1118082.html',
  58         'md5': '2424c74948a62e5f31988438979c5ad1',
  59         'info_dict': {
  60             'id': '1118082',
  61             'ext': 'mp4',
  62             'title': '与龙共舞 完整版',
  63             'description': 'md5:7506a5eeb1722bb9d4068f85024e3986',
  64         },
  65         'params': {
  66             'hls_prefer_native': True,
  67         },
  68     }, {
  69         'url': 'http://sports.le.com/video/25737697.html',
  70         'only_matching': True,
  71     }, {
  72         'url': 'http://www.lesports.com/match/1023203003.html',
  73         'only_matching': True,
  74     }, {
  75         'url': 'http://sports.le.com/match/1023203003.html',
  76         'only_matching': True,
  77     }]
  78
  79     # ror() and calc_time_key() are reversed from a embedded swf file in LetvPlayer.swf
  80     def ror(self, param1, param2):
  81         _loc3_ = 0
  82         while _loc3_ < param2:
  83             param1 = urshift(param1, 1) + ((param1 & 1) << 31)
  84             _loc3_ += 1
  85         return param1
  86
  87     def calc_time_key(self, param1):
  88         _loc2_ = 185025305
  89         return self.ror(param1, _loc2_ % 17) ^ _loc2_
  90
  91     # see M3U8Encryption class in KLetvPlayer.swf
  92     @staticmethod
  93     def decrypt_m3u8(encrypted_data):
  94         if encrypted_data[:5].decode('utf-8').lower() != 'vc_01':
  95             return encrypted_data
  96         encrypted_data = encrypted_data[5:]
  97
  98         _loc4_ = bytearray(2 * len(encrypted_data))
  99         for idx, val in enumerate(encrypted_data):
 100             b = compat_ord(val)
 101             _loc4_[2 * idx] = b // 16
 102             _loc4_[2 * idx + 1] = b % 16
 103         idx = len(_loc4_) - 11
 104         _loc4_ = _loc4_[idx:] + _loc4_[:idx]
 105         _loc7_ = bytearray(len(encrypted_data))
 106         for i in range(len(encrypted_data)):
 107             _loc7_[i] = _loc4_[2 * i] * 16 + _loc4_[2 * i + 1]
 108
 109         return bytes(_loc7_)
 110
 111     def _check_errors(self, play_json):
 112         # Check for errors
 113         playstatus = play_json['msgs']['playstatus']
 114         if playstatus['status'] == 0:
 115             flag = playstatus['flag']
 116             if flag == 1:
 117                 self.raise_geo_restricted()
 118             else:
 119                 raise ExtractorError('Generic error. flag = %d' % flag, expected=True)
 120
 121     def _real_extract(self, url):
 122         media_id = self._match_id(url)
 123         page = self._download_webpage(url, media_id)
 124
 125         play_json_flash = self._download_json(
 126             'http://player-pc.le.com/mms/out/video/playJson',
 127             media_id, 'Downloading flash playJson data', query={
 128                 'id': media_id,
 129                 'platid': 1,
 130                 'splatid': 105,
 131                 'format': 1,
 132                 'source': 1000,
 133                 'tkey': self.calc_time_key(int(time.time())),
 134                 'domain': 'www.le.com',
 135                 'region': 'cn',
 136             },
 137             headers=self.geo_verification_headers())
 138         self._check_errors(play_json_flash)
 139
 140         def get_flash_urls(media_url, format_id):
 141             nodes_data = self._download_json(
 142                 media_url, media_id,
 143                 'Download JSON metadata for format %s' % format_id,
 144                 query={
 145                     'm3v': 1,
 146                     'format': 1,
 147                     'expect': 3,
 148                     'tss': 'ios',
 149                 })
 150
 151             req = self._request_webpage(
 152                 nodes_data['nodelist'][0]['location'], media_id,
 153                 note='Downloading m3u8 information for format %s' % format_id)
 154
 155             m3u8_data = self.decrypt_m3u8(req.read())
 156
 157             return {
 158                 'hls': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
 159             }
 160
 161         extracted_formats = []
 162         formats = []
 163         playurl = play_json_flash['msgs']['playurl']
 164         play_domain = playurl['domain'][0]
 165
 166         for format_id, format_data in playurl.get('dispatch', []).items():
 167             if format_id in extracted_formats:
 168                 continue
 169             extracted_formats.append(format_id)
 170
 171             media_url = play_domain + format_data[0]
 172             for protocol, format_url in get_flash_urls(media_url, format_id).items():
 173                 f = {
 174                     'url': format_url,
 175                     'ext': determine_ext(format_data[1]),
 176                     'format_id': '%s-%s' % (protocol, format_id),
 177                     'protocol': 'm3u8_native' if protocol == 'hls' else 'http',
 178                     'quality': int_or_none(format_id),
 179                 }
 180
 181                 if format_id[-1:] == 'p':
 182                     f['height'] = int_or_none(format_id[:-1])
 183
 184                 formats.append(f)
 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': formats,
 194             'title': playurl['title'],
 195             'thumbnail': playurl['pic'],
 196             'description': description,
 197             'timestamp': publish_time,
 198             '_format_sort_fields': ('res', 'quality'),
 199         }
 200
 201
 202 class LePlaylistIE(InfoExtractor):
 203     _VALID_URL = r'https?://[a-z]+\.le\.com/(?!video)[a-z]+/(?P<id>[a-z0-9_]+)'
 204
 205     _TESTS = [{
 206         'url': 'http://www.le.com/tv/46177.html',
 207         'info_dict': {
 208             'id': '46177',
 209             'title': '美人天下',
 210             'description': 'md5:395666ff41b44080396e59570dbac01c'
 211         },
 212         'playlist_count': 35
 213     }, {
 214         'url': 'http://tv.le.com/izt/wuzetian/index.html',
 215         'info_dict': {
 216             'id': 'wuzetian',
 217             'title': '武媚娘传奇',
 218             'description': 'md5:e12499475ab3d50219e5bba00b3cb248'
 219         },
 220         # This playlist contains some extra videos other than the drama itself
 221         'playlist_mincount': 96
 222     }, {
 223         'url': 'http://tv.le.com/pzt/lswjzzjc/index.shtml',
 224         # This series is moved to http://www.le.com/tv/10005297.html
 225         'only_matching': True,
 226     }, {
 227         'url': 'http://www.le.com/comic/92063.html',
 228         'only_matching': True,
 229     }, {
 230         'url': 'http://list.le.com/listn/c1009_sc532002_d2_p1_o1.html',
 231         'only_matching': True,
 232     }]
 233
 234     @classmethod
 235     def suitable(cls, url):
 236         return False if LeIE.suitable(url) else super(LePlaylistIE, cls).suitable(url)
 237
 238     def _real_extract(self, url):
 239         playlist_id = self._match_id(url)
 240         page = self._download_webpage(url, playlist_id)
 241
 242         # Currently old domain names are still used in playlists
 243         media_ids = orderedSet(re.findall(
 244             r'<a[^>]+href="http://www\.letv\.com/ptv/vplay/(\d+)\.html', page))
 245         entries = [self.url_result(LeIE._URL_TEMPLATE % media_id, ie='Le')
 246                    for media_id in media_ids]
 247
 248         title = self._html_search_meta('keywords', page,
 249                                        fatal=False).split('，')[0]
 250         description = self._html_search_meta('description', page, fatal=False)
 251
 252         return self.playlist_result(entries, playlist_id, playlist_title=title,
 253                                     playlist_description=description)
 254
 255
 256 class LetvCloudIE(InfoExtractor):
 257     # Most of *.letv.com is changed to *.le.com on 2016/01/02
 258     # but yuntv.letv.com is kept, so also keep the extractor name
 259     IE_DESC = '乐视云'
 260     _VALID_URL = r'https?://yuntv\.letv\.com/bcloud.html\?.+'
 261
 262     _TESTS = [{
 263         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=467623dedf',
 264         'md5': '26450599afd64c513bc77030ad15db44',
 265         'info_dict': {
 266             'id': 'p7jnfw5hw9_467623dedf',
 267             'ext': 'mp4',
 268             'title': 'Video p7jnfw5hw9_467623dedf',
 269         },
 270     }, {
 271         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=ec93197892&pu=2c7cd40209&auto_play=1&gpcflag=1&width=640&height=360',
 272         'md5': 'e03d9cc8d9c13191e1caf277e42dbd31',
 273         'info_dict': {
 274             'id': 'p7jnfw5hw9_ec93197892',
 275             'ext': 'mp4',
 276             'title': 'Video p7jnfw5hw9_ec93197892',
 277         },
 278     }, {
 279         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=187060b6fd',
 280         'md5': 'cb988699a776b22d4a41b9d43acfb3ac',
 281         'info_dict': {
 282             'id': 'p7jnfw5hw9_187060b6fd',
 283             'ext': 'mp4',
 284             'title': 'Video p7jnfw5hw9_187060b6fd',
 285         },
 286     }]
 287
 288     @staticmethod
 289     def sign_data(obj):
 290         if obj['cf'] == 'flash':
 291             salt = '2f9d6924b33a165a6d8b5d3d42f4f987'
 292             items = ['cf', 'format', 'ran', 'uu', 'ver', 'vu']
 293         elif obj['cf'] == 'html5':
 294             salt = 'fbeh5player12c43eccf2bec3300344'
 295             items = ['cf', 'ran', 'uu', 'bver', 'vu']
 296         input_data = ''.join([item + obj[item] for item in items]) + salt
 297         obj['sign'] = hashlib.md5(input_data.encode('utf-8')).hexdigest()
 298
 299     def _get_formats(self, cf, uu, vu, media_id):
 300         def get_play_json(cf, timestamp):
 301             data = {
 302                 'cf': cf,
 303                 'ver': '2.2',
 304                 'bver': 'firefox44.0',
 305                 'format': 'json',
 306                 'uu': uu,
 307                 'vu': vu,
 308                 'ran': compat_str(timestamp),
 309             }
 310             self.sign_data(data)
 311             return self._download_json(
 312                 'http://api.letvcloud.com/gpc.php?' + compat_urllib_parse_urlencode(data),
 313                 media_id, 'Downloading playJson data for type %s' % cf)
 314
 315         play_json = get_play_json(cf, time.time())
 316         # The server time may be different from local time
 317         if play_json.get('code') == 10071:
 318             play_json = get_play_json(cf, play_json['timestamp'])
 319
 320         if not play_json.get('data'):
 321             if play_json.get('message'):
 322                 raise ExtractorError('Letv cloud said: %s' % play_json['message'], expected=True)
 323             elif play_json.get('code'):
 324                 raise ExtractorError('Letv cloud returned error %d' % play_json['code'], expected=True)
 325             else:
 326                 raise ExtractorError('Letv cloud returned an unknown error')
 327
 328         def b64decode(s):
 329             return compat_b64decode(s).decode('utf-8')
 330
 331         formats = []
 332         for media in play_json['data']['video_info']['media'].values():
 333             play_url = media['play_url']
 334             url = b64decode(play_url['main_url'])
 335             decoded_url = b64decode(url_basename(url))
 336             formats.append({
 337                 'url': url,
 338                 'ext': determine_ext(decoded_url),
 339                 'format_id': str_or_none(play_url.get('vtype')),
 340                 'format_note': str_or_none(play_url.get('definition')),
 341                 'width': int_or_none(play_url.get('vwidth')),
 342                 'height': int_or_none(play_url.get('vheight')),
 343             })
 344
 345         return formats
 346
 347     def _real_extract(self, url):
 348         uu_mobj = re.search(r'uu=([\w]+)', url)
 349         vu_mobj = re.search(r'vu=([\w]+)', url)
 350
 351         if not uu_mobj or not vu_mobj:
 352             raise ExtractorError('Invalid URL: %s' % url, expected=True)
 353
 354         uu = uu_mobj.group(1)
 355         vu = vu_mobj.group(1)
 356         media_id = uu + '_' + vu
 357
 358         formats = self._get_formats('flash', uu, vu, media_id) + self._get_formats('html5', uu, vu, media_id)
 359
 360         return {
 361             'id': media_id,
 362             'title': 'Video %s' % media_id,
 363             'formats': formats,
 364         }