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 )
  27
  28
  29 class LeIE(InfoExtractor):
  30     IE_DESC = '乐视网'
  31     _VALID_URL = r'https?://(?:www\.le\.com/ptv/vplay|(?:sports\.le|(?:www\.)?lesports)\.com/(?:match|video))/(?P<id>\d+)\.html'
  32     _GEO_COUNTRIES = ['CN']
  33     _URL_TEMPLATE = 'http://www.le.com/ptv/vplay/%s.html'
  34
  35     _TESTS = [{
  36         'url': 'http://www.le.com/ptv/vplay/22005890.html',
  37         'md5': 'edadcfe5406976f42f9f266057ee5e40',
  38         'info_dict': {
  39             'id': '22005890',
  40             'ext': 'mp4',
  41             'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家',
  42             'description': 'md5:a9cb175fd753e2962176b7beca21a47c',
  43         },
  44         'params': {
  45             'hls_prefer_native': True,
  46         },
  47     }, {
  48         'url': 'http://www.le.com/ptv/vplay/1415246.html',
  49         'info_dict': {
  50             'id': '1415246',
  51             'ext': 'mp4',
  52             'title': '美人天下01',
  53             'description': 'md5:28942e650e82ed4fcc8e4de919ee854d',
  54         },
  55         'params': {
  56             'hls_prefer_native': True,
  57         },
  58     }, {
  59         'note': 'This video is available only in Mainland China, thus a proxy is needed',
  60         'url': 'http://www.le.com/ptv/vplay/1118082.html',
  61         'md5': '2424c74948a62e5f31988438979c5ad1',
  62         'info_dict': {
  63             'id': '1118082',
  64             'ext': 'mp4',
  65             'title': '与龙共舞 完整版',
  66             'description': 'md5:7506a5eeb1722bb9d4068f85024e3986',
  67         },
  68         'params': {
  69             'hls_prefer_native': True,
  70         },
  71     }, {
  72         'url': 'http://sports.le.com/video/25737697.html',
  73         'only_matching': True,
  74     }, {
  75         'url': 'http://www.lesports.com/match/1023203003.html',
  76         'only_matching': True,
  77     }, {
  78         'url': 'http://sports.le.com/match/1023203003.html',
  79         'only_matching': True,
  80     }]
  81
  82     # ror() and calc_time_key() are reversed from a embedded swf file in LetvPlayer.swf
  83     def ror(self, param1, param2):
  84         _loc3_ = 0
  85         while _loc3_ < param2:
  86             param1 = urshift(param1, 1) + ((param1 & 1) << 31)
  87             _loc3_ += 1
  88         return param1
  89
  90     def calc_time_key(self, param1):
  91         _loc2_ = 185025305
  92         return self.ror(param1, _loc2_ % 17) ^ _loc2_
  93
  94     # see M3U8Encryption class in KLetvPlayer.swf
  95     @staticmethod
  96     def decrypt_m3u8(encrypted_data):
  97         if encrypted_data[:5].decode('utf-8').lower() != 'vc_01':
  98             return encrypted_data
  99         encrypted_data = encrypted_data[5:]
 100
 101         _loc4_ = bytearray(2 * len(encrypted_data))
 102         for idx, val in enumerate(encrypted_data):
 103             b = compat_ord(val)
 104             _loc4_[2 * idx] = b // 16
 105             _loc4_[2 * idx + 1] = b % 16
 106         idx = len(_loc4_) - 11
 107         _loc4_ = _loc4_[idx:] + _loc4_[:idx]
 108         _loc7_ = bytearray(len(encrypted_data))
 109         for i in range(len(encrypted_data)):
 110             _loc7_[i] = _loc4_[2 * i] * 16 + _loc4_[2 * i + 1]
 111
 112         return bytes(_loc7_)
 113
 114     def _check_errors(self, play_json):
 115         # Check for errors
 116         playstatus = play_json['msgs']['playstatus']
 117         if playstatus['status'] == 0:
 118             flag = playstatus['flag']
 119             if flag == 1:
 120                 self.raise_geo_restricted()
 121             else:
 122                 raise ExtractorError('Generic error. flag = %d' % flag, expected=True)
 123
 124     def _real_extract(self, url):
 125         media_id = self._match_id(url)
 126         page = self._download_webpage(url, media_id)
 127
 128         play_json_flash = self._download_json(
 129             'http://player-pc.le.com/mms/out/video/playJson',
 130             media_id, 'Downloading flash playJson data', query={
 131                 'id': media_id,
 132                 'platid': 1,
 133                 'splatid': 101,
 134                 'format': 1,
 135                 'source': 1000,
 136                 'tkey': self.calc_time_key(int(time.time())),
 137                 'domain': 'www.le.com',
 138                 'region': 'cn',
 139             },
 140             headers=self.geo_verification_headers())
 141         self._check_errors(play_json_flash)
 142
 143         def get_flash_urls(media_url, format_id):
 144             nodes_data = self._download_json(
 145                 media_url, media_id,
 146                 'Download JSON metadata for format %s' % format_id,
 147                 query={
 148                     'm3v': 1,
 149                     'format': 1,
 150                     'expect': 3,
 151                     'tss': 'ios',
 152                 })
 153
 154             req = self._request_webpage(
 155                 nodes_data['nodelist'][0]['location'], media_id,
 156                 note='Downloading m3u8 information for format %s' % format_id)
 157
 158             m3u8_data = self.decrypt_m3u8(req.read())
 159
 160             return {
 161                 'hls': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
 162             }
 163
 164         extracted_formats = []
 165         formats = []
 166         playurl = play_json_flash['msgs']['playurl']
 167         play_domain = playurl['domain'][0]
 168
 169         for format_id, format_data in playurl.get('dispatch', []).items():
 170             if format_id in extracted_formats:
 171                 continue
 172             extracted_formats.append(format_id)
 173
 174             media_url = play_domain + format_data[0]
 175             for protocol, format_url in get_flash_urls(media_url, format_id).items():
 176                 f = {
 177                     'url': format_url,
 178                     'ext': determine_ext(format_data[1]),
 179                     'format_id': '%s-%s' % (protocol, format_id),
 180                     'protocol': 'm3u8_native' if protocol == 'hls' else 'http',
 181                     'quality': int_or_none(format_id),
 182                 }
 183
 184                 if format_id[-1:] == 'p':
 185                     f['height'] = int_or_none(format_id[:-1])
 186
 187                 formats.append(f)
 188         self._sort_formats(formats, ('height', 'quality', 'format_id'))
 189
 190         publish_time = parse_iso8601(self._html_search_regex(
 191             r'发布时间&nbsp;([^<>]+) ', page, 'publish time', default=None),
 192             delimiter=' ', timezone=datetime.timedelta(hours=8))
 193         description = self._html_search_meta('description', page, fatal=False)
 194
 195         return {
 196             'id': media_id,
 197             'formats': formats,
 198             'title': playurl['title'],
 199             'thumbnail': playurl['pic'],
 200             'description': description,
 201             'timestamp': publish_time,
 202         }
 203
 204
 205 class LePlaylistIE(InfoExtractor):
 206     _VALID_URL = r'https?://[a-z]+\.le\.com/(?!video)[a-z]+/(?P<id>[a-z0-9_]+)'
 207
 208     _TESTS = [{
 209         'url': 'http://www.le.com/tv/46177.html',
 210         'info_dict': {
 211             'id': '46177',
 212             'title': '美人天下',
 213             'description': 'md5:395666ff41b44080396e59570dbac01c'
 214         },
 215         'playlist_count': 35
 216     }, {
 217         'url': 'http://tv.le.com/izt/wuzetian/index.html',
 218         'info_dict': {
 219             'id': 'wuzetian',
 220             'title': '武媚娘传奇',
 221             'description': 'md5:e12499475ab3d50219e5bba00b3cb248'
 222         },
 223         # This playlist contains some extra videos other than the drama itself
 224         'playlist_mincount': 96
 225     }, {
 226         'url': 'http://tv.le.com/pzt/lswjzzjc/index.shtml',
 227         # This series is moved to http://www.le.com/tv/10005297.html
 228         'only_matching': True,
 229     }, {
 230         'url': 'http://www.le.com/comic/92063.html',
 231         'only_matching': True,
 232     }, {
 233         'url': 'http://list.le.com/listn/c1009_sc532002_d2_p1_o1.html',
 234         'only_matching': True,
 235     }]
 236
 237     @classmethod
 238     def suitable(cls, url):
 239         return False if LeIE.suitable(url) else super(LePlaylistIE, cls).suitable(url)
 240
 241     def _real_extract(self, url):
 242         playlist_id = self._match_id(url)
 243         page = self._download_webpage(url, playlist_id)
 244
 245         # Currently old domain names are still used in playlists
 246         media_ids = orderedSet(re.findall(
 247             r'<a[^>]+href="http://www\.letv\.com/ptv/vplay/(\d+)\.html', page))
 248         entries = [self.url_result(LeIE._URL_TEMPLATE % media_id, ie='Le')
 249                    for media_id in media_ids]
 250
 251         title = self._html_search_meta('keywords', page,
 252                                        fatal=False).split('，')[0]
 253         description = self._html_search_meta('description', page, fatal=False)
 254
 255         return self.playlist_result(entries, playlist_id, playlist_title=title,
 256                                     playlist_description=description)
 257
 258
 259 class LetvCloudIE(InfoExtractor):
 260     # Most of *.letv.com is changed to *.le.com on 2016/01/02
 261     # but yuntv.letv.com is kept, so also keep the extractor name
 262     IE_DESC = '乐视云'
 263     _VALID_URL = r'https?://yuntv\.letv\.com/bcloud.html\?.+'
 264
 265     _TESTS = [{
 266         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=467623dedf',
 267         'md5': '26450599afd64c513bc77030ad15db44',
 268         'info_dict': {
 269             'id': 'p7jnfw5hw9_467623dedf',
 270             'ext': 'mp4',
 271             'title': 'Video p7jnfw5hw9_467623dedf',
 272         },
 273     }, {
 274         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=ec93197892&pu=2c7cd40209&auto_play=1&gpcflag=1&width=640&height=360',
 275         'md5': 'e03d9cc8d9c13191e1caf277e42dbd31',
 276         'info_dict': {
 277             'id': 'p7jnfw5hw9_ec93197892',
 278             'ext': 'mp4',
 279             'title': 'Video p7jnfw5hw9_ec93197892',
 280         },
 281     }, {
 282         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=187060b6fd',
 283         'md5': 'cb988699a776b22d4a41b9d43acfb3ac',
 284         'info_dict': {
 285             'id': 'p7jnfw5hw9_187060b6fd',
 286             'ext': 'mp4',
 287             'title': 'Video p7jnfw5hw9_187060b6fd',
 288         },
 289     }]
 290
 291     @staticmethod
 292     def sign_data(obj):
 293         if obj['cf'] == 'flash':
 294             salt = '2f9d6924b33a165a6d8b5d3d42f4f987'
 295             items = ['cf', 'format', 'ran', 'uu', 'ver', 'vu']
 296         elif obj['cf'] == 'html5':
 297             salt = 'fbeh5player12c43eccf2bec3300344'
 298             items = ['cf', 'ran', 'uu', 'bver', 'vu']
 299         input_data = ''.join([item + obj[item] for item in items]) + salt
 300         obj['sign'] = hashlib.md5(input_data.encode('utf-8')).hexdigest()
 301
 302     def _get_formats(self, cf, uu, vu, media_id):
 303         def get_play_json(cf, timestamp):
 304             data = {
 305                 'cf': cf,
 306                 'ver': '2.2',
 307                 'bver': 'firefox44.0',
 308                 'format': 'json',
 309                 'uu': uu,
 310                 'vu': vu,
 311                 'ran': compat_str(timestamp),
 312             }
 313             self.sign_data(data)
 314             return self._download_json(
 315                 'http://api.letvcloud.com/gpc.php?' + compat_urllib_parse_urlencode(data),
 316                 media_id, 'Downloading playJson data for type %s' % cf)
 317
 318         play_json = get_play_json(cf, time.time())
 319         # The server time may be different from local time
 320         if play_json.get('code') == 10071:
 321             play_json = get_play_json(cf, play_json['timestamp'])
 322
 323         if not play_json.get('data'):
 324             if play_json.get('message'):
 325                 raise ExtractorError('Letv cloud said: %s' % play_json['message'], expected=True)
 326             elif play_json.get('code'):
 327                 raise ExtractorError('Letv cloud returned error %d' % play_json['code'], expected=True)
 328             else:
 329                 raise ExtractorError('Letv cloud returned an unknwon error')
 330
 331         def b64decode(s):
 332             return base64.b64decode(s.encode('utf-8')).decode('utf-8')
 333
 334         formats = []
 335         for media in play_json['data']['video_info']['media'].values():
 336             play_url = media['play_url']
 337             url = b64decode(play_url['main_url'])
 338             decoded_url = b64decode(url_basename(url))
 339             formats.append({
 340                 'url': url,
 341                 'ext': determine_ext(decoded_url),
 342                 'format_id': str_or_none(play_url.get('vtype')),
 343                 'format_note': str_or_none(play_url.get('definition')),
 344                 'width': int_or_none(play_url.get('vwidth')),
 345                 'height': int_or_none(play_url.get('vheight')),
 346             })
 347
 348         return formats
 349
 350     def _real_extract(self, url):
 351         uu_mobj = re.search(r'uu=([\w]+)', url)
 352         vu_mobj = re.search(r'vu=([\w]+)', url)
 353
 354         if not uu_mobj or not vu_mobj:
 355             raise ExtractorError('Invalid URL: %s' % url, expected=True)
 356
 357         uu = uu_mobj.group(1)
 358         vu = vu_mobj.group(1)
 359         media_id = uu + '_' + vu
 360
 361         formats = self._get_formats('flash', uu, vu, media_id) + self._get_formats('html5', uu, vu, media_id)
 362         self._sort_formats(formats)
 363
 364         return {
 365             'id': media_id,
 366             'title': 'Video %s' % media_id,
 367             'formats': formats,
 368         }