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\.com/video)/(?P<id>\d+)\.html'
  32
  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:f88573d9d7225ada1359eaf0dbf8bcda',
  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         'skip': 'Only available in China',
  72     }, {
  73         'url': 'http://sports.le.com/video/25737697.html',
  74         'only_matching': True,
  75     }]
  76
  77     # ror() and calc_time_key() are reversed from a embedded swf file in KLetvPlayer.swf
  78     def ror(self, param1, param2):
  79         _loc3_ = 0
  80         while _loc3_ < param2:
  81             param1 = urshift(param1, 1) + ((param1 & 1) << 31)
  82             _loc3_ += 1
  83         return param1
  84
  85     def calc_time_key(self, param1):
  86         _loc2_ = 773625421
  87         _loc3_ = self.ror(param1, _loc2_ % 13)
  88         _loc3_ = _loc3_ ^ _loc2_
  89         _loc3_ = self.ror(_loc3_, _loc2_ % 17)
  90         return _loc3_
  91
  92     # see M3U8Encryption class in KLetvPlayer.swf
  93     @staticmethod
  94     def decrypt_m3u8(encrypted_data):
  95         if encrypted_data[:5].decode('utf-8').lower() != 'vc_01':
  96             return encrypted_data
  97         encrypted_data = encrypted_data[5:]
  98
  99         _loc4_ = bytearray(2 * len(encrypted_data))
 100         for idx, val in enumerate(encrypted_data):
 101             b = compat_ord(val)
 102             _loc4_[2 * idx] = b // 16
 103             _loc4_[2 * idx + 1] = b % 16
 104         idx = len(_loc4_) - 11
 105         _loc4_ = _loc4_[idx:] + _loc4_[:idx]
 106         _loc7_ = bytearray(len(encrypted_data))
 107         for i in range(len(encrypted_data)):
 108             _loc7_[i] = _loc4_[2 * i] * 16 + _loc4_[2 * i + 1]
 109
 110         return bytes(_loc7_)
 111
 112     def _real_extract(self, url):
 113         media_id = self._match_id(url)
 114         page = self._download_webpage(url, media_id)
 115         params = {
 116             'id': media_id,
 117             'platid': 1,
 118             'splatid': 101,
 119             'format': 1,
 120             'tkey': self.calc_time_key(int(time.time())),
 121             'domain': 'www.le.com'
 122         }
 123
 124         play_json = self._download_json(
 125             'http://api.le.com/mms/out/video/playJson',
 126             media_id, 'Downloading playJson data', query=params,
 127             headers=self.geo_verification_headers())
 128
 129         # Check for errors
 130         playstatus = play_json['playstatus']
 131         if playstatus['status'] == 0:
 132             flag = playstatus['flag']
 133             if flag == 1:
 134                 msg = 'Country %s auth error' % playstatus['country']
 135             else:
 136                 msg = 'Generic error. flag = %d' % flag
 137             raise ExtractorError(msg, expected=True)
 138
 139         playurl = play_json['playurl']
 140
 141         formats = ['350', '1000', '1300', '720p', '1080p']
 142         dispatch = playurl['dispatch']
 143
 144         urls = []
 145         for format_id in formats:
 146             if format_id in dispatch:
 147                 media_url = playurl['domain'][0] + dispatch[format_id][0]
 148                 media_url += '&' + compat_urllib_parse_urlencode({
 149                     'm3v': 1,
 150                     'format': 1,
 151                     'expect': 3,
 152                     'rateid': format_id,
 153                 })
 154
 155                 nodes_data = self._download_json(
 156                     media_url, media_id,
 157                     'Download JSON metadata for format %s' % format_id)
 158
 159                 req = self._request_webpage(
 160                     nodes_data['nodelist'][0]['location'], media_id,
 161                     note='Downloading m3u8 information for format %s' % format_id)
 162
 163                 m3u8_data = self.decrypt_m3u8(req.read())
 164
 165                 url_info_dict = {
 166                     'url': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
 167                     'ext': determine_ext(dispatch[format_id][1]),
 168                     'format_id': format_id,
 169                     'protocol': 'm3u8',
 170                 }
 171
 172                 if format_id[-1:] == 'p':
 173                     url_info_dict['height'] = int_or_none(format_id[:-1])
 174
 175                 urls.append(url_info_dict)
 176
 177         publish_time = parse_iso8601(self._html_search_regex(
 178             r'发布时间&nbsp;([^<>]+) ', page, 'publish time', default=None),
 179             delimiter=' ', timezone=datetime.timedelta(hours=8))
 180         description = self._html_search_meta('description', page, fatal=False)
 181
 182         return {
 183             'id': media_id,
 184             'formats': urls,
 185             'title': playurl['title'],
 186             'thumbnail': playurl['pic'],
 187             'description': description,
 188             'timestamp': publish_time,
 189         }
 190
 191
 192 class LePlaylistIE(InfoExtractor):
 193     _VALID_URL = r'https?://[a-z]+\.le\.com/(?!video)[a-z]+/(?P<id>[a-z0-9_]+)'
 194
 195     _TESTS = [{
 196         'url': 'http://www.le.com/tv/46177.html',
 197         'info_dict': {
 198             'id': '46177',
 199             'title': '美人天下',
 200             'description': 'md5:395666ff41b44080396e59570dbac01c'
 201         },
 202         'playlist_count': 35
 203     }, {
 204         'url': 'http://tv.le.com/izt/wuzetian/index.html',
 205         'info_dict': {
 206             'id': 'wuzetian',
 207             'title': '武媚娘传奇',
 208             'description': 'md5:e12499475ab3d50219e5bba00b3cb248'
 209         },
 210         # This playlist contains some extra videos other than the drama itself
 211         'playlist_mincount': 96
 212     }, {
 213         'url': 'http://tv.le.com/pzt/lswjzzjc/index.shtml',
 214         # This series is moved to http://www.le.com/tv/10005297.html
 215         'only_matching': True,
 216     }, {
 217         'url': 'http://www.le.com/comic/92063.html',
 218         'only_matching': True,
 219     }, {
 220         'url': 'http://list.le.com/listn/c1009_sc532002_d2_p1_o1.html',
 221         'only_matching': True,
 222     }]
 223
 224     @classmethod
 225     def suitable(cls, url):
 226         return False if LeIE.suitable(url) else super(LePlaylistIE, cls).suitable(url)
 227
 228     def _real_extract(self, url):
 229         playlist_id = self._match_id(url)
 230         page = self._download_webpage(url, playlist_id)
 231
 232         # Currently old domain names are still used in playlists
 233         media_ids = orderedSet(re.findall(
 234             r'<a[^>]+href="http://www\.letv\.com/ptv/vplay/(\d+)\.html', page))
 235         entries = [self.url_result(LeIE._URL_TEMPLATE % media_id, ie='Le')
 236                    for media_id in media_ids]
 237
 238         title = self._html_search_meta('keywords', page,
 239                                        fatal=False).split('，')[0]
 240         description = self._html_search_meta('description', page, fatal=False)
 241
 242         return self.playlist_result(entries, playlist_id, playlist_title=title,
 243                                     playlist_description=description)
 244
 245
 246 class LetvCloudIE(InfoExtractor):
 247     # Most of *.letv.com is changed to *.le.com on 2016/01/02
 248     # but yuntv.letv.com is kept, so also keep the extractor name
 249     IE_DESC = '乐视云'
 250     _VALID_URL = r'https?://yuntv\.letv\.com/bcloud.html\?.+'
 251
 252     _TESTS = [{
 253         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=467623dedf',
 254         'md5': '26450599afd64c513bc77030ad15db44',
 255         'info_dict': {
 256             'id': 'p7jnfw5hw9_467623dedf',
 257             'ext': 'mp4',
 258             'title': 'Video p7jnfw5hw9_467623dedf',
 259         },
 260     }, {
 261         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=ec93197892&pu=2c7cd40209&auto_play=1&gpcflag=1&width=640&height=360',
 262         'md5': 'e03d9cc8d9c13191e1caf277e42dbd31',
 263         'info_dict': {
 264             'id': 'p7jnfw5hw9_ec93197892',
 265             'ext': 'mp4',
 266             'title': 'Video p7jnfw5hw9_ec93197892',
 267         },
 268     }, {
 269         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=187060b6fd',
 270         'md5': 'cb988699a776b22d4a41b9d43acfb3ac',
 271         'info_dict': {
 272             'id': 'p7jnfw5hw9_187060b6fd',
 273             'ext': 'mp4',
 274             'title': 'Video p7jnfw5hw9_187060b6fd',
 275         },
 276     }]
 277
 278     @staticmethod
 279     def sign_data(obj):
 280         if obj['cf'] == 'flash':
 281             salt = '2f9d6924b33a165a6d8b5d3d42f4f987'
 282             items = ['cf', 'format', 'ran', 'uu', 'ver', 'vu']
 283         elif obj['cf'] == 'html5':
 284             salt = 'fbeh5player12c43eccf2bec3300344'
 285             items = ['cf', 'ran', 'uu', 'bver', 'vu']
 286         input_data = ''.join([item + obj[item] for item in items]) + salt
 287         obj['sign'] = hashlib.md5(input_data.encode('utf-8')).hexdigest()
 288
 289     def _get_formats(self, cf, uu, vu, media_id):
 290         def get_play_json(cf, timestamp):
 291             data = {
 292                 'cf': cf,
 293                 'ver': '2.2',
 294                 'bver': 'firefox44.0',
 295                 'format': 'json',
 296                 'uu': uu,
 297                 'vu': vu,
 298                 'ran': compat_str(timestamp),
 299             }
 300             self.sign_data(data)
 301             return self._download_json(
 302                 'http://api.letvcloud.com/gpc.php?' + compat_urllib_parse_urlencode(data),
 303                 media_id, 'Downloading playJson data for type %s' % cf)
 304
 305         play_json = get_play_json(cf, time.time())
 306         # The server time may be different from local time
 307         if play_json.get('code') == 10071:
 308             play_json = get_play_json(cf, play_json['timestamp'])
 309
 310         if not play_json.get('data'):
 311             if play_json.get('message'):
 312                 raise ExtractorError('Letv cloud said: %s' % play_json['message'], expected=True)
 313             elif play_json.get('code'):
 314                 raise ExtractorError('Letv cloud returned error %d' % play_json['code'], expected=True)
 315             else:
 316                 raise ExtractorError('Letv cloud returned an unknwon error')
 317
 318         def b64decode(s):
 319             return base64.b64decode(s.encode('utf-8')).decode('utf-8')
 320
 321         formats = []
 322         for media in play_json['data']['video_info']['media'].values():
 323             play_url = media['play_url']
 324             url = b64decode(play_url['main_url'])
 325             decoded_url = b64decode(url_basename(url))
 326             formats.append({
 327                 'url': url,
 328                 'ext': determine_ext(decoded_url),
 329                 'format_id': str_or_none(play_url.get('vtype')),
 330                 'format_note': str_or_none(play_url.get('definition')),
 331                 'width': int_or_none(play_url.get('vwidth')),
 332                 'height': int_or_none(play_url.get('vheight')),
 333             })
 334
 335         return formats
 336
 337     def _real_extract(self, url):
 338         uu_mobj = re.search('uu=([\w]+)', url)
 339         vu_mobj = re.search('vu=([\w]+)', url)
 340
 341         if not uu_mobj or not vu_mobj:
 342             raise ExtractorError('Invalid URL: %s' % url, expected=True)
 343
 344         uu = uu_mobj.group(1)
 345         vu = vu_mobj.group(1)
 346         media_id = uu + '_' + vu
 347
 348         formats = self._get_formats('flash', uu, vu, media_id) + self._get_formats('html5', uu, vu, media_id)
 349         self._sort_formats(formats)
 350
 351         return {
 352             'id': media_id,
 353             'title': 'Video %s' % media_id,
 354             'formats': formats,
 355         }