Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/letv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import datetime
   5 import re
   6 import time
   7 import base64
   8 import hashlib
   9
  10 from .common import InfoExtractor
  11 from ..compat import (
  12     compat_urllib_parse,
  13     compat_ord,
  14     compat_str,
  15 )
  16 from ..utils import (
  17     determine_ext,
  18     ExtractorError,
  19     parse_iso8601,
  20     sanitized_Request,
  21     int_or_none,
  22     str_or_none,
  23     encode_data_uri,
  24     url_basename,
  25 )
  26
  27
  28 class LetvIE(InfoExtractor):
  29     IE_DESC = '乐视网'
  30     _VALID_URL = r'http://www\.letv\.com/ptv/vplay/(?P<id>\d+).html'
  31
  32     _TESTS = [{
  33         'url': 'http://www.letv.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.letv.com/ptv/vplay/1415246.html',
  46         'info_dict': {
  47             'id': '1415246',
  48             'ext': 'mp4',
  49             'title': '美人天下01',
  50             'description': 'md5:f88573d9d7225ada1359eaf0dbf8bcda',
  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.letv.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         'skip': 'Only available in China',
  69     }]
  70
  71     @staticmethod
  72     def urshift(val, n):
  73         return val >> n if val >= 0 else (val + 0x100000000) >> n
  74
  75     # ror() and calc_time_key() are reversed from a embedded swf file in KLetvPlayer.swf
  76     def ror(self, param1, param2):
  77         _loc3_ = 0
  78         while _loc3_ < param2:
  79             param1 = self.urshift(param1, 1) + ((param1 & 1) << 31)
  80             _loc3_ += 1
  81         return param1
  82
  83     def calc_time_key(self, param1):
  84         _loc2_ = 773625421
  85         _loc3_ = self.ror(param1, _loc2_ % 13)
  86         _loc3_ = _loc3_ ^ _loc2_
  87         _loc3_ = self.ror(_loc3_, _loc2_ % 17)
  88         return _loc3_
  89
  90     # see M3U8Encryption class in KLetvPlayer.swf
  91     @staticmethod
  92     def decrypt_m3u8(encrypted_data):
  93         if encrypted_data[:5].decode('utf-8').lower() != 'vc_01':
  94             return encrypted_data
  95         encrypted_data = encrypted_data[5:]
  96
  97         _loc4_ = bytearray()
  98         while encrypted_data:
  99             b = compat_ord(encrypted_data[0])
 100             _loc4_.extend([b // 16, b & 0x0f])
 101             encrypted_data = encrypted_data[1:]
 102         idx = len(_loc4_) - 11
 103         _loc4_ = _loc4_[idx:] + _loc4_[:idx]
 104         _loc7_ = bytearray()
 105         while _loc4_:
 106             _loc7_.append(_loc4_[0] * 16 + _loc4_[1])
 107             _loc4_ = _loc4_[2:]
 108
 109         return bytes(_loc7_)
 110
 111     def _real_extract(self, url):
 112         media_id = self._match_id(url)
 113         page = self._download_webpage(url, media_id)
 114         params = {
 115             'id': media_id,
 116             'platid': 1,
 117             'splatid': 101,
 118             'format': 1,
 119             'tkey': self.calc_time_key(int(time.time())),
 120             'domain': 'www.letv.com'
 121         }
 122         play_json_req = sanitized_Request(
 123             'http://api.letv.com/mms/out/video/playJson?' + compat_urllib_parse.urlencode(params)
 124         )
 125         cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
 126         if cn_verification_proxy:
 127             play_json_req.add_header('Ytdl-request-proxy', cn_verification_proxy)
 128
 129         play_json = self._download_json(
 130             play_json_req,
 131             media_id, 'Downloading playJson data')
 132
 133         # Check for errors
 134         playstatus = play_json['playstatus']
 135         if playstatus['status'] == 0:
 136             flag = playstatus['flag']
 137             if flag == 1:
 138                 msg = 'Country %s auth error' % playstatus['country']
 139             else:
 140                 msg = 'Generic error. flag = %d' % flag
 141             raise ExtractorError(msg, expected=True)
 142
 143         playurl = play_json['playurl']
 144
 145         formats = ['350', '1000', '1300', '720p', '1080p']
 146         dispatch = playurl['dispatch']
 147
 148         urls = []
 149         for format_id in formats:
 150             if format_id in dispatch:
 151                 media_url = playurl['domain'][0] + dispatch[format_id][0]
 152                 media_url += '&' + compat_urllib_parse.urlencode({
 153                     'm3v': 1,
 154                     'format': 1,
 155                     'expect': 3,
 156                     'rateid': format_id,
 157                 })
 158
 159                 nodes_data = self._download_json(
 160                     media_url, media_id,
 161                     'Download JSON metadata for format %s' % format_id)
 162
 163                 req = self._request_webpage(
 164                     nodes_data['nodelist'][0]['location'], media_id,
 165                     note='Downloading m3u8 information for format %s' % format_id)
 166
 167                 m3u8_data = self.decrypt_m3u8(req.read())
 168
 169                 url_info_dict = {
 170                     'url': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
 171                     'ext': determine_ext(dispatch[format_id][1]),
 172                     'format_id': format_id,
 173                     'protocol': 'm3u8',
 174                 }
 175
 176                 if format_id[-1:] == 'p':
 177                     url_info_dict['height'] = int_or_none(format_id[:-1])
 178
 179                 urls.append(url_info_dict)
 180
 181         publish_time = parse_iso8601(self._html_search_regex(
 182             r'发布时间&nbsp;([^<>]+) ', page, 'publish time', default=None),
 183             delimiter=' ', timezone=datetime.timedelta(hours=8))
 184         description = self._html_search_meta('description', page, fatal=False)
 185
 186         return {
 187             'id': media_id,
 188             'formats': urls,
 189             'title': playurl['title'],
 190             'thumbnail': playurl['pic'],
 191             'description': description,
 192             'timestamp': publish_time,
 193         }
 194
 195
 196 class LetvTvIE(InfoExtractor):
 197     _VALID_URL = r'http://www.letv.com/tv/(?P<id>\d+).html'
 198     _TESTS = [{
 199         'url': 'http://www.letv.com/tv/46177.html',
 200         'info_dict': {
 201             'id': '46177',
 202             'title': '美人天下',
 203             'description': 'md5:395666ff41b44080396e59570dbac01c'
 204         },
 205         'playlist_count': 35
 206     }]
 207
 208     def _real_extract(self, url):
 209         playlist_id = self._match_id(url)
 210         page = self._download_webpage(url, playlist_id)
 211
 212         media_urls = list(set(re.findall(
 213             r'http://www.letv.com/ptv/vplay/\d+.html', page)))
 214         entries = [self.url_result(media_url, ie='Letv')
 215                    for media_url in media_urls]
 216
 217         title = self._html_search_meta('keywords', page,
 218                                        fatal=False).split('，')[0]
 219         description = self._html_search_meta('description', page, fatal=False)
 220
 221         return self.playlist_result(entries, playlist_id, playlist_title=title,
 222                                     playlist_description=description)
 223
 224
 225 class LetvPlaylistIE(LetvTvIE):
 226     _VALID_URL = r'http://tv.letv.com/[a-z]+/(?P<id>[a-z]+)/index.s?html'
 227     _TESTS = [{
 228         'url': 'http://tv.letv.com/izt/wuzetian/index.html',
 229         'info_dict': {
 230             'id': 'wuzetian',
 231             'title': '武媚娘传奇',
 232             'description': 'md5:e12499475ab3d50219e5bba00b3cb248'
 233         },
 234         # This playlist contains some extra videos other than the drama itself
 235         'playlist_mincount': 96
 236     }, {
 237         'url': 'http://tv.letv.com/pzt/lswjzzjc/index.shtml',
 238         'info_dict': {
 239             'id': 'lswjzzjc',
 240             # The title should be "劲舞青春", but I can't find a simple way to
 241             # determine the playlist title
 242             'title': '乐视午间自制剧场',
 243             'description': 'md5:b1eef244f45589a7b5b1af9ff25a4489'
 244         },
 245         'playlist_mincount': 7
 246     }]
 247
 248
 249 class LetvCloudIE(InfoExtractor):
 250     IE_DESC = '乐视云'
 251     _VALID_URL = r'https?://yuntv\.letv\.com/bcloud.html\?.+'
 252
 253     _TESTS = [{
 254         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=467623dedf',
 255         'md5': '26450599afd64c513bc77030ad15db44',
 256         'info_dict': {
 257             'id': 'p7jnfw5hw9_467623dedf',
 258             'ext': 'mp4',
 259             'title': 'Video p7jnfw5hw9_467623dedf',
 260         },
 261     }, {
 262         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=ec93197892&pu=2c7cd40209&auto_play=1&gpcflag=1&width=640&height=360',
 263         'md5': 'e03d9cc8d9c13191e1caf277e42dbd31',
 264         'info_dict': {
 265             'id': 'p7jnfw5hw9_ec93197892',
 266             'ext': 'mp4',
 267             'title': 'Video p7jnfw5hw9_ec93197892',
 268         },
 269     }, {
 270         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=187060b6fd',
 271         'md5': 'cb988699a776b22d4a41b9d43acfb3ac',
 272         'info_dict': {
 273             'id': 'p7jnfw5hw9_187060b6fd',
 274             'ext': 'mp4',
 275             'title': 'Video p7jnfw5hw9_187060b6fd',
 276         },
 277     }]
 278
 279     @staticmethod
 280     def sign_data(obj):
 281         if obj['cf'] == 'flash':
 282             salt = '2f9d6924b33a165a6d8b5d3d42f4f987'
 283             items = ['cf', 'format', 'ran', 'uu', 'ver', 'vu']
 284         elif obj['cf'] == 'html5':
 285             salt = 'fbeh5player12c43eccf2bec3300344'
 286             items = ['cf', 'ran', 'uu', 'bver', 'vu']
 287         input_data = ''.join([item + obj[item] for item in items]) + salt
 288         obj['sign'] = hashlib.md5(input_data.encode('utf-8')).hexdigest()
 289
 290     def _get_formats(self, cf, uu, vu, media_id):
 291         def get_play_json(cf, timestamp):
 292             data = {
 293                 'cf': cf,
 294                 'ver': '2.2',
 295                 'bver': 'firefox44.0',
 296                 'format': 'json',
 297                 'uu': uu,
 298                 'vu': vu,
 299                 'ran': compat_str(timestamp),
 300             }
 301             self.sign_data(data)
 302             return self._download_json(
 303                 'http://api.letvcloud.com/gpc.php?' + compat_urllib_parse.urlencode(data),
 304                 media_id, 'Downloading playJson data for type %s' % cf)
 305
 306         play_json = get_play_json(cf, time.time())
 307         # The server time may be different from local time
 308         if play_json.get('code') == 10071:
 309             play_json = get_play_json(cf, play_json['timestamp'])
 310
 311         if not play_json.get('data'):
 312             if play_json.get('message'):
 313                 raise ExtractorError('Letv cloud said: %s' % play_json['message'], expected=True)
 314             elif play_json.get('code'):
 315                 raise ExtractorError('Letv cloud returned error %d' % play_json['code'], expected=True)
 316             else:
 317                 raise ExtractorError('Letv cloud returned an unknwon error')
 318
 319         def b64decode(s):
 320             return base64.b64decode(s.encode('utf-8')).decode('utf-8')
 321
 322         formats = []
 323         for media in play_json['data']['video_info']['media'].values():
 324             play_url = media['play_url']
 325             url = b64decode(play_url['main_url'])
 326             decoded_url = b64decode(url_basename(url))
 327             formats.append({
 328                 'url': url,
 329                 'ext': determine_ext(decoded_url),
 330                 'format_id': int_or_none(play_url.get('vtype')),
 331                 'format_note': str_or_none(play_url.get('definition')),
 332                 'width': int_or_none(play_url.get('vwidth')),
 333                 'height': int_or_none(play_url.get('vheight')),
 334             })
 335
 336         return formats
 337
 338     def _real_extract(self, url):
 339         uu_mobj = re.search('uu=([\w]+)', url)
 340         vu_mobj = re.search('vu=([\w]+)', url)
 341
 342         if not uu_mobj or not vu_mobj:
 343             raise ExtractorError('Invalid URL: %s' % url, expected=True)
 344
 345         uu = uu_mobj.group(1)
 346         vu = vu_mobj.group(1)
 347         media_id = uu + '_' + vu
 348
 349         formats = self._get_formats('flash', uu, vu, media_id) + self._get_formats('html5', uu, vu, media_id)
 350         self._sort_formats(formats)
 351
 352         return {
 353             'id': media_id,
 354             'title': 'Video %s' % media_id,
 355             'formats': formats,
 356         }