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
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_urllib_parse,
  11     compat_urllib_request,
  12     compat_urlparse,
  13 )
  14 from ..utils import (
  15     determine_ext,
  16     ExtractorError,
  17     parse_iso8601,
  18 )
  19
  20
  21 class LetvIE(InfoExtractor):
  22     _VALID_URL = r'http://www\.letv\.com/ptv/vplay/(?P<id>\d+).html'
  23
  24     _TESTS = [{
  25         'url': 'http://www.letv.com/ptv/vplay/22005890.html',
  26         'md5': 'cab23bd68d5a8db9be31c9a222c1e8df',
  27         'info_dict': {
  28             'id': '22005890',
  29             'ext': 'mp4',
  30             'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家',
  31             'timestamp': 1424747397,
  32             'upload_date': '20150224',
  33             'description': 'md5:a9cb175fd753e2962176b7beca21a47c',
  34         }
  35     }, {
  36         'url': 'http://www.letv.com/ptv/vplay/1415246.html',
  37         'info_dict': {
  38             'id': '1415246',
  39             'ext': 'mp4',
  40             'title': '美人天下01',
  41             'description': 'md5:f88573d9d7225ada1359eaf0dbf8bcda',
  42         },
  43     }, {
  44         'note': 'This video is available only in Mainland China, thus a proxy is needed',
  45         'url': 'http://www.letv.com/ptv/vplay/1118082.html',
  46         'md5': 'f80936fbe20fb2f58648e81386ff7927',
  47         'info_dict': {
  48             'id': '1118082',
  49             'ext': 'mp4',
  50             'title': '与龙共舞 完整版',
  51             'description': 'md5:7506a5eeb1722bb9d4068f85024e3986',
  52         },
  53         'skip': 'Only available in China',
  54     }]
  55
  56     @staticmethod
  57     def urshift(val, n):
  58         return val >> n if val >= 0 else (val + 0x100000000) >> n
  59
  60     # ror() and calc_time_key() are reversed from a embedded swf file in KLetvPlayer.swf
  61     def ror(self, param1, param2):
  62         _loc3_ = 0
  63         while _loc3_ < param2:
  64             param1 = self.urshift(param1, 1) + ((param1 & 1) << 31)
  65             _loc3_ += 1
  66         return param1
  67
  68     def calc_time_key(self, param1):
  69         _loc2_ = 773625421
  70         _loc3_ = self.ror(param1, _loc2_ % 13)
  71         _loc3_ = _loc3_ ^ _loc2_
  72         _loc3_ = self.ror(_loc3_, _loc2_ % 17)
  73         return _loc3_
  74
  75     def _real_extract(self, url):
  76         media_id = self._match_id(url)
  77         page = self._download_webpage(url, media_id)
  78         params = {
  79             'id': media_id,
  80             'platid': 1,
  81             'splatid': 101,
  82             'format': 1,
  83             'tkey': self.calc_time_key(int(time.time())),
  84             'domain': 'www.letv.com'
  85         }
  86         play_json_req = compat_urllib_request.Request(
  87             'http://api.letv.com/mms/out/video/playJson?' + compat_urllib_parse.urlencode(params)
  88         )
  89         cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
  90         if cn_verification_proxy:
  91             play_json_req.add_header('Ytdl-request-proxy', cn_verification_proxy)
  92
  93         play_json = self._download_json(
  94             play_json_req,
  95             media_id, 'Downloading playJson data')
  96
  97         # Check for errors
  98         playstatus = play_json['playstatus']
  99         if playstatus['status'] == 0:
 100             flag = playstatus['flag']
 101             if flag == 1:
 102                 msg = 'Country %s auth error' % playstatus['country']
 103             else:
 104                 msg = 'Generic error. flag = %d' % flag
 105             raise ExtractorError(msg, expected=True)
 106
 107         playurl = play_json['playurl']
 108
 109         formats = ['350', '1000', '1300', '720p', '1080p']
 110         dispatch = playurl['dispatch']
 111
 112         urls = []
 113         for format_id in formats:
 114             if format_id in dispatch:
 115                 media_url = playurl['domain'][0] + dispatch[format_id][0]
 116
 117                 # Mimic what flvxz.com do
 118                 url_parts = list(compat_urlparse.urlparse(media_url))
 119                 qs = dict(compat_urlparse.parse_qs(url_parts[4]))
 120                 qs.update({
 121                     'platid': '14',
 122                     'splatid': '1401',
 123                     'tss': 'no',
 124                     'retry': 1
 125                 })
 126                 url_parts[4] = compat_urllib_parse.urlencode(qs)
 127                 media_url = compat_urlparse.urlunparse(url_parts)
 128
 129                 url_info_dict = {
 130                     'url': media_url,
 131                     'ext': determine_ext(dispatch[format_id][1]),
 132                     'format_id': format_id,
 133                 }
 134
 135                 if format_id[-1:] == 'p':
 136                     url_info_dict['height'] = format_id[:-1]
 137
 138                 urls.append(url_info_dict)
 139
 140         publish_time = parse_iso8601(self._html_search_regex(
 141             r'发布时间&nbsp;([^<>]+) ', page, 'publish time', default=None),
 142             delimiter=' ', timezone=datetime.timedelta(hours=8))
 143         description = self._html_search_meta('description', page, fatal=False)
 144
 145         return {
 146             'id': media_id,
 147             'formats': urls,
 148             'title': playurl['title'],
 149             'thumbnail': playurl['pic'],
 150             'description': description,
 151             'timestamp': publish_time,
 152         }
 153
 154
 155 class LetvTvIE(InfoExtractor):
 156     _VALID_URL = r'http://www.letv.com/tv/(?P<id>\d+).html'
 157     _TESTS = [{
 158         'url': 'http://www.letv.com/tv/46177.html',
 159         'info_dict': {
 160             'id': '46177',
 161             'title': '美人天下',
 162             'description': 'md5:395666ff41b44080396e59570dbac01c'
 163         },
 164         'playlist_count': 35
 165     }]
 166
 167     def _real_extract(self, url):
 168         playlist_id = self._match_id(url)
 169         page = self._download_webpage(url, playlist_id)
 170
 171         media_urls = list(set(re.findall(
 172             r'http://www.letv.com/ptv/vplay/\d+.html', page)))
 173         entries = [self.url_result(media_url, ie='Letv')
 174                    for media_url in media_urls]
 175
 176         title = self._html_search_meta('keywords', page,
 177                                        fatal=False).split('，')[0]
 178         description = self._html_search_meta('description', page, fatal=False)
 179
 180         return self.playlist_result(entries, playlist_id, playlist_title=title,
 181                                     playlist_description=description)
 182
 183
 184 class LetvPlaylistIE(LetvTvIE):
 185     _VALID_URL = r'http://tv.letv.com/[a-z]+/(?P<id>[a-z]+)/index.s?html'
 186     _TESTS = [{
 187         'url': 'http://tv.letv.com/izt/wuzetian/index.html',
 188         'info_dict': {
 189             'id': 'wuzetian',
 190             'title': '武媚娘传奇',
 191             'description': 'md5:e12499475ab3d50219e5bba00b3cb248'
 192         },
 193         # This playlist contains some extra videos other than the drama itself
 194         'playlist_mincount': 96
 195     }, {
 196         'url': 'http://tv.letv.com/pzt/lswjzzjc/index.shtml',
 197         'info_dict': {
 198             'id': 'lswjzzjc',
 199             # The title should be "劲舞青春", but I can't find a simple way to
 200             # determine the playlist title
 201             'title': '乐视午间自制剧场',
 202             'description': 'md5:b1eef244f45589a7b5b1af9ff25a4489'
 203         },
 204         'playlist_mincount': 7
 205     }]