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_urlparse,
  11     compat_urllib_parse,
  12 )
  13 from ..utils import (
  14     determine_ext,
  15     ExtractorError,
  16     parse_iso8601,
  17 )
  18
  19
  20 class LetvIE(InfoExtractor):
  21     _VALID_URL = r'http://www\.letv\.com/ptv/vplay/(?P<id>\d+).html'
  22
  23     _TESTS = [{
  24         'url': 'http://www.letv.com/ptv/vplay/22005890.html',
  25         'md5': 'cab23bd68d5a8db9be31c9a222c1e8df',
  26         'info_dict': {
  27             'id': '22005890',
  28             'ext': 'mp4',
  29             'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家',
  30             'timestamp': 1424747397,
  31             'upload_date': '20150224',
  32             'description': 'md5:a9cb175fd753e2962176b7beca21a47c',
  33         }
  34     }, {
  35         'url': 'http://www.letv.com/ptv/vplay/1415246.html',
  36         'info_dict': {
  37             'id': '1415246',
  38             'ext': 'mp4',
  39             'title': '美人天下01',
  40             'description': 'md5:f88573d9d7225ada1359eaf0dbf8bcda',
  41         },
  42         'expected_warnings': [
  43             'publish time'
  44         ]
  45     }]
  46     # http://www.letv.com/ptv/vplay/1118082.html
  47     # This video is available only in Mainland China
  48
  49     @staticmethod
  50     def urshift(val, n):
  51         return val >> n if val >= 0 else (val + 0x100000000) >> n
  52
  53     # ror() and calc_time_key() are reversed from a embedded swf file in KLetvPlayer.swf
  54     def ror(self, param1, param2):
  55         _loc3_ = 0
  56         while _loc3_ < param2:
  57             param1 = self.urshift(param1, 1) + ((param1 & 1) << 31)
  58             _loc3_ += 1
  59         return param1
  60
  61     def calc_time_key(self, param1):
  62         _loc2_ = 773625421
  63         _loc3_ = self.ror(param1, _loc2_ % 13)
  64         _loc3_ = _loc3_ ^ _loc2_
  65         _loc3_ = self.ror(_loc3_, _loc2_ % 17)
  66         return _loc3_
  67
  68     def _real_extract(self, url):
  69         media_id = self._match_id(url)
  70         page = self._download_webpage(url, media_id)
  71         params = {
  72             'id': media_id,
  73             'platid': 1,
  74             'splatid': 101,
  75             'format': 1,
  76             'tkey': self.calc_time_key(int(time.time())),
  77             'domain': 'www.letv.com'
  78         }
  79         play_json = self._download_json(
  80             'http://api.letv.com/mms/out/video/playJson?' + compat_urllib_parse.urlencode(params),
  81             media_id, 'playJson data')
  82
  83         # Check for errors
  84         playstatus = play_json['playstatus']
  85         if playstatus['status'] == 0:
  86             flag = playstatus['flag']
  87             if flag == 1:
  88                 msg = 'Country %s auth error' % playstatus['country']
  89             else:
  90                 msg = 'Generic error. flag = %d' % flag
  91             raise ExtractorError(msg, expected=True)
  92
  93         playurl = play_json['playurl']
  94
  95         formats = ['350', '1000', '1300', '720p', '1080p']
  96         dispatch = playurl['dispatch']
  97
  98         urls = []
  99         for format_id in formats:
 100             if format_id in dispatch:
 101                 media_url = playurl['domain'][0] + dispatch[format_id][0]
 102
 103                 # Mimic what flvxz.com do
 104                 url_parts = list(compat_urlparse.urlparse(media_url))
 105                 qs = dict(compat_urlparse.parse_qs(url_parts[4]))
 106                 qs.update({
 107                     'platid': '14',
 108                     'splatid': '1401',
 109                     'tss': 'no',
 110                     'retry': 1
 111                 })
 112                 url_parts[4] = compat_urllib_parse.urlencode(qs)
 113                 media_url = compat_urlparse.urlunparse(url_parts)
 114
 115                 url_info_dict = {
 116                     'url': media_url,
 117                     'ext': determine_ext(dispatch[format_id][1])
 118                 }
 119
 120                 if format_id[-1:] == 'p':
 121                     url_info_dict['height'] = format_id[:-1]
 122
 123                 urls.append(url_info_dict)
 124
 125         publish_time = parse_iso8601(self._html_search_regex(
 126             r'发布时间&nbsp;([^<>]+) ', page, 'publish time', fatal=False),
 127             delimiter=' ', timezone=datetime.timedelta(hours=8))
 128         description = self._html_search_meta('description', page, fatal=False)
 129
 130         return {
 131             'id': media_id,
 132             'formats': urls,
 133             'title': playurl['title'],
 134             'thumbnail': playurl['pic'],
 135             'description': description,
 136             'timestamp': publish_time,
 137         }
 138
 139
 140 class LetvTvIE(InfoExtractor):
 141     _VALID_URL = r'http://www.letv.com/tv/(?P<id>\d+).html'
 142     _TESTS = [{
 143         'url': 'http://www.letv.com/tv/46177.html',
 144         'info_dict': {
 145             'id': '46177',
 146             'title': '美人天下',
 147             'description': 'md5:395666ff41b44080396e59570dbac01c'
 148         },
 149         'playlist_count': 35
 150     }]
 151
 152     def _real_extract(self, url):
 153         playlist_id = self._match_id(url)
 154         page = self._download_webpage(url, playlist_id)
 155
 156         media_urls = list(set(re.findall(
 157             r'http://www.letv.com/ptv/vplay/\d+.html', page)))
 158         entries = [self.url_result(media_url, ie='Letv')
 159                    for media_url in media_urls]
 160
 161         title = self._html_search_meta('keywords', page,
 162                                        fatal=False).split('，')[0]
 163         description = self._html_search_meta('description', page, fatal=False)
 164
 165         return self.playlist_result(entries, playlist_id, playlist_title=title,
 166                                     playlist_description=description)
 167
 168
 169 class LetvPlaylistIE(LetvTvIE):
 170     _VALID_URL = r'http://tv.letv.com/[a-z]+/(?P<id>[a-z]+)/index.s?html'
 171     _TESTS = [{
 172         'url': 'http://tv.letv.com/izt/wuzetian/index.html',
 173         'info_dict': {
 174             'id': 'wuzetian',
 175             'title': '武媚娘传奇',
 176             'description': 'md5:e12499475ab3d50219e5bba00b3cb248'
 177         },
 178         # This playlist contains some extra videos other than the drama itself
 179         'playlist_mincount': 96
 180     }, {
 181         'url': 'http://tv.letv.com/pzt/lswjzzjc/index.shtml',
 182         'info_dict': {
 183             'id': 'lswjzzjc',
 184             # The title should be "劲舞青春", but I can't find a simple way to
 185             # determine the playlist title
 186             'title': '乐视午间自制剧场',
 187             'description': 'md5:b1eef244f45589a7b5b1af9ff25a4489'
 188         },
 189         'playlist_mincount': 7
 190     }]