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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import hashlib
   5 import time
   6 from .common import InfoExtractor
   7 from ..utils import (ExtractorError, unescapeHTML)
   8 from ..compat import (compat_str, compat_basestring)
   9
  10
  11 class DouyuTVIE(InfoExtractor):
  12     IE_DESC = '斗鱼'
  13     _VALID_URL = r'https?://(?:www\.)?douyu(?:tv)?\.com/(?P<id>[A-Za-z0-9]+)'
  14     _TESTS = [{
  15         'url': 'http://www.douyutv.com/iseven',
  16         'info_dict': {
  17             'id': '17732',
  18             'display_id': 'iseven',
  19             'ext': 'flv',
  20             'title': 're:^清晨醒脑！T-ara根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
  21             'description': 're:.*m7show@163\.com.*',
  22             'thumbnail': 're:^https?://.*\.jpg$',
  23             'uploader': '7师傅',
  24             'uploader_id': '431925',
  25             'is_live': True,
  26         },
  27         'params': {
  28             'skip_download': True,
  29         },
  30     }, {
  31         'url': 'http://www.douyutv.com/85982',
  32         'info_dict': {
  33             'id': '85982',
  34             'display_id': '85982',
  35             'ext': 'flv',
  36             'title': 're:^小漠从零单排记！——CSOL2躲猫猫 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
  37             'description': 'md5:746a2f7a253966a06755a912f0acc0d2',
  38             'thumbnail': 're:^https?://.*\.jpg$',
  39             'uploader': 'douyu小漠',
  40             'uploader_id': '3769985',
  41             'is_live': True,
  42         },
  43         'params': {
  44             'skip_download': True,
  45         },
  46         'skip': 'Room not found',
  47     }, {
  48         'url': 'http://www.douyutv.com/17732',
  49         'info_dict': {
  50             'id': '17732',
  51             'display_id': '17732',
  52             'ext': 'flv',
  53             'title': 're:^清晨醒脑！T-ara根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
  54             'description': 're:.*m7show@163\.com.*',
  55             'thumbnail': 're:^https?://.*\.jpg$',
  56             'uploader': '7师傅',
  57             'uploader_id': '431925',
  58             'is_live': True,
  59         },
  60         'params': {
  61             'skip_download': True,
  62         },
  63     }, {
  64         'url': 'http://www.douyu.com/xiaocang',
  65         'only_matching': True,
  66     }]
  67
  68     def _real_extract(self, url):
  69         video_id = self._match_id(url)
  70
  71         if video_id.isdigit():
  72             room_id = video_id
  73         else:
  74             page = self._download_webpage(url, video_id)
  75             room_id = self._html_search_regex(
  76                 r'"room_id"\s*:\s*(\d+),', page, 'room id')
  77
  78         config = None
  79         # Douyu API sometimes returns error "Unable to load the requested class: eticket_redis_cache"
  80         # Retry with different parameters - same parameters cause same errors
  81         for i in range(5):
  82             prefix = 'room/%s?aid=android&client_sys=android&time=%d' % (
  83                 room_id, int(time.time()))
  84             auth = hashlib.md5((prefix + '1231').encode('ascii')).hexdigest()
  85
  86             config_page = self._download_webpage(
  87                 'http://www.douyutv.com/api/v1/%s&auth=%s' % (prefix, auth),
  88                 video_id)
  89             try:
  90                 config = self._parse_json(config_page, video_id, fatal=False)
  91             except ExtractorError:
  92                 # Wait some time before retrying to get a different time() value
  93                 self._sleep(1, video_id, msg_template='%(video_id)s: Error occurs. '
  94                                                       'Waiting for %(timeout)s seconds before retrying')
  95                 continue
  96             else:
  97                 break
  98         if config is None:
  99             raise ExtractorError('Unable to fetch API result')
 100
 101         data = config['data']
 102
 103         error_code = config.get('error', 0)
 104         if error_code is not 0:
 105             error_desc = 'Server reported error %i' % error_code
 106             if isinstance(data, (compat_str, compat_basestring)):
 107                 error_desc += ': ' + data
 108             raise ExtractorError(error_desc, expected=True)
 109
 110         show_status = data.get('show_status')
 111         # 1 = live, 2 = offline
 112         if show_status == '2':
 113             raise ExtractorError(
 114                 'Live stream is offline', expected=True)
 115
 116         base_url = data['rtmp_url']
 117         live_path = data['rtmp_live']
 118
 119         title = self._live_title(unescapeHTML(data['room_name']))
 120         description = data.get('show_details')
 121         thumbnail = data.get('room_src')
 122
 123         uploader = data.get('nickname')
 124         uploader_id = data.get('owner_uid')
 125
 126         multi_formats = data.get('rtmp_multi_bitrate')
 127         if not isinstance(multi_formats, dict):
 128             multi_formats = {}
 129         multi_formats['live'] = live_path
 130
 131         formats = [{
 132             'url': '%s/%s' % (base_url, format_path),
 133             'format_id': format_id,
 134             'preference': 1 if format_id == 'live' else 0,
 135         } for format_id, format_path in multi_formats.items()]
 136         self._sort_formats(formats)
 137
 138         return {
 139             'id': room_id,
 140             'display_id': video_id,
 141             'title': title,
 142             'description': description,
 143             'thumbnail': thumbnail,
 144             'uploader': uploader,
 145             'uploader_id': uploader_id,
 146             'formats': formats,
 147             'is_live': True,
 148         }