_ Git - youtube-dl/blob - youtube_dl/extractor/showroomlive.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import ExtractorError, compat_urlparse
   6
   7
   8 class ShowroomLiveIE(InfoExtractor):
   9     _VALID_URL = r'https?://(?:www\.)?showroom-live\.com/(?P<id>[0-9a-zA-Z_]+)'
  10     _TEST = {
  11         'url': 'https://www.showroom-live.com/48_Nana_Okada',
  12         'skip': 'Only live broadcasts, can\'t predict test case.',
  13         'info_dict': {
  14             'id': '48_Nana_Okada',
  15             'ext': 'mp4',
  16             'uploader_id': '48_Nana_Okada',
  17         }
  18     }
  19
  20     def _real_extract(self, url):
  21         broadcaster_id = self._match_id(url)
  22
  23         # There is no showroom on these pages.
  24         if broadcaster_id in ['onlive', 'timetable', 'event', 'campaign', 'news', 'ranking']:
  25             raise ExtractorError('URL %s does not contain a showroom' % url)
  26
  27         # Retrieve the information we need
  28         webpage = self._download_webpage(url, broadcaster_id)
  29         room_id = self._search_regex(r'profile\?room_id\=(\d+)', webpage, 'room_id')
  30         room_url = compat_urlparse.urljoin(url, "/api/room/profile?room_id=%s") % room_id
  31         room = self._download_json(room_url, broadcaster_id)
  32
  33         is_live = room.get('is_onlive')
  34         if not is_live:
  35             raise ExtractorError('%s their showroom is not live' % broadcaster_id)
  36
  37         # Prepare and return the information
  38         uploader = room.get('performer_name') or broadcaster_id  # performer_name can be an empty string.
  39         title = room.get('room_name', room.get('main_name', "%s's Showroom" % uploader))
  40
  41         return {
  42             'is_live': is_live,
  43             'id': str(room.get('live_id')),
  44             'timestamp': room.get('current_live_started_at'),
  45             'uploader': uploader,
  46             'uploader_id': broadcaster_id,
  47             'title': title,
  48             'description': room.get('description'),
  49             'formats': self._extract_formats(url, broadcaster_id, room_id)
  50         }
  51
  52     def _extract_formats(self, url, broadcaster_id, room_id):
  53         formats = []
  54
  55         stream_url = compat_urlparse.urljoin(url, "/api/live/streaming_url?room_id=%s") % room_id
  56         streaming_url_list = self._download_json(stream_url, broadcaster_id).get('streaming_url_list', [])
  57
  58         for stream in streaming_url_list:
  59             if stream.get('type') == "hls":
  60                 formats.extend(self._extract_m3u8_formats(
  61                     stream.get('url'),
  62                     broadcaster_id,
  63                     ext='mp4',
  64                     m3u8_id='hls',
  65                     preference=stream.get('quality', 100),
  66                     live=True
  67                 ))
  68             elif stream.get('type') == 'rtmp':
  69                 url = stream.get('url') + '/' + stream.get('stream_name')
  70                 formats.append({
  71                     'url': url,
  72                     'format_id': 'rtmp',
  73                     'protocol': 'rtmp',
  74                     'ext': 'flv',
  75                     'preference': stream.get('quality', 100),
  76                     'format_note': stream.get('label')
  77                 })
  78
  79         self._sort_formats(formats)
  80         return formats