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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5
   6 from ..utils import (
   7     int_or_none,
   8     parse_iso8601
   9 )
  10
  11 class FreshliveIE(InfoExtractor):
  12     _VALID_URL = r'https?://freshlive\.tv/(?P<streamer>[^/]+)/(?P<id>[0-9]+)'
  13     _TEST = {
  14         'url': 'https://freshlive.tv/satotv/74712',
  15         'md5': '224f50d268b6b9f94e4198deccd55d6d',
  16         'info_dict': {
  17             'description': 'テスト',
  18             'duration': 1511,
  19             'id': '74712',
  20             'ext': 'mp4',
  21             'timestamp': 1483621764,
  22             'title': 'テスト',
  23             'thumbnail': r're:^https?://.*\.jpg$',
  24             'upload_date': '20170105',
  25             'uploader': 'サトTV',
  26             'uploader_id': 'satotv',
  27             'view_count': int,
  28         }
  29     }
  30
  31     def _real_extract(self, url):
  32         video_id = self._match_id(url)
  33         webpage = self._download_webpage(url, video_id)
  34
  35         options = self._parse_json(
  36             self._search_regex(
  37                 r'window\.__CONTEXT__\s*=\s*({.+?});\s*</script>',
  38                 webpage, 'initial context'),
  39             video_id)
  40
  41         programs = options['context']['dispatcher']['stores']['ProgramStore']['programs']
  42         info = programs.get(video_id, {})
  43
  44         video_url = info.get('liveStreamUrl') or info.get('archiveStreamUrl')
  45         if not video_url:
  46             raise ExtractorError('%s not a valid broadcast ID' % video_id, expected=True)
  47
  48         formats = self._extract_m3u8_formats(
  49             video_url, video_id, ext='mp4', m3u8_id='hls')
  50
  51         return {
  52             'id': video_id,
  53             'formats': formats,
  54             'title': info.get('title'),
  55             'description': info.get('description'),
  56             'duration': int_or_none(info.get('airTime')),
  57             'is_live': int_or_none(info.get('airTime')) == None,
  58             'thumbnail': info.get('thumbnailUrl'),
  59             'uploader': info.get('channel', {}).get('title'),
  60             'uploader_id': info.get('channel', {}).get('code'),
  61             'uploader_url': info.get('channel', {}).get('permalink'),
  62             'timestamp': parse_iso8601(info.get('startAt')),
  63             'view_count': int_or_none(info.get('viewCount')),
  64         }