youtube_dl/extractor/freshlive.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import compat_str
   6 from ..utils import (
   7     ExtractorError,
   8     int_or_none,
   9     try_get,
  10     unified_timestamp,
  11 )
  12
  13
  14 class FreshLiveIE(InfoExtractor):
  15     _VALID_URL = r'https?://freshlive\.tv/[^/]+/(?P<id>\d+)'
  16     _TEST = {
  17         'url': 'https://freshlive.tv/satotv/74712',
  18         'md5': '9f0cf5516979c4454ce982df3d97f352',
  19         'info_dict': {
  20             'id': '74712',
  21             'ext': 'mp4',
  22             'title': 'テスト',
  23             'description': 'テスト',
  24             'thumbnail': r're:^https?://.*\.jpg$',
  25             'duration': 1511,
  26             'timestamp': 1483619655,
  27             'upload_date': '20170105',
  28             'uploader': 'サトTV',
  29             'uploader_id': 'satotv',
  30             'view_count': int,
  31             'comment_count': int,
  32             'is_live': False,
  33         }
  34     }
  35
  36     def _real_extract(self, url):
  37         video_id = self._match_id(url)
  38
  39         webpage = self._download_webpage(url, video_id)
  40
  41         options = self._parse_json(
  42             self._search_regex(
  43                 r'window\.__CONTEXT__\s*=\s*({.+?});\s*</script>',
  44                 webpage, 'initial context'),
  45             video_id)
  46
  47         info = options['context']['dispatcher']['stores']['ProgramStore']['programs'][video_id]
  48
  49         title = info['title']
  50
  51         if info.get('status') == 'upcoming':
  52             raise ExtractorError('Stream %s is upcoming' % video_id, expected=True)
  53
  54         stream_url = info.get('liveStreamUrl') or info['archiveStreamUrl']
  55
  56         is_live = info.get('liveStreamUrl') is not None
  57
  58         formats = self._extract_m3u8_formats(
  59             stream_url, video_id, ext='mp4',
  60             entry_protocol='m3u8' if is_live else 'm3u8_native',
  61             m3u8_id='hls')
  62
  63         if is_live:
  64             title = self._live_title(title)
  65
  66         return {
  67             'id': video_id,
  68             'formats': formats,
  69             'title': title,
  70             'description': info.get('description'),
  71             'thumbnail': info.get('thumbnailUrl'),
  72             'duration': int_or_none(info.get('airTime')),
  73             'timestamp': unified_timestamp(info.get('createdAt')),
  74             'uploader': try_get(
  75                 info, lambda x: x['channel']['title'], compat_str),
  76             'uploader_id': try_get(
  77                 info, lambda x: x['channel']['code'], compat_str),
  78             'uploader_url': try_get(
  79                 info, lambda x: x['channel']['permalink'], compat_str),
  80             'view_count': int_or_none(info.get('viewCount')),
  81             'comment_count': int_or_none(info.get('commentCount')),
  82             'tags': info.get('tags', []),
  83             'is_live': is_live,
  84         }