youtube_dl/extractor/beampro.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     clean_html,
   8     compat_str,
   9     float_or_none,
  10     int_or_none,
  11     parse_iso8601,
  12     try_get,
  13     urljoin,
  14 )
  15
  16
  17 class BeamProBaseIE(InfoExtractor):
  18     _API_BASE = 'https://mixer.com/api/v1'
  19     _RATINGS = {'family': 0, 'teen': 13, '18+': 18}
  20
  21     def _extract_channel_info(self, chan):
  22         user_id = chan.get('userId') or try_get(chan, lambda x: x['user']['id'])
  23         return {
  24             'uploader': chan.get('token') or try_get(
  25                 chan, lambda x: x['user']['username'], compat_str),
  26             'uploader_id': compat_str(user_id) if user_id else None,
  27             'age_limit': self._RATINGS.get(chan.get('audience')),
  28         }
  29
  30
  31 class BeamProLiveIE(BeamProBaseIE):
  32     IE_NAME = 'Mixer:live'
  33     _VALID_URL = r'https?://(?:\w+\.)?(?:beam\.pro|mixer\.com)/(?P<id>[^/?#&]+)'
  34     _TEST = {
  35         'url': 'http://mixer.com/niterhayven',
  36         'info_dict': {
  37             'id': '261562',
  38             'ext': 'mp4',
  39             'title': 'Introducing The Witcher 3 //  The Grind Starts Now!',
  40             'description': 'md5:0b161ac080f15fe05d18a07adb44a74d',
  41             'thumbnail': r're:https://.*\.jpg$',
  42             'timestamp': 1483477281,
  43             'upload_date': '20170103',
  44             'uploader': 'niterhayven',
  45             'uploader_id': '373396',
  46             'age_limit': 18,
  47             'is_live': True,
  48             'view_count': int,
  49         },
  50         'skip': 'niterhayven is offline',
  51         'params': {
  52             'skip_download': True,
  53         },
  54     }
  55
  56     _MANIFEST_URL_TEMPLATE = '%s/channels/%%s/manifest.%%s' % BeamProBaseIE._API_BASE
  57
  58     @classmethod
  59     def suitable(cls, url):
  60         return False if BeamProVodIE.suitable(url) else super(BeamProLiveIE, cls).suitable(url)
  61
  62     def _real_extract(self, url):
  63         channel_name = self._match_id(url)
  64
  65         chan = self._download_json(
  66             '%s/channels/%s' % (self._API_BASE, channel_name), channel_name)
  67
  68         if chan.get('online') is False:
  69             raise ExtractorError(
  70                 '{0} is offline'.format(channel_name), expected=True)
  71
  72         channel_id = chan['id']
  73
  74         def manifest_url(kind):
  75             return self._MANIFEST_URL_TEMPLATE % (channel_id, kind)
  76
  77         formats = self._extract_m3u8_formats(
  78             manifest_url('m3u8'), channel_name, ext='mp4', m3u8_id='hls',
  79             fatal=False)
  80         formats.extend(self._extract_smil_formats(
  81             manifest_url('smil'), channel_name, fatal=False))
  82         self._sort_formats(formats)
  83
  84         info = {
  85             'id': compat_str(chan.get('id') or channel_name),
  86             'title': self._live_title(chan.get('name') or channel_name),
  87             'description': clean_html(chan.get('description')),
  88             'thumbnail': try_get(
  89                 chan, lambda x: x['thumbnail']['url'], compat_str),
  90             'timestamp': parse_iso8601(chan.get('updatedAt')),
  91             'is_live': True,
  92             'view_count': int_or_none(chan.get('viewersTotal')),
  93             'formats': formats,
  94         }
  95         info.update(self._extract_channel_info(chan))
  96
  97         return info
  98
  99
 100 class BeamProVodIE(BeamProBaseIE):
 101     IE_NAME = 'Mixer:vod'
 102     _VALID_URL = r'https?://(?:\w+\.)?(?:beam\.pro|mixer\.com)/[^/?#&]+\?.*?\bvod=(?P<id>\d+)'
 103     _TEST = {
 104         'url': 'https://mixer.com/willow8714?vod=2259830',
 105         'md5': 'b2431e6e8347dc92ebafb565d368b76b',
 106         'info_dict': {
 107             'id': '2259830',
 108             'ext': 'mp4',
 109             'title': 'willow8714\'s Channel',
 110             'duration': 6828.15,
 111             'thumbnail': r're:https://.*source\.png$',
 112             'timestamp': 1494046474,
 113             'upload_date': '20170506',
 114             'uploader': 'willow8714',
 115             'uploader_id': '6085379',
 116             'age_limit': 13,
 117             'view_count': int,
 118         },
 119         'params': {
 120             'skip_download': True,
 121         },
 122     }
 123
 124     @staticmethod
 125     def _extract_format(vod, vod_type):
 126         if not vod.get('baseUrl'):
 127             return []
 128
 129         if vod_type == 'hls':
 130             filename, protocol = 'manifest.m3u8', 'm3u8_native'
 131         elif vod_type == 'raw':
 132             filename, protocol = 'source.mp4', 'https'
 133         else:
 134             assert False
 135
 136         data = vod.get('data') if isinstance(vod.get('data'), dict) else {}
 137
 138         format_id = [vod_type]
 139         if isinstance(data.get('Height'), compat_str):
 140             format_id.append('%sp' % data['Height'])
 141
 142         return [{
 143             'url': urljoin(vod['baseUrl'], filename),
 144             'format_id': '-'.join(format_id),
 145             'ext': 'mp4',
 146             'protocol': protocol,
 147             'width': int_or_none(data.get('Width')),
 148             'height': int_or_none(data.get('Height')),
 149             'fps': int_or_none(data.get('Fps')),
 150             'tbr': int_or_none(data.get('Bitrate'), 1000),
 151         }]
 152
 153     def _real_extract(self, url):
 154         vod_id = self._match_id(url)
 155
 156         vod_info = self._download_json(
 157             '%s/recordings/%s' % (self._API_BASE, vod_id), vod_id)
 158
 159         state = vod_info.get('state')
 160         if state != 'AVAILABLE':
 161             raise ExtractorError(
 162                 'VOD %s is not available (state: %s)' % (vod_id, state),
 163                 expected=True)
 164
 165         formats = []
 166         thumbnail_url = None
 167
 168         for vod in vod_info['vods']:
 169             vod_type = vod.get('format')
 170             if vod_type in ('hls', 'raw'):
 171                 formats.extend(self._extract_format(vod, vod_type))
 172             elif vod_type == 'thumbnail':
 173                 thumbnail_url = urljoin(vod.get('baseUrl'), 'source.png')
 174
 175         self._sort_formats(formats)
 176
 177         info = {
 178             'id': vod_id,
 179             'title': vod_info.get('name') or vod_id,
 180             'duration': float_or_none(vod_info.get('duration')),
 181             'thumbnail': thumbnail_url,
 182             'timestamp': parse_iso8601(vod_info.get('createdAt')),
 183             'view_count': int_or_none(vod_info.get('viewsTotal')),
 184             'formats': formats,
 185         }
 186         info.update(self._extract_channel_info(vod_info.get('channel') or {}))
 187
 188         return info