1
0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-15 14:47:44 +01:00
yt-dlp/youtube_dl/extractor/beampro.py

189 lines
6.1 KiB
Python
Raw Normal View History

2017-01-03 23:51:08 +01:00
# coding: utf-8
from __future__ import unicode_literals
from .common import InfoExtractor
from ..utils import (
ExtractorError,
clean_html,
compat_str,
2017-05-14 10:04:42 +02:00
float_or_none,
2017-01-03 23:51:08 +01:00
int_or_none,
parse_iso8601,
try_get,
2017-05-14 10:04:42 +02:00
urljoin,
2017-01-03 23:51:08 +01:00
)
2017-05-14 10:04:42 +02:00
class BeamProBaseIE(InfoExtractor):
_API_BASE = 'https://mixer.com/api/v1'
2017-05-14 10:04:42 +02:00
_RATINGS = {'family': 0, 'teen': 13, '18+': 18}
def _extract_channel_info(self, chan):
user_id = chan.get('userId') or try_get(chan, lambda x: x['user']['id'])
return {
'uploader': chan.get('token') or try_get(
chan, lambda x: x['user']['username'], compat_str),
'uploader_id': compat_str(user_id) if user_id else None,
'age_limit': self._RATINGS.get(chan.get('audience')),
}
class BeamProLiveIE(BeamProBaseIE):
IE_NAME = 'Mixer:live'
_VALID_URL = r'https?://(?:\w+\.)?(?:beam\.pro|mixer\.com)/(?P<id>[^/?#&]+)'
2017-01-03 23:51:08 +01:00
_TEST = {
'url': 'http://mixer.com/niterhayven',
2017-01-03 23:51:08 +01:00
'info_dict': {
'id': '261562',
'ext': 'mp4',
'title': 'Introducing The Witcher 3 // The Grind Starts Now!',
'description': 'md5:0b161ac080f15fe05d18a07adb44a74d',
2017-01-03 23:51:08 +01:00
'thumbnail': r're:https://.*\.jpg$',
'timestamp': 1483477281,
2017-01-03 23:51:08 +01:00
'upload_date': '20170103',
'uploader': 'niterhayven',
'uploader_id': '373396',
'age_limit': 18,
2017-01-03 23:51:08 +01:00
'is_live': True,
'view_count': int,
2017-01-03 23:51:08 +01:00
},
'skip': 'niterhayven is offline',
'params': {
'skip_download': True,
},
}
_MANIFEST_URL_TEMPLATE = '%s/channels/%%s/manifest.%%s' % BeamProBaseIE._API_BASE
2017-05-14 10:04:42 +02:00
@classmethod
def suitable(cls, url):
return False if BeamProVodIE.suitable(url) else super(BeamProLiveIE, cls).suitable(url)
2017-01-03 23:51:08 +01:00
def _real_extract(self, url):
channel_name = self._match_id(url)
2017-01-03 23:51:08 +01:00
chan = self._download_json(
'%s/channels/%s' % (self._API_BASE, channel_name), channel_name)
2017-01-03 23:51:08 +01:00
if chan.get('online') is False:
raise ExtractorError(
'{0} is offline'.format(channel_name), expected=True)
2017-01-03 23:51:08 +01:00
channel_id = chan['id']
2017-01-03 23:51:08 +01:00
def manifest_url(kind):
return self._MANIFEST_URL_TEMPLATE % (channel_id, kind)
formats = self._extract_m3u8_formats(
manifest_url('m3u8'), channel_name, ext='mp4', m3u8_id='hls',
fatal=False)
formats.extend(self._extract_smil_formats(
manifest_url('smil'), channel_name, fatal=False))
self._sort_formats(formats)
2017-01-03 23:51:08 +01:00
2017-05-14 10:04:42 +02:00
info = {
'id': compat_str(chan.get('id') or channel_name),
'title': self._live_title(chan.get('name') or channel_name),
'description': clean_html(chan.get('description')),
'thumbnail': try_get(
chan, lambda x: x['thumbnail']['url'], compat_str),
'timestamp': parse_iso8601(chan.get('updatedAt')),
'is_live': True,
'view_count': int_or_none(chan.get('viewersTotal')),
'formats': formats,
2017-01-03 23:51:08 +01:00
}
2017-05-14 10:04:42 +02:00
info.update(self._extract_channel_info(chan))
return info
class BeamProVodIE(BeamProBaseIE):
IE_NAME = 'Mixer:vod'
_VALID_URL = r'https?://(?:\w+\.)?(?:beam\.pro|mixer\.com)/[^/?#&]+\?.*?\bvod=(?P<id>\d+)'
2017-05-14 10:04:42 +02:00
_TEST = {
'url': 'https://mixer.com/willow8714?vod=2259830',
2017-05-14 10:04:42 +02:00
'md5': 'b2431e6e8347dc92ebafb565d368b76b',
'info_dict': {
'id': '2259830',
'ext': 'mp4',
'title': 'willow8714\'s Channel',
'duration': 6828.15,
'thumbnail': r're:https://.*source\.png$',
'timestamp': 1494046474,
'upload_date': '20170506',
'uploader': 'willow8714',
'uploader_id': '6085379',
'age_limit': 13,
'view_count': int,
},
'params': {
'skip_download': True,
},
2017-05-14 10:04:42 +02:00
}
@staticmethod
def _extract_format(vod, vod_type):
2017-05-14 10:04:42 +02:00
if not vod.get('baseUrl'):
return []
if vod_type == 'hls':
filename, protocol = 'manifest.m3u8', 'm3u8_native'
2017-05-14 10:04:42 +02:00
elif vod_type == 'raw':
filename, protocol = 'source.mp4', 'https'
else:
assert False
2017-05-14 10:04:42 +02:00
data = vod.get('data') if isinstance(vod.get('data'), dict) else {}
2017-05-14 10:04:42 +02:00
format_id = [vod_type]
if isinstance(data.get('Height'), compat_str):
2017-05-14 10:04:42 +02:00
format_id.append('%sp' % data['Height'])
return [{
'url': urljoin(vod['baseUrl'], filename),
'format_id': '-'.join(format_id),
'ext': 'mp4',
'protocol': protocol,
'width': int_or_none(data.get('Width')),
'height': int_or_none(data.get('Height')),
'fps': int_or_none(data.get('Fps')),
'tbr': int_or_none(data.get('Bitrate'), 1000),
}]
def _real_extract(self, url):
vod_id = self._match_id(url)
vod_info = self._download_json(
'%s/recordings/%s' % (self._API_BASE, vod_id), vod_id)
2017-05-14 10:04:42 +02:00
state = vod_info.get('state')
if state != 'AVAILABLE':
raise ExtractorError(
'VOD %s is not available (state: %s)' % (vod_id, state),
expected=True)
2017-05-14 10:04:42 +02:00
formats = []
thumbnail_url = None
for vod in vod_info['vods']:
vod_type = vod.get('format')
if vod_type in ('hls', 'raw'):
formats.extend(self._extract_format(vod, vod_type))
elif vod_type == 'thumbnail':
thumbnail_url = urljoin(vod.get('baseUrl'), 'source.png')
self._sort_formats(formats)
info = {
'id': vod_id,
'title': vod_info.get('name') or vod_id,
'duration': float_or_none(vod_info.get('duration')),
'thumbnail': thumbnail_url,
'timestamp': parse_iso8601(vod_info.get('createdAt')),
'view_count': int_or_none(vod_info.get('viewsTotal')),
'formats': formats,
}
info.update(self._extract_channel_info(vod_info.get('channel') or {}))
2017-05-14 10:04:42 +02:00
return info