[extractor/tvp] Support stream.tvp.pl (#6139)

Authored by: selfisekai
This commit is contained in:
lauren n. liberda 2023-02-12 05:43:10 +01:00 committed by GitHub
parent cc2389c8ac
commit a31d0fa6c3
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23

View file

@ -268,8 +268,11 @@ def _real_extract(self, url):
class TVPStreamIE(InfoExtractor):
IE_NAME = 'tvp:stream'
_VALID_URL = r'(?:tvpstream:|https?://tvpstream\.vod\.tvp\.pl/(?:\?(?:[^&]+[&;])*channel_id=)?)(?P<id>\d*)'
_VALID_URL = r'(?:tvpstream:|https?://(?:tvpstream\.vod|stream)\.tvp\.pl/(?:\?(?:[^&]+[&;])*channel_id=)?)(?P<id>\d*)'
_TESTS = [{
'url': 'https://stream.tvp.pl/?channel_id=56969941',
'only_matching': True,
}, {
# untestable as "video" id changes many times across a day
'url': 'https://tvpstream.vod.tvp.pl/?channel_id=1455',
'only_matching': True,
@ -285,28 +288,21 @@ class TVPStreamIE(InfoExtractor):
'only_matching': True,
}]
_PLAYER_BOX_RE = r'<div\s[^>]*id\s*=\s*["\']?tvp_player_box["\']?[^>]+data-%s-id\s*=\s*["\']?(\d+)'
_BUTTON_RE = r'<div\s[^>]*data-channel-id=["\']?%s["\']?[^>]*\sdata-title=(?:"([^"]*)"|\'([^\']*)\')[^>]*\sdata-stationname=(?:"([^"]*)"|\'([^\']*)\')'
def _real_extract(self, url):
channel_id = self._match_id(url)
channel_url = self._proto_relative_url('//tvpstream.vod.tvp.pl/?channel_id=%s' % channel_id or 'default')
webpage = self._download_webpage(channel_url, channel_id, 'Downloading channel webpage')
if not channel_id:
channel_id = self._search_regex(self._PLAYER_BOX_RE % 'channel',
webpage, 'default channel id')
video_id = self._search_regex(self._PLAYER_BOX_RE % 'video',
webpage, 'video id')
audition_title, station_name = self._search_regex(
self._BUTTON_RE % (re.escape(channel_id)), webpage,
'audition title and station name',
group=(1, 2))
channel_url = self._proto_relative_url('//stream.tvp.pl/?channel_id=%s' % channel_id or 'default')
webpage = self._download_webpage(channel_url, channel_id or 'default', 'Downloading channel webpage')
channels = self._search_json(
r'window\.__channels\s*=', webpage, 'channel list', channel_id,
contains_pattern=r'\[\s*{(?s:.+)}\s*]')
channel = traverse_obj(channels, (lambda _, v: channel_id == str(v['id'])), get_all=False) if channel_id else channels[0]
audition = traverse_obj(channel, ('items', lambda _, v: v['is_live'] is True), get_all=False)
return {
'_type': 'url_transparent',
'id': channel_id,
'url': 'tvp:%s' % video_id,
'title': audition_title,
'alt_title': station_name,
'id': channel_id or channel['id'],
'url': 'tvp:%s' % audition['video_id'],
'title': audition.get('title'),
'alt_title': channel.get('title'),
'is_live': True,
'ie_key': 'TVPEmbed',
}