yt-dlp/yt_dlp/extractor/stripchat.py

from .common import InfoExtractor
from ..utils import ExtractorError, lowercase_escape, traverse_obj


class StripchatIE(InfoExtractor):
    _VALID_URL = r'https?://stripchat\.com/(?P<id>[^/?#]+)'
    _TESTS = [{
        'url': 'https://stripchat.com/Joselin_Flower',
        'info_dict': {
            'id': 'Joselin_Flower',
            'ext': 'mp4',
            'title': 're:^Joselin_Flower [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
            'description': str,
            'is_live': True,
            'age_limit': 18,
        },
        'skip': 'Room is offline',
    }, {
        'url': 'https://stripchat.com/Rakhijaan@xh',
        'only_matching': True
    }]

    def _real_extract(self, url):
        video_id = self._match_id(url)
        webpage = self._download_webpage(url, video_id, headers=self.geo_verification_headers())

        data = self._parse_json(
            self._search_regex(
                r'<script\b[^>]*>\s*window\.__PRELOADED_STATE__\s*=(?P<value>.*?)<\/script>',
                webpage, 'data', default='{}', group='value'),
            video_id, transform_source=lowercase_escape, fatal=False)
        if not data:
            raise ExtractorError('Unable to find configuration for stream.')

        if traverse_obj(data, ('viewCam', 'show'), expected_type=dict):
            raise ExtractorError('Model is in private show', expected=True)
        elif not traverse_obj(data, ('viewCam', 'model', 'isLive'), expected_type=bool):
            raise ExtractorError('Model is offline', expected=True)

        server = traverse_obj(data, ('viewCam', 'viewServers', 'flashphoner-hls'), expected_type=str)
        model_id = traverse_obj(data, ('viewCam', 'model', 'id'), expected_type=int)

        formats = []
        for host in traverse_obj(data, (
                'config', 'data', (('featuresV2', 'hlsFallback', 'fallbackDomains', ...), 'hlsStreamHost'))):
            formats = self._extract_m3u8_formats(
                f'https://b-{server}.{host}/hls/{model_id}/{model_id}.m3u8',
                video_id, ext='mp4', m3u8_id='hls', fatal=False, live=True)
            if formats:
                break
        if not formats:
            self.raise_no_formats('No active streams found', expected=True)

        self._sort_formats(formats)

        return {
            'id': video_id,
            'title': video_id,
            'description': self._og_search_description(webpage),
            'is_live': True,
            'formats': formats,
            # Stripchat declares the RTA meta-tag, but in an non-standard format so _rta_search() can't be used
            'age_limit': 18,
        }