[ie/orf:on] Improve extraction (#9677)

[yt-dlp.git] / yt_dlp / extractor / podbayfm.py

from .common import InfoExtractor
from ..utils import OnDemandPagedList, int_or_none, jwt_decode_hs256, try_call


def result_from_props(props, episode_id=None):
    return {
        'id': props.get('podcast_id') or episode_id,
        'title': props.get('title'),
        'url': props['mediaURL'],
        'ext': 'mp3',
        'thumbnail': try_call(lambda: jwt_decode_hs256(props['image'])['url']),
        'timestamp': props.get('timestamp'),
        'duration': int_or_none(props.get('duration')),
    }


class PodbayFMIE(InfoExtractor):
    _VALID_URL = r'https?://podbay\.fm/p/[^/]*/e/(?P<id>[^/]*)/?(?:[\?#].*)?$'
    _TESTS = [{
        'url': 'https://podbay.fm/p/behind-the-bastards/e/1647338400',
        'md5': '98b41285dcf7989d105a4ed0404054cf',
        'info_dict': {
            'id': '1647338400',
            'title': 'Part One: Kissinger',
            'ext': 'mp3',
            'thumbnail': r're:^https?://.*\.jpg',
            'timestamp': 1647338400,
            'duration': 5001,
            'upload_date': '20220315',
        },
    }]

    def _real_extract(self, url):
        episode_id = self._match_id(url)
        webpage = self._download_webpage(url, episode_id)
        data = self._search_nextjs_data(webpage, episode_id)
        return result_from_props(data['props']['pageProps']['episode'], episode_id)


class PodbayFMChannelIE(InfoExtractor):
    _VALID_URL = r'https?://podbay\.fm/p/(?P<id>[^/]*)/?(?:[\?#].*)?$'
    _TESTS = [{
        'url': 'https://podbay.fm/p/behind-the-bastards',
        'info_dict': {
            'id': 'behind-the-bastards',
            'title': 'Behind the Bastards',
        },
    }]
    _PAGE_SIZE = 10

    def _fetch_page(self, channel_id, pagenum):
        return self._download_json(
            f'https://podbay.fm/api/podcast?reverse=true&page={pagenum}&slug={channel_id}',
            channel_id)['podcast']

    @staticmethod
    def _results_from_page(channel_id, page):
        return [{
            **result_from_props(e),
            'extractor': PodbayFMIE.IE_NAME,
            'extractor_key': PodbayFMIE.ie_key(),
            # somehow they use timestamps as the episode identifier
            'webpage_url': f'https://podbay.fm/p/{channel_id}/e/{e["timestamp"]}',
        } for e in page['episodes']]

    def _real_extract(self, url):
        channel_id = self._match_id(url)

        first_page = self._fetch_page(channel_id, 0)
        entries = OnDemandPagedList(
            lambda pagenum: self._results_from_page(
                channel_id, self._fetch_page(channel_id, pagenum) if pagenum else first_page),
            self._PAGE_SIZE)

        return self.playlist_result(entries, channel_id, first_page.get('title'))