[extractor/jwplatform] Look for `data-video-jw-id`

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

import re

from .common import InfoExtractor
from ..utils import unsmuggle_url


class JWPlatformIE(InfoExtractor):
    _VALID_URL = r'(?:https?://(?:content\.jwplatform|cdn\.jwplayer)\.com/(?:(?:feed|player|thumb|preview)s|jw6|v2/media)/|jwplatform:)(?P<id>[a-zA-Z0-9]{8})'
    _TESTS = [{
        'url': 'http://content.jwplatform.com/players/nPripu9l-ALJ3XQCI.js',
        'md5': 'fa8899fa601eb7c83a64e9d568bdf325',
        'info_dict': {
            'id': 'nPripu9l',
            'ext': 'mov',
            'title': 'Big Buck Bunny Trailer',
            'description': 'Big Buck Bunny is a short animated film by the Blender Institute. It is made using free and open source software.',
            'upload_date': '20081127',
            'timestamp': 1227796140,
        }
    }, {
        'url': 'https://cdn.jwplayer.com/players/nPripu9l-ALJ3XQCI.js',
        'only_matching': True,
    }]

    @staticmethod
    def _extract_url(webpage):
        urls = JWPlatformIE._extract_urls(webpage)
        return urls[0] if urls else None

    @staticmethod
    def _extract_urls(webpage):
        for tag, key in ((r'(?:script|iframe)', 'src'), ('input', 'value')):
            # <input value=URL> is used by hyland.com
            # if we find <iframe>, dont look for <input>
            ret = re.findall(
                r'<%s[^>]+?%s=["\']((?:https?:)?//(?:content\.jwplatform|cdn\.jwplayer)\.com/players/[a-zA-Z0-9]{8})' % (tag, key),
                webpage)
            if ret:
                return ret
        mobj = re.search(r'<div\b[^>]* data-video-jw-id="([a-zA-Z0-9]{8})"', webpage)
        if mobj:
            return [f'jwplatform:{mobj.group(1)}']

    def _real_extract(self, url):
        url, smuggled_data = unsmuggle_url(url, {})
        self._initialize_geo_bypass({
            'countries': smuggled_data.get('geo_countries'),
        })
        video_id = self._match_id(url)
        json_data = self._download_json('https://cdn.jwplayer.com/v2/media/' + video_id, video_id)
        return self._parse_jwplayer_data(json_data, video_id)