+ _VALID_URL = r'https?://mediazone\.vrt\.be/api/v1/(?P<site_id>canvas|een|ketnet|vrt(?:video|nieuws)|sporza)/assets/(?P<id>[^/?#&]+)'
+ _TESTS = [{
+ 'url': 'https://mediazone.vrt.be/api/v1/ketnet/assets/md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
+ 'md5': '68993eda72ef62386a15ea2cf3c93107',
+ 'info_dict': {
+ 'id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
+ 'display_id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
+ 'ext': 'mp4',
+ 'title': 'Nachtwacht: De Greystook',
+ 'description': 'Nachtwacht: De Greystook',
+ 'thumbnail': r're:^https?://.*\.jpg$',
+ 'duration': 1468.04,
+ },
+ 'expected_warnings': ['is not a supported codec', 'Unknown MIME type'],
+ }, {
+ 'url': 'https://mediazone.vrt.be/api/v1/canvas/assets/mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e',
+ 'only_matching': True,
+ }]
+ _HLS_ENTRY_PROTOCOLS_MAP = {
+ 'HLS': 'm3u8_native',
+ 'HLS_AES': 'm3u8',
+ }
+ _REST_API_BASE = 'https://media-services-public.vrt.be/vualto-video-aggregator-web/rest/external/v1'
+
+ def _real_extract(self, url):
+ mobj = re.match(self._VALID_URL, url)
+ site_id, video_id = mobj.group('site_id'), mobj.group('id')
+
+ # Old API endpoint, serves more formats but may fail for some videos
+ data = self._download_json(
+ 'https://mediazone.vrt.be/api/v1/%s/assets/%s'
+ % (site_id, video_id), video_id, 'Downloading asset JSON',
+ 'Unable to download asset JSON', fatal=False)
+
+ # New API endpoint
+ if not data:
+ token = self._download_json(
+ '%s/tokens' % self._REST_API_BASE, video_id,
+ 'Downloading token', data=b'',
+ headers={'Content-Type': 'application/json'})['vrtPlayerToken']
+ data = self._download_json(
+ '%s/videos/%s' % (self._REST_API_BASE, video_id),
+ video_id, 'Downloading video JSON', fatal=False, query={
+ 'vrtPlayerToken': token,
+ 'client': '%s@PROD' % site_id,
+ }, expected_status=400)
+ message = data.get('message')
+ if message and not data.get('title'):
+ if data.get('code') == 'AUTHENTICATION_REQUIRED':
+ self.raise_login_required(message)
+ raise ExtractorError(message, expected=True)
+
+ title = data['title']
+ description = data.get('description')
+
+ formats = []
+ for target in data['targetUrls']:
+ format_url, format_type = url_or_none(target.get('url')), str_or_none(target.get('type'))
+ if not format_url or not format_type:
+ continue
+ format_type = format_type.upper()
+ if format_type in self._HLS_ENTRY_PROTOCOLS_MAP:
+ formats.extend(self._extract_m3u8_formats(
+ format_url, video_id, 'mp4', self._HLS_ENTRY_PROTOCOLS_MAP[format_type],
+ m3u8_id=format_type, fatal=False))
+ elif format_type == 'HDS':
+ formats.extend(self._extract_f4m_formats(
+ format_url, video_id, f4m_id=format_type, fatal=False))
+ elif format_type == 'MPEG_DASH':
+ formats.extend(self._extract_mpd_formats(
+ format_url, video_id, mpd_id=format_type, fatal=False))
+ elif format_type == 'HSS':
+ formats.extend(self._extract_ism_formats(
+ format_url, video_id, ism_id='mss', fatal=False))
+ else:
+ formats.append({
+ 'format_id': format_type,
+ 'url': format_url,
+ })
+ self._sort_formats(formats)
+
+ subtitles = {}
+ subtitle_urls = data.get('subtitleUrls')
+ if isinstance(subtitle_urls, list):
+ for subtitle in subtitle_urls:
+ subtitle_url = subtitle.get('url')
+ if subtitle_url and subtitle.get('type') == 'CLOSED':
+ subtitles.setdefault('nl', []).append({'url': subtitle_url})
+
+ return {
+ 'id': video_id,
+ 'display_id': video_id,
+ 'title': title,
+ 'description': description,
+ 'formats': formats,
+ 'duration': float_or_none(data.get('duration'), 1000),
+ 'thumbnail': data.get('posterImageUrl'),
+ 'subtitles': subtitles,
+ }
+
+
+class CanvasEenIE(InfoExtractor):