+ base_url, display_id = re.search(self._VALID_URL, url).groups()
+ page = self._download_webpage(url, display_id)
+ xml_url = self._search_regex(
+ r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/(?:[a-z0-9\-]+/)+[a-z0-9/~_.-]+)'}\)\);", page, 'XMLURL')
+ xml = self._download_xml(base_url + xml_url, display_id)
+
+ medias = []
+
+ for xml_media in xml.findall('video') + xml.findall('audio'):
+ media_id = xml_media.get('externalId')
+ media = {
+ 'id': media_id,
+ 'title': xpath_text(xml_media, 'title', 'title', True),
+ 'duration': parse_duration(xpath_text(xml_media, 'duration')),
+ 'formats': self._extract_formats(xpath_element(
+ xml_media, 'assets'), media_id),
+ 'thumbnails': self._extract_thumbnails(xpath_element(
+ xml_media, 'teaserImage/variants'), base_url),
+ 'description': xpath_text(xml_media, 'desc'),
+ 'webpage_url': xpath_text(xml_media, 'permalink'),
+ 'uploader': xpath_text(xml_media, 'author'),
+ }
+ broadcast_date = xpath_text(xml_media, 'broadcastDate')
+ if broadcast_date:
+ media['upload_date'] = ''.join(reversed(broadcast_date.split('.')))
+ medias.append(media)
+
+ if len(medias) > 1:
+ self._downloader.report_warning(
+ 'found multiple medias; please '
+ 'report this with the video URL to http://yt-dl.org/bug')
+ if not medias:
+ raise ExtractorError('No media entries found')
+ return medias[0]
+
+ def _extract_formats(self, assets, media_id):
+ formats = []
+ for asset in assets.findall('asset'):
+ format_url = xpath_text(asset, ['downloadUrl', 'url'])
+ asset_type = asset.get('type')
+ if asset_type.startswith('HDS'):
+ formats.extend(self._extract_f4m_formats(
+ format_url + '?hdcore=3.2.0', media_id, f4m_id='hds', fatal=False))
+ elif asset_type.startswith('HLS'):
+ formats.extend(self._extract_m3u8_formats(
+ format_url, media_id, 'mp4', 'm3u8_native', m3u8_id='hds', fatal=False))
+ else:
+ format_info = {
+ 'ext': xpath_text(asset, 'mediaType'),
+ 'width': int_or_none(xpath_text(asset, 'frameWidth')),
+ 'height': int_or_none(xpath_text(asset, 'frameHeight')),
+ 'tbr': int_or_none(xpath_text(asset, 'bitrateVideo')),
+ 'abr': int_or_none(xpath_text(asset, 'bitrateAudio')),
+ 'vcodec': xpath_text(asset, 'codecVideo'),
+ 'acodec': xpath_text(asset, 'codecAudio'),
+ 'container': xpath_text(asset, 'mediaType'),
+ 'filesize': int_or_none(xpath_text(asset, 'size')),
+ }
+ format_url = self._proto_relative_url(format_url)
+ if format_url:
+ http_format_info = format_info.copy()
+ http_format_info.update({
+ 'url': format_url,
+ 'format_id': 'http-%s' % asset_type,
+ })
+ formats.append(http_format_info)
+ server_prefix = xpath_text(asset, 'serverPrefix')
+ if server_prefix:
+ rtmp_format_info = format_info.copy()
+ rtmp_format_info.update({
+ 'url': server_prefix,
+ 'play_path': xpath_text(asset, 'fileName'),
+ 'format_id': 'rtmp-%s' % asset_type,
+ })
+ formats.append(rtmp_format_info)
+ self._sort_formats(formats)
+ return formats
+
+ def _extract_thumbnails(self, variants, base_url):
+ thumbnails = [{
+ 'url': base_url + xpath_text(variant, 'url'),
+ 'width': int_or_none(xpath_text(variant, 'width')),
+ 'height': int_or_none(xpath_text(variant, 'height')),
+ } for variant in variants.findall('variant') if xpath_text(variant, 'url')]
+ thumbnails.sort(key=lambda x: x['width'] * x['height'], reverse=True)
+ return thumbnails
+
+
+class BRMediathekIE(InfoExtractor):
+ IE_DESC = 'Bayerischer Rundfunk Mediathek'
+ _VALID_URL = r'https?://(?:www\.)?br\.de/mediathek/video/[^/?&#]*?-(?P<id>av:[0-9a-f]{24})'
+
+ _TESTS = [{
+ 'url': 'https://www.br.de/mediathek/video/gesundheit-die-sendung-vom-28112017-av:5a1e6a6e8fce6d001871cc8e',
+ 'md5': 'fdc3d485835966d1622587d08ba632ec',
+ 'info_dict': {
+ 'id': 'av:5a1e6a6e8fce6d001871cc8e',
+ 'ext': 'mp4',
+ 'title': 'Die Sendung vom 28.11.2017',
+ 'description': 'md5:6000cdca5912ab2277e5b7339f201ccc',
+ 'timestamp': 1511942766,
+ 'upload_date': '20171129',
+ }
+ }]
+
+ def _real_extract(self, url):
+ clip_id = self._match_id(url)
+
+ clip = self._download_json(
+ 'https://proxy-base.master.mango.express/graphql',
+ clip_id, data=json.dumps({
+ "query": """{
+ viewer {
+ clip(id: "%s") {
+ title
+ description
+ duration
+ createdAt
+ ageRestriction
+ videoFiles {
+ edges {
+ node {
+ publicLocation
+ fileSize
+ videoProfile {
+ width
+ height
+ bitrate
+ encoding
+ }
+ }
+ }
+ }
+ captionFiles {
+ edges {
+ node {
+ publicLocation
+ }
+ }
+ }
+ teaserImages {
+ edges {
+ node {
+ imageFiles {
+ edges {
+ node {
+ publicLocation
+ width
+ height
+ }
+ }
+ }
+ }
+ }
+ }
+ }
+ }
+}""" % clip_id}).encode(), headers={
+ 'Content-Type': 'application/json',
+ })['data']['viewer']['clip']
+ title = clip['title']
+
+ formats = []
+ for edge in clip.get('videoFiles', {}).get('edges', []):
+ node = edge.get('node', {})
+ n_url = node.get('publicLocation')
+ if not n_url:
+ continue
+ ext = determine_ext(n_url)
+ if ext == 'm3u8':
+ formats.extend(self._extract_m3u8_formats(
+ n_url, clip_id, 'mp4', 'm3u8_native',
+ m3u8_id='hls', fatal=False))
+ else:
+ video_profile = node.get('videoProfile', {})
+ tbr = int_or_none(video_profile.get('bitrate'))
+ format_id = 'http'
+ if tbr:
+ format_id += '-%d' % tbr
+ formats.append({
+ 'format_id': format_id,
+ 'url': n_url,
+ 'width': int_or_none(video_profile.get('width')),
+ 'height': int_or_none(video_profile.get('height')),
+ 'tbr': tbr,
+ 'filesize': int_or_none(node.get('fileSize')),
+ })
+ self._sort_formats(formats)
+
+ subtitles = {}
+ for edge in clip.get('captionFiles', {}).get('edges', []):
+ node = edge.get('node', {})
+ n_url = node.get('publicLocation')
+ if not n_url: