+ webpage = self._download_webpage(url, video_id)
+
+ ERRORS = (
+ ('>Leider liegt eine Störung vor.', 'Video %s is unavailable'),
+ ('>Der gewünschte Beitrag ist nicht mehr verfügbar.<',
+ 'Video %s is no longer available'),
+ )
+
+ for pattern, message in ERRORS:
+ if pattern in webpage:
+ raise ExtractorError(message % video_id, expected=True)
+
+ if re.search(r'[\?&]rss($|[=&])', url):
+ doc = compat_etree_fromstring(webpage.encode('utf-8'))
+ if doc.tag == 'rss':
+ return GenericIE()._extract_rss(url, video_id, doc)
+
+ title = self._html_search_regex(
+ [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
+ r'<meta name="dcterms\.title" content="(.*?)"/>',
+ r'<h4 class="headline">(.*?)</h4>',
+ r'<title[^>]*>(.*?)</title>'],
+ webpage, 'title')
+ description = self._html_search_meta(
+ 'dcterms.abstract', webpage, 'description', default=None)
+ if description is None:
+ description = self._html_search_meta(
+ 'description', webpage, 'meta description', default=None)
+ if description is None:
+ description = self._html_search_regex(
+ r'<p\s+class="teasertext">(.+?)</p>',
+ webpage, 'teaser text', default=None)
+
+ # Thumbnail is sometimes not present.
+ # It is in the mobile version, but that seems to use a different URL
+ # structure altogether.
+ thumbnail = self._og_search_thumbnail(webpage, default=None)
+
+ media_streams = re.findall(r'''(?x)
+ mediaCollection\.addMediaStream\([0-9]+,\s*[0-9]+,\s*"[^"]*",\s*
+ "([^"]+)"''', webpage)
+
+ if media_streams:
+ QUALITIES = qualities(['lo', 'hi', 'hq'])
+ formats = []
+ for furl in set(media_streams):
+ if furl.endswith('.f4m'):
+ fid = 'f4m'
+ else:
+ fid_m = re.match(r'.*\.([^.]+)\.[^.]+$', furl)
+ fid = fid_m.group(1) if fid_m else None
+ formats.append({
+ 'quality': QUALITIES(fid),
+ 'format_id': fid,
+ 'url': furl,
+ })
+ self._sort_formats(formats)
+ info = {
+ 'formats': formats,
+ }
+ else: # request JSON file
+ if not document_id:
+ video_id = self._search_regex(
+ r'/play/(?:config|media)/(\d+)', webpage, 'media id')
+ info = self._extract_media_info(
+ 'http://www.ardmediathek.de/play/media/%s' % video_id,
+ webpage, video_id)
+
+ info.update({
+ 'id': video_id,
+ 'title': self._live_title(title) if info.get('is_live') else title,
+ 'description': description,
+ 'thumbnail': thumbnail,
+ })
+
+ return info
+
+
+class ARDIE(InfoExtractor):
+ _VALID_URL = r'(?P<mainurl>https?://(www\.)?daserste\.de/[^?#]+/videos/(?P<display_id>[^/?#]+)-(?P<id>[0-9]+))\.html'
+ _TESTS = [{
+ # available till 14.02.2019
+ 'url': 'http://www.daserste.de/information/talk/maischberger/videos/das-groko-drama-zerlegen-sich-die-volksparteien-video-102.html',
+ 'md5': '8e4ec85f31be7c7fc08a26cdbc5a1f49',
+ 'info_dict': {
+ 'display_id': 'das-groko-drama-zerlegen-sich-die-volksparteien-video',
+ 'id': '102',
+ 'ext': 'mp4',
+ 'duration': 4435.0,
+ 'title': 'Das GroKo-Drama: Zerlegen sich die Volksparteien?',
+ 'upload_date': '20180214',
+ 'thumbnail': r're:^https?://.*\.jpg$',
+ },
+ }, {
+ 'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html',
+ 'only_matching': True,
+ }]
+
+ def _real_extract(self, url):
+ mobj = re.match(self._VALID_URL, url)
+ display_id = mobj.group('display_id')
+
+ player_url = mobj.group('mainurl') + '~playerXml.xml'
+ doc = self._download_xml(player_url, display_id)
+ video_node = doc.find('./video')
+ upload_date = unified_strdate(xpath_text(
+ video_node, './broadcastDate'))
+ thumbnail = xpath_text(video_node, './/teaserImage//variant/url')
+
+ formats = []
+ for a in video_node.findall('.//asset'):
+ f = {
+ 'format_id': a.attrib['type'],
+ 'width': int_or_none(a.find('./frameWidth').text),
+ 'height': int_or_none(a.find('./frameHeight').text),
+ 'vbr': int_or_none(a.find('./bitrateVideo').text),
+ 'abr': int_or_none(a.find('./bitrateAudio').text),
+ 'vcodec': a.find('./codecVideo').text,
+ 'tbr': int_or_none(a.find('./totalBitrate').text),
+ }
+ if a.find('./serverPrefix').text:
+ f['url'] = a.find('./serverPrefix').text
+ f['playpath'] = a.find('./fileName').text
+ else:
+ f['url'] = a.find('./fileName').text
+ formats.append(f)
+ self._sort_formats(formats)
+
+ return {
+ 'id': mobj.group('id'),
+ 'formats': formats,
+ 'display_id': display_id,
+ 'title': video_node.find('./title').text,
+ 'duration': parse_duration(video_node.find('./duration').text),
+ 'upload_date': upload_date,
+ 'thumbnail': thumbnail,
+ }
+
+
+class ARDBetaMediathekIE(InfoExtractor):
+ _VALID_URL = r'https://(?:beta|www)\.ardmediathek\.de/[^/]+/(?:player|live)/(?P<video_id>[a-zA-Z0-9]+)(?:/(?P<display_id>[^/?#]+))?'
+ _TESTS = [{
+ 'url': 'https://beta.ardmediathek.de/ard/player/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE/die-robuste-roswita',
+ 'md5': '2d02d996156ea3c397cfc5036b5d7f8f',
+ 'info_dict': {
+ 'display_id': 'die-robuste-roswita',
+ 'id': 'Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE',
+ 'title': 'Tatort: Die robuste Roswita',
+ 'description': r're:^Der Mord.*trüber ist als die Ilm.',
+ 'duration': 5316,
+ 'thumbnail': 'https://img.ardmediathek.de/standard/00/55/43/59/34/-1774185891/16x9/960?mandant=ard',
+ 'upload_date': '20180826',
+ 'ext': 'mp4',
+ },
+ }, {
+ 'url': 'https://www.ardmediathek.de/ard/player/Y3JpZDovL3N3ci5kZS9hZXgvbzEwNzE5MTU/',
+ 'only_matching': True,
+ }, {
+ 'url': 'https://www.ardmediathek.de/swr/live/Y3JpZDovL3N3ci5kZS8xMzQ4MTA0Mg',
+ 'only_matching': True,
+ }]
+
+ def _real_extract(self, url):
+ mobj = re.match(self._VALID_URL, url)
+ video_id = mobj.group('video_id')
+ display_id = mobj.group('display_id') or video_id
+
+ webpage = self._download_webpage(url, display_id)
+ data_json = self._search_regex(r'window\.__APOLLO_STATE__\s*=\s*(\{.*);\n', webpage, 'json')
+ data = self._parse_json(data_json, display_id)
+
+ res = {
+ 'id': video_id,
+ 'display_id': display_id,
+ }
+ formats = []
+ subtitles = {}
+ geoblocked = False
+ for widget in data.values():
+ if widget.get('_geoblocked') is True:
+ geoblocked = True
+ if '_duration' in widget:
+ res['duration'] = int_or_none(widget['_duration'])
+ if 'clipTitle' in widget:
+ res['title'] = widget['clipTitle']
+ if '_previewImage' in widget:
+ res['thumbnail'] = widget['_previewImage']
+ if 'broadcastedOn' in widget:
+ res['timestamp'] = unified_timestamp(widget['broadcastedOn'])
+ if 'synopsis' in widget:
+ res['description'] = widget['synopsis']
+ subtitle_url = url_or_none(widget.get('_subtitleUrl'))
+ if subtitle_url:
+ subtitles.setdefault('de', []).append({
+ 'ext': 'ttml',
+ 'url': subtitle_url,
+ })
+ if '_quality' in widget:
+ format_url = url_or_none(try_get(
+ widget, lambda x: x['_stream']['json'][0]))
+ if not format_url:
+ continue
+ ext = determine_ext(format_url)
+ if ext == 'f4m':
+ formats.extend(self._extract_f4m_formats(
+ format_url + '?hdcore=3.11.0',
+ video_id, f4m_id='hds', fatal=False))
+ elif ext == 'm3u8':
+ formats.extend(self._extract_m3u8_formats(
+ format_url, video_id, 'mp4', m3u8_id='hls',
+ fatal=False))
+ else:
+ # HTTP formats are not available when geoblocked is True,
+ # other formats are fine though
+ if geoblocked:
+ continue
+ quality = str_or_none(widget.get('_quality'))
+ formats.append({
+ 'format_id': ('http-' + quality) if quality else 'http',
+ 'url': format_url,
+ 'preference': 10, # Plain HTTP, that's nice
+ })
+
+ if not formats and geoblocked:
+ self.raise_geo_restricted(
+ msg='This video is not available due to geoblocking',
+ countries=['DE'])
+
+ self._sort_formats(formats)
+ res.update({
+ 'subtitles': subtitles,
+ 'formats': formats,
+ })
+
+ return res
Rapsys Git / youtubedl / blobdiff