2 from __future__ import unicode_literals
6 from .common import InfoExtractor
7 from .generic import GenericIE
19 from ..compat import compat_etree_fromstring
22 class ARDMediathekIE(InfoExtractor):
23 IE_NAME = 'ARD:mediathek'
24 _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.(?:daserste|rbb-online)\.de|one\.ard\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
27 # available till 26.07.2022
28 'url': 'http://www.ardmediathek.de/tv/S%C3%9CDLICHT/Was-ist-die-Kunst-der-Zukunft-liebe-Ann/BR-Fernsehen/Video?bcastId=34633636&documentId=44726822',
32 'title': 'Was ist die Kunst der Zukunft, liebe Anna McCarthy?',
33 'description': 'md5:4ada28b3e3b5df01647310e41f3a62f5',
38 'skip_download': True,
41 'url': 'https://one.ard.de/tv/Mord-mit-Aussicht/Mord-mit-Aussicht-6-39-T%C3%B6dliche-Nach/ONE/Video?bcastId=46384294&documentId=55586872',
42 'only_matching': True,
45 'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086',
46 'only_matching': True,
48 'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
49 'only_matching': True,
52 'url': 'http://mediathek.rbb-online.de/radio/Hörspiel/Vor-dem-Fest/kulturradio/Audio?documentId=30796318&topRessort=radio&bcastId=9839158',
53 'only_matching': True,
56 def _extract_media_info(self, media_info_url, webpage, video_id):
57 media_info = self._download_json(
58 media_info_url, video_id, 'Downloading media JSON')
60 formats = self._extract_formats(media_info, video_id)
63 if '"fsk"' in webpage:
65 'This video is only available after 20:00', expected=True)
66 elif media_info.get('_geoblocked'):
67 raise ExtractorError('This video is not available due to geo restriction', expected=True)
69 self._sort_formats(formats)
71 duration = int_or_none(media_info.get('_duration'))
72 thumbnail = media_info.get('_previewImage')
73 is_live = media_info.get('_isLive') is True
76 subtitle_url = media_info.get('_subtitleUrl')
86 'thumbnail': thumbnail,
89 'subtitles': subtitles,
92 def _extract_formats(self, media_info, video_id):
93 type_ = media_info.get('_type')
94 media_array = media_info.get('_mediaArray', [])
96 for num, media in enumerate(media_array):
97 for stream in media.get('_mediaStreamArray', []):
98 stream_urls = stream.get('_stream')
101 if not isinstance(stream_urls, list):
102 stream_urls = [stream_urls]
103 quality = stream.get('_quality')
104 server = stream.get('_server')
105 for stream_url in stream_urls:
106 if not url_or_none(stream_url):
108 ext = determine_ext(stream_url)
109 if quality != 'auto' and ext in ('f4m', 'm3u8'):
112 formats.extend(self._extract_f4m_formats(
113 update_url_query(stream_url, {
115 'plugin': 'aasp-3.1.1.69.124'
117 video_id, f4m_id='hds', fatal=False))
119 formats.extend(self._extract_m3u8_formats(
120 stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
122 if server and server.startswith('rtmp'):
125 'play_path': stream_url,
126 'format_id': 'a%s-rtmp-%s' % (num, quality),
131 'format_id': 'a%s-%s-%s' % (num, ext, quality)
133 m = re.search(r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$', stream_url)
136 'width': int(m.group('width')),
137 'height': int(m.group('height')),
144 def _real_extract(self, url):
145 # determine video id from url
146 m = re.match(self._VALID_URL, url)
150 numid = re.search(r'documentId=([0-9]+)', url)
152 document_id = video_id = numid.group(1)
154 video_id = m.group('video_id')
156 webpage = self._download_webpage(url, video_id)
159 ('>Leider liegt eine Störung vor.', 'Video %s is unavailable'),
160 ('>Der gewünschte Beitrag ist nicht mehr verfügbar.<',
161 'Video %s is no longer available'),
164 for pattern, message in ERRORS:
165 if pattern in webpage:
166 raise ExtractorError(message % video_id, expected=True)
168 if re.search(r'[\?&]rss($|[=&])', url):
169 doc = compat_etree_fromstring(webpage.encode('utf-8'))
171 return GenericIE()._extract_rss(url, video_id, doc)
173 title = self._html_search_regex(
174 [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
175 r'<meta name="dcterms\.title" content="(.*?)"/>',
176 r'<h4 class="headline">(.*?)</h4>',
177 r'<title[^>]*>(.*?)</title>'],
179 description = self._html_search_meta(
180 'dcterms.abstract', webpage, 'description', default=None)
181 if description is None:
182 description = self._html_search_meta(
183 'description', webpage, 'meta description', default=None)
184 if description is None:
185 description = self._html_search_regex(
186 r'<p\s+class="teasertext">(.+?)</p>',
187 webpage, 'teaser text', default=None)
189 # Thumbnail is sometimes not present.
190 # It is in the mobile version, but that seems to use a different URL
191 # structure altogether.
192 thumbnail = self._og_search_thumbnail(webpage, default=None)
194 media_streams = re.findall(r'''(?x)
195 mediaCollection\.addMediaStream\([0-9]+,\s*[0-9]+,\s*"[^"]*",\s*
196 "([^"]+)"''', webpage)
199 QUALITIES = qualities(['lo', 'hi', 'hq'])
201 for furl in set(media_streams):
202 if furl.endswith('.f4m'):
205 fid_m = re.match(r'.*\.([^.]+)\.[^.]+$', furl)
206 fid = fid_m.group(1) if fid_m else None
208 'quality': QUALITIES(fid),
212 self._sort_formats(formats)
216 else: # request JSON file
218 video_id = self._search_regex(
219 r'/play/(?:config|media)/(\d+)', webpage, 'media id')
220 info = self._extract_media_info(
221 'http://www.ardmediathek.de/play/media/%s' % video_id,
226 'title': self._live_title(title) if info.get('is_live') else title,
227 'description': description,
228 'thumbnail': thumbnail,
234 class ARDIE(InfoExtractor):
235 _VALID_URL = r'(?P<mainurl>https?://(www\.)?daserste\.de/[^?#]+/videos/(?P<display_id>[^/?#]+)-(?P<id>[0-9]+))\.html'
237 # available till 14.02.2019
238 'url': 'http://www.daserste.de/information/talk/maischberger/videos/das-groko-drama-zerlegen-sich-die-volksparteien-video-102.html',
239 'md5': '8e4ec85f31be7c7fc08a26cdbc5a1f49',
241 'display_id': 'das-groko-drama-zerlegen-sich-die-volksparteien-video',
245 'title': 'Das GroKo-Drama: Zerlegen sich die Volksparteien?',
246 'upload_date': '20180214',
247 'thumbnail': r're:^https?://.*\.jpg$',
250 'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html',
251 'only_matching': True,
254 def _real_extract(self, url):
255 mobj = re.match(self._VALID_URL, url)
256 display_id = mobj.group('display_id')
258 player_url = mobj.group('mainurl') + '~playerXml.xml'
259 doc = self._download_xml(player_url, display_id)
260 video_node = doc.find('./video')
261 upload_date = unified_strdate(xpath_text(
262 video_node, './broadcastDate'))
263 thumbnail = xpath_text(video_node, './/teaserImage//variant/url')
266 for a in video_node.findall('.//asset'):
268 'format_id': a.attrib['type'],
269 'width': int_or_none(a.find('./frameWidth').text),
270 'height': int_or_none(a.find('./frameHeight').text),
271 'vbr': int_or_none(a.find('./bitrateVideo').text),
272 'abr': int_or_none(a.find('./bitrateAudio').text),
273 'vcodec': a.find('./codecVideo').text,
274 'tbr': int_or_none(a.find('./totalBitrate').text),
276 if a.find('./serverPrefix').text:
277 f['url'] = a.find('./serverPrefix').text
278 f['playpath'] = a.find('./fileName').text
280 f['url'] = a.find('./fileName').text
282 self._sort_formats(formats)
285 'id': mobj.group('id'),
287 'display_id': display_id,
288 'title': video_node.find('./title').text,
289 'duration': parse_duration(video_node.find('./duration').text),
290 'upload_date': upload_date,
291 'thumbnail': thumbnail,
295 class ARDBetaMediathekIE(InfoExtractor):
296 _VALID_URL = r'https://beta\.ardmediathek\.de/[a-z]+/player/(?P<video_id>[a-zA-Z0-9]+)/(?P<display_id>[^/?#]+)'
298 'url': 'https://beta.ardmediathek.de/ard/player/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE/die-robuste-roswita',
299 'md5': '2d02d996156ea3c397cfc5036b5d7f8f',
301 'display_id': 'die-robuste-roswita',
302 'id': 'Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE',
303 'title': 'Tatort: Die robuste Roswita',
304 'description': r're:^Der Mord.*trüber ist als die Ilm.',
306 'thumbnail': 'https://img.ardmediathek.de/standard/00/55/43/59/34/-1774185891/16x9/960?mandant=ard',
307 'upload_date': '20180826',
312 def _real_extract(self, url):
313 mobj = re.match(self._VALID_URL, url)
314 video_id = mobj.group('video_id')
315 display_id = mobj.group('display_id')
317 webpage = self._download_webpage(url, display_id)
318 data_json = self._search_regex(r'window\.__APOLLO_STATE__\s*=\s*(\{.*);\n', webpage, 'json')
319 data = self._parse_json(data_json, display_id)
323 'display_id': display_id,
326 for widget in data.values():
327 if widget.get('_geoblocked'):
328 raise ExtractorError('This video is not available due to geoblocking', expected=True)
330 if '_duration' in widget:
331 res['duration'] = widget['_duration']
332 if 'clipTitle' in widget:
333 res['title'] = widget['clipTitle']
334 if '_previewImage' in widget:
335 res['thumbnail'] = widget['_previewImage']
336 if 'broadcastedOn' in widget:
337 res['upload_date'] = unified_strdate(widget['broadcastedOn'])
338 if 'synopsis' in widget:
339 res['description'] = widget['synopsis']
340 if '_subtitleUrl' in widget:
341 res['subtitles'] = {'de': [{
343 'url': widget['_subtitleUrl'],
345 if '_quality' in widget:
346 format_url = widget['_stream']['json'][0]
348 if format_url.endswith('.f4m'):
349 formats.extend(self._extract_f4m_formats(
350 format_url + '?hdcore=3.11.0',
351 video_id, f4m_id='hds', fatal=False))
352 elif format_url.endswith('m3u8'):
353 formats.extend(self._extract_m3u8_formats(
354 format_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
357 'format_id': 'http-' + widget['_quality'],
359 'preference': 10, # Plain HTTP, that's nice
362 self._sort_formats(formats)
363 res['formats'] = formats