2 from __future__ import unicode_literals
6 from .common import InfoExtractor
7 from .generic import GenericIE
18 from ..compat import compat_etree_fromstring
21 class ARDMediathekIE(InfoExtractor):
22 IE_NAME = 'ARD:mediathek'
23 _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.(?:daserste|rbb-online)\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
26 'url': 'http://www.ardmediathek.de/tv/Dokumentation-und-Reportage/Ich-liebe-das-Leben-trotzdem/rbb-Fernsehen/Video?documentId=29582122&bcastId=3822114',
30 'title': 'Ich liebe das Leben trotzdem',
31 'description': 'md5:45e4c225c72b27993314b31a84a5261c',
36 'skip_download': True,
38 'skip': 'HTTP Error 404: Not Found',
40 'url': 'http://www.ardmediathek.de/tv/Tatort/Tatort-Scheinwelten-H%C3%B6rfassung-Video/Das-Erste/Video?documentId=29522730&bcastId=602916',
41 'md5': 'f4d98b10759ac06c0072bbcd1f0b9e3e',
45 'title': 'Tatort: Scheinwelten - Hörfassung (Video tgl. ab 20 Uhr)',
46 'description': 'md5:196392e79876d0ac94c94e8cdb2875f1',
49 'skip': 'HTTP Error 404: Not Found',
52 'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086',
53 'md5': '219d94d8980b4f538c7fcb0865eb7f2c',
57 'title': 'Tod eines Fußballers',
58 'description': 'md5:f6e39f3461f0e1f54bfa48c8875c86ef',
61 'skip': 'HTTP Error 404: Not Found',
63 'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
64 'only_matching': True,
67 'url': 'http://mediathek.rbb-online.de/radio/Hörspiel/Vor-dem-Fest/kulturradio/Audio?documentId=30796318&topRessort=radio&bcastId=9839158',
68 'md5': '4e8f00631aac0395fee17368ac0e9867',
72 'title': 'Vor dem Fest',
73 'description': 'md5:c0c1c8048514deaed2a73b3a60eecacb',
76 'skip': 'Video is no longer available',
79 def _extract_media_info(self, media_info_url, webpage, video_id):
80 media_info = self._download_json(
81 media_info_url, video_id, 'Downloading media JSON')
83 formats = self._extract_formats(media_info, video_id)
86 if '"fsk"' in webpage:
88 'This video is only available after 20:00', expected=True)
89 elif media_info.get('_geoblocked'):
90 raise ExtractorError('This video is not available due to geo restriction', expected=True)
92 self._sort_formats(formats)
94 duration = int_or_none(media_info.get('_duration'))
95 thumbnail = media_info.get('_previewImage')
96 is_live = media_info.get('_isLive') is True
99 subtitle_url = media_info.get('_subtitleUrl')
108 'duration': duration,
109 'thumbnail': thumbnail,
112 'subtitles': subtitles,
115 def _extract_formats(self, media_info, video_id):
116 type_ = media_info.get('_type')
117 media_array = media_info.get('_mediaArray', [])
119 for num, media in enumerate(media_array):
120 for stream in media.get('_mediaStreamArray', []):
121 stream_urls = stream.get('_stream')
124 if not isinstance(stream_urls, list):
125 stream_urls = [stream_urls]
126 quality = stream.get('_quality')
127 server = stream.get('_server')
128 for stream_url in stream_urls:
129 ext = determine_ext(stream_url)
130 if quality != 'auto' and ext in ('f4m', 'm3u8'):
133 formats.extend(self._extract_f4m_formats(
134 update_url_query(stream_url, {
136 'plugin': 'aasp-3.1.1.69.124'
138 video_id, f4m_id='hds', fatal=False))
140 formats.extend(self._extract_m3u8_formats(
141 stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
143 if server and server.startswith('rtmp'):
146 'play_path': stream_url,
147 'format_id': 'a%s-rtmp-%s' % (num, quality),
149 elif stream_url.startswith('http'):
152 'format_id': 'a%s-%s-%s' % (num, ext, quality)
156 m = re.search(r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$', stream_url)
159 'width': int(m.group('width')),
160 'height': int(m.group('height')),
167 def _real_extract(self, url):
168 # determine video id from url
169 m = re.match(self._VALID_URL, url)
173 numid = re.search(r'documentId=([0-9]+)', url)
175 document_id = video_id = numid.group(1)
177 video_id = m.group('video_id')
179 webpage = self._download_webpage(url, video_id)
182 ('>Leider liegt eine Störung vor.', 'Video %s is unavailable'),
183 ('>Der gewünschte Beitrag ist nicht mehr verfügbar.<',
184 'Video %s is no longer available'),
187 for pattern, message in ERRORS:
188 if pattern in webpage:
189 raise ExtractorError(message % video_id, expected=True)
191 if re.search(r'[\?&]rss($|[=&])', url):
192 doc = compat_etree_fromstring(webpage.encode('utf-8'))
194 return GenericIE()._extract_rss(url, video_id, doc)
196 title = self._html_search_regex(
197 [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
198 r'<meta name="dcterms\.title" content="(.*?)"/>',
199 r'<h4 class="headline">(.*?)</h4>'],
201 description = self._html_search_meta(
202 'dcterms.abstract', webpage, 'description', default=None)
203 if description is None:
204 description = self._html_search_meta(
205 'description', webpage, 'meta description')
207 # Thumbnail is sometimes not present.
208 # It is in the mobile version, but that seems to use a different URL
209 # structure altogether.
210 thumbnail = self._og_search_thumbnail(webpage, default=None)
212 media_streams = re.findall(r'''(?x)
213 mediaCollection\.addMediaStream\([0-9]+,\s*[0-9]+,\s*"[^"]*",\s*
214 "([^"]+)"''', webpage)
217 QUALITIES = qualities(['lo', 'hi', 'hq'])
219 for furl in set(media_streams):
220 if furl.endswith('.f4m'):
223 fid_m = re.match(r'.*\.([^.]+)\.[^.]+$', furl)
224 fid = fid_m.group(1) if fid_m else None
226 'quality': QUALITIES(fid),
230 self._sort_formats(formats)
234 else: # request JSON file
236 video_id = self._search_regex(
237 r'/play/(?:config|media)/(\d+)', webpage, 'media id')
238 info = self._extract_media_info(
239 'http://www.ardmediathek.de/play/media/%s' % video_id,
244 'title': self._live_title(title) if info.get('is_live') else title,
245 'description': description,
246 'thumbnail': thumbnail,
252 class ARDIE(InfoExtractor):
253 _VALID_URL = r'(?P<mainurl>https?://(www\.)?daserste\.de/[^?#]+/videos/(?P<display_id>[^/?#]+)-(?P<id>[0-9]+))\.html'
255 'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html',
256 'md5': 'd216c3a86493f9322545e045ddc3eb35',
258 'display_id': 'die-story-im-ersten-mission-unter-falscher-flagge',
262 'title': 'Die Story im Ersten: Mission unter falscher Flagge',
263 'upload_date': '20140804',
264 'thumbnail': r're:^https?://.*\.jpg$',
266 'skip': 'HTTP Error 404: Not Found',
269 def _real_extract(self, url):
270 mobj = re.match(self._VALID_URL, url)
271 display_id = mobj.group('display_id')
273 player_url = mobj.group('mainurl') + '~playerXml.xml'
274 doc = self._download_xml(player_url, display_id)
275 video_node = doc.find('./video')
276 upload_date = unified_strdate(xpath_text(
277 video_node, './broadcastDate'))
278 thumbnail = xpath_text(video_node, './/teaserImage//variant/url')
281 for a in video_node.findall('.//asset'):
283 'format_id': a.attrib['type'],
284 'width': int_or_none(a.find('./frameWidth').text),
285 'height': int_or_none(a.find('./frameHeight').text),
286 'vbr': int_or_none(a.find('./bitrateVideo').text),
287 'abr': int_or_none(a.find('./bitrateAudio').text),
288 'vcodec': a.find('./codecVideo').text,
289 'tbr': int_or_none(a.find('./totalBitrate').text),
291 if a.find('./serverPrefix').text:
292 f['url'] = a.find('./serverPrefix').text
293 f['playpath'] = a.find('./fileName').text
295 f['url'] = a.find('./fileName').text
297 self._sort_formats(formats)
300 'id': mobj.group('id'),
302 'display_id': display_id,
303 'title': video_node.find('./title').text,
304 'duration': parse_duration(video_node.find('./duration').text),
305 'upload_date': upload_date,
306 'thumbnail': thumbnail,