+
+ def _extract_video(self, url, video_id, lang):
+ """Extract from videos.arte.tv"""
+ ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
+ ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
+ ref_xml_doc = self._download_xml(ref_xml_url, video_id, note=u'Downloading metadata')
+ config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)
+ config_xml_url = config_node.attrib['ref']
+ config_xml = self._download_webpage(config_xml_url, video_id, note=u'Downloading configuration')
+
+ video_urls = list(re.finditer(r'<url quality="(?P<quality>.*?)">(?P<url>.*?)</url>', config_xml))
+ def _key(m):
+ quality = m.group('quality')
+ if quality == 'hd':
+ return 2
+ else:
+ return 1
+ # We pick the best quality
+ video_urls = sorted(video_urls, key=_key)
+ video_url = list(video_urls)[-1].group('url')
+
+ title = self._html_search_regex(r'<name>(.*?)</name>', config_xml, 'title')
+ thumbnail = self._html_search_regex(r'<firstThumbnailUrl>(.*?)</firstThumbnailUrl>',
+ config_xml, 'thumbnail')
+ return {'id': video_id,
+ 'title': title,
+ 'thumbnail': thumbnail,
+ 'url': video_url,
+ 'ext': 'flv',
+ }
+
+ def _extract_liveweb(self, url, name, lang):
+ """Extract form http://liveweb.arte.tv/"""
+ webpage = self._download_webpage(url, name)
+ video_id = self._search_regex(r'eventId=(\d+?)("|&)', webpage, 'event id')
+ config_doc = self._download_xml('http://download.liveweb.arte.tv/o21/liveweb/events/event-%s.xml' % video_id,
+ video_id, 'Downloading information')
+ event_doc = config_doc.find('event')
+ url_node = event_doc.find('video').find('urlHd')
+ if url_node is None:
+ url_node = event_doc.find('urlSd')
+
+ return {'id': video_id,
+ 'title': event_doc.find('name%s' % lang.capitalize()).text,
+ 'url': url_node.text.replace('MP4', 'mp4'),
+ 'ext': 'flv',
+ 'thumbnail': self._og_search_thumbnail(webpage),
+ }
+
+
+class ArteTVPlus7IE(InfoExtractor):
+ IE_NAME = 'arte.tv:+7'
+ _VALID_URL = r'https?://www\.arte.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
+
+ @classmethod
+ def _extract_url_info(cls, url):
+ mobj = re.match(cls._VALID_URL, url)
+ lang = mobj.group('lang')
+ # This is not a real id, it can be for example AJT for the news
+ # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
+ video_id = mobj.group('id')
+ return video_id, lang
+
+ def _real_extract(self, url):
+ video_id, lang = self._extract_url_info(url)
+ webpage = self._download_webpage(url, video_id)
+ return self._extract_from_webpage(webpage, video_id, lang)
+
+ def _extract_from_webpage(self, webpage, video_id, lang):
+ json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url')
+ return self._extract_from_json_url(json_url, video_id, lang)
+
+ def _extract_from_json_url(self, json_url, video_id, lang):
+ json_info = self._download_webpage(json_url, video_id, 'Downloading info json')
+ self.report_extraction(video_id)
+ info = json.loads(json_info)
+ player_info = info['videoJsonPlayer']
+
+ info_dict = {
+ 'id': player_info['VID'],
+ 'title': player_info['VTI'],
+ 'description': player_info.get('VDE'),
+ 'upload_date': unified_strdate(player_info.get('VDA', '').split(' ')[0]),
+ 'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
+ }
+
+ all_formats = player_info['VSR'].values()
+ # Some formats use the m3u8 protocol
+ all_formats = list(filter(lambda f: f.get('videoFormat') != 'M3U8', all_formats))
+ def _match_lang(f):
+ if f.get('versionCode') is None:
+ return True
+ # Return true if that format is in the language of the url
+ if lang == 'fr':
+ l = 'F'
+ elif lang == 'de':
+ l = 'A'
+ regexes = [r'VO?%s' % l, r'VO?.-ST%s' % l]
+ return any(re.match(r, f['versionCode']) for r in regexes)
+ # Some formats may not be in the same language as the url
+ formats = filter(_match_lang, all_formats)
+ formats = list(formats) # in python3 filter returns an iterator
+ if not formats:
+ # Some videos are only available in the 'Originalversion'
+ # they aren't tagged as being in French or German
+ if all(f['versionCode'] == 'VO' for f in all_formats):
+ formats = all_formats
+ else:
+ raise ExtractorError(u'The formats list is empty')
+
+ if re.match(r'[A-Z]Q', formats[0]['quality']) is not None:
+ def sort_key(f):
+ return ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality'])
Rapsys Git / youtubedl / blobdiff