+ # 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):
+ patterns_templates = (r'arte_vp_url=["\'](.*?%s.*?)["\']', r'data-url=["\']([^"]+%s[^"]+)["\']')
+ ids = (video_id, '')
+ # some pages contain multiple videos (like
+ # http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D),
+ # so we first try to look for json URLs that contain the video id from
+ # the 'vid' parameter.
+ patterns = [t % re.escape(_id) for _id in ids for t in patterns_templates]
+ json_url = self._html_search_regex(
+ patterns, webpage, 'json vp url', default=None)
+ if not json_url:
+ def find_iframe_url(webpage, default=NO_DEFAULT):
+ return self._html_search_regex(
+ r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1',
+ webpage, 'iframe url', group='url', default=default)
+
+ iframe_url = find_iframe_url(webpage, None)
+ if not iframe_url:
+ embed_url = self._html_search_regex(
+ r'arte_vp_url_oembed=\'([^\']+?)\'', webpage, 'embed url', default=None)
+ if embed_url:
+ player = self._download_json(
+ embed_url, video_id, 'Downloading player page')
+ iframe_url = find_iframe_url(player['html'])
+ # en and es URLs produce react-based pages with different layout (e.g.
+ # http://www.arte.tv/guide/en/053330-002-A/carnival-italy?zone=world)
+ if not iframe_url:
+ program = self._search_regex(
+ r'program\s*:\s*({.+?["\']embed_html["\'].+?}),?\s*\n',
+ webpage, 'program', default=None)
+ if program:
+ embed_html = self._parse_json(program, video_id)
+ if embed_html:
+ iframe_url = find_iframe_url(embed_html['embed_html'])
+ if iframe_url:
+ json_url = compat_parse_qs(
+ compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0]
+ if json_url:
+ return self._extract_from_json_url(json_url, video_id, lang)
+ # Differend kind of embed URL (e.g.
+ # http://www.arte.tv/magazine/trepalium/fr/episode-0406-replay-trepalium)
+ embed_url = self._search_regex(
+ r'<iframe[^>]+src=(["\'])(?P<url>.+?)\1',
+ webpage, 'embed url', group='url')
+ return self.url_result(embed_url)
+
+ def _extract_from_json_url(self, json_url, video_id, lang):
+ info = self._download_json(json_url, video_id)
+ player_info = info['videoJsonPlayer']
+
+ upload_date_str = player_info.get('shootingDate')
+ if not upload_date_str:
+ upload_date_str = (player_info.get('VRA') or player_info.get('VDA') or '').split(' ')[0]
+
+ title = player_info['VTI'].strip()
+ subtitle = player_info.get('VSU', '').strip()
+ if subtitle:
+ title += ' - %s' % subtitle
+
+ info_dict = {
+ 'id': player_info['VID'],
+ 'title': title,
+ 'description': player_info.get('VDE'),
+ 'upload_date': unified_strdate(upload_date_str),
+ 'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
+ }
+ qfunc = qualities(['HQ', 'MQ', 'EQ', 'SQ'])
+
+ LANGS = {
+ 'fr': 'F',
+ 'de': 'A',
+ 'en': 'E[ANG]',
+ 'es': 'E[ESP]',
+ }
+
+ formats = []
+ for format_id, format_dict in player_info['VSR'].items():
+ f = dict(format_dict)
+ versionCode = f.get('versionCode')
+ langcode = LANGS.get(lang, lang)
+ lang_rexs = [r'VO?%s-' % re.escape(langcode), r'VO?.-ST%s$' % re.escape(langcode)]
+ lang_pref = None
+ if versionCode:
+ matched_lang_rexs = [r for r in lang_rexs if re.match(r, versionCode)]
+ lang_pref = -10 if not matched_lang_rexs else 10 * len(matched_lang_rexs)
+ source_pref = 0
+ if versionCode is not None:
+ # The original version with subtitles has lower relevance
+ if re.match(r'VO-ST(F|A|E)', versionCode):
+ source_pref -= 10
+ # The version with sourds/mal subtitles has also lower relevance
+ elif re.match(r'VO?(F|A|E)-STM\1', versionCode):
+ source_pref -= 9
+ format = {
+ 'format_id': format_id,
+ 'preference': -10 if f.get('videoFormat') == 'M3U8' else None,
+ 'language_preference': lang_pref,
+ 'format_note': '%s, %s' % (f.get('versionCode'), f.get('versionLibelle')),
+ 'width': int_or_none(f.get('width')),
+ 'height': int_or_none(f.get('height')),
+ 'tbr': int_or_none(f.get('bitrate')),
+ 'quality': qfunc(f.get('quality')),
+ 'source_preference': source_pref,
+ }
+
+ if f.get('mediaType') == 'rtmp':
+ format['url'] = f['streamer']
+ format['play_path'] = 'mp4:' + f['url']
+ format['ext'] = 'flv'
+ else:
+ format['url'] = f['url']
+
+ formats.append(format)
+
+ self._check_formats(formats, video_id)
+ self._sort_formats(formats)
+
+ info_dict['formats'] = formats
+ return info_dict
+
+
+# It also uses the arte_vp_url url from the webpage to extract the information
+class ArteTVCreativeIE(ArteTVPlus7IE):
+ IE_NAME = 'arte.tv:creative'
+ _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de|en|es)/(?:magazine?/)?(?P<id>[^/?#&]+)'
+
+ _TESTS = [{
+ 'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
+ 'info_dict': {
+ 'id': '72176',
+ 'ext': 'mp4',
+ 'title': 'Folge 2 - Corporate Design',
+ 'upload_date': '20131004',
+ },
+ }, {
+ 'url': 'http://creative.arte.tv/fr/Monty-Python-Reunion',
+ 'info_dict': {
+ 'id': '160676',
+ 'ext': 'mp4',
+ 'title': 'Monty Python live (mostly)',
+ 'description': 'Événement ! Quarante-cinq ans après leurs premiers succès, les légendaires Monty Python remontent sur scène.\n',
+ 'upload_date': '20140805',
+ }
+ }]
+
+
+class ArteTVFutureIE(ArteTVPlus7IE):
+ IE_NAME = 'arte.tv:future'
+ _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)'
+
+ _TESTS = [{
+ 'url': 'http://future.arte.tv/fr/info-sciences/les-ecrevisses-aussi-sont-anxieuses',
+ 'info_dict': {
+ 'id': '050940-028-A',
+ 'ext': 'mp4',
+ 'title': 'Les écrevisses aussi peuvent être anxieuses',
+ 'upload_date': '20140902',
+ },
+ }, {
+ 'url': 'http://future.arte.tv/fr/la-science-est-elle-responsable',
+ 'only_matching': True,
+ }]
+
+
+class ArteTVDDCIE(ArteTVPlus7IE):
+ IE_NAME = 'arte.tv:ddc'
+ _VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>[^/?#&]+)'
+
+ def _real_extract(self, url):
+ video_id, lang = self._extract_url_info(url)
+ if lang == 'folge':
+ lang = 'de'
+ elif lang == 'emission':
+ lang = 'fr'
+ webpage = self._download_webpage(url, video_id)
+ scriptElement = get_element_by_attribute('class', 'visu_video_block', webpage)
+ script_url = self._html_search_regex(r'src="(.*?)"', scriptElement, 'script url')
+ javascriptPlayerGenerator = self._download_webpage(script_url, video_id, 'Download javascript player generator')
+ json_url = self._search_regex(r"json_url=(.*)&rendering_place.*", javascriptPlayerGenerator, 'json url')
+ return self._extract_from_json_url(json_url, video_id, lang)
+
+
+class ArteTVConcertIE(ArteTVPlus7IE):
+ IE_NAME = 'arte.tv:concert'
+ _VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)'
+
+ _TEST = {
+ 'url': 'http://concert.arte.tv/de/notwist-im-pariser-konzertclub-divan-du-monde',
+ 'md5': '9ea035b7bd69696b67aa2ccaaa218161',
+ 'info_dict': {
+ 'id': '186',
+ 'ext': 'mp4',
+ 'title': 'The Notwist im Pariser Konzertclub "Divan du Monde"',
+ 'upload_date': '20140128',
+ 'description': 'md5:486eb08f991552ade77439fe6d82c305',
+ },
+ }
+
+
+class ArteTVCinemaIE(ArteTVPlus7IE):
+ IE_NAME = 'arte.tv:cinema'
+ _VALID_URL = r'https?://cinema\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>.+)'
+
+ _TEST = {
+ 'url': 'http://cinema.arte.tv/de/node/38291',
+ 'md5': '6b275511a5107c60bacbeeda368c3aa1',
+ 'info_dict': {
+ 'id': '055876-000_PWA12025-D',
+ 'ext': 'mp4',
+ 'title': 'Tod auf dem Nil',
+ 'upload_date': '20160122',
+ 'description': 'md5:7f749bbb77d800ef2be11d54529b96bc',
+ },
+ }
Rapsys Git / youtubedl / blobdiff