-class ArteTvIE(InfoExtractor):
- _VALID_URL = r'http://videos\.arte\.tv/(?P<lang>fr|de|en|es)/.*-(?P<id>.*?)\.html'
- IE_NAME = 'arte.tv'
-
- def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
- lang = mobj.group('lang')
- video_id = mobj.group('id')
-
- 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='Downloading metadata')
- config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)
- config_xml_url = config_node.attrib['ref']
- config = self._download_xml(
- config_xml_url, video_id, note='Downloading configuration')
-
- formats = [{
- 'format_id': q.attrib['quality'],
- # The playpath starts at 'mp4:', if we don't manually
- # split the url, rtmpdump will incorrectly parse them
- 'url': q.text.split('mp4:', 1)[0],
- 'play_path': 'mp4:' + q.text.split('mp4:', 1)[1],
- 'ext': 'flv',
- 'quality': 2 if q.attrib['quality'] == 'hd' else 1,
- } for q in config.findall('./urls/url')]
- self._sort_formats(formats)
-
- title = config.find('.//name').text
- thumbnail = config.find('.//firstThumbnailUrl').text
- return {
- 'id': video_id,
- 'title': title,
- 'thumbnail': thumbnail,
- 'formats': formats,
- }
-
-
-class ArteTVPlus7IE(InfoExtractor):
- IE_NAME = 'arte.tv:+7'
- _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/(?:(?:sendungen|emissions|embed)/)?(?P<id>[^/]+)/(?P<name>[^/?#&+])'
-
- @classmethod
- def _extract_url_info(cls, url):
- mobj = re.match(cls._VALID_URL, url)
- lang = mobj.group('lang')
- query = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
- if 'vid' in query:
- video_id = query['vid'][0]
- else:
- # 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):
Rapsys Git / youtubedl / blobdiff