+ }, {
+ # From http://www.nbc.com/the-blacklist/video/sir-crispin-crandall/2928790?onid=137781#vc137781=1
+ # geo-restricted (US), HLS encrypted with AES-128
+ 'url': 'http://player.theplatform.com/p/NnzsPC/onsite_universal/select/media/guid/2410887629/2928790?fwsitesection=nbc_the_blacklist_video_library&autoPlay=true&carouselID=137781',
+ 'only_matching': True,
+ }]
+
+ @classmethod
+ def _extract_urls(cls, webpage):
+ m = re.search(
+ r'''(?x)
+ <meta\s+
+ property=(["'])(?:og:video(?::(?:secure_)?url)?|twitter:player)\1\s+
+ content=(["'])(?P<url>https?://player\.theplatform\.com/p/.+?)\2
+ ''', webpage)
+ if m:
+ return [m.group('url')]
+
+ # Are whitesapces ignored in URLs?
+ # https://github.com/ytdl-org/youtube-dl/issues/12044
+ matches = re.findall(
+ r'(?s)<(?:iframe|script)[^>]+src=(["\'])((?:https?:)?//player\.theplatform\.com/p/.+?)\1', webpage)
+ if matches:
+ return [re.sub(r'\s', '', list(zip(*matches))[1][0])]
+
+ @staticmethod
+ def _sign_url(url, sig_key, sig_secret, life=600, include_qs=False):
+ flags = '10' if include_qs else '00'
+ expiration_date = '%x' % (int(time.time()) + life)
+
+ def str_to_hex(str):
+ return binascii.b2a_hex(str.encode('ascii')).decode('ascii')
+
+ def hex_to_bytes(hex):
+ return binascii.a2b_hex(hex.encode('ascii'))
+
+ relative_path = re.match(r'https?://link\.theplatform\.com/s/([^?]+)', url).group(1)
+ clear_text = hex_to_bytes(flags + expiration_date + str_to_hex(relative_path))
+ checksum = hmac.new(sig_key.encode('ascii'), clear_text, hashlib.sha1).hexdigest()
+ sig = flags + expiration_date + checksum + str_to_hex(sig_secret)
+ return '%s&sig=%s' % (url, sig)
+
+ def _real_extract(self, url):
+ url, smuggled_data = unsmuggle_url(url, {})
+
+ mobj = re.match(self._VALID_URL, url)
+ provider_id = mobj.group('provider_id')
+ video_id = mobj.group('id')
+
+ if not provider_id:
+ provider_id = 'dJ5BDC'
+
+ path = provider_id + '/'
+ if mobj.group('media'):
+ path += mobj.group('media')
+ path += video_id
+
+ qs_dict = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
+ if 'guid' in qs_dict:
+ webpage = self._download_webpage(url, video_id)
+ scripts = re.findall(r'<script[^>]+src="([^"]+)"', webpage)
+ feed_id = None
+ # feed id usually locates in the last script.
+ # Seems there's no pattern for the interested script filename, so
+ # I try one by one
+ for script in reversed(scripts):
+ feed_script = self._download_webpage(
+ self._proto_relative_url(script, 'http:'),
+ video_id, 'Downloading feed script')
+ feed_id = self._search_regex(
+ r'defaultFeedId\s*:\s*"([^"]+)"', feed_script,
+ 'default feed id', default=None)
+ if feed_id is not None:
+ break
+ if feed_id is None:
+ raise ExtractorError('Unable to find feed id')
+ return self.url_result('http://feed.theplatform.com/f/%s/%s?byGuid=%s' % (
+ provider_id, feed_id, qs_dict['guid'][0]))
+
+ if smuggled_data.get('force_smil_url', False):
+ smil_url = url
+ # Explicitly specified SMIL (see https://github.com/ytdl-org/youtube-dl/issues/7385)
+ elif '/guid/' in url:
+ headers = {}
+ source_url = smuggled_data.get('source_url')
+ if source_url:
+ headers['Referer'] = source_url
+ request = sanitized_Request(url, headers=headers)
+ webpage = self._download_webpage(request, video_id)
+ smil_url = self._search_regex(
+ r'<link[^>]+href=(["\'])(?P<url>.+?)\1[^>]+type=["\']application/smil\+xml',
+ webpage, 'smil url', group='url')
+ path = self._search_regex(
+ r'link\.theplatform\.com/s/((?:[^/?#&]+/)+[^/?#&]+)', smil_url, 'path')
+ smil_url += '?' if '?' not in smil_url else '&' + 'formats=m3u,mpeg4'
+ elif mobj.group('config'):
+ config_url = url + '&form=json'
+ config_url = config_url.replace('swf/', 'config/')
+ config_url = config_url.replace('onsite/', 'onsite/config/')
+ config = self._download_json(config_url, video_id, 'Downloading config')
+ if 'releaseUrl' in config:
+ release_url = config['releaseUrl']
+ else:
+ release_url = 'http://link.theplatform.com/s/%s?mbr=true' % path
+ smil_url = release_url + '&formats=MPEG4&manifest=f4m'
Rapsys Git / youtubedl / blobdiff