- encrypted_play_info = self._search_regex(
- r'm-play-info="([^"]+)"', webpage, 'play info')
- play_info = self._parse_json(
- self._decrypt_play_info(encrypted_play_info), track_id)
-
- if message and 'stream_url' not in play_info:
- raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
-
- song_url = play_info['stream_url']
-
- title = self._html_search_regex(r'm-title="([^"]+)"', webpage, 'title')
- thumbnail = self._proto_relative_url(self._html_search_regex(
- r'm-thumbnail-url="([^"]+)"', webpage, 'thumbnail', fatal=False))
- uploader = self._html_search_regex(
- r'm-owner-name="([^"]+)"', webpage, 'uploader', fatal=False)
- uploader_id = self._search_regex(
- r'\s+"profile": "([^"]+)",', webpage, 'uploader id', fatal=False)
- description = self._og_search_description(webpage)
- view_count = str_to_int(self._search_regex(
- [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
- r'/listeners/?">([0-9,.]+)</a>',
- r'm-tooltip=["\']([\d,.]+) plays'],
- webpage, 'play count', default=None))
+ js_url = self._search_regex(
+ r'<script[^>]+\bsrc=["\"](https://(?:www\.)?mixcloud\.com/media/(?:js2/www_js_4|js/www)\.[^>]+\.js)',
+ webpage, 'js url')
+ js = self._download_webpage(js_url, track_id, 'Downloading JS')
+ # Known plaintext attack
+ if encrypted_play_info:
+ kps = ['{"stream_url":']
+ kpa_target = encrypted_play_info
+ else:
+ kps = ['https://', 'http://']
+ kpa_target = compat_b64decode(info_json['streamInfo']['url'])
+ for kp in kps:
+ partial_key = self._decrypt_xor_cipher(kpa_target, kp)
+ for quote in ["'", '"']:
+ key = self._search_regex(
+ r'{0}({1}[^{0}]*){0}'.format(quote, re.escape(partial_key)),
+ js, 'encryption key', default=None)
+ if key is not None:
+ break
+ else:
+ continue
+ break
+ else:
+ raise ExtractorError('Failed to extract encryption key')
+
+ if encrypted_play_info is not None:
+ play_info = self._parse_json(self._decrypt_xor_cipher(key, encrypted_play_info), 'play info')
+ if message and 'stream_url' not in play_info:
+ raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
+ song_url = play_info['stream_url']
+ formats = [{
+ 'format_id': 'normal',
+ 'url': song_url
+ }]
+
+ title = self._html_search_regex(r'm-title="([^"]+)"', webpage, 'title')
+ thumbnail = self._proto_relative_url(self._html_search_regex(
+ r'm-thumbnail-url="([^"]+)"', webpage, 'thumbnail', fatal=False))
+ uploader = self._html_search_regex(
+ r'm-owner-name="([^"]+)"', webpage, 'uploader', fatal=False)
+ uploader_id = self._search_regex(
+ r'\s+"profile": "([^"]+)",', webpage, 'uploader id', fatal=False)
+ description = self._og_search_description(webpage)
+ view_count = str_to_int(self._search_regex(
+ [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
+ r'/listeners/?">([0-9,.]+)</a>',
+ r'(?:m|data)-tooltip=["\']([\d,.]+) plays'],
+ webpage, 'play count', default=None))
+
+ else:
+ title = info_json['name']
+ thumbnail = urljoin(
+ 'https://thumbnailer.mixcloud.com/unsafe/600x600/',
+ try_get(info_json, lambda x: x['picture']['urlRoot'], compat_str))
+ uploader = try_get(info_json, lambda x: x['owner']['displayName'])
+ uploader_id = try_get(info_json, lambda x: x['owner']['username'])
+ description = try_get(info_json, lambda x: x['description'])
+ view_count = int_or_none(try_get(info_json, lambda x: x['plays']))
+
+ stream_info = info_json['streamInfo']
+ formats = []
+
+ def decrypt_url(f_url):
+ for k in (key, 'IFYOUWANTTHEARTISTSTOGETPAIDDONOTDOWNLOADFROMMIXCLOUD'):
+ decrypted_url = self._decrypt_xor_cipher(k, f_url)
+ if re.search(r'^https?://[0-9a-z.]+/[0-9A-Za-z/.?=&_-]+$', decrypted_url):
+ return decrypted_url
+
+ for url_key in ('url', 'hlsUrl', 'dashUrl'):
+ format_url = stream_info.get(url_key)
+ if not format_url:
+ continue
+ decrypted = decrypt_url(compat_b64decode(format_url))
+ if not decrypted:
+ continue
+ if url_key == 'hlsUrl':
+ formats.extend(self._extract_m3u8_formats(
+ decrypted, track_id, 'mp4', entry_protocol='m3u8_native',
+ m3u8_id='hls', fatal=False))
+ elif url_key == 'dashUrl':
+ formats.extend(self._extract_mpd_formats(
+ decrypted, track_id, mpd_id='dash', fatal=False))
+ else:
+ formats.append({
+ 'format_id': 'http',
+ 'url': decrypted,
+ 'downloader_options': {
+ # Mixcloud starts throttling at >~5M
+ 'http_chunk_size': 5242880,
+ },
+ })
+ self._sort_formats(formats)
Rapsys Git / youtubedl / blobdiff