[mixcloud] Extract decrypt key

[youtube-dl] / youtube_dl / extractor / mixcloud.py

diff --git a/youtube_dl/extractor/mixcloud.py b/youtube_dl/extractor/mixcloud.py
index bc38a86638f025dad1a24b5a0cdbd284fc4c3848..fcf7beeb25af5e5b9394e448519a891ce46d1f2d 100644 (file)
--- a/youtube_dl/extractor/mixcloud.py
+++ b/youtube_dl/extractor/mixcloud.py
@@ -9,6 +9,7 @@ from .common import InfoExtractor
 from ..compat import (
     compat_chr,
     compat_ord,
+    compat_str,
     compat_urllib_parse_unquote,
     compat_urlparse,
 )
@@ -53,16 +54,27 @@ class MixcloudIE(InfoExtractor):
         'only_matching': True,
     }]
 
-    # See https://www.mixcloud.com/media/js2/www_js_2.9e23256562c080482435196ca3975ab5.js
-    @staticmethod
-    def _decrypt_play_info(play_info):
-        KEY = 'pleasedontdownloadourmusictheartistswontgetpaid'
+    _keys = [
+        'return { requestAnimationFrame: function(callback) { callback(); }, innerHeight: 500 };',
+        'pleasedontdownloadourmusictheartistswontgetpaid',
+        'window.addEventListener = window.addEventListener || function() {};',
+        '(function() { return new Date().toLocaleDateString(); })()'
+    ]
+    _current_key = None
 
+    # See https://www.mixcloud.com/media/js2/www_js_2.9e23256562c080482435196ca3975ab5.js
+    def _decrypt_play_info(self, play_info, video_id):
         play_info = base64.b64decode(play_info.encode('ascii'))
-
-        return ''.join([
-            compat_chr(compat_ord(ch) ^ compat_ord(KEY[idx % len(KEY)]))
-            for idx, ch in enumerate(play_info)])
+        for num, key in enumerate(self._keys, start=1):
+            try:
+                return self._parse_json(
+                    ''.join([
+                        compat_chr(compat_ord(ch) ^ compat_ord(key[idx % len(key)]))
+                        for idx, ch in enumerate(play_info)]),
+                    video_id)
+            except ExtractorError:
+                if num == len(self._keys):
+                    raise
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -72,14 +84,28 @@ class MixcloudIE(InfoExtractor):
 
         webpage = self._download_webpage(url, track_id)
 
+        if not self._current_key:
+            js_url = self._search_regex(
+                r'<script[^>]+\bsrc=["\"](https://(?:www\.)?mixcloud\.com/media/js2/www_js_4\.[^>]+\.js)',
+                webpage, 'js url', default=None)
+            if js_url:
+                js = self._download_webpage(js_url, track_id, fatal=False)
+                if js:
+                    key = self._search_regex(
+                        r'player\s*:\s*{.*?\bvalue\s*:\s*(["\'])(?P<key>(?:(?!\1).)+)\1',
+                        js, 'key', default=None, group='key')
+                    if key and isinstance(key, compat_str):
+                        self._keys.insert(0, key)
+                        self._current_key = key
+
         message = self._html_search_regex(
             r'(?s)<div[^>]+class="global-message cloudcast-disabled-notice-light"[^>]*>(.+?)<(?:a|/div)',
             webpage, 'error message', default=None)
 
         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)
+
+        play_info = 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)
@@ -97,7 +123,7 @@ class MixcloudIE(InfoExtractor):
         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'],
+             r'(?:m|data)-tooltip=["\']([\d,.]+) plays'],
             webpage, 'play count', default=None))
 
         return {
@@ -143,7 +169,7 @@ class MixcloudPlaylistBaseIE(InfoExtractor):
 
 
 class MixcloudUserIE(MixcloudPlaylistBaseIE):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/(?P<type>uploads|favorites|listens)?/?$'
+    _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/(?P<type>uploads|favorites|listens)?/?$'
     IE_NAME = 'mixcloud:user'
 
     _TESTS = [{
@@ -216,7 +242,7 @@ class MixcloudUserIE(MixcloudPlaylistBaseIE):
 
 
 class MixcloudPlaylistIE(MixcloudPlaylistBaseIE):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/playlists/(?P<playlist>[^/]+)/?$'
+    _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/playlists/(?P<playlist>[^/]+)/?$'
     IE_NAME = 'mixcloud:playlist'
 
     _TESTS = [{
@@ -259,7 +285,7 @@ class MixcloudPlaylistIE(MixcloudPlaylistBaseIE):
 
 
 class MixcloudStreamIE(MixcloudPlaylistBaseIE):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P<id>[^/]+)/stream/?$'
+    _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<id>[^/]+)/stream/?$'
     IE_NAME = 'mixcloud:stream'
 
     _TEST = {