Merge 'ytdl-org/youtube-dl/master' release 2020.11.19

[yt-dlp.git] / youtube_dlc / extractor / mtv.py

diff --git a/youtube_dlc/extractor/mtv.py b/youtube_dlc/extractor/mtv.py
index e545a9ef3bd1ccd094c05f654e781201d87b0d81..d31f53137a84f2fcde19b5b2f03e0768e65d89f1 100644 (file)
--- a/youtube_dlc/extractor/mtv.py
+++ b/youtube_dlc/extractor/mtv.py
@@ -45,7 +45,7 @@ def _remove_template_parameter(url):
         # Remove the templates, like &device={device}
         return re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', url)
 
-    def _get_feed_url(self, uri):
+    def _get_feed_url(self, uri, url=None):
         return self._FEED_URL
 
     def _get_thumbnail_url(self, uri, itemdoc):
@@ -211,9 +211,9 @@ def _get_feed_query(self, uri):
             data['lang'] = self._LANG
         return data
 
-    def _get_videos_info(self, uri, use_hls=True):
+    def _get_videos_info(self, uri, use_hls=True, url=None):
         video_id = self._id_from_uri(uri)
-        feed_url = self._get_feed_url(uri)
+        feed_url = self._get_feed_url(uri, url)
         info_url = update_url_query(feed_url, self._get_feed_query(uri))
         return self._get_videos_info_from_url(info_url, video_id, use_hls)
 
@@ -256,7 +256,6 @@ def _extract_triforce_mgid(self, webpage, data_zone=None, video_id=None):
         return try_get(feed, lambda x: x['result']['data']['id'], compat_str)
 
     def _extract_new_triforce_mgid(self, webpage, url='', video_id=None):
-        # print(compat_urlparse.urlparse(url).netloc)
         if url == '':
             return
         domain = get_domain(url)
@@ -281,7 +280,7 @@ def _extract_new_triforce_mgid(self, webpage, url='', video_id=None):
 
         item_id = try_get(manifest, lambda x: x['manifest']['reporting']['itemId'], compat_str)
         if not item_id:
-            self.to_screen('Found no id!')
+            self.to_screen('No id found!')
             return
 
         # 'episode' can be anything. 'content' is used often as well
@@ -290,7 +289,7 @@ def _extract_new_triforce_mgid(self, webpage, url='', video_id=None):
 
         return mgid
 
-    def _extract_mgid(self, webpage, url, data_zone=None):
+    def _extract_mgid(self, webpage, url, title=None, data_zone=None):
         try:
             # the url can be http://media.mtvnservices.com/fb/{mgid}.swf
             # or http://media.mtvnservices.com/{mgid}
@@ -301,6 +300,21 @@ def _extract_mgid(self, webpage, url, data_zone=None):
         except RegexNotFoundError:
             mgid = None
 
+        if not title:
+            title = url_basename(url)
+
+        try:
+            window_data = self._parse_json(self._search_regex(
+                r'(?s)window.__DATA__ = (?P<json>{.+});', webpage,
+                'JSON Window Data', default=None, fatal=False, group='json'), title, fatal=False)
+            main_container = None
+            for i in range(len(window_data['children'])):
+                if window_data['children'][i]['type'] == 'MainContainer':
+                    main_container = window_data['children'][i]
+            mgid = main_container['children'][0]['props']['media']['video']['config']['uri']
+        except (KeyError, IndexError, TypeError):
+            pass
+
         if mgid is None or ':' not in mgid:
             mgid = self._search_regex(
                 [r'data-mgid="(.*?)"', r'swfobject\.embedSWF\(".*?(mgid:.*?)"'],
@@ -323,8 +337,8 @@ def _extract_mgid(self, webpage, url, data_zone=None):
     def _real_extract(self, url):
         title = url_basename(url)
         webpage = self._download_webpage(url, title)
-        mgid = self._extract_mgid(webpage, url)
-        videos_info = self._get_videos_info(mgid)
+        mgid = self._extract_mgid(webpage, url, title=title)
+        videos_info = self._get_videos_info(mgid, url=url)
         return videos_info
 
 
@@ -389,6 +403,18 @@ class MTVIE(MTVServicesInfoExtractor):
         'only_matching': True,
     }]
 
+    @staticmethod
+    def extract_child_with_type(parent, t):
+        children = parent['children']
+        return next(c for c in children if c.get('type') == t)
+
+    def _extract_mgid(self, webpage):
+        data = self._parse_json(self._search_regex(
+            r'__DATA__\s*=\s*({.+?});', webpage, 'data'), None)
+        main_container = self.extract_child_with_type(data, 'MainContainer')
+        video_player = self.extract_child_with_type(main_container, 'VideoPlayer')
+        return video_player['props']['media']['video']['config']['uri']
+
 
 class MTVJapanIE(MTVServicesInfoExtractor):
     IE_NAME = 'mtvjapan'