Merge pull request #4598 from 0xced/encrypted-age-gate

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

diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index 3319d6bff95afeef6222cbbda941b6d71c83134d..175e43272e8875e9efc6f91ae33f88d9f118e0b3 100644 (file)
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -527,8 +527,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 return 's[%s%s%s]' % (starts, ends, steps)
 
             step = None
-            start = '(Never used)'  # Quelch pyflakes warnings - start will be
-                                    # set as soon as step is set
+            # Quelch pyflakes warnings - start will be set when step is set
+            start = '(Never used)'
             for i, prev in zip(idxs[1:], idxs[:-1]):
                 if step is not None:
                     if i - prev == step:
@@ -766,11 +766,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             age_gate = True
             # We simulate the access to the video from www.youtube.com/v/{video_id}
             # this can be viewed without login into Youtube
+            url = proto + '://www.youtube.com/embed/%s' % video_id
+            embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
             data = compat_urllib_parse.urlencode({
                 'video_id': video_id,
                 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
                 'sts': self._search_regex(
-                    r'"sts"\s*:\s*(\d+)', video_webpage, 'sts', default=''),
+                    r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
             })
             video_info_url = proto + '://www.youtube.com/get_video_info?' + data
             video_info_webpage = self._download_webpage(
@@ -968,11 +970,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 elif 's' in url_data:
                     encrypted_sig = url_data['s'][0]
 
-                    if not age_gate:
-                        jsplayer_url_json = self._search_regex(
-                            r'"assets":.+?"js":\s*("[^"]+")',
-                            video_webpage, 'JS player URL')
-                        player_url = json.loads(jsplayer_url_json)
+                    jsplayer_url_json = self._search_regex(
+                        r'"assets":.+?"js":\s*("[^"]+")',
+                        embed_webpage if age_gate else video_webpage, 'JS player URL')
+                    player_url = json.loads(jsplayer_url_json)
                     if player_url is None:
                         player_url_json = self._search_regex(
                             r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
@@ -1128,6 +1129,13 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'info_dict': {
             'title': 'JODA7',
         }
+    }, {
+        'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
+        'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
+        'info_dict': {
+                'title': 'Uploads from Interstellar Movie',
+        },
+        'playlist_mincout': 21,
     }]
 
     def _real_initialize(self):
@@ -1212,6 +1220,10 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
                 'Downloading page #%s' % page_num,
                 transform_source=uppercase_escape)
             content_html = more['content_html']
+            if not content_html.strip():
+                # Some webpages show a "Load more" button but they don't
+                # have more videos
+                break
             more_widget_html = more['load_more_widget_html']
 
         playlist_title = self._html_search_regex(
@@ -1555,9 +1567,11 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
         feed_entries = []
         paging = 0
         for i in itertools.count(1):
-            info = self._download_json(self._FEED_TEMPLATE % paging,
-                                       '%s feed' % self._FEED_NAME,
-                                       'Downloading page %s' % i)
+            info = self._download_json(
+                self._FEED_TEMPLATE % paging,
+                '%s feed' % self._FEED_NAME,
+                'Downloading page %s' % i,
+                transform_source=uppercase_escape)
             feed_html = info.get('feed_html') or info.get('content_html')
             load_more_widget_html = info.get('load_more_widget_html') or feed_html
             m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)