Update upstream source from tag 'upstream/2019.09.28'

[youtubedl] / youtube_dl / extractor / generic.py

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 77e2174609d817fa2934a2c56fb11cf22bd9193f..d1725d98b0c63f72031068a001ea150b54fd23ec 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -2075,6 +2075,22 @@ class GenericIE(InfoExtractor):
             },
             'playlist_count': 6,
         },
+        {
+            # Squarespace video embed, 2019-08-28
+            'url': 'http://ootboxford.com',
+            'info_dict': {
+                'id': 'Tc7b_JGdZfw',
+                'title': 'Out of the Blue, at Childish Things 10',
+                'ext': 'mp4',
+                'description': 'md5:a83d0026666cf5ee970f8bd1cfd69c7f',
+                'uploader_id': 'helendouglashouse',
+                'uploader': 'Helen & Douglas House',
+                'upload_date': '20140328',
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
         {
             # Zype embed
             'url': 'https://www.cookscountry.com/episode/554-smoky-barbecue-favorites',
@@ -2226,7 +2242,7 @@ class GenericIE(InfoExtractor):
                 default_search = 'fixup_error'
 
             if default_search in ('auto', 'auto_warning', 'fixup_error'):
-                if '/' in url:
+                if re.match(r'^[^\s/]+\.[^\s/]+/', url):
                     self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
                     return self.url_result('http://' + url)
                 elif default_search != 'fixup_error':
@@ -2395,6 +2411,12 @@ class GenericIE(InfoExtractor):
         # Unescaping the whole page allows to handle those cases in a generic way
         webpage = compat_urllib_parse_unquote(webpage)
 
+        # Unescape squarespace embeds to be detected by generic extractor,
+        # see https://github.com/ytdl-org/youtube-dl/issues/21294
+        webpage = re.sub(
+            r'<div[^>]+class=[^>]*?\bsqs-video-wrapper\b[^>]*>',
+            lambda x: unescapeHTML(x.group(0)), webpage)
+
         # it's tempting to parse this further, but you would
         # have to take into account all the variations like
         #   Video Title - Site Name