[extractor/common] Extract timestamp from Last-Modified header

youtube_dl/extractor/common.py

@@ -2513,16 +2513,18 @@ class InfoExtractor(object):
         # amp-video and amp-audio are very similar to their HTML5 counterparts
         # so we wll include them right here (see
         # https://www.ampproject.org/docs/reference/components/amp-video)
+        # For dl8-* tags see https://delight-vr.com/documentation/dl8-video/
+        _MEDIA_TAG_NAME_RE = r'(?:(?:amp|dl8(?:-live)?)-)?(video|audio)'
         media_tags = [(media_tag, media_type, '')
                       for media_tag, media_type
-                      in re.findall(r'(?s)(<(?:amp-)?(video|audio)[^>]*/>)', webpage)]
+                      in re.findall(r'(?s)(<%s[^>]*/>)' % _MEDIA_TAG_NAME_RE, webpage)]
         media_tags.extend(re.findall(
             # We only allow video|audio followed by a whitespace or '>'.
             # Allowing more characters may end up in significant slow down (see
             # https://github.com/ytdl-org/youtube-dl/issues/11979, example URL:
             # http://www.porntrex.com/maps/videositemap.xml).
-            r'(?s)(<(?P<tag>(?:amp-)?(?:video|audio))(?:\s+[^>]*)?>)(.*?)</(?P=tag)>', webpage))
-        for media_tag, media_type, media_content in media_tags:
+            r'(?s)(<(?P<tag>%s)(?:\s+[^>]*)?>)(.*?)</(?P=tag)>' % _MEDIA_TAG_NAME_RE, webpage))
+        for media_tag, _, media_type, media_content in media_tags:
             media_info = {
                 'formats': [],
                 'subtitles': {},

youtube_dl/extractor/generic.py

@@ -2360,7 +2360,7 @@ class GenericIE(InfoExtractor):
         info_dict = {
             'id': video_id,
             'title': self._generic_title(url),
-            'upload_date': unified_strdate(head_response.headers.get('Last-Modified'))
+            'timestamp': unified_timestamp(head_response.headers.get('Last-Modified'))
         }
 
         # Check for direct link to a video
@@ -2466,7 +2466,9 @@ class GenericIE(InfoExtractor):
         # Sometimes embedded video player is hidden behind percent encoding
         # (e.g. https://github.com/ytdl-org/youtube-dl/issues/2448)
         # Unescaping the whole page allows to handle those cases in a generic way
-        webpage = compat_urllib_parse_unquote(webpage)
+        # FIXME: unescaping the whole page may break URLs, commenting out for now.
+        # There probably should be a second run of generic extractor on unescaped webpage.
+        # 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