From f98709af3100b562a053331fd52612acc99e1d6d Mon Sep 17 00:00:00 2001
From: Lauren Liberda <laura@selfisekai.rocks>
Date: Sat, 23 Oct 2021 04:02:23 +0200
Subject: [PATCH] [extractor] Add `_search_nextjs_data` (#1386) Authored by:
selfisekai
---
yt_dlp/extractor/common.py | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/yt_dlp/extractor/common.py b/yt_dlp/extractor/common.py
index 7500402fa..5c6e59901 100644
--- a/yt_dlp/extractor/common.py
+++ b/yt_dlp/extractor/common.py
@@ -1496,6 +1496,13 @@ class InfoExtractor(object):
break
return dict((k, v) for k, v in info.items() if v is not None)
+ def _search_nextjs_data(self, webpage, video_id, **kw):
+ return self._parse_json(
+ self._search_regex(
+ r'(?s)<script[^>]+id=[\'"]__NEXT_DATA__[\'"][^>]*>([^<]+)</script>',
+ webpage, 'next.js data', **kw),
+ video_id, **kw)
+
@staticmethod
def _hidden_inputs(html):
html = re.sub(r'<!--(?:(?!<!--).)*-->', '', html)