From b6497f65de36d1fdca8734c4af14141dc455a54d Mon Sep 17 00:00:00 2001
From: jvoisin <julien.voisin@dustri.org>
Date: Sun, 15 Dec 2019 14:18:50 +0100
Subject: [PATCH] Improve the robustness of the HTML parser
---
libmat2/web.py | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/libmat2/web.py b/libmat2/web.py
index b770200..75633a3 100644
--- a/libmat2/web.py
+++ b/libmat2/web.py
@@ -96,6 +96,15 @@ class _HTMLParser(parser.HTMLParser):
self.tag_required_blocklist = required_blocklisted_tags
self.tag_blocklist = blocklisted_tags
+ def error(self, msg):
+ """ Amusingly, Python's documentation doesn't mention that this
+ function needs to be implemented in subclasses of the parent class
+ of parser.HTMLParser. This was found by fuzzing,
+ triggering the following exception:
+ NotImplementedError: subclasses of ParserBase must override error()
+ """
+ raise ValueError(msg)
+
def handle_starttag(self, tag: str, attrs: List[Tuple[str, Optional[str]]]):
# Ignore the type, because mypy is too stupid to infer
# that get_starttag_text() can't return None.
--
GitLab