From 43496ec3cd21113f5332bb2664d9f9d37c40270b Mon Sep 17 00:00:00 2001
From: Simon Cross <hodgestar@gmail.com>
Date: Sun, 25 Aug 2024 13:06:35 +0200
Subject: [PATCH] Fix HTMLParser error handling. (#87)

* Fix HTMLParser error handling which referenced Python's html.HTMLParseError which was never raised and removed in Python 3.5.

* Don't use assertRaisesRegex on Python 2.

* Fix typo in HTMLParser error handling comment.
---
 genshi/input.py            | 10 +++++++---
 genshi/tests/test_input.py | 16 +++++++++++++++-
 2 files changed, 22 insertions(+), 4 deletions(-)

diff --git a/genshi/input.py b/genshi/input.py
index fa18c38..c21990a 100644
--- a/genshi/input.py
+++ b/genshi/input.py
@@ -346,9 +346,13 @@ def _generate():
                         for tag in open_tags:
                             yield END, QName(tag), pos
                         break
-            except html.HTMLParseError as e:
-                msg = '%s: line %d, column %d' % (e.msg, e.lineno, e.offset)
-                raise ParseError(msg, self.filename, e.lineno, e.offset)
+            except Exception as e:
+                # Python's simple HTMLParser does not raise detailed
+                # errors except in strict mode which was deprecated
+                # in Python 3.3 and removed in Python 3.5 and which in
+                # any case is not used is this code.
+                msg = str(e)
+                raise ParseError(msg, self.filename)
         return Stream(_generate()).filter(_coalesce)
 
     def __iter__(self):
diff --git a/genshi/tests/test_input.py b/genshi/tests/test_input.py
index 44b7442..e68515d 100644
--- a/genshi/tests/test_input.py
+++ b/genshi/tests/test_input.py
@@ -15,7 +15,7 @@
 
 from genshi.core import Attrs, QName, Stream
 from genshi.input import XMLParser, HTMLParser, ParseError, ET
-from genshi.compat import StringIO, BytesIO
+from genshi.compat import IS_PYTHON2, StringIO, BytesIO
 from genshi.tests.utils import doctest_suite
 from xml.etree import ElementTree
 
@@ -294,6 +294,20 @@ def test_convert_ElementTree_to_markup_stream(self):
         self.assertEqual((Stream.END, QName("span")), events[4][:2])
         self.assertEqual((Stream.END, QName("div")), events[5][:2])
 
+    def test_parsing_error(self):
+        text = u'<div></div>'.encode('utf-8')
+        events = HTMLParser(BytesIO(text))
+        if IS_PYTHON2:
+            self.assertRaises(ParseError, list, events)
+        else:
+            self.assertRaisesRegex(
+                ParseError,
+                r"source returned bytes, but no encoding specified",
+                list,
+                events,
+            )
+
+
 def suite():
     suite = unittest.TestSuite()
     suite.addTest(doctest_suite(XMLParser.__module__))