gh-99064: Added warnings for file objects with non-UTF8 content

Doc/library/xml.etree.elementtree.rst

@@ -665,6 +665,11 @@ Functions
    not given, the standard :class:`XMLParser` parser is used.  Returns an
    :class:`ElementTree` instance.
 
+   .. note::
+
+      When the *source* encoding is ``ISO-8859-1`` and the mode is ``r``,
+      a warning is emited.
+
 
 .. function:: ProcessingInstruction(target, text=None)
 

Lib/test/test_xml_etree.py

@@ -464,6 +464,13 @@ def test_makeelement(self):
         elem[:] = tuple([subelem])
         self.serialize_check(elem, '<tag><subtag key="value" /></tag>')
 
+    def test_parse_encoding_warn(self):
+        with self.assertWarns(RuntimeWarning) as cm:
+            with open(SIMPLE_XMLFILE, 'r', encoding='ISO-8859-1') as fp:
+                ET.parse(fp)
+                self.assertIn("ISO-8859-1 encoding should be read using 'rb' mode.",
+                              str(cm.warnings[0].message))
+
     def test_parsefile(self):
         # Test parsing from file.
 

Lib/xml/etree/ElementTree.py

@@ -1200,6 +1200,9 @@ def parse(source, parser=None):
     Return an ElementTree instance.
 
     """
+    if getattr(source, 'encoding', None) == 'ISO-8859-1' and source.mode == 'r':
+        import warnings
+        warnings.warn("ISO-8859-1 encoding should be read using 'rb' mode.", category=RuntimeWarning)
     tree = ElementTree()
     tree.parse(source, parser)
     return tree

Misc/NEWS.d/next/Library/2024-09-10-11-19-05.gh-issue-99064.iHNbcr.rst

@@ -0,0 +1,2 @@
+Added a warning in the :func:`xml.etree.ElementTree.parse` for
+files opened in ``r`` mode with ``'ISO-8859-1'`` encoding. Patch by RUANG.