[3.14] gh-149489: Fix ElementTree serialization to HTML (GH-149490) (GH-150596)

* The content of elements "xmp", "iframe", "noembed", "noframes",
  and "plaintext" is no longer escaped.
* The "plaintext" element no longer have the closing tag.

(cherry picked from commit bcd29e466f55d8b4e3849ed6ada8ce86a46f5072)

diff --git a/Lib/test/test_xml_etree.py b/Lib/test/test_xml_etree.py
index eb5ca80b1aaca159de12eaa2735caa737c4c8b21..8c693bfbdb39d924caa6259f322c310814bf585f 100644 (file)
--- a/Lib/test/test_xml_etree.py
+++ b/Lib/test/test_xml_etree.py
@@ -1247,7 +1247,12 @@ class ElementTreeTest(unittest.TestCase):
               {'': 'http://www.w3.org/2001/XMLSchema',
                'ns': 'http://www.w3.org/2001/XMLSchema'})
 
-    def test_processinginstruction(self):
+    def test_comment_serialization(self):
+        comm = ET.Comment('<spam> & ham')
+        # comments are not escaped
+        self.assertEqual(ET.tostring(comm), b'<!--<spam> & ham-->')
+
+    def test_processinginstruction_serialization(self):
         # Test ProcessingInstruction directly
 
         self.assertEqual(ET.tostring(ET.ProcessingInstruction('test', 'instruction')),
@@ -1256,13 +1261,22 @@ class ElementTreeTest(unittest.TestCase):
                 b'<?test instruction?>')
 
         # Issue #2746
-
+        # processing instructions are not escaped
         self.assertEqual(ET.tostring(ET.PI('test', '<testing&>')),
                 b'<?test <testing&>?>')
         self.assertEqual(ET.tostring(ET.PI('test', '<testing&>\xe3'), 'latin-1'),
                 b"<?xml version='1.0' encoding='latin-1'?>\n"
                 b"<?test <testing&>\xe3?>")
 
+    @support.subTests('tag', ("script", "style", "xmp", "iframe", "noembed", "noframes"))
+    def test_html_cdata_elems_serialization(self, tag):
+        # content of raw text elements is not escaped in html
+        tag = tag.title()
+        elem = ET.Element(tag)
+        elem.text = '<spam>&ham'
+        self.assertEqual(ET.tostring(elem, method='html'),
+                         ('<%s><spam>&ham</%s>' % (tag, tag)).encode())
+
     def test_html_empty_elems_serialization(self):
         # issue 15970
         # from http://www.w3.org/TR/html401/index/elements.html
@@ -1277,6 +1291,14 @@ class ElementTreeTest(unittest.TestCase):
                                        method='html')
                 self.assertEqual(serialized, expected)
 
+    def test_html_plaintext_serialization(self):
+        # content of plaintext is not escaped in html
+        # no end tag for plaintext
+        elem = ET.Element('PlainText')
+        elem.text = '<spam>&ham'
+        self.assertEqual(ET.tostring(elem, method='html'),
+                         b'<PlainText><spam>&ham')
+
     def test_dump_attribute_order(self):
         # See BPO 34160
         e = ET.Element('cirriculum', status='public', company='example')

diff --git a/Lib/xml/etree/ElementTree.py b/Lib/xml/etree/ElementTree.py
index 5d8b22ffb62c0dd60f3c60a7c578f343d6c73d0c..0a4203d372ce9914d6319ae2a32e8b535399f9a6 100644 (file)
--- a/Lib/xml/etree/ElementTree.py
+++ b/Lib/xml/etree/ElementTree.py
@@ -917,9 +917,12 @@ def _serialize_xml(write, elem, qnames, namespaces,
     if elem.tail:
         write(_escape_cdata(elem.tail))
 
+_CDATA_CONTENT_ELEMENTS = {"script", "style", "xmp", "iframe", "noembed",
+                           "noframes", "plaintext"}
+
 HTML_EMPTY = {"area", "base", "basefont", "br", "col", "embed", "frame", "hr",
               "img", "input", "isindex", "link", "meta", "param", "source",
-              "track", "wbr"}
+              "track", "wbr", "plaintext"}
 
 def _serialize_html(write, elem, qnames, namespaces, **kwargs):
     tag = elem.tag
@@ -960,7 +963,7 @@ def _serialize_html(write, elem, qnames, namespaces, **kwargs):
             write(">")
             ltag = tag.lower()
             if text:
-                if ltag == "script" or ltag == "style":
+                if ltag in _CDATA_CONTENT_ELEMENTS:
                     write(text)
                 else:
                     write(_escape_cdata(text))

diff --git a/Misc/NEWS.d/next/Library/2026-05-07-14-18-47.gh-issue-149489.bX9iHe.rst b/Misc/NEWS.d/next/Library/2026-05-07-14-18-47.gh-issue-149489.bX9iHe.rst
new file mode 100644 (file)
index 0000000..4f47d36
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2026-05-07-14-18-47.gh-issue-149489.bX9iHe.rst
@@ -0,0 +1,3 @@
+Fix :mod:`~xml.etree.ElementTree` serialization to HTML. The content of
+elements "xmp", "iframe", "noembed", "noframes", and "plaintext" is no longer
+escaped. The "plaintext" element no longer have the closing tag.