Sync from SUSE:SLFO:1.1 python-xmltodict revision 8233a6b5f71b35f2de6e50ec1997dd83

CVE-2025-9375.patch

@@ -0,0 +1,222 @@
+From 6df43e8d0133f700c328e3da4028315fb9f873da Mon Sep 17 00:00:00 2001
+From: Martin Blech <78768+martinblech@users.noreply.github.com>
+Date: Thu, 4 Sep 2025 17:25:39 -0700
+Subject: [PATCH 2/2] Merge the following three upstream commits as a combined
+ fix for CVE-2025-9375
+
+- Prevent XML injection: reject '<'/'>' in element/attr names (incl. @xmlns) (ecd456a)
+- Enhance unparse() XML name validation with stricter rules and tests (f98c90f)
+- Harden XML name validation: reject quotes and '='; add tests (8860b0e)
+---
+ tests/test_dicttoxml.py | 106 ++++++++++++++++++++++++++++++++++++++++
+ xmltodict.py            |  56 ++++++++++++++++++++-
+ 2 files changed, 160 insertions(+), 2 deletions(-)
+
+diff --git a/tests/test_dicttoxml.py b/tests/test_dicttoxml.py
+index 7fc2171..9830e2a 100644
+--- a/tests/test_dicttoxml.py
++++ b/tests/test_dicttoxml.py
+@@ -213,3 +213,109 @@ xmlns:b="http://b.com/"><x a:attr="val">1</x><a:y>2</a:y><b:z>3</b:z></root>'''
+         expected_xml = '<?xml version="1.0" encoding="utf-8"?>\n<x>false</x>'
+         xml = unparse(dict(x=False))
+         self.assertEqual(xml, expected_xml)
++
++    def test_rejects_tag_name_with_angle_brackets(self):
++        # Minimal guard: disallow '<' or '>' to prevent breaking tag context
++        with self.assertRaises(ValueError):
++            unparse({"m><tag>content</tag": "unsafe"}, full_document=False)
++
++    def test_rejects_attribute_name_with_angle_brackets(self):
++        # Now we expect bad attribute names to be rejected
++        with self.assertRaises(ValueError):
++            unparse(
++                {"a": {"@m><tag>content</tag": "unsafe", "#text": "x"}},
++                full_document=False,
++            )
++
++    def test_rejects_malicious_xmlns_prefix(self):
++        # xmlns prefixes go under @xmlns mapping; reject angle brackets in prefix
++        with self.assertRaises(ValueError):
++            unparse(
++                {
++                    "a": {
++                        "@xmlns": {"m><bad": "http://example.com/"},
++                        "#text": "x",
++                    }
++                },
++                full_document=False,
++            )
++
++    def test_attribute_values_with_angle_brackets_are_escaped(self):
++        # Attribute values should be escaped by XMLGenerator
++        xml = unparse({"a": {"@attr": "1<middle>2", "#text": "x"}}, full_document=False)
++        # The generated XML should contain escaped '<' and '>' within the attribute value
++        self.assertIn('attr="1&lt;middle&gt;2"', xml)
++
++    def test_rejects_tag_name_starting_with_question(self):
++        with self.assertRaises(ValueError):
++            unparse({"?pi": "data"}, full_document=False)
++
++    def test_rejects_tag_name_starting_with_bang(self):
++        with self.assertRaises(ValueError):
++            unparse({"!decl": "data"}, full_document=False)
++
++    def test_rejects_attribute_name_starting_with_question(self):
++        with self.assertRaises(ValueError):
++            unparse({"a": {"@?weird": "x"}}, full_document=False)
++
++    def test_rejects_attribute_name_starting_with_bang(self):
++        with self.assertRaises(ValueError):
++            unparse({"a": {"@!weird": "x"}}, full_document=False)
++
++    def test_rejects_xmlns_prefix_starting_with_question_or_bang(self):
++        with self.assertRaises(ValueError):
++            unparse({"a": {"@xmlns": {"?p": "http://e/"}}}, full_document=False)
++        with self.assertRaises(ValueError):
++            unparse({"a": {"@xmlns": {"!p": "http://e/"}}}, full_document=False)
++
++    def test_rejects_non_string_names(self):
++        class Weird:
++            def __str__(self):
++                return "bad>name"
++
++        # Non-string element key
++        with self.assertRaises(ValueError):
++            unparse({Weird(): "x"}, full_document=False)
++        # Non-string attribute key
++        with self.assertRaises(ValueError):
++            unparse({"a": {Weird(): "x"}}, full_document=False)
++
++    def test_rejects_tag_name_with_slash(self):
++        with self.assertRaises(ValueError):
++            unparse({"bad/name": "x"}, full_document=False)
++
++    def test_rejects_tag_name_with_whitespace(self):
++        for name in ["bad name", "bad\tname", "bad\nname"]:
++            with self.assertRaises(ValueError):
++                unparse({name: "x"}, full_document=False)
++
++    def test_rejects_attribute_name_with_slash(self):
++        with self.assertRaises(ValueError):
++            unparse({"a": {"@bad/name": "x"}}, full_document=False)
++
++    def test_rejects_attribute_name_with_whitespace(self):
++        for name in ["@bad name", "@bad\tname", "@bad\nname"]:
++            with self.assertRaises(ValueError):
++                unparse({"a": {name: "x"}}, full_document=False)
++
++    def test_rejects_xmlns_prefix_with_slash_or_whitespace(self):
++        # Slash
++        with self.assertRaises(ValueError):
++            unparse({"a": {"@xmlns": {"bad/prefix": "http://e/"}}}, full_document=False)
++        # Whitespace
++        with self.assertRaises(ValueError):
++            unparse({"a": {"@xmlns": {"bad prefix": "http://e/"}}}, full_document=False)
++
++    def test_rejects_names_with_quotes_and_equals(self):
++        # Element names
++        for name in ['a"b', "a'b", "a=b"]:
++            with self.assertRaises(ValueError):
++                unparse({name: "x"}, full_document=False)
++        # Attribute names
++        for name in ['@a"b', "@a'b", "@a=b"]:
++            with self.assertRaises(ValueError):
++                unparse({"a": {name: "x"}}, full_document=False)
++        # xmlns prefixes
++        for prefix in ['a"b', "a'b", "a=b"]:
++            with self.assertRaises(ValueError):
++                unparse({"a": {"@xmlns": {prefix: "http://e/"}}}, full_document=False)
+diff --git a/xmltodict.py b/xmltodict.py
+index a0ba0de..67c248a 100755
+--- a/xmltodict.py
++++ b/xmltodict.py
+@@ -369,7 +369,54 @@ def parse(xml_input, encoding=None, expat=expat, process_namespaces=False,
+     return handler.item
+ 
+ 
++def _has_angle_brackets(value):
++    """Return True if value (a str) contains '<' or '>'.
++
++    Non-string values return False. Uses fast substring checks implemented in C.
++    """
++    return isinstance(value, str) and ("<" in value or ">" in value)
++
++
++def _has_invalid_name_chars(value):
++    """Return True if value (a str) contains any disallowed name characters.
++
++    Disallowed: '<', '>', '/', or any whitespace character.
++    Non-string values return False.
++    """
++    if not isinstance(value, str):
++        return False
++    if "<" in value or ">" in value or "/" in value:
++        return True
++    # Check for any whitespace (spaces, tabs, newlines, etc.)
++    return any(ch.isspace() for ch in value)
++
++
++def _validate_name(value, kind):
++    """Validate an element/attribute name for XML safety.
++
++    Raises ValueError with a specific reason when invalid.
++
++    kind: 'element' or 'attribute' (used in error messages)
++    """
++    if not isinstance(value, str):
++        raise ValueError(f"{kind} name must be a string")
++    if value.startswith("?") or value.startswith("!"):
++        raise ValueError(f'Invalid {kind} name: cannot start with "?" or "!"')
++    if "<" in value or ">" in value:
++        raise ValueError(f'Invalid {kind} name: "<" or ">" not allowed')
++    if "/" in value:
++        raise ValueError(f'Invalid {kind} name: "/" not allowed')
++    if '"' in value or "'" in value:
++        raise ValueError(f"Invalid {kind} name: quotes not allowed")
++    if "=" in value:
++        raise ValueError(f'Invalid {kind} name: "=" not allowed')
++    if any(ch.isspace() for ch in value):
++        raise ValueError(f"Invalid {kind} name: whitespace not allowed")
++
++
+ def _process_namespace(name, namespaces, ns_sep=':', attr_prefix='@'):
++    if not isinstance(name, str):
++        return name
+     if not namespaces:
+         return name
+     try:
+@@ -402,6 +449,8 @@ def _emit(key, value, content_handler,
+         if result is None:
+             return
+         key, value = result
++    # Minimal validation to avoid breaking out of tag context
++    _validate_name(key, "element")
+     if not hasattr(value, '__iter__') or isinstance(value, (str, dict)):
+         value = [value]
+     for index, v in enumerate(value):
+@@ -425,17 +474,20 @@ def _emit(key, value, content_handler,
+             if ik == cdata_key:
+                 cdata = iv
+                 continue
+-            if ik.startswith(attr_prefix):
++            if isinstance(ik, str) and ik.startswith(attr_prefix):
+                 ik = _process_namespace(ik, namespaces, namespace_separator,
+                                         attr_prefix)
+                 if ik == '@xmlns' and isinstance(iv, dict):
+                     for k, v in iv.items():
++                        _validate_name(k, "attribute")
+                         attr = 'xmlns{}'.format(f':{k}' if k else '')
+                         attrs[attr] = str(v)
+                     continue
+                 if not isinstance(iv, str):
+                     iv = str(iv)
+-                attrs[ik[len(attr_prefix):]] = iv
++                attr_name = ik[len(attr_prefix) :]
++                _validate_name(attr_name, "attribute")
++                attrs[attr_name] = iv
+                 continue
+             children.append((ik, iv))
+         if pretty:
+-- 
+2.51.0
+

python-xmltodict.changes

@@ -1,7 +1,9 @@
 -------------------------------------------------------------------
-Mon Feb 12 15:02:30 UTC 2024 - Matej Cepl <mcepl@cepl.eu>
+Tue Sep  9 08:21:40 UTC 2025 - John Paul Adrian Glaubitz <adrian.glaubitz@suse.com>
 
-- Clean up the SPEC file.
+- Cherry-pick security-fix-prereqs.patch to allow backport of CVE fix
+- Cherry-pick CVE-2025-9375.patch to fix multiple XML Injection
+  vulnerabilities in XML parser (bsc#1249036, CVE-2025-9375)
 
 -------------------------------------------------------------------
 Tue May  9 13:41:52 UTC 2023 - Johannes Kastl <kastl@b1-systems.de>

python-xmltodict.spec

@@ -1,7 +1,7 @@
 #
 # spec file for package python-xmltodict
 #
-# Copyright (c) 2024 SUSE LLC
+# Copyright (c) 2023 SUSE LLC
 #
 # All modifications and additions to the file contributed by third parties
 # remain the property of their copyright owners, unless otherwise agreed
@@ -25,15 +25,15 @@ License:        MIT
 URL:            https://github.com/martinblech/xmltodict
 Source:         https://files.pythonhosted.org/packages/source/x/xmltodict/xmltodict-%{version}.tar.gz
 Patch0:         skip-tests-expat-245.patch
-BuildRequires:  %{python_module pip}
+Patch1:         security-fix-prereqs.patch
-BuildRequires:  %{python_module pytest}
+Patch2:         CVE-2025-9375.patch
 BuildRequires:  %{python_module setuptools}
-BuildRequires:  %{python_module wheel}
 BuildRequires:  %{python_module xml}
 BuildRequires:  fdupes
 BuildRequires:  python-rpm-macros
 Requires:       python-xml
 BuildArch:      noarch
+BuildRequires:  %{python_module pytest}
 %python_subpackages
 
 %description
@@ -47,11 +47,11 @@ sed -i '1{\@^#!%{_bindir}/env python@d}' xmltodict.py
 
 %build
 export LANG=en_US.UTF-8
-%pyproject_wheel
+%python_build
 
 %install
 export LANG=en_US.UTF-8
-%pyproject_install
+%python_install
 %python_expand %fdupes %{buildroot}%{$python_sitelib}
 
 %check
@@ -63,6 +63,6 @@ export LANG=en_US.UTF-8
 %doc README.md
 %{python_sitelib}/xmltodict.py*
 %pycache_only %{python_sitelib}/__pycache__/xmltodict.*.py*
-%{python_sitelib}/xmltodict-%{version}*-info
+%{python_sitelib}/xmltodict-%{version}-py*.egg-info
 
 %changelog

security-fix-prereqs.patch

@@ -0,0 +1,154 @@
+From 239ef35e545ef6333c0b5dff6484659ffd62c9b0 Mon Sep 17 00:00:00 2001
+From: Dimitri Papadopoulos
+ <3234522+DimitriPapadopoulos@users.noreply.github.com>
+Date: Sun, 17 Jul 2022 09:28:21 +0200
+Subject: [PATCH 1/2] Merge the following four upstream commits required for
+ CVE-2025-9375
+
+- Merge `isinstance` calls (9c3ec3c)
+- Apply ruff/pyupgrade rule UP031 (d4a50f5)
+- Apply ruff/pyupgrade rule UP032 (3df6d23)
+- Get rid of Python 2 basestring and unicode (#346) (5b1b511)
+---
+ ez_setup.py             |  7 +++----
+ tests/test_xmltodict.py |  4 ++--
+ xmltodict.py            | 39 ++++++++++++---------------------------
+ 3 files changed, 17 insertions(+), 33 deletions(-)
+
+diff --git a/ez_setup.py b/ez_setup.py
+index 800c31e..b5cc352 100644
+--- a/ez_setup.py
++++ b/ez_setup.py
+@@ -131,7 +131,7 @@ def archive_context(filename):
+ 
+ def _do_download(version, download_base, to_dir, download_delay):
+     """Download Setuptools."""
+-    py_desig = 'py{sys.version_info[0]}.{sys.version_info[1]}'.format(sys=sys)
++    py_desig = f'py{sys.version_info[0]}.{sys.version_info[1]}'
+     tp = 'setuptools-{version}-{py_desig}.egg'
+     egg = os.path.join(to_dir, tp.format(**locals()))
+     if not os.path.exists(egg):
+@@ -245,8 +245,7 @@ def download_file_powershell(url, target):
+     ps_cmd = (
+         "[System.Net.WebRequest]::DefaultWebProxy.Credentials = "
+         "[System.Net.CredentialCache]::DefaultCredentials; "
+-        '(new-object System.Net.WebClient).DownloadFile("%(url)s", "%(target)s")'
+-        % locals()
++        '(new-object System.Net.WebClient).DownloadFile("{url}", "{target}")'.format(**locals())
+     )
+     cmd = [
+         'powershell',
+@@ -346,7 +345,7 @@ def download_setuptools(
+     """
+     # making sure we use the absolute path
+     to_dir = os.path.abspath(to_dir)
+-    zip_name = "setuptools-%s.zip" % version
++    zip_name = f"setuptools-{version}.zip"
+     url = download_base + zip_name
+     saveto = os.path.join(to_dir, zip_name)
+     if not os.path.exists(saveto):  # Avoid repeated downloads
+diff --git a/tests/test_xmltodict.py b/tests/test_xmltodict.py
+index 04137f9..aa7a1a4 100644
+--- a/tests/test_xmltodict.py
++++ b/tests/test_xmltodict.py
+@@ -168,14 +168,14 @@ class XMLToDictTestCase(unittest.TestCase):
+         except NameError:
+             value = chr(39321)
+         self.assertEqual({'a': value},
+-                         parse('<a>%s</a>' % value))
++                         parse(f'<a>{value}</a>'))
+ 
+     def test_encoded_string(self):
+         try:
+             value = unichr(39321)
+         except NameError:
+             value = chr(39321)
+-        xml = '<a>%s</a>' % value
++        xml = f'<a>{value}</a>'
+         self.assertEqual(parse(xml),
+                          parse(xml.encode('utf-8')))
+ 
+diff --git a/xmltodict.py b/xmltodict.py
+index ca760aa..a0ba0de 100755
+--- a/xmltodict.py
++++ b/xmltodict.py
+@@ -22,15 +22,6 @@ if tuple(map(int, platform.python_version_tuple()[:2])) < (3, 7):
+ 
+ from inspect import isgenerator
+ 
+-try:  # pragma no cover
+-    _basestring = basestring
+-except NameError:  # pragma no cover
+-    _basestring = str
+-try:  # pragma no cover
+-    _unicode = unicode
+-except NameError:  # pragma no cover
+-    _unicode = str
+-
+ __author__ = 'Martin Blech'
+ __version__ = '0.13.0'
+ __license__ = 'MIT'
+@@ -335,9 +326,8 @@ def parse(xml_input, encoding=None, expat=expat, process_namespaces=False,
+     """
+     handler = _DictSAXHandler(namespace_separator=namespace_separator,
+                               **kwargs)
+-    if isinstance(xml_input, _unicode):
+-        if not encoding:
+-            encoding = 'utf-8'
++    if isinstance(xml_input, str):
++        encoding = encoding or 'utf-8'
+         xml_input = xml_input.encode(encoding)
+     if not process_namespaces:
+         namespace_separator = None
+@@ -412,9 +402,7 @@ def _emit(key, value, content_handler,
+         if result is None:
+             return
+         key, value = result
+-    if (not hasattr(value, '__iter__')
+-            or isinstance(value, _basestring)
+-            or isinstance(value, dict)):
++    if not hasattr(value, '__iter__') or isinstance(value, (str, dict)):
+         value = [value]
+     for index, v in enumerate(value):
+         if full_document and depth == 0 and index > 0:
+@@ -422,16 +410,13 @@ def _emit(key, value, content_handler,
+         if v is None:
+             v = _dict()
+         elif isinstance(v, bool):
+-            if v:
+-                v = _unicode('true')
+-            else:
+-                v = _unicode('false')
+-        elif not isinstance(v, dict):
+-            if expand_iter and hasattr(v, '__iter__') and not isinstance(v, _basestring):
++            v = 'true' if v else 'false'
++        elif not isinstance(v, (dict, str)):
++            if expand_iter and hasattr(v, '__iter__'):
+                 v = _dict(((expand_iter, v),))
+             else:
+-                v = _unicode(v)
+-        if isinstance(v, _basestring):
++                v = str(v)
++        if isinstance(v, str):
+             v = _dict(((cdata_key, v),))
+         cdata = None
+         attrs = _dict()
+@@ -445,11 +430,11 @@ def _emit(key, value, content_handler,
+                                         attr_prefix)
+                 if ik == '@xmlns' and isinstance(iv, dict):
+                     for k, v in iv.items():
+-                        attr = 'xmlns{}'.format(':{}'.format(k) if k else '')
+-                        attrs[attr] = _unicode(v)
++                        attr = 'xmlns{}'.format(f':{k}' if k else '')
++                        attrs[attr] = str(v)
+                     continue
+-                if not isinstance(iv, _unicode):
+-                    iv = _unicode(iv)
++                if not isinstance(iv, str):
++                    iv = str(iv)
+                 attrs[ik[len(attr_prefix):]] = iv
+                 continue
+             children.append((ik, iv))
+-- 
+2.51.0
+