fix junitxml bin_xml_escape: supplementary plane characters incorrectly escaped

changelog/14483.bugfix.rst

@@ -0,0 +1 @@
+Fixed ``bin_xml_escape`` in junitxml incorrectly escaping supplementary plane characters (U+10000 and above, including emoji) due to using ``\u`` instead of ``\U`` for the supplementary plane range in the ``illegal_xml_re`` regex.

src/_pytest/junitxml.py

@@ -55,9 +55,7 @@ def repl(matchobj: re.Match[str]) -> str:
     # The spec range of valid chars is:
     # Char ::= #x9 | #xA | #xD | [#x20-#xD7FF] | [#xE000-#xFFFD] | [#x10000-#x10FFFF]
     # For an unknown(?) reason, we disallow #x7F (DEL) as well.
-    illegal_xml_re = (
-        "[^\u0009\u000a\u000d\u0020-\u007e\u0080-\ud7ff\ue000-\ufffd\u10000-\u10ffff]"
-    )
+    illegal_xml_re = "[^\u0009\u000a\u000d\u0020-\u007e\u0080-\ud7ff\ue000-\ufffd\U00010000-\U0010ffff]"
     return re.sub(illegal_xml_re, repl, str(arg))
 
 

testing/test_junitxml.py

@@ -1122,8 +1122,7 @@ def test_invalid_xml_escape() -> None:
         0xFFFE,
         0x0FFFF,
     )  # , 0x110000)
-    valid = (0x9, 0xA, 0x20)
-    # 0xD, 0xD7FF, 0xE000, 0xFFFD, 0x10000, 0x10FFFF)
+    valid = (0x9, 0xA, 0x20, 0xD, 0xD7FF, 0xE000, 0xFFFD, 0x10000, 0x10FFFF)
 
     for i in invalid:
         got = bin_xml_escape(chr(i))
@@ -1136,6 +1135,13 @@ def test_invalid_xml_escape() -> None:
         assert chr(i) == bin_xml_escape(chr(i))
 
 
+def test_bin_xml_escape_supplementary_plane() -> None:
+    assert bin_xml_escape(chr(0x1F600)) == chr(0x1F600)
+    assert bin_xml_escape("test_😀") == "test_😀"
+    assert bin_xml_escape("test_𠀀") == "test_𠀀"
+    assert bin_xml_escape("test_𝄞") == "test_𝄞"
+
+
 def test_logxml_path_expansion(tmp_path: Path, monkeypatch: MonkeyPatch) -> None:
     home_tilde = Path(os.path.expanduser("~")).joinpath("test.xml")
     xml_tilde = LogXML(Path("~", "test.xml"), None)