PyObject_Dump() encodes unicode objects to utf8 with backslashreplace (instead

of strict) error handler to escape surrogates

3 files changed, 16 insertions, 1 deletions

diff --git a/Lib/test/test_sys.py b/Lib/test/test_sys.py
index abbd759..ecbc9db 100644
--- a/Lib/test/test_sys.py
+++ b/Lib/test/test_sys.py
@@ -145,6 +145,16 @@ class SysModuleTest(unittest.TestCase):
                               "raise SystemExit(47)"])
         self.assertEqual(rc, 47)
 
+        # test that the exit message is written with backslashreplace error
+        # handler to stderr
+        import subprocess
+        code = r'import sys; sys.exit("surrogates:\uDCFF")'
+        process = subprocess.Popen([sys.executable, "-c", code],
+                                   stderr=subprocess.PIPE)
+        stdout, stderr = process.communicate()
+        self.assertEqual(process.returncode, 1)
+        self.assertTrue(stderr.startswith(b"surrogates:\\udcff"), stderr)
+
     def test_getdefaultencoding(self):
         self.assertRaises(TypeError, sys.getdefaultencoding, 42)
         # can't check more than the type, as the user might have changed it
diff --git a/Misc/NEWS b/Misc/NEWS
index f801e3e..6b38425 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -12,6 +12,9 @@ What's New in Python 3.2 Alpha 1?
 Core and Builtins
 -----------------
 
+- PyObject_Dump() encodes unicode objects to utf8 with backslashreplace
+  (instead of strict) error handler to escape surrogates
+
 - Issue #8715: Create PyUnicode_EncodeFSDefault() function: Encode a Unicode
   object to Py_FileSystemDefaultEncoding with the "surrogateescape" error
   handler, and return bytes. If Py_FileSystemDefaultEncoding is not set, fall
diff --git a/Objects/object.c b/Objects/object.c
index 8ddc7ec..1f4e3dd 100644
--- a/Objects/object.c
+++ b/Objects/object.c
@@ -303,7 +303,9 @@ internal_print(PyObject *op, FILE *fp, int flags, int nesting)
             }
             else if (PyUnicode_Check(s)) {
                 PyObject *t;
-                t = _PyUnicode_AsDefaultEncodedString(s, NULL);
+                t = PyUnicode_EncodeUTF8(PyUnicode_AS_UNICODE(s),
+                                         PyUnicode_GET_SIZE(s),
+                                         "backslashreplace");
                 if (t == NULL)
                     ret = 0;
                 else {