make 'c' only accept bytes and 'C' only unicode #5499

4 files changed, 15 insertions, 14 deletions

diff --git a/Doc/c-api/arg.rst b/Doc/c-api/arg.rst
index dfedf972fe..a554bc9300 100644
--- a/Doc/c-api/arg.rst
+++ b/Doc/c-api/arg.rst
@@ -208,9 +208,13 @@ variable(s) whose address should be passed.
    Convert a Python integer to a C :ctype:`Py_ssize_t`.
 
 ``c`` (string of length 1) [char]
-   Convert a Python character, represented as a string of length 1, to a C
+   Convert a Python character, represented as a byte string of length 1, to a C
    :ctype:`char`.
 
+``C`` (string of length 1) [int]
+   Covert a Python character, represented as a unicode string of length 1, to a
+   C :ctype:`int`.
+
 ``f`` (float) [float]
    Convert a Python floating point number to a C :ctype:`float`.
 
diff --git a/Lib/test/buffer_tests.py b/Lib/test/buffer_tests.py
index 8696eab61d..58689cdb34 100644
--- a/Lib/test/buffer_tests.py
+++ b/Lib/test/buffer_tests.py
@@ -112,7 +112,7 @@ class MixinBytesBufferCommonTests(object):
         self.assertEqual(b'abc   ', self.marshal(b'abc').ljust(6))
         self.assertEqual(b'abc', self.marshal(b'abc').ljust(3))
         self.assertEqual(b'abc', self.marshal(b'abc').ljust(2))
-        self.assertEqual(b'abc*******', self.marshal(b'abc').ljust(10, '*'))
+        self.assertEqual(b'abc*******', self.marshal(b'abc').ljust(10, b'*'))
         self.assertRaises(TypeError, self.marshal(b'abc').ljust)
 
     def test_rjust(self):
@@ -120,7 +120,7 @@ class MixinBytesBufferCommonTests(object):
         self.assertEqual(b'   abc', self.marshal(b'abc').rjust(6))
         self.assertEqual(b'abc', self.marshal(b'abc').rjust(3))
         self.assertEqual(b'abc', self.marshal(b'abc').rjust(2))
-        self.assertEqual(b'*******abc', self.marshal(b'abc').rjust(10, '*'))
+        self.assertEqual(b'*******abc', self.marshal(b'abc').rjust(10, b'*'))
         self.assertRaises(TypeError, self.marshal(b'abc').rjust)
 
     def test_center(self):
@@ -128,7 +128,7 @@ class MixinBytesBufferCommonTests(object):
         self.assertEqual(b' abc  ', self.marshal(b'abc').center(6))
         self.assertEqual(b'abc', self.marshal(b'abc').center(3))
         self.assertEqual(b'abc', self.marshal(b'abc').center(2))
-        self.assertEqual(b'***abc****', self.marshal(b'abc').center(10, '*'))
+        self.assertEqual(b'***abc****', self.marshal(b'abc').center(10, b'*'))
         self.assertRaises(TypeError, self.marshal(b'abc').center)
 
     def test_swapcase(self):
diff --git a/Misc/NEWS b/Misc/NEWS
index b02e785684..8e9fbf455b 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -12,6 +12,9 @@ What's New in Python 3.1 alpha 2?
 Core and Builtins
 -----------------
 
+- Issue #5499: The 'c' code for argument parsing functions now only accepts a
+  byte, and the 'C' code only accepts a unicode character.
+
 - Issue #1665206: Remove the last eager import in _warnings.c and make it lazy.
 
 - Fix a segfault when running test_exceptions with coverage, caused by
diff --git a/Python/getargs.c b/Python/getargs.c
index 3ab59b3e42..a89ee6f858 100644
--- a/Python/getargs.c
+++ b/Python/getargs.c
@@ -776,24 +776,18 @@ convertsimple(PyObject *arg, const char **p_format, va_list *p_va, int flags,
 		char *p = va_arg(*p_va, char *);
 		if (PyBytes_Check(arg) && PyBytes_Size(arg) == 1)
 			*p = PyBytes_AS_STRING(arg)[0];
-		else if (PyUnicode_Check(arg) &&
-			 PyUnicode_GET_SIZE(arg) == 1 &&
-			 PyUnicode_AS_UNICODE(arg)[0] < 256)
-			*p = (char)PyUnicode_AS_UNICODE(arg)[0];
 		else
-			return converterr("char < 256", arg, msgbuf, bufsize);
+			return converterr("a byte string of length 1", arg, msgbuf, bufsize);
 		break;
 	}
 
 	case 'C': {/* unicode char */
 		int *p = va_arg(*p_va, int *);
-		if (PyBytes_Check(arg) && PyBytes_Size(arg) == 1)
-			*p = PyBytes_AS_STRING(arg)[0];
-		else if (PyUnicode_Check(arg) &&
-			 PyUnicode_GET_SIZE(arg) == 1)
+		if (PyUnicode_Check(arg) &&
+                    PyUnicode_GET_SIZE(arg) == 1)
 			*p = PyUnicode_AS_UNICODE(arg)[0];
 		else
-			return converterr("char", arg, msgbuf, bufsize);
+			return converterr("a unicode character", arg, msgbuf, bufsize);
 		break;
 	}