Merged revisions 78449 via svnmerge from

svn+ssh://pythondev@svn.python.org/python/trunk

........
  r78449 | ezio.melotti | 2010-02-25 19:36:04 +0200 (Thu, 25 Feb 2010) | 1 line

  #7649: "u'%c' % char" now behaves like "u'%s' % char" and raises a UnicodeDecodeError if 'char' is a byte string that can't be decoded using the default encoding.
........

2 files changed, 21 insertions, 11 deletions

diff --git a/Misc/NEWS b/Misc/NEWS
index 4292d8d3c0..33601a8d58 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -16,8 +16,9 @@ Core and Builtins
   UnicodeEncodeError, UnicodeDecodeError, and UnicodeTranslateError to
   strings.
 
-- Issue #7649: Fix u'%c' % char for character in range 0x80..0xFF, raise an
-  UnicodeDecodeError.
+- Issue #7649: "u'%c' % char" now behaves like "u'%s' % char" and raises a
+  UnicodeDecodeError if 'char' is a byte string that can't be decoded using
+  the default encoding.
 
 - Issue #5677: Explicitly forbid write operations on read-only file objects,
   and read operations on write-only file objects.  On Windows, the system C
diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
index 667afaef28..133cae5d5f 100644
--- a/Objects/unicodeobject.c
+++ b/Objects/unicodeobject.c
@@ -8357,7 +8357,8 @@ formatchar(Py_UNICODE *buf,
            size_t buflen,
            PyObject *v)
 {
-    PyObject *s;
+    PyObject *unistr;
+    char *str;
     /* presume that the buffer is at least 2 characters long */
     if (PyUnicode_Check(v)) {
         if (PyUnicode_GET_SIZE(v) != 1)
@@ -8368,14 +8369,22 @@ formatchar(Py_UNICODE *buf,
     else if (PyString_Check(v)) {
         if (PyString_GET_SIZE(v) != 1)
             goto onError;
-        /* #7649: if the char is a non-ascii (i.e. in range(0x80,0x100)) byte
-           string, "u'%c' % char" should fail with a UnicodeDecodeError */
-        s = PyUnicode_FromStringAndSize(PyString_AS_STRING(v), 1);
-        /* if the char is not decodable return -1 */
-        if (s == NULL)
-            return -1;
-        buf[0] = PyUnicode_AS_UNICODE(s)[0];
-        Py_DECREF(s);
+        /* #7649: "u'%c' % char" should behave like "u'%s' % char" and fail
+           with a UnicodeDecodeError if 'char' is not decodable with the
+           default encoding (usually ASCII, but it might be something else) */
+        str = PyString_AS_STRING(v);
+        if ((unsigned char)str[0] > 0x7F) {
+            /* the char is not ASCII; try to decode the string using the
+               default encoding and return -1 to let the UnicodeDecodeError
+               be raised if the string can't be decoded */
+            unistr = PyUnicode_Decode(str, 1, NULL, "strict");
+            if (unistr == NULL)
+                return -1;
+            buf[0] = PyUnicode_AS_UNICODE(unistr)[0];
+            Py_DECREF(unistr);
+        }
+        else
+            buf[0] = (Py_UNICODE)str[0];
     }
 
     else {