Issue #4874: Most builtin decoders now reject unicode input.

diff --git a/Lib/test/test_codecs.py b/Lib/test/test_codecs.py
index e485fdd2fe89953935f8a6f24d228505187e219c..1730dbe5939b9dea9fd4bd76392954477d81c424 100644 (file)
--- a/Lib/test/test_codecs.py
+++ b/Lib/test/test_codecs.py
@@ -1475,6 +1475,36 @@ class WithStmtTest(unittest.TestCase):
                                        info.streamwriter, 'strict') as srw:
             self.assertEquals(srw.read(), "\xfc")
 
+class TypesTest(unittest.TestCase):
+    def test_decode_unicode(self):
+        # Most decoders don't accept unicode input
+        decoders = [
+            codecs.utf_7_decode,
+            codecs.utf_8_decode,
+            codecs.utf_16_le_decode,
+            codecs.utf_16_be_decode,
+            codecs.utf_16_ex_decode,
+            codecs.utf_32_decode,
+            codecs.utf_32_le_decode,
+            codecs.utf_32_be_decode,
+            codecs.utf_32_ex_decode,
+            codecs.latin_1_decode,
+            codecs.ascii_decode,
+            codecs.charmap_decode,
+        ]
+        if hasattr(codecs, "mbcs_decode"):
+            decoders.append(codecs.mbcs_decode)
+        for decoder in decoders:
+            self.assertRaises(TypeError, decoder, "xxx")
+
+    def test_unicode_escape(self):
+        # Escape-decoding an unicode string is supported ang gives the same
+        # result as decoding the equivalent ASCII bytes string.
+        self.assertEquals(codecs.unicode_escape_decode(r"\u1234"), ("\u1234", 6))
+        self.assertEquals(codecs.unicode_escape_decode(br"\u1234"), ("\u1234", 6))
+        self.assertEquals(codecs.raw_unicode_escape_decode(r"\u1234"), ("\u1234", 6))
+        self.assertEquals(codecs.raw_unicode_escape_decode(br"\u1234"), ("\u1234", 6))
+
 
 def test_main():
     support.run_unittest(
@@ -1501,6 +1531,7 @@ def test_main():
         BasicUnicodeTest,
         CharmapTest,
         WithStmtTest,
+        TypesTest,
     )
 
 

diff --git a/Misc/NEWS b/Misc/NEWS
index b92048c2c6d59d97082adec36734e8263cf52c46..2d18011d5142066aba4c5290314782a22c0d85fa 100644 (file)
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -12,6 +12,8 @@ What's New in Python 3.1 alpha 0
 Core and Builtins
 -----------------
 
+- Issue #4874: Most builtin decoders now reject unicode input.
+
 - Issue #4842: Don't allow trailing 'L' when constructing an integer
   from a string.
 

diff --git a/Modules/_codecsmodule.c b/Modules/_codecsmodule.c
index fc3e3f9bbd66add4a115d409048f5c951a235719..e7dbf89499ee96c41884ea5ed6b5989dfb9a9a32 100644 (file)
--- a/Modules/_codecsmodule.c
+++ b/Modules/_codecsmodule.c
@@ -258,7 +258,7 @@ utf_7_decode(PyObject *self,
     Py_ssize_t consumed;
     PyObject *decoded = NULL;
 
-    if (!PyArg_ParseTuple(args, "s*|zi:utf_7_decode",
+    if (!PyArg_ParseTuple(args, "y*|zi:utf_7_decode",
                          &pbuf, &errors, &final))
        return NULL;
     consumed = pbuf.len;
@@ -281,7 +281,7 @@ utf_8_decode(PyObject *self,
     Py_ssize_t consumed;
     PyObject *decoded = NULL;
 
-    if (!PyArg_ParseTuple(args, "s*|zi:utf_8_decode",
+    if (!PyArg_ParseTuple(args, "y*|zi:utf_8_decode",
                          &pbuf, &errors, &final))
        return NULL;
     consumed = pbuf.len;
@@ -305,7 +305,7 @@ utf_16_decode(PyObject *self,
     Py_ssize_t consumed;
     PyObject *decoded;
 
-    if (!PyArg_ParseTuple(args, "s*|zi:utf_16_decode",
+    if (!PyArg_ParseTuple(args, "y*|zi:utf_16_decode",
                          &pbuf, &errors, &final))
        return NULL;
     consumed = pbuf.len; /* This is overwritten unless final is true. */
@@ -328,7 +328,7 @@ utf_16_le_decode(PyObject *self,
     Py_ssize_t consumed;
     PyObject *decoded = NULL;
 
-    if (!PyArg_ParseTuple(args, "s*|zi:utf_16_le_decode",
+    if (!PyArg_ParseTuple(args, "y*|zi:utf_16_le_decode",
                          &pbuf, &errors, &final))
        return NULL;
 
@@ -352,7 +352,7 @@ utf_16_be_decode(PyObject *self,
     Py_ssize_t consumed;
     PyObject *decoded = NULL;
 
-    if (!PyArg_ParseTuple(args, "s*|zi:utf_16_be_decode",
+    if (!PyArg_ParseTuple(args, "y*|zi:utf_16_be_decode",
                          &pbuf, &errors, &final))
        return NULL;
 
@@ -384,7 +384,7 @@ utf_16_ex_decode(PyObject *self,
     int final = 0;
     Py_ssize_t consumed;
 
-    if (!PyArg_ParseTuple(args, "s*|zii:utf_16_ex_decode",
+    if (!PyArg_ParseTuple(args, "y*|zii:utf_16_ex_decode",
                          &pbuf, &errors, &byteorder, &final))
        return NULL;
     consumed = pbuf.len; /* This is overwritten unless final is true. */
@@ -409,7 +409,7 @@ utf_32_decode(PyObject *self,
     Py_ssize_t consumed;
     PyObject *decoded;
 
-    if (!PyArg_ParseTuple(args, "s*|zi:utf_32_decode",
+    if (!PyArg_ParseTuple(args, "y*|zi:utf_32_decode",
                          &pbuf, &errors, &final))
        return NULL;
     consumed = pbuf.len; /* This is overwritten unless final is true. */
@@ -432,7 +432,7 @@ utf_32_le_decode(PyObject *self,
     Py_ssize_t consumed;
     PyObject *decoded;
 
-    if (!PyArg_ParseTuple(args, "s*|zi:utf_32_le_decode",
+    if (!PyArg_ParseTuple(args, "y*|zi:utf_32_le_decode",
                          &pbuf, &errors, &final))
        return NULL;
     consumed = pbuf.len; /* This is overwritten unless final is true. */
@@ -455,7 +455,7 @@ utf_32_be_decode(PyObject *self,
     Py_ssize_t consumed;
     PyObject *decoded;
 
-    if (!PyArg_ParseTuple(args, "s*|zi:utf_32_be_decode",
+    if (!PyArg_ParseTuple(args, "y*|zi:utf_32_be_decode",
                          &pbuf, &errors, &final))
        return NULL;
     consumed = pbuf.len; /* This is overwritten unless final is true. */
@@ -486,7 +486,7 @@ utf_32_ex_decode(PyObject *self,
     int final = 0;
     Py_ssize_t consumed;
 
-    if (!PyArg_ParseTuple(args, "s*|zii:utf_32_ex_decode",
+    if (!PyArg_ParseTuple(args, "y*|zii:utf_32_ex_decode",
                          &pbuf, &errors, &byteorder, &final))
        return NULL;
     consumed = pbuf.len; /* This is overwritten unless final is true. */
@@ -542,7 +542,7 @@ latin_1_decode(PyObject *self,
        PyObject *unicode;
     const char *errors = NULL;
 
-    if (!PyArg_ParseTuple(args, "s*|z:latin_1_decode",
+    if (!PyArg_ParseTuple(args, "y*|z:latin_1_decode",
                          &pbuf, &errors))
        return NULL;
 
@@ -559,7 +559,7 @@ ascii_decode(PyObject *self,
        PyObject *unicode;
     const char *errors = NULL;
 
-    if (!PyArg_ParseTuple(args, "s*|z:ascii_decode",
+    if (!PyArg_ParseTuple(args, "y*|z:ascii_decode",
                          &pbuf, &errors))
        return NULL;
 
@@ -577,7 +577,7 @@ charmap_decode(PyObject *self,
     const char *errors = NULL;
     PyObject *mapping = NULL;
 
-    if (!PyArg_ParseTuple(args, "s*|zO:charmap_decode",
+    if (!PyArg_ParseTuple(args, "y*|zO:charmap_decode",
                          &pbuf, &errors, &mapping))
        return NULL;
     if (mapping == Py_None)
@@ -600,7 +600,7 @@ mbcs_decode(PyObject *self,
     Py_ssize_t consumed;
     PyObject *decoded = NULL;
 
-    if (!PyArg_ParseTuple(args, "s*|zi:mbcs_decode",
+    if (!PyArg_ParseTuple(args, "y*|zi:mbcs_decode",
                          &pbuf, &errors, &final))
        return NULL;
     consumed = pbuf.len;