Issue #15379: Fix passing of non-BMP characters as integers for the charmap decoder...

Patch by Serhiy Storchaka.

diff --git a/Lib/test/test_codeccallbacks.py b/Lib/test/test_codeccallbacks.py
index c5b1e25c208daa3f14792d79b63fcf223011ba4e..e656d2fe8f79cc4e539df498883c50e9dfd867d9 100644 (file)
--- a/Lib/test/test_codeccallbacks.py
+++ b/Lib/test/test_codeccallbacks.py
@@ -744,7 +744,7 @@ class CodecCallbackTest(unittest.TestCase):
                 raise ValueError
         self.assertRaises(UnicodeError, codecs.charmap_decode, b"\xff", "strict", {0xff: None})
         self.assertRaises(ValueError, codecs.charmap_decode, b"\xff", "strict", D())
-        self.assertRaises(TypeError, codecs.charmap_decode, b"\xff", "strict", {0xff: sys.maxunicode+1})
+        self.assertRaises(TypeError, codecs.charmap_decode, b"\xff", "strict", {0xff: 0x110000})
 
     def test_encodehelper(self):
         # enhance coverage of:

diff --git a/Lib/test/test_codecs.py b/Lib/test/test_codecs.py
index 3426a4dde9a9a98d2ff7f9fc768a65156466a227..f342d88b9f6a9306f72a9a4feb955422d4edc2eb 100644 (file)
--- a/Lib/test/test_codecs.py
+++ b/Lib/test/test_codecs.py
@@ -1546,6 +1546,10 @@ class CharmapTest(unittest.TestCase):
             ("abc", 3)
         )
 
+        self.assertRaises(UnicodeDecodeError,
+            codecs.charmap_decode, b"\x00\x01\x02", "strict", "ab"
+        )
+
         self.assertEqual(
             codecs.charmap_decode(b"\x00\x01\x02", "replace", "ab"),
             ("ab\ufffd", 3)
@@ -1572,6 +1576,107 @@ class CharmapTest(unittest.TestCase):
             ("", len(allbytes))
         )
 
+    def test_decode_with_int2str_map(self):
+        self.assertEqual(
+            codecs.charmap_decode(b"\x00\x01\x02", "strict",
+                                  {0: 'a', 1: 'b', 2: 'c'}),
+            ("abc", 3)
+        )
+
+        self.assertEqual(
+            codecs.charmap_decode(b"\x00\x01\x02", "strict",
+                                  {0: 'Aa', 1: 'Bb', 2: 'Cc'}),
+            ("AaBbCc", 3)
+        )
+
+        self.assertEqual(
+            codecs.charmap_decode(b"\x00\x01\x02", "strict",
+                                  {0: '\U0010FFFF', 1: 'b', 2: 'c'}),
+            ("\U0010FFFFbc", 3)
+        )
+
+        self.assertEqual(
+            codecs.charmap_decode(b"\x00\x01\x02", "strict",
+                                  {0: 'a', 1: 'b', 2: ''}),
+            ("ab", 3)
+        )
+
+        self.assertRaises(UnicodeDecodeError,
+            codecs.charmap_decode, b"\x00\x01\x02", "strict",
+                                   {0: 'a', 1: 'b'}
+        )
+
+        self.assertEqual(
+            codecs.charmap_decode(b"\x00\x01\x02", "replace",
+                                  {0: 'a', 1: 'b'}),
+            ("ab\ufffd", 3)
+        )
+
+        self.assertEqual(
+            codecs.charmap_decode(b"\x00\x01\x02", "replace",
+                                  {0: 'a', 1: 'b', 2: None}),
+            ("ab\ufffd", 3)
+        )
+
+        self.assertEqual(
+            codecs.charmap_decode(b"\x00\x01\x02", "ignore",
+                                  {0: 'a', 1: 'b'}),
+            ("ab", 3)
+        )
+
+        self.assertEqual(
+            codecs.charmap_decode(b"\x00\x01\x02", "ignore",
+                                  {0: 'a', 1: 'b', 2: None}),
+            ("ab", 3)
+        )
+
+        allbytes = bytes(range(256))
+        self.assertEqual(
+            codecs.charmap_decode(allbytes, "ignore", {}),
+            ("", len(allbytes))
+        )
+
+    def test_decode_with_int2int_map(self):
+        a = ord('a')
+        b = ord('b')
+        c = ord('c')
+
+        self.assertEqual(
+            codecs.charmap_decode(b"\x00\x01\x02", "strict",
+                                  {0: a, 1: b, 2: c}),
+            ("abc", 3)
+        )
+
+        # Issue #15379
+        self.assertEqual(
+            codecs.charmap_decode(b"\x00\x01\x02", "strict",
+                                  {0: 0x10FFFF, 1: b, 2: c}),
+            ("\U0010FFFFbc", 3)
+        )
+
+        self.assertRaises(TypeError,
+            codecs.charmap_decode, b"\x00\x01\x02", "strict",
+                                   {0: 0x110000, 1: b, 2: c}
+        )
+
+        self.assertRaises(UnicodeDecodeError,
+            codecs.charmap_decode, b"\x00\x01\x02", "strict",
+                                   {0: a, 1: b},
+        )
+
+        self.assertEqual(
+            codecs.charmap_decode(b"\x00\x01\x02", "replace",
+                                  {0: a, 1: b}),
+            ("ab\ufffd", 3)
+        )
+
+        self.assertEqual(
+            codecs.charmap_decode(b"\x00\x01\x02", "ignore",
+                                  {0: a, 1: b}),
+            ("ab", 3)
+        )
+
+
 class WithStmtTest(unittest.TestCase):
     def test_encodedfile(self):
         f = io.BytesIO(b"\xc3\xbc")

diff --git a/Misc/NEWS b/Misc/NEWS
index 74782aac68c27a24f0cebce6c25dc0a111ae5d4e..57ed006fa762729bbb47b1fde7d206d918315be1 100644 (file)
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -10,6 +10,9 @@ What's New in Python 3.2.4
 Core and Builtins
 -----------------
 
+- Issue #15379: Fix passing of non-BMP characters as integers for the charmap
+  decoder (already working as unicode strings).  Patch by Serhiy Storchaka.
+
 - Issue #13992: The trashcan mechanism is now thread-safe.  This eliminates
   sporadic crashes in multi-thread programs when several long deallocator
   chains ran concurrently and involved subclasses of built-in container

diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
index 8b782b4065e55ffbadc8d9cd60ee97e9c7077b7b..f59db36dd3d455c241aecfca5fa02300d490b43b 100644 (file)
--- a/Objects/unicodeobject.c
+++ b/Objects/unicodeobject.c
@@ -5250,12 +5250,36 @@ PyObject *PyUnicode_DecodeCharmap(const char *s,
             /* Apply mapping */
             if (PyLong_Check(x)) {
                 long value = PyLong_AS_LONG(x);
-                if (value < 0 || value > 65535) {
+                if (value < 0 || value > 0x10FFFF) {
                     PyErr_SetString(PyExc_TypeError,
-                                    "character mapping must be in range(65536)");
+                                    "character mapping must be in range(0x110000)");
                     Py_DECREF(x);
                     goto onError;
                 }
+
+#ifndef Py_UNICODE_WIDE
+                if (value > 0xFFFF) {
+                    /* see the code for 1-n mapping below */
+                    if (extrachars < 2) {
+                        /* resize first */
+                        Py_ssize_t oldpos = p - PyUnicode_AS_UNICODE(v);
+                        Py_ssize_t needed = 10 - extrachars;
+                        extrachars += needed;
+                        /* XXX overflow detection missing */
+                        if (_PyUnicode_Resize(&v,
+                                              PyUnicode_GET_SIZE(v) + needed) < 0) {
+                            Py_DECREF(x);
+                            goto onError;
+                        }
+                        p = PyUnicode_AS_UNICODE(v) + oldpos;
+                    }
+                    value -= 0x10000;
+                    *p++ = 0xD800 | (value >> 10);
+                    *p++ = 0xDC00 | (value & 0x3FF);
+                    extrachars -= 2;
+                }
+                else
+#endif
                 *p++ = (Py_UNICODE)value;
             }
             else if (x == Py_None) {