Author: christian.heimes
Date: Sun Nov 4 13:32:02 2007
New Revision: 58839
Modified:
python/branches/py3k-pep3137/ (props changed)
python/branches/py3k-pep3137/Doc/reference/datamodel.rst
python/branches/py3k-pep3137/Include/unicodeobject.h
python/branches/py3k-pep3137/Lib/test/test_bytes.py
python/branches/py3k-pep3137/Lib/test/test_import.py
python/branches/py3k-pep3137/Objects/bytesobject.c
python/branches/py3k-pep3137/Objects/unicodeobject.c
python/branches/py3k-pep3137/Python/import.c
python/branches/py3k-pep3137/Python/sysmodule.c
Log:
Merged revisions 58818-58838 via svnmerge from
svn+ssh://[EMAIL PROTECTED]/python/branches/py3k
........
r58819 | guido.van.rossum | 2007-11-03 01:24:24 +0100 (Sat, 03 Nov 2007) | 4
lines
Patch 1171 by mfenniak -- allow subclassing of bytes.
I suspect this has some problems when the subclass is evil,
but that's for later.
........
r58824 | georg.brandl | 2007-11-03 09:44:43 +0100 (Sat, 03 Nov 2007) | 2 lines
Re-add two paragraphs that seem to have been lost during the merge from trunk.
........
r58836 | christian.heimes | 2007-11-04 12:37:22 +0100 (Sun, 04 Nov 2007) | 2
lines
Disabled test_sys_path_with_unicode for now
Fixed encoding of author name in tarfile.py.
........
r58837 | christian.heimes | 2007-11-04 12:43:14 +0100 (Sun, 04 Nov 2007) | 3
lines
Fixed a bug in PyUnicode_DecodeFSDefault. strcmp() returns 0 on success.
Added PyUnicode_DecodeFSDefaultAndSize
Fixed a problem with the sys.path code that caused a segfault on Windows when
the path contains non ASCII chars. The code for sys.executable, exec_prefix and
prefix should be fixed, too.
........
r58838 | christian.heimes | 2007-11-04 13:10:01 +0100 (Sun, 04 Nov 2007) | 2
lines
Applied patch #1379 with a minor tweak.
PyModule_GetName() returns a char* from PyUnicode_AsString but the code in
import.c was using PyString_FromString on it.
........
Modified: python/branches/py3k-pep3137/Doc/reference/datamodel.rst
==============================================================================
--- python/branches/py3k-pep3137/Doc/reference/datamodel.rst (original)
+++ python/branches/py3k-pep3137/Doc/reference/datamodel.rst Sun Nov 4
13:32:02 2007
@@ -1289,8 +1289,7 @@
instances are compared by object identity ("address"). See also the
description of :meth:`__hash__` for some important notes on creating
:term:`hashable` objects which support custom comparison operations and are
- usable as dictionary keys. (Note: the restriction that exceptions are not
- propagated by :meth:`__cmp__` has been removed since Python 1.5.)
+ usable as dictionary keys.
.. method:: object.__hash__(self)
@@ -1307,8 +1306,18 @@
(e.g., using exclusive or) the hash values for the components of the object
that
also play a part in comparison of objects.
- :meth:`__hash__` may also return a long integer object; the 32-bit integer
is
- then derived from the hash of that object.
+ If a class does not define a :meth:`__cmp__` or :meth:`__eq__` method it
+ should not define a :meth:`__hash__` operation either; if it defines
+ :meth:`__cmp__` or :meth:`__eq__` but not :meth:`__hash__`, its instances
+ will not be usable as dictionary keys. If a class defines mutable objects
+ and implements a :meth:`__cmp__` or :meth:`__eq__` method, it should not
+ implement :meth:`__hash__`, since the dictionary implementation requires
that
+ a key's hash value is immutable (if the object's hash value changes, it will
+ be in the wrong hash bucket).
+
+ User-defined classes have :meth:`__cmp__` and :meth:`__hash__` methods
+ by default; with them, all objects compare unequal and ``x.__hash__()``
+ returns ``id(x)``.
.. method:: object.__bool__(self)
Modified: python/branches/py3k-pep3137/Include/unicodeobject.h
==============================================================================
--- python/branches/py3k-pep3137/Include/unicodeobject.h (original)
+++ python/branches/py3k-pep3137/Include/unicodeobject.h Sun Nov 4
13:32:02 2007
@@ -155,6 +155,7 @@
# define PyUnicode_DecodeCharmap PyUnicodeUCS2_DecodeCharmap
# define PyUnicode_DecodeLatin1 PyUnicodeUCS2_DecodeLatin1
# define PyUnicode_DecodeFSDefault PyUnicodeUCS2_DecodeFSDefault
+# define PyUnicode_DecodeFSDefaultAndSize PyUnicodeUCS2_DecodeFSDefaultAndSize
# define PyUnicode_DecodeRawUnicodeEscape PyUnicodeUCS2_DecodeRawUnicodeEscape
# define PyUnicode_DecodeUTF32 PyUnicodeUCS2_DecodeUTF32
# define PyUnicode_DecodeUTF32Stateful PyUnicodeUCS2_DecodeUTF32Stateful
@@ -247,6 +248,7 @@
# define PyUnicode_DecodeCharmap PyUnicodeUCS4_DecodeCharmap
# define PyUnicode_DecodeLatin1 PyUnicodeUCS4_DecodeLatin1
# define PyUnicode_DecodeFSDefault PyUnicodeUCS4_DecodeFSDefault
+# define PyUnicode_DecodeFSDefaultAndSize PyUnicodeUCS4_DecodeFSDefaultAndSize
# define PyUnicode_DecodeRawUnicodeEscape PyUnicodeUCS4_DecodeRawUnicodeEscape
# define PyUnicode_DecodeUTF32 PyUnicodeUCS4_DecodeUTF32
# define PyUnicode_DecodeUTF32Stateful PyUnicodeUCS4_DecodeUTF32Stateful
@@ -657,6 +659,12 @@
const char *s /* encoded string */
);
+PyAPI_FUNC(PyObject*) PyUnicode_DecodeFSDefaultAndSize(
+ const char *s, /* encoded string */
+ Py_ssize_t size /* size */
+ );
+
+
/* Return a char* holding the UTF-8 encoded value of the
Unicode object.
Modified: python/branches/py3k-pep3137/Lib/test/test_bytes.py
==============================================================================
--- python/branches/py3k-pep3137/Lib/test/test_bytes.py (original)
+++ python/branches/py3k-pep3137/Lib/test/test_bytes.py Sun Nov 4 13:32:02 2007
@@ -8,6 +8,7 @@
import os
import re
import sys
+import copy
import pickle
import tempfile
import unittest
@@ -804,11 +805,89 @@
pass
+class BytesSubclass(bytes):
+ pass
+
+class BytesSubclassTest(unittest.TestCase):
+
+ def test_basic(self):
+ self.assert_(issubclass(BytesSubclass, bytes))
+ self.assert_(isinstance(BytesSubclass(), bytes))
+
+ a, b = b"abcd", b"efgh"
+ _a, _b = BytesSubclass(a), BytesSubclass(b)
+
+ # test comparison operators with subclass instances
+ self.assert_(_a == _a)
+ self.assert_(_a != _b)
+ self.assert_(_a < _b)
+ self.assert_(_a <= _b)
+ self.assert_(_b >= _a)
+ self.assert_(_b > _a)
+ self.assert_(_a is not a)
+
+ # test concat of subclass instances
+ self.assertEqual(a + b, _a + _b)
+ self.assertEqual(a + b, a + _b)
+ self.assertEqual(a + b, _a + b)
+
+ # test repeat
+ self.assert_(a*5 == _a*5)
+
+ def test_join(self):
+ # Make sure join returns a NEW object for single item sequences
+ # involving a subclass.
+ # Make sure that it is of the appropriate type.
+ s1 = BytesSubclass(b"abcd")
+ s2 = b"".join([s1])
+ self.assert_(s1 is not s2)
+ self.assert_(type(s2) is bytes)
+
+ # Test reverse, calling join on subclass
+ s3 = s1.join([b"abcd"])
+ self.assert_(type(s3) is bytes)
+
+ def test_pickle(self):
+ a = BytesSubclass(b"abcd")
+ a.x = 10
+ a.y = BytesSubclass(b"efgh")
+ for proto in range(pickle.HIGHEST_PROTOCOL):
+ b = pickle.loads(pickle.dumps(a, proto))
+ self.assertNotEqual(id(a), id(b))
+ self.assertEqual(a, b)
+ self.assertEqual(a.x, b.x)
+ self.assertEqual(a.y, b.y)
+ self.assertEqual(type(a), type(b))
+ self.assertEqual(type(a.y), type(b.y))
+
+ def test_copy(self):
+ a = BytesSubclass(b"abcd")
+ a.x = 10
+ a.y = BytesSubclass(b"efgh")
+ for copy_method in (copy.copy, copy.deepcopy):
+ b = copy_method(a)
+ self.assertNotEqual(id(a), id(b))
+ self.assertEqual(a, b)
+ self.assertEqual(a.x, b.x)
+ self.assertEqual(a.y, b.y)
+ self.assertEqual(type(a), type(b))
+ self.assertEqual(type(a.y), type(b.y))
+
+ def test_init_override(self):
+ class subclass(bytes):
+ def __init__(self, newarg=1, *args, **kwargs):
+ bytes.__init__(self, *args, **kwargs)
+ x = subclass(4, source=b"abcd")
+ self.assertEqual(x, b"abcd")
+ x = subclass(newarg=4, source=b"abcd")
+ self.assertEqual(x, b"abcd")
+
+
def test_main():
test.test_support.run_unittest(BytesTest)
test.test_support.run_unittest(BytesAsStringTest)
+ test.test_support.run_unittest(BytesSubclassTest)
test.test_support.run_unittest(BufferPEP3137Test)
if __name__ == "__main__":
- ##test_main()
- unittest.main()
+ test_main()
Modified: python/branches/py3k-pep3137/Lib/test/test_import.py
==============================================================================
--- python/branches/py3k-pep3137/Lib/test/test_import.py (original)
+++ python/branches/py3k-pep3137/Lib/test/test_import.py Sun Nov 4
13:32:02 2007
@@ -171,7 +171,7 @@
shutil.rmtree(self.path)
sys.path = self.syspath
- def test_sys_path_with_unicode(self):
+ def XXX_test_sys_path_with_unicode(self):
for i, subpath in enumerate(self.SAMPLES):
path = os.path.join(self.path, subpath)
os.mkdir(path)
Modified: python/branches/py3k-pep3137/Objects/bytesobject.c
==============================================================================
--- python/branches/py3k-pep3137/Objects/bytesobject.c (original)
+++ python/branches/py3k-pep3137/Objects/bytesobject.c Sun Nov 4 13:32:02 2007
@@ -2923,13 +2923,21 @@
static PyObject *
bytes_reduce(PyBytesObject *self)
{
- PyObject *latin1;
+ PyObject *latin1, *dict;
if (self->ob_bytes)
latin1 = PyUnicode_DecodeLatin1(self->ob_bytes,
Py_Size(self), NULL);
else
latin1 = PyUnicode_FromString("");
- return Py_BuildValue("(O(Ns))", Py_Type(self), latin1, "latin-1");
+
+ dict = PyObject_GetAttrString((PyObject *)self, "__dict__");
+ if (dict == NULL) {
+ PyErr_Clear();
+ dict = Py_None;
+ Py_INCREF(dict);
+ }
+
+ return Py_BuildValue("(O(Ns)N)", Py_Type(self), latin1, "latin-1", dict);
}
static PySequenceMethods bytes_as_sequence = {
@@ -3059,8 +3067,7 @@
PyObject_GenericGetAttr, /* tp_getattro */
0, /* tp_setattro */
&bytes_as_buffer, /* tp_as_buffer */
- /* bytes is 'final' or 'sealed' */
- Py_TPFLAGS_DEFAULT, /* tp_flags */
+ Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE, /* tp_flags */
bytes_doc, /* tp_doc */
0, /* tp_traverse */
0, /* tp_clear */
Modified: python/branches/py3k-pep3137/Objects/unicodeobject.c
==============================================================================
--- python/branches/py3k-pep3137/Objects/unicodeobject.c (original)
+++ python/branches/py3k-pep3137/Objects/unicodeobject.c Sun Nov 4
13:32:02 2007
@@ -1256,10 +1256,14 @@
}
PyObject*
-PyUnicode_DecodeFSDefault(const char *s)
-{
+PyUnicode_DecodeFSDefault(const char *s) {
Py_ssize_t size = (Py_ssize_t)strlen(s);
+ return PyUnicode_DecodeFSDefaultAndSize(s, size);
+}
+PyObject*
+PyUnicode_DecodeFSDefaultAndSize(const char *s, Py_ssize_t size)
+{
/* During the early bootstrapping process, Py_FileSystemDefaultEncoding
can be undefined. If it is case, decode using UTF-8. The following
assumes
that Py_FileSystemDefaultEncoding is set to a built-in encoding during
the
@@ -1267,11 +1271,11 @@
*/
if (Py_FileSystemDefaultEncoding) {
#if defined(MS_WINDOWS) && defined(HAVE_USABLE_WCHAR_T)
- if (strcmp(Py_FileSystemDefaultEncoding, "mbcs")) {
+ if (strcmp(Py_FileSystemDefaultEncoding, "mbcs") == 0) {
return PyUnicode_DecodeMBCS(s, size, "replace");
}
#elif defined(__APPLE__)
- if (strcmp(Py_FileSystemDefaultEncoding, "utf-8")) {
+ if (strcmp(Py_FileSystemDefaultEncoding, "utf-8") == 0) {
return PyUnicode_DecodeUTF8(s, size, "replace");
}
#endif
Modified: python/branches/py3k-pep3137/Python/import.c
==============================================================================
--- python/branches/py3k-pep3137/Python/import.c (original)
+++ python/branches/py3k-pep3137/Python/import.c Sun Nov 4 13:32:02 2007
@@ -2385,7 +2385,7 @@
subname = name;
else {
PyObject *parentname, *parent;
- parentname = PyString_FromStringAndSize(name, (subname-name));
+ parentname = PyUnicode_FromStringAndSize(name, (subname-name));
if (parentname == NULL) {
imp_modules_reloading_clear();
return NULL;
@@ -2394,7 +2394,7 @@
if (parent == NULL) {
PyErr_Format(PyExc_ImportError,
"reload(): parent %.200s not in sys.modules",
- PyString_AS_STRING(parentname));
+ PyUnicode_AsString(parentname));
Py_DECREF(parentname);
imp_modules_reloading_clear();
return NULL;
Modified: python/branches/py3k-pep3137/Python/sysmodule.c
==============================================================================
--- python/branches/py3k-pep3137/Python/sysmodule.c (original)
+++ python/branches/py3k-pep3137/Python/sysmodule.c Sun Nov 4 13:32:02 2007
@@ -1145,7 +1145,7 @@
p = strchr(path, delim);
if (p == NULL)
p = strchr(path, '\0'); /* End of string */
- w = PyUnicode_FromStringAndSize(path, (Py_ssize_t) (p - path));
+ w = PyUnicode_DecodeFSDefaultAndSize(path, (Py_ssize_t) (p -
path));
if (w == NULL) {
Py_DECREF(v);
return NULL;
_______________________________________________
Python-3000-checkins mailing list
[email protected]
http://mail.python.org/mailman/listinfo/python-3000-checkins
