cpython: c10a3ddba483 (original) (raw)

Mercurial > cpython

changeset 82059:c10a3ddba483

(Merge 3.3) Issue #17137: When an Unicode string is resized, the internal wide character string (wstr) format is now cleared. [#17137]

line wrap: on

line diff

--- a/Lib/test/test_unicode.py +++ b/Lib/test/test_unicode.py @@ -2191,6 +2191,21 @@ class UnicodeTest(string_tests.CommonTes self.assertEqual(args[0], text) self.assertEqual(len(args), 1)

• def test_resize(self):

•    for length in range(1, 100, 7):[](#l1.8)

•        # generate a fresh string (refcount=1)[](#l1.9)

•        text = 'a' * length + 'b'[](#l1.10)

+

•        # fill wstr internal field[](#l1.12)

•        abc = text.encode('unicode_internal')[](#l1.13)

•        self.assertEqual(abc.decode('unicode_internal'), text)[](#l1.14)

+

•        # resize text: wstr field must be cleared and then recomputed[](#l1.16)

•        text += 'c'[](#l1.17)

•        abcdef = text.encode('unicode_internal')[](#l1.18)

•        self.assertNotEqual(abc, abcdef)[](#l1.19)

•        self.assertEqual(abcdef.decode('unicode_internal'), text)[](#l1.20)

+ class StringModuleTest(unittest.TestCase): def test_formatter_parser(self):

--- a/Misc/NEWS +++ b/Misc/NEWS @@ -10,6 +10,9 @@ What's New in Python 3.4.0 Alpha 1? Core and Builtins ----------------- +- Issue #17137: When an Unicode string is resized, the internal wide character

• string (wstr) format is now cleared. +

• Issue #17043: The unicode-internal decoder no longer read past the end of input buffer.

--- a/Objects/unicodeobject.c +++ b/Objects/unicodeobject.c @@ -717,6 +717,10 @@ resize_compact(PyObject *unicode, Py_ssi if (!PyUnicode_IS_ASCII(unicode)) _PyUnicode_WSTR_LENGTH(unicode) = length; }

• else if (_PyUnicode_HAS_WSTR_MEMORY(unicode)) {

•    PyObject_DEL(_PyUnicode_WSTR(unicode));[](#l3.8)

•    _PyUnicode_WSTR(unicode) = NULL;[](#l3.9)

• }

#ifdef Py_DEBUG unicode_fill_invalid(unicode, old_length); #endif