From 6cf1f31b1738166be03464bf077bf0477ce3f1dc Mon Sep 17 00:00:00 2001 From: Serhiy Storchaka Date: Thu, 3 Dec 2015 01:02:03 +0200 Subject: Issue #25709: Fixed problem with in-place string concatenation and utf-8 cache. --- Lib/test/test_unicode.py | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) (limited to 'Lib/test') diff --git a/Lib/test/test_unicode.py b/Lib/test/test_unicode.py index f359082267..ef0fd1c5f2 100644 --- a/Lib/test/test_unicode.py +++ b/Lib/test/test_unicode.py @@ -2303,6 +2303,23 @@ class UnicodeTest(string_tests.CommonTest, self.assertNotEqual(abc, abcdef) self.assertEqual(abcdef.decode('unicode_internal'), text) + @support.cpython_only + def test_pep393_utf8_caching_bug(self): + # Issue #25709: Problem with string concatenation and utf-8 cache + from _testcapi import getargs_s_hash + for k in 0x24, 0xa4, 0x20ac, 0x1f40d: + s = '' + for i in range(5): + # Due to CPython specific optimization the 's' string can be + # resized in-place. + s += chr(k) + # Parsing with the "s#" format code calls indirectly + # PyUnicode_AsUTF8AndSize() which creates the UTF-8 + # encoded string cached in the Unicode object. + self.assertEqual(getargs_s_hash(s), chr(k).encode() * (i + 1)) + # Check that the second call returns the same result + self.assertEqual(getargs_s_hash(s), chr(k).encode() * (i + 1)) + class StringModuleTest(unittest.TestCase): def test_formatter_parser(self): -- cgit v1.2.1