diff options
| -rw-r--r-- | Lib/test/test_string.py | 23 | ||||
| -rw-r--r-- | Objects/stringobject.c | 20 |
2 files changed, 31 insertions, 12 deletions
diff --git a/Lib/test/test_string.py b/Lib/test/test_string.py index 859dd4e..ba9d9d3 100644 --- a/Lib/test/test_string.py +++ b/Lib/test/test_string.py @@ -52,6 +52,29 @@ class StringTest( self.checkraises(TypeError, string_tests.BadSeq1(), 'join', ' ') self.checkequal('a b c', string_tests.BadSeq2(), 'join', ' ') + def test_bug1001011(self): + # Make sure join returns a NEW object for single item sequences + # involving a subclass + # Make sure that it is of the appropriate type + # Check the optimisation still occurs for standard objects + class str_subclass(str): pass + s1 = str_subclass('abcd') + s2 = ''.join([s1]) + self.failIf(s1 is s2) + self.assertEqual(type(s2), type('')) + s3 = 'abcd' + s4 = ''.join([s3]) + self.failUnless(s3 is s4) + if test_support.have_unicode: + class unicode_subclass(unicode): pass + u1 = unicode_subclass(u'abcd') + u2 = ''.join([u1]) + self.failIf(u1 is u2) + self.assertEqual(type(u2), type(u'')) + u3 = u'abcd' + u4 = ''.join([u3]) + self.failUnless(u3 is u4) + class ModuleTest(unittest.TestCase): def test_attrs(self): diff --git a/Objects/stringobject.c b/Objects/stringobject.c index b40351a..c87b688 100644 --- a/Objects/stringobject.c +++ b/Objects/stringobject.c @@ -1618,22 +1618,18 @@ string_join(PyStringObject *self, PyObject *orig) } if (seqlen == 1) { item = PySequence_Fast_GET_ITEM(seq, 0); - if (!PyString_Check(item) && !PyUnicode_Check(item)) { - PyErr_Format(PyExc_TypeError, - "sequence item 0: expected string," - " %.80s found", - item->ob_type->tp_name); + if (PyString_CheckExact(item) || PyUnicode_CheckExact(item)) { + Py_INCREF(item); Py_DECREF(seq); - return NULL; + return item; } - Py_INCREF(item); - Py_DECREF(seq); - return item; } - /* There are at least two things to join. Do a pre-pass to figure out - * the total amount of space we'll need (sz), see whether any argument - * is absurd, and defer to the Unicode join if appropriate. + /* There are at least two things to join, or else we have a subclass + * of the builtin types in the sequence. + * Do a pre-pass to figure out the total amount of space we'll + * need (sz), see whether any argument is absurd, and defer to + * the Unicode join if appropriate. */ for (i = 0; i < seqlen; i++) { const size_t old_sz = sz; |