diff --git a/Lib/test/string_tests.py b/Lib/test/string_tests.py index 1cc36d836b2..24a4a57e59d 100644 --- a/Lib/test/string_tests.py +++ b/Lib/test/string_tests.py @@ -1183,6 +1183,63 @@ class MixinStrUnicodeUserStringTest: self.checkraises(ValueError, S, 'rpartition', '') self.checkraises(TypeError, S, 'rpartition', None) + def test_none_arguments(self): + # issue 11828 + s = 'hello' + self.checkequal(2, s, 'find', 'l', None) + self.checkequal(3, s, 'find', 'l', -2, None) + self.checkequal(2, s, 'find', 'l', None, -2) + self.checkequal(0, s, 'find', 'h', None, None) + + self.checkequal(3, s, 'rfind', 'l', None) + self.checkequal(3, s, 'rfind', 'l', -2, None) + self.checkequal(2, s, 'rfind', 'l', None, -2) + self.checkequal(0, s, 'rfind', 'h', None, None) + + self.checkequal(2, s, 'index', 'l', None) + self.checkequal(3, s, 'index', 'l', -2, None) + self.checkequal(2, s, 'index', 'l', None, -2) + self.checkequal(0, s, 'index', 'h', None, None) + + self.checkequal(3, s, 'rindex', 'l', None) + self.checkequal(3, s, 'rindex', 'l', -2, None) + self.checkequal(2, s, 'rindex', 'l', None, -2) + self.checkequal(0, s, 'rindex', 'h', None, None) + + self.checkequal(2, s, 'count', 'l', None) + self.checkequal(1, s, 'count', 'l', -2, None) + self.checkequal(1, s, 'count', 'l', None, -2) + self.checkequal(0, s, 'count', 'x', None, None) + + self.checkequal(True, s, 'endswith', 'o', None) + self.checkequal(True, s, 'endswith', 'lo', -2, None) + self.checkequal(True, s, 'endswith', 'l', None, -2) + self.checkequal(False, s, 'endswith', 'x', None, None) + + self.checkequal(True, s, 'startswith', 'h', None) + self.checkequal(True, s, 'startswith', 'l', -2, None) + self.checkequal(True, s, 'startswith', 'h', None, -2) + self.checkequal(False, s, 'startswith', 'x', None, None) + + def test_find_etc_raise_correct_error_messages(self): + # issue 11828 + s = 'hello' + x = 'x' + self.assertRaisesRegexp(TypeError, r'^find\(', s.find, + x, None, None, None) + self.assertRaisesRegexp(TypeError, r'^rfind\(', s.rfind, + x, None, None, None) + self.assertRaisesRegexp(TypeError, r'^index\(', s.index, + x, None, None, None) + self.assertRaisesRegexp(TypeError, r'^rindex\(', s.rindex, + x, None, None, None) + self.assertRaisesRegexp(TypeError, r'^count\(', s.count, + x, None, None, None) + self.assertRaisesRegexp(TypeError, r'^startswith\(', s.startswith, + x, None, None, None) + self.assertRaisesRegexp(TypeError, r'^endswith\(', s.endswith, + x, None, None, None) + class MixinStrUnicodeTest: # Additional tests that only work with str and unicode. diff --git a/Lib/test/test_bytes.py b/Lib/test/test_bytes.py index 9be1008c364..4e33a42b2b2 100644 --- a/Lib/test/test_bytes.py +++ b/Lib/test/test_bytes.py @@ -476,6 +476,68 @@ class BaseBytesTest(unittest.TestCase): self.assertRaises(ValueError, self.type2test.maketrans, b'abc', b'xyzq') self.assertRaises(TypeError, self.type2test.maketrans, 'abc', 'def') + def test_none_arguments(self): + # issue 11828 + b = self.type2test(b'hello') + l = self.type2test(b'l') + h = self.type2test(b'h') + x = self.type2test(b'x') + o = self.type2test(b'o') + + self.assertEqual(2, b.find(l, None)) + self.assertEqual(3, b.find(l, -2, None)) + self.assertEqual(2, b.find(l, None, -2)) + self.assertEqual(0, b.find(h, None, None)) + + self.assertEqual(3, b.rfind(l, None)) + self.assertEqual(3, b.rfind(l, -2, None)) + self.assertEqual(2, b.rfind(l, None, -2)) + self.assertEqual(0, b.rfind(h, None, None)) + + self.assertEqual(2, b.index(l, None)) + self.assertEqual(3, b.index(l, -2, None)) + self.assertEqual(2, b.index(l, None, -2)) + self.assertEqual(0, b.index(h, None, None)) + + self.assertEqual(3, b.rindex(l, None)) + self.assertEqual(3, b.rindex(l, -2, None)) + self.assertEqual(2, b.rindex(l, None, -2)) + self.assertEqual(0, b.rindex(h, None, None)) + + self.assertEqual(2, b.count(l, None)) + self.assertEqual(1, b.count(l, -2, None)) + self.assertEqual(1, b.count(l, None, -2)) + self.assertEqual(0, b.count(x, None, None)) + + self.assertEqual(True, b.endswith(o, None)) + self.assertEqual(True, b.endswith(o, -2, None)) + self.assertEqual(True, b.endswith(l, None, -2)) + self.assertEqual(False, b.endswith(x, None, None)) + + self.assertEqual(True, b.startswith(h, None)) + self.assertEqual(True, b.startswith(l, -2, None)) + self.assertEqual(True, b.startswith(h, None, -2)) + self.assertEqual(False, b.startswith(x, None, None)) + + def test_find_etc_raise_correct_error_messages(self): + # issue 11828 + b = self.type2test(b'hello') + x = self.type2test(b'x') + self.assertRaisesRegexp(TypeError, r'\bfind\b', b.find, + x, None, None, None) + self.assertRaisesRegexp(TypeError, r'\brfind\b', b.rfind, + x, None, None, None) + self.assertRaisesRegexp(TypeError, r'\bindex\b', b.index, + x, None, None, None) + self.assertRaisesRegexp(TypeError, r'\brindex\b', b.rindex, + x, None, None, None) + self.assertRaisesRegexp(TypeError, r'\bcount\b', b.count, + x, None, None, None) + self.assertRaisesRegexp(TypeError, r'\bstartswith\b', b.startswith, + x, None, None, None) + self.assertRaisesRegexp(TypeError, r'\bendswith\b', b.endswith, + x, None, None, None) + class BytesTest(BaseBytesTest): type2test = bytes diff --git a/Misc/ACKS b/Misc/ACKS index a98bab7ed2d..2b9595b996e 100644 --- a/Misc/ACKS +++ b/Misc/ACKS @@ -66,6 +66,7 @@ Donald Beaudry David Beazley Robin Becker Neal Becker +Torsten Becker Bill Bedford Stefan Behnel Reimer Behrends diff --git a/Misc/NEWS b/Misc/NEWS index fe6ba0194b0..36f34e66f34 100644 --- a/Misc/NEWS +++ b/Misc/NEWS @@ -49,6 +49,9 @@ Core and Builtins - Issue #11272: On Windows, input() strips '\r' (and not only '\n'), and sys.stdin uses universal newline (replace '\r\n' by '\n'). +- issue #11828: startswith and endswith don't accept None as slice index. + Patch by Torsten Becker. + - Issue #10830: Fix PyUnicode_FromFormatV("%c") for non-BMP characters on narrow build. diff --git a/Objects/bytearrayobject.c b/Objects/bytearrayobject.c index 0fb76582a9b..827fdedb43d 100644 --- a/Objects/bytearrayobject.c +++ b/Objects/bytearrayobject.c @@ -1081,8 +1081,8 @@ bytearray_find_internal(PyByteArrayObject *self, PyObject *args, int dir) Py_ssize_t start=0, end=PY_SSIZE_T_MAX; Py_ssize_t res; - if (!PyArg_ParseTuple(args, "O|O&O&:find/rfind/index/rindex", &subobj, - _PyEval_SliceIndex, &start, _PyEval_SliceIndex, &end)) + if (!stringlib_parse_args_finds("find/rfind/index/rindex", + args, &subobj, &start, &end)) return -2; if (_getbuffer(subobj, &subbuf) < 0) return -2; @@ -1132,8 +1132,7 @@ bytearray_count(PyByteArrayObject *self, PyObject *args) Py_buffer vsub; PyObject *count_obj; - if (!PyArg_ParseTuple(args, "O|O&O&:count", &sub_obj, - _PyEval_SliceIndex, &start, _PyEval_SliceIndex, &end)) + if (!stringlib_parse_args_finds("count", args, &sub_obj, &start, &end)) return NULL; if (_getbuffer(sub_obj, &vsub) < 0) @@ -1291,8 +1290,7 @@ bytearray_startswith(PyByteArrayObject *self, PyObject *args) PyObject *subobj; int result; - if (!PyArg_ParseTuple(args, "O|O&O&:startswith", &subobj, - _PyEval_SliceIndex, &start, _PyEval_SliceIndex, &end)) + if (!stringlib_parse_args_finds("startswith", args, &subobj, &start, &end)) return NULL; if (PyTuple_Check(subobj)) { Py_ssize_t i; @@ -1331,8 +1329,7 @@ bytearray_endswith(PyByteArrayObject *self, PyObject *args) PyObject *subobj; int result; - if (!PyArg_ParseTuple(args, "O|O&O&:endswith", &subobj, - _PyEval_SliceIndex, &start, _PyEval_SliceIndex, &end)) + if (!stringlib_parse_args_finds("endswith", args, &subobj, &start, &end)) return NULL; if (PyTuple_Check(subobj)) { Py_ssize_t i; diff --git a/Objects/bytesobject.c b/Objects/bytesobject.c index fcc499e4b7b..97512017e38 100644 --- a/Objects/bytesobject.c +++ b/Objects/bytesobject.c @@ -1244,19 +1244,9 @@ bytes_find_internal(PyBytesObject *self, PyObject *args, int dir) const char *sub; Py_ssize_t sub_len; Py_ssize_t start=0, end=PY_SSIZE_T_MAX; - PyObject *obj_start=Py_None, *obj_end=Py_None; - if (!PyArg_ParseTuple(args, "O|OO:find/rfind/index/rindex", &subobj, - &obj_start, &obj_end)) - return -2; - /* To support None in "start" and "end" arguments, meaning - the same as if they were not passed. - */ - if (obj_start != Py_None) - if (!_PyEval_SliceIndex(obj_start, &start)) - return -2; - if (obj_end != Py_None) - if (!_PyEval_SliceIndex(obj_end, &end)) + if (!stringlib_parse_args_finds("find/rfind/index/rindex", + args, &subobj, &start, &end)) return -2; if (PyBytes_Check(subobj)) { @@ -1503,8 +1493,7 @@ bytes_count(PyBytesObject *self, PyObject *args) Py_ssize_t sub_len; Py_ssize_t start = 0, end = PY_SSIZE_T_MAX; - if (!PyArg_ParseTuple(args, "O|O&O&:count", &sub_obj, - _PyEval_SliceIndex, &start, _PyEval_SliceIndex, &end)) + if (!stringlib_parse_args_finds("count", args, &sub_obj, &start, &end)) return NULL; if (PyBytes_Check(sub_obj)) { @@ -2222,8 +2211,7 @@ bytes_startswith(PyBytesObject *self, PyObject *args) PyObject *subobj; int result; - if (!PyArg_ParseTuple(args, "O|O&O&:startswith", &subobj, - _PyEval_SliceIndex, &start, _PyEval_SliceIndex, &end)) + if (!stringlib_parse_args_finds("startswith", args, &subobj, &start, &end)) return NULL; if (PyTuple_Check(subobj)) { Py_ssize_t i; @@ -2263,8 +2251,7 @@ bytes_endswith(PyBytesObject *self, PyObject *args) PyObject *subobj; int result; - if (!PyArg_ParseTuple(args, "O|O&O&:endswith", &subobj, - _PyEval_SliceIndex, &start, _PyEval_SliceIndex, &end)) + if (!stringlib_parse_args_finds("endswith", args, &subobj, &start, &end)) return NULL; if (PyTuple_Check(subobj)) { Py_ssize_t i; diff --git a/Objects/stringlib/find.h b/Objects/stringlib/find.h index f915296cae6..ce615dcb8a4 100644 --- a/Objects/stringlib/find.h +++ b/Objects/stringlib/find.h @@ -93,32 +93,33 @@ stringlib_contains_obj(PyObject* str, PyObject* sub) #endif /* STRINGLIB_WANT_CONTAINS_OBJ */ -#if STRINGLIB_IS_UNICODE - /* This function is a helper for the "find" family (find, rfind, index, -rindex) of unicodeobject.c file, because they all have the same -behaviour for the arguments. +rindex) and for count, startswith and endswith, because they all have +the same behaviour for the arguments. It does not touch the variables received until it knows everything is ok. - -Note that we receive a pointer to the pointer of the substring object, -so when we create that object in this function we don't DECREF it, -because it continues living in the caller functions (those functions, -after finishing using the substring, must DECREF it). */ +#define FORMAT_BUFFER_SIZE 50 + Py_LOCAL_INLINE(int) -_ParseTupleFinds (PyObject *args, PyObject **substring, - Py_ssize_t *start, Py_ssize_t *end) { - PyObject *tmp_substring; +stringlib_parse_args_finds(const char * function_name, PyObject *args, + PyObject **subobj, + Py_ssize_t *start, Py_ssize_t *end) +{ + PyObject *tmp_subobj; Py_ssize_t tmp_start = 0; Py_ssize_t tmp_end = PY_SSIZE_T_MAX; PyObject *obj_start=Py_None, *obj_end=Py_None; + char format[FORMAT_BUFFER_SIZE] = "O|OO:"; + size_t len = strlen(format); - if (!PyArg_ParseTuple(args, "O|OO:find", &tmp_substring, - &obj_start, &obj_end)) + strncpy(format + len, function_name, FORMAT_BUFFER_SIZE - len - 1); + format[FORMAT_BUFFER_SIZE - 1] = '\0'; + + if (!PyArg_ParseTuple(args, format, &tmp_subobj, &obj_start, &obj_end)) return 0; /* To support None in "start" and "end" arguments, meaning @@ -131,16 +132,44 @@ _ParseTupleFinds (PyObject *args, PyObject **substring, if (!_PyEval_SliceIndex(obj_end, &tmp_end)) return 0; - tmp_substring = PyUnicode_FromObject(tmp_substring); - if (!tmp_substring) - return 0; - *start = tmp_start; *end = tmp_end; - *substring = tmp_substring; + *subobj = tmp_subobj; return 1; } +#undef FORMAT_BUFFER_SIZE + +#if STRINGLIB_IS_UNICODE + +/* +Wraps stringlib_parse_args_finds() and additionally ensures that the +first argument is a unicode object. + +Note that we receive a pointer to the pointer of the substring object, +so when we create that object in this function we don't DECREF it, +because it continues living in the caller functions (those functions, +after finishing using the substring, must DECREF it). +*/ + +Py_LOCAL_INLINE(int) +stringlib_parse_args_finds_unicode(const char * function_name, PyObject *args, + PyUnicodeObject **substring, + Py_ssize_t *start, Py_ssize_t *end) +{ + PyObject *tmp_substring; + + if(stringlib_parse_args_finds(function_name, args, &tmp_substring, + start, end)) { + tmp_substring = PyUnicode_FromObject(tmp_substring); + if (!tmp_substring) + return 0; + *substring = (PyUnicodeObject *)tmp_substring; + return 1; + } + return 0; +} + #endif /* STRINGLIB_IS_UNICODE */ #endif /* STRINGLIB_FIND_H */ diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c index cbda72532d8..ddf6f2e20d4 100644 --- a/Objects/unicodeobject.c +++ b/Objects/unicodeobject.c @@ -7442,13 +7442,8 @@ unicode_count(PyUnicodeObject *self, PyObject *args) Py_ssize_t end = PY_SSIZE_T_MAX; PyObject *result; - if (!PyArg_ParseTuple(args, "O|O&O&:count", &substring, - _PyEval_SliceIndex, &start, _PyEval_SliceIndex, &end)) - return NULL; - - substring = (PyUnicodeObject *)PyUnicode_FromObject( - (PyObject *)substring); - if (substring == NULL) + if (!stringlib_parse_args_finds_unicode("count", args, &substring, + &start, &end)) return NULL; ADJUST_INDICES(start, end, self->length); @@ -7585,12 +7580,13 @@ Return -1 on failure."); static PyObject * unicode_find(PyUnicodeObject *self, PyObject *args) { - PyObject *substring; + PyUnicodeObject *substring; Py_ssize_t start; Py_ssize_t end; Py_ssize_t result; - if (!_ParseTupleFinds(args, &substring, &start, &end)) + if (!stringlib_parse_args_finds_unicode("find", args, &substring, + &start, &end)) return NULL; result = stringlib_find_slice( @@ -7647,11 +7643,12 @@ static PyObject * unicode_index(PyUnicodeObject *self, PyObject *args) { Py_ssize_t result; - PyObject *substring; + PyUnicodeObject *substring; Py_ssize_t start; Py_ssize_t end; - if (!_ParseTupleFinds(args, &substring, &start, &end)) + if (!stringlib_parse_args_finds_unicode("index", args, &substring, + &start, &end)) return NULL; result = stringlib_find_slice( @@ -8509,12 +8506,13 @@ Return -1 on failure."); static PyObject * unicode_rfind(PyUnicodeObject *self, PyObject *args) { - PyObject *substring; + PyUnicodeObject *substring; Py_ssize_t start; Py_ssize_t end; Py_ssize_t result; - if (!_ParseTupleFinds(args, &substring, &start, &end)) + if (!stringlib_parse_args_finds_unicode("rfind", args, &substring, + &start, &end)) return NULL; result = stringlib_rfind_slice( @@ -8536,12 +8534,13 @@ Like S.rfind() but raise ValueError when the substring is not found."); static PyObject * unicode_rindex(PyUnicodeObject *self, PyObject *args) { - PyObject *substring; + PyUnicodeObject *substring; Py_ssize_t start; Py_ssize_t end; Py_ssize_t result; - if (!_ParseTupleFinds(args, &substring, &start, &end)) + if (!stringlib_parse_args_finds_unicode("rindex", args, &substring, + &start, &end)) return NULL; result = stringlib_rfind_slice( @@ -9011,8 +9010,7 @@ unicode_startswith(PyUnicodeObject *self, Py_ssize_t end = PY_SSIZE_T_MAX; int result; - if (!PyArg_ParseTuple(args, "O|O&O&:startswith", &subobj, - _PyEval_SliceIndex, &start, _PyEval_SliceIndex, &end)) + if (!stringlib_parse_args_finds("startswith", args, &subobj, &start, &end)) return NULL; if (PyTuple_Check(subobj)) { Py_ssize_t i; @@ -9057,8 +9055,7 @@ unicode_endswith(PyUnicodeObject *self, Py_ssize_t end = PY_SSIZE_T_MAX; int result; - if (!PyArg_ParseTuple(args, "O|O&O&:endswith", &subobj, - _PyEval_SliceIndex, &start, _PyEval_SliceIndex, &end)) + if (!stringlib_parse_args_finds("endswith", args, &subobj, &start, &end)) return NULL; if (PyTuple_Check(subobj)) { Py_ssize_t i;