[Python-checkins] cpython: Issue #22615: Argument Clinic now supports the "type" argument for the

[larry.hastings]

https://hg.python.org/cpython/rev/c0224ff67cdd
changeset: 93023:c0224ff67cdd
parent: 93021:1d5485471457
user: Larry Hastings <larry at hastings.org>
date: Mon Oct 13 10:39:41 2014 +0100
summary:
Issue #22615: Argument Clinic now supports the "type" argument for the
int converter. This permits using the int converter with enums and
typedefs.

files:
Misc/NEWS | 4 +
Modules/arraymodule.c | 111 ++++++++++----------
Modules/clinic/arraymodule.c.h | 6 +-
Objects/bytesobject.c | 2 +-
Tools/clinic/clinic.py | 4 +-
5 files changed, 68 insertions(+), 59 deletions(-)


diff --git a/Misc/NEWS b/Misc/NEWS
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -1263,6 +1263,10 @@
Tools/Demos
-----------

+- Issue #22615: Argument Clinic now supports the "type" argument for the
+ int converter. This permits using the int converter with enums and
+ typedefs.
+
- Issue #20076: The makelocalealias.py script no longer ignores UTF-8 mapping.

- Issue #20079: The makelocalealias.py script now can parse the SUPPORTED file
diff --git a/Modules/arraymodule.c b/Modules/arraymodule.c
--- a/Modules/arraymodule.c
+++ b/Modules/arraymodule.c
@@ -59,7 +59,50 @@

#define PyArrayIter_Check(op) PyObject_TypeCheck(op, &PyArrayIter_Type)

-/* Must come after arrayobject and arrayiterobject definitions. */
+enum machine_format_code {
+ UNKNOWN_FORMAT = -1,
+ /* UNKNOWN_FORMAT is used to indicate that the machine format for an
+ * array type code cannot be interpreted. When this occurs, a list of
+ * Python objects is used to represent the content of the array
+ * instead of using the memory content of the array directly. In that
+ * case, the array_reconstructor mechanism is bypassed completely, and
+ * the standard array constructor is used instead.
+ *
+ * This is will most likely occur when the machine doesn't use IEEE
+ * floating-point numbers.
+ */
+
+ UNSIGNED_INT8 = 0,
+ SIGNED_INT8 = 1,
+ UNSIGNED_INT16_LE = 2,
+ UNSIGNED_INT16_BE = 3,
+ SIGNED_INT16_LE = 4,
+ SIGNED_INT16_BE = 5,
+ UNSIGNED_INT32_LE = 6,
+ UNSIGNED_INT32_BE = 7,
+ SIGNED_INT32_LE = 8,
+ SIGNED_INT32_BE = 9,
+ UNSIGNED_INT64_LE = 10,
+ UNSIGNED_INT64_BE = 11,
+ SIGNED_INT64_LE = 12,
+ SIGNED_INT64_BE = 13,
+ IEEE_754_FLOAT_LE = 14,
+ IEEE_754_FLOAT_BE = 15,
+ IEEE_754_DOUBLE_LE = 16,
+ IEEE_754_DOUBLE_BE = 17,
+ UTF16_LE = 18,
+ UTF16_BE = 19,
+ UTF32_LE = 20,
+ UTF32_BE = 21
+};
+#define MACHINE_FORMAT_CODE_MIN 0
+#define MACHINE_FORMAT_CODE_MAX 21
+
+
+/*
+ * Must come after arrayobject, arrayiterobject,
+ * and enum machine_code_type definitions.
+ */
#include "clinic/arraymodule.c.h"

#define array_Check(op) PyObject_TypeCheck(op, &Arraytype)
@@ -1712,45 +1755,6 @@

/*********************** Pickling support ************************/

-enum machine_format_code {
- UNKNOWN_FORMAT = -1,
- /* UNKNOWN_FORMAT is used to indicate that the machine format for an
- * array type code cannot be interpreted. When this occurs, a list of
- * Python objects is used to represent the content of the array
- * instead of using the memory content of the array directly. In that
- * case, the array_reconstructor mechanism is bypassed completely, and
- * the standard array constructor is used instead.
- *
- * This is will most likely occur when the machine doesn't use IEEE
- * floating-point numbers.
- */
-
- UNSIGNED_INT8 = 0,
- SIGNED_INT8 = 1,
- UNSIGNED_INT16_LE = 2,
- UNSIGNED_INT16_BE = 3,
- SIGNED_INT16_LE = 4,
- SIGNED_INT16_BE = 5,
- UNSIGNED_INT32_LE = 6,
- UNSIGNED_INT32_BE = 7,
- SIGNED_INT32_LE = 8,
- SIGNED_INT32_BE = 9,
- UNSIGNED_INT64_LE = 10,
- UNSIGNED_INT64_BE = 11,
- SIGNED_INT64_LE = 12,
- SIGNED_INT64_BE = 13,
- IEEE_754_FLOAT_LE = 14,
- IEEE_754_FLOAT_BE = 15,
- IEEE_754_DOUBLE_LE = 16,
- IEEE_754_DOUBLE_BE = 17,
- UTF16_LE = 18,
- UTF16_BE = 19,
- UTF32_LE = 20,
- UTF32_BE = 21
-};
-#define MACHINE_FORMAT_CODE_MIN 0
-#define MACHINE_FORMAT_CODE_MAX 21
-
static const struct mformatdescr {
size_t size;
int is_signed;
@@ -1939,13 +1943,12 @@
[clinic start generated code]*/

static PyObject *
-array__array_reconstructor_impl(PyModuleDef *module, PyTypeObject *arraytype, int typecode, int mformat_code, PyObject *items)
-/*[clinic end generated code: output=a0a4ab61c2fbc17a input=450d59a5373c4eea]*/
+array__array_reconstructor_impl(PyModuleDef *module, PyTypeObject *arraytype, int typecode, enum machine_format_code mformat_code, PyObject *items)
+/*[clinic end generated code: output=c51081ec91caf7e9 input=f72492708c0a1d50]*/
{
PyObject *converted_items;
PyObject *result;
struct arraydescr *descr;
- enum machine_format_code mformat_code_enum = mformat_code;

if (!PyType_Check(arraytype)) {
PyErr_Format(PyExc_TypeError,
@@ -1968,8 +1971,8 @@
"second argument must be a valid type code");
return NULL;
}
- if (mformat_code_enum < MACHINE_FORMAT_CODE_MIN ||
- mformat_code_enum > MACHINE_FORMAT_CODE_MAX) {
+ if (mformat_code < MACHINE_FORMAT_CODE_MIN ||
+ mformat_code > MACHINE_FORMAT_CODE_MAX) {
PyErr_SetString(PyExc_ValueError,
"third argument must be a valid machine format code.");
return NULL;
@@ -1982,8 +1985,8 @@
}

/* Fast path: No decoding has to be done. */
- if (mformat_code_enum == typecode_to_mformat_code((char)typecode) ||
- mformat_code_enum == UNKNOWN_FORMAT) {
+ if (mformat_code == typecode_to_mformat_code((char)typecode) ||
+ mformat_code == UNKNOWN_FORMAT) {
return make_array(arraytype, (char)typecode, items);
}

@@ -1992,16 +1995,16 @@
* object is architecturally different from the one that pickled the
* array.
*/
- if (Py_SIZE(items) % mformat_descriptors[mformat_code_enum].size != 0) {
+ if (Py_SIZE(items) % mformat_descriptors[mformat_code].size != 0) {
PyErr_SetString(PyExc_ValueError,
"string length not a multiple of item size");
return NULL;
}
- switch (mformat_code_enum) {
+ switch (mformat_code) {
case IEEE_754_FLOAT_LE:
case IEEE_754_FLOAT_BE: {
int i;
- int le = (mformat_code_enum == IEEE_754_FLOAT_LE) ? 1 : 0;
+ int le = (mformat_code == IEEE_754_FLOAT_LE) ? 1 : 0;
Py_ssize_t itemcount = Py_SIZE(items) / 4;
const unsigned char *memstr =
(unsigned char *)PyBytes_AS_STRING(items);
@@ -2023,7 +2026,7 @@
case IEEE_754_DOUBLE_LE:
case IEEE_754_DOUBLE_BE: {
int i;
- int le = (mformat_code_enum == IEEE_754_DOUBLE_LE) ? 1 : 0;
+ int le = (mformat_code == IEEE_754_DOUBLE_LE) ? 1 : 0;
Py_ssize_t itemcount = Py_SIZE(items) / 8;
const unsigned char *memstr =
(unsigned char *)PyBytes_AS_STRING(items);
@@ -2044,7 +2047,7 @@
}
case UTF16_LE:
case UTF16_BE: {
- int byteorder = (mformat_code_enum == UTF16_LE) ? -1 : 1;
+ int byteorder = (mformat_code == UTF16_LE) ? -1 : 1;
converted_items = PyUnicode_DecodeUTF16(
PyBytes_AS_STRING(items), Py_SIZE(items),
"strict", &byteorder);
@@ -2054,7 +2057,7 @@
}
case UTF32_LE:
case UTF32_BE: {
- int byteorder = (mformat_code_enum == UTF32_LE) ? -1 : 1;
+ int byteorder = (mformat_code == UTF32_LE) ? -1 : 1;
converted_items = PyUnicode_DecodeUTF32(
PyBytes_AS_STRING(items), Py_SIZE(items),
"strict", &byteorder);
@@ -2079,7 +2082,7 @@
case SIGNED_INT64_BE: {
int i;
const struct mformatdescr mf_descr =
- mformat_descriptors[mformat_code_enum];
+ mformat_descriptors[mformat_code];
Py_ssize_t itemcount = Py_SIZE(items) / mf_descr.size;
const unsigned char *memstr =
(unsigned char *)PyBytes_AS_STRING(items);
diff --git a/Modules/clinic/arraymodule.c.h b/Modules/clinic/arraymodule.c.h
--- a/Modules/clinic/arraymodule.c.h
+++ b/Modules/clinic/arraymodule.c.h
@@ -446,7 +446,7 @@
{"_array_reconstructor", (PyCFunction)array__array_reconstructor, METH_VARARGS, array__array_reconstructor__doc__},

static PyObject *
-array__array_reconstructor_impl(PyModuleDef *module, PyTypeObject *arraytype, int typecode, int mformat_code, PyObject *items);
+array__array_reconstructor_impl(PyModuleDef *module, PyTypeObject *arraytype, int typecode, enum machine_format_code mformat_code, PyObject *items);

static PyObject *
array__array_reconstructor(PyModuleDef *module, PyObject *args)
@@ -454,7 +454,7 @@
PyObject *return_value = NULL;
PyTypeObject *arraytype;
int typecode;
- int mformat_code;
+ enum machine_format_code mformat_code;
PyObject *items;

if (!PyArg_ParseTuple(args,
@@ -502,4 +502,4 @@

#define ARRAY_ARRAYITERATOR___SETSTATE___METHODDEF \
{"__setstate__", (PyCFunction)array_arrayiterator___setstate__, METH_O, array_arrayiterator___setstate____doc__},
-/*[clinic end generated code: output=dff8eae01f0ab208 input=a9049054013a1b77]*/
+/*[clinic end generated code: output=e1deb61c6a3bc8c8 input=a9049054013a1b77]*/
diff --git a/Objects/bytesobject.c b/Objects/bytesobject.c
--- a/Objects/bytesobject.c
+++ b/Objects/bytesobject.c
@@ -1819,7 +1819,7 @@

static PyObject *
bytes_translate_impl(PyBytesObject *self, PyObject *table, int group_right_1, PyObject *deletechars)
-/*[clinic end generated code: output=f0f29a57f41df5d8 input=a90fad893c3c88d7]*/
+/*[clinic end generated code: output=f0f29a57f41df5d8 input=d8fa5519d7cc4be7]*/
{
char *input, *output;
const char *table_chars;
diff --git a/Tools/clinic/clinic.py b/Tools/clinic/clinic.py
--- a/Tools/clinic/clinic.py
+++ b/Tools/clinic/clinic.py
@@ -2426,11 +2426,13 @@
format_unit = 'i'
c_ignored_default = "0"

- def converter_init(self, *, types='int'):
+ def converter_init(self, *, types='int', type=None):
if types == 'str':
self.format_unit = 'C'
elif types != 'int':
fail("int_converter: illegal 'types' argument")
+ if type != None:
+ self.type = type

class unsigned_int_converter(CConverter):
type = 'unsigned int'