@@ -174,25 +174,24 @@ BB_setitem(arrayobject *ap, Py_ssize_t i, PyObject *v)
174 174
static PyObject *
175 175
u_getitem(arrayobject *ap, Py_ssize_t i)
176 176
{
177 -
return PyUnicode_FromOrdinal(((Py_UCS4 *) ap->ob_item)[i]);
177 +
return PyUnicode_FromUnicode(&((Py_UNICODE *) ap->ob_item)[i], 1);
178 178
}
179 179 180 180
static int
181 181
u_setitem(arrayobject *ap, Py_ssize_t i, PyObject *v)
182 182
{
183 -
PyObject *p;
183 +
Py_UNICODE *p;
184 +
Py_ssize_t len;
184 185 185 -
if (!PyArg_Parse(v, "U;array item must be unicode character", &p))
186 -
return -1;
187 -
if (PyUnicode_READY(p))
186 +
if (!PyArg_Parse(v, "u#;array item must be unicode character", &p, &len))
188 187
return -1;
189 -
if (PyUnicode_GET_LENGTH(p) != 1) {
188 +
if (len != 1) {
190 189
PyErr_SetString(PyExc_TypeError,
191 190
"array item must be unicode character");
192 191
return -1;
193 192
}
194 193
if (i >= 0)
195 -
((Py_UCS4 *)ap->ob_item)[i] = PyUnicode_READ_CHAR(p, 0);
194 +
((Py_UNICODE *)ap->ob_item)[i] = p[0];
196 195
return 0;
197 196
}
198 197
@@ -444,13 +443,6 @@ d_setitem(arrayobject *ap, Py_ssize_t i, PyObject *v)
444 443
return 0;
445 444
}
446 445 447 -
#if SIZEOF_INT == 4
448 -
# define STRUCT_LONG_FORMAT "I"
449 -
#elif SIZEOF_LONG == 4
450 -
# define STRUCT_LONG_FORMAT "L"
451 -
#else
452 -
# error "Unable to get struct format for Py_UCS4"
453 -
#endif
454 446 455 447
/* Description of types.
456 448
*
@@ -460,7 +452,7 @@ d_setitem(arrayobject *ap, Py_ssize_t i, PyObject *v)
460 452
static struct arraydescr descriptors[] = {
461 453
{'b', 1, b_getitem, b_setitem, "b", 1, 1},
462 454
{'B', 1, BB_getitem, BB_setitem, "B", 1, 0},
463 -
{'u', sizeof(Py_UCS4), u_getitem, u_setitem, STRUCT_LONG_FORMAT, 0, 0},
455 +
{'u', sizeof(Py_UNICODE), u_getitem, u_setitem, "u", 0, 0},
464 456
{'h', sizeof(short), h_getitem, h_setitem, "h", 1, 1},
465 457
{'H', sizeof(short), HH_getitem, HH_setitem, "H", 1, 0},
466 458
{'i', sizeof(int), i_getitem, i_setitem, "i", 1, 1},
@@ -1519,26 +1511,25 @@ This method is deprecated. Use tobytes instead.");
1519 1511
static PyObject *
1520 1512
array_fromunicode(arrayobject *self, PyObject *args)
1521 1513
{
1522 -
PyObject *ustr;
1514 +
Py_UNICODE *ustr;
1523 1515
Py_ssize_t n;
1516 +
char typecode;
1524 1517 1525 -
if (!PyArg_ParseTuple(args, "U:fromunicode", &ustr))
1518 +
if (!PyArg_ParseTuple(args, "u#:fromunicode", &ustr, &n))
1526 1519
return NULL;
1527 -
if (self->ob_descr->typecode != 'u') {
1520 +
typecode = self->ob_descr->typecode;
1521 +
if ((typecode != 'u')) {
1528 1522
PyErr_SetString(PyExc_ValueError,
1529 1523
"fromunicode() may only be called on "
1530 1524
"unicode type arrays");
1531 1525
return NULL;
1532 1526
}
1533 -
if (PyUnicode_READY(ustr))
1534 -
return NULL;
1535 -
n = PyUnicode_GET_LENGTH(ustr);
1536 1527
if (n > 0) {
1537 1528
Py_ssize_t old_size = Py_SIZE(self);
1538 1529
if (array_resize(self, old_size + n) == -1)
1539 1530
return NULL;
1540 -
if (!PyUnicode_AsUCS4(ustr, (Py_UCS4 *)self->ob_item + old_size, n, 0))
1541 -
return NULL;
1531 +
memcpy(self->ob_item + old_size * sizeof(Py_UNICODE),
1532 +
ustr, n * sizeof(Py_UNICODE));
1542 1533
}
1543 1534 1544 1535
Py_INCREF(Py_None);
@@ -1557,14 +1548,14 @@ append Unicode data to an array of some other type.");
1557 1548
static PyObject *
1558 1549
array_tounicode(arrayobject *self, PyObject *unused)
1559 1550
{
1560 -
if (self->ob_descr->typecode != 'u') {
1551 +
char typecode;
1552 +
typecode = self->ob_descr->typecode;
1553 +
if ((typecode != 'u')) {
1561 1554
PyErr_SetString(PyExc_ValueError,
1562 1555
"tounicode() may only be called on unicode type arrays");
1563 1556
return NULL;
1564 1557
}
1565 -
return PyUnicode_FromKindAndData(PyUnicode_4BYTE_KIND,
1566 -
(Py_UCS4 *) self->ob_item,
1567 -
Py_SIZE(self));
1558 +
return PyUnicode_FromUnicode((Py_UNICODE *) self->ob_item, Py_SIZE(self));
1568 1559
}
1569 1560 1570 1561
PyDoc_STRVAR(tounicode_doc,
@@ -1671,7 +1662,13 @@ typecode_to_mformat_code(char typecode)
1671 1662
return UNSIGNED_INT8;
1672 1663 1673 1664
case 'u':
1674 -
return UTF32_LE + is_big_endian;
1665 +
if (sizeof(Py_UNICODE) == 2) {
1666 +
return UTF16_LE + is_big_endian;
1667 +
}
1668 +
if (sizeof(Py_UNICODE) == 4) {
1669 +
return UTF32_LE + is_big_endian;
1670 +
}
1671 +
return UNKNOWN_FORMAT;
1675 1672 1676 1673
case 'f':
1677 1674
if (sizeof(float) == 4) {
@@ -2419,8 +2416,14 @@ array_buffer_getbuf(arrayobject *self, Py_buffer *view, int flags)
2419 2416
view->strides = &(view->itemsize);
2420 2417
view->format = NULL;
2421 2418
view->internal = NULL;
2422 -
if ((flags & PyBUF_FORMAT) == PyBUF_FORMAT)
2419 +
if ((flags & PyBUF_FORMAT) == PyBUF_FORMAT) {
2423 2420
view->format = self->ob_descr->formats;
2421 +
#ifdef Py_UNICODE_WIDE
2422 +
if (self->ob_descr->typecode == 'u') {
2423 +
view->format = "w";
2424 +
}
2425 +
#endif
2426 +
}
2424 2427 2425 2428
finish:
2426 2429
self->ob_exports++;
@@ -2534,25 +2537,29 @@ array_new(PyTypeObject *type, PyObject *args, PyObject *kwds)
2534 2537
Py_DECREF(v);
2535 2538
}
2536 2539
else if (initial != NULL && PyUnicode_Check(initial)) {
2540 +
Py_UNICODE *ustr;
2537 2541
Py_ssize_t n;
2538 -
if (PyUnicode_READY(initial)) {
2542 + 2543 +
ustr = PyUnicode_AsUnicode(initial);
2544 +
if (ustr == NULL) {
2545 +
PyErr_NoMemory();
2539 2546
Py_DECREF(a);
2540 2547
return NULL;
2541 2548
}
2542 -
n = PyUnicode_GET_LENGTH(initial);
2549 + 2550 +
n = PyUnicode_GET_DATA_SIZE(initial);
2543 2551
if (n > 0) {
2544 2552
arrayobject *self = (arrayobject *)a;
2545 -
Py_UCS4 *item = (Py_UCS4 *)self->ob_item;
2546 -
item = (Py_UCS4 *)PyMem_Realloc(item, n * sizeof(Py_UCS4));
2553 +
char *item = self->ob_item;
2554 +
item = (char *)PyMem_Realloc(item, n);
2547 2555
if (item == NULL) {
2548 2556
PyErr_NoMemory();
2549 2557
Py_DECREF(a);
2550 2558
return NULL;
2551 2559
}
2552 -
self->ob_item = (char*)item;
2553 -
Py_SIZE(self) = n;
2554 -
if (!PyUnicode_AsUCS4(initial, item, n, 0))
2555 -
return NULL;
2560 +
self->ob_item = item;
2561 +
Py_SIZE(self) = n / sizeof(Py_UNICODE);
2562 +
memcpy(item, ustr, n);
2556 2563
self->allocated = Py_SIZE(self);
2557 2564
}
2558 2565
}
@@ -2593,7 +2600,7 @@ is a single character. The following type codes are defined:\n\
2593 2600
Type code C Type Minimum size in bytes \n\
2594 2601
'b' signed integer 1 \n\
2595 2602
'B' unsigned integer 1 \n\
2596 -
'u' Unicode character 4 \n\
2603 +
'u' Unicode character 2 (see note) \n\
2597 2604
'h' signed integer 2 \n\
2598 2605
'H' unsigned integer 2 \n\
2599 2606
'i' signed integer 2 \n\
@@ -2605,6 +2612,9 @@ is a single character. The following type codes are defined:\n\
2605 2612
'f' floating point 4 \n\
2606 2613
'd' floating point 8 \n\
2607 2614
\n\
2615 +
NOTE: The 'u' typecode corresponds to Python's unicode character. On \n\
2616 +
narrow builds this is 2-bytes on wide builds this is 4-bytes.\n\
2617 +
\n\
2608 2618
NOTE: The 'q' and 'Q' type codes are only available if the platform \n\
2609 2619
C compiler used to build Python supports 'long long', or, on Windows, \n\
2610 2620
'__int64'.\n\
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4