| ︙ | | |
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
|
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
|
-
+
|
if ((unsigned)(ch - 1) < (UNICODE_SELF - 1)) {
return 1;
}
if (ch <= 0x7FF) {
return 2;
}
#if TCL_UTF_MAX > 3
if (((unsigned)(ch - 0x10000) <= 0xfffff)) {
if (((unsigned)(ch - 0x10000) <= 0xFFFFF)) {
return 4;
}
#endif
return 3;
}
/*
|
| ︙ | | |
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
|
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
|
-
+
-
+
-
+
-
+
-
+
|
} else if (byte < 0xE0) {
if ((src[1] & 0xC0) == 0x80) {
/*
* Two-byte-character lead-byte followed by a trail-byte.
*/
*chPtr = (Tcl_UniChar) (((byte & 0x1F) << 6) | (src[1] & 0x3F));
if ((*chPtr == 0) || (*chPtr > 0x7f)) {
if ((unsigned)(*chPtr - 1) >= (UNICODE_SELF - 1)) {
return 2;
}
}
/*
* A two-byte-character lead-byte not followed by trail-byte
* represents itself.
*/
} else if (byte < 0xF0) {
if (((src[1] & 0xC0) == 0x80) && ((src[2] & 0xC0) == 0x80)) {
/*
* Three-byte-character lead byte followed by two trail bytes.
*/
*chPtr = (Tcl_UniChar) (((byte & 0x0F) << 12)
| ((src[1] & 0x3F) << 6) | (src[2] & 0x3F));
if (*chPtr > 0x7ff) {
if (*chPtr > 0x7FF) {
return 3;
}
}
/*
* A three-byte-character lead-byte not followed by two trail-bytes
* represents itself.
*/
}
#if TCL_UTF_MAX > 3
else if (byte < 0xF8) {
if (((src[1] & 0xC0) == 0x80) && ((src[2] & 0xC0) == 0x80) && ((src[3] & 0xC0) == 0x80)) {
/*
* Four-byte-character lead byte followed by three trail bytes.
*/
*chPtr = (Tcl_UniChar) (((byte & 0x0E) << 18) | ((src[1] & 0x3F) << 12)
*chPtr = (Tcl_UniChar) (((byte & 0x07) << 18) | ((src[1] & 0x3F) << 12)
| ((src[2] & 0x3F) << 6) | (src[3] & 0x3F));
if ((*chPtr <= 0x10ffff) && (*chPtr > 0xffff)) {
if ((unsigned)(*chPtr - 0x10000) <= 0xFFFFF) {
return 4;
}
}
/*
* A three-byte-character lead-byte not followed by two trail-bytes
* A four-byte-character lead-byte not followed by two trail-bytes
* represents itself.
*/
}
#endif
*chPtr = (Tcl_UniChar) byte;
return 1;
|
| ︙ | | |
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
|
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
|
-
+
|
const char *ct, /* UTF string cs is compared to. */
unsigned long numChars) /* Number of UTF chars to compare. */
{
Tcl_UniChar ch1, ch2;
/*
* Cannot use 'memcmp(cs, ct, n);' as byte representation of \u0000 (the
* pair of bytes 0xc0,0x80) is larger than byte representation of \u0001
* pair of bytes 0xC0,0x80) is larger than byte representation of \u0001
* (the byte 0x01.)
*/
while (numChars-- > 0) {
/*
* n must be interpreted as chars, not bytes. This should be called
* only when both strings are of at least n chars long (no need for \0
|
| ︙ | | |
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
|
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
|
-
-
+
+
-
+
|
int
Tcl_UniCharIsControl(
int ch) /* Unicode character to test. */
{
#if TCL_UTF_MAX > 3
if (UNICODE_OUT_OF_RANGE(ch)) {
ch &= 0x1fffff;
if ((ch == 0xe0001) || ((ch >= 0xe0020) && (ch <= 0xe007f))) {
ch &= 0x1FFFFF;
if ((ch == 0xE0001) || ((ch >= 0xE0020) && (ch <= 0xE007f))) {
return 1;
}
if ((ch >= 0xf0000) && ((ch & 0xffff) <= 0xfffd)) {
if ((ch >= 0xF0000) && ((ch & 0xFFFF) <= 0xFFFD)) {
return 1;
}
return 0;
}
#endif
return ((CONTROL_BITS >> GetCategory(ch)) & 1);
}
|
| ︙ | | |
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
|
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
|
-
-
+
+
|
int
Tcl_UniCharIsGraph(
int ch) /* Unicode character to test. */
{
#if TCL_UTF_MAX > 3
if (UNICODE_OUT_OF_RANGE(ch)) {
ch &= 0x1fffff;
return (ch >= 0xe0100) && (ch <= 0xe01ef);
ch &= 0x1FFFFF;
return (ch >= 0xE0100) && (ch <= 0xE01EF);
}
#endif
return ((GRAPH_BITS >> GetCategory(ch)) & 1);
}
/*
*----------------------------------------------------------------------
|
| ︙ | | |
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
|
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
|
-
-
+
+
|
int
Tcl_UniCharIsPrint(
int ch) /* Unicode character to test. */
{
#if TCL_UTF_MAX > 3
if (UNICODE_OUT_OF_RANGE(ch)) {
ch &= 0x1fffff;
return (ch >= 0xe0100) && (ch <= 0xe01ef);
ch &= 0x1FFFFF;
return (ch >= 0xE0100) && (ch <= 0xE01EF);
}
#endif
return (((GRAPH_BITS|SPACE_BITS) >> GetCategory(ch)) & 1);
}
/*
*----------------------------------------------------------------------
|
| ︙ | | |
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
|
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
|
-
+
-
+
-
-
+
+
|
int
Tcl_UniCharIsSpace(
int ch) /* Unicode character to test. */
{
#if TCL_UTF_MAX > 3
/* Ignore upper 11 bits. */
ch &= 0x1fffff;
ch &= 0x1FFFFF;
#else
/* Ignore upper 16 bits. */
ch &= 0xffff;
ch &= 0xFFFF;
#endif
/*
* If the character is within the first 127 characters, just use the
* standard C function, otherwise consult the Unicode table.
*/
if (ch < 0x80) {
return TclIsSpaceProc((char) ch);
#if TCL_UTF_MAX > 3
} else if (UNICODE_OUT_OF_RANGE(ch)) {
return 0;
#endif
} else if (ch == 0x0085 || ch == 0x180e || ch == 0x200b
|| ch == 0x202f || ch == 0x2060 || ch == 0xfeff) {
} else if (ch == 0x0085 || ch == 0x180E || ch == 0x200B
|| ch == 0x202F || ch == 0x2060 || ch == 0xFEFF) {
return 1;
} else {
return ((SPACE_BITS >> GetCategory(ch)) & 1);
}
}
/*
|
| ︙ | | |