aboutsummaryrefslogtreecommitdiffstats
path: root/contrib/libs/icu/include/unicode/uniset.h
blob: 84774d9f36ecf675bc5d3eaa6713fe4cbd709af7 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
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
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
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
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
***************************************************************************
* Copyright (C) 1999-2016, International Business Machines Corporation
* and others. All Rights Reserved.
***************************************************************************
*   Date        Name        Description
*   10/20/99    alan        Creation.
***************************************************************************
*/

#ifndef UNICODESET_H
#define UNICODESET_H

#include "unicode/utypes.h"

#if U_SHOW_CPLUSPLUS_API

#include "unicode/ucpmap.h"
#include "unicode/unifilt.h"
#include "unicode/unistr.h"
#include "unicode/uset.h"

/**
 * \file
 * \brief C++ API: Unicode Set
 */

U_NAMESPACE_BEGIN

// Forward Declarations.
class BMPSet;
class ParsePosition;
class RBBIRuleScanner;
class SymbolTable;
class UnicodeSetStringSpan;
class UVector;
class RuleCharacterIterator;

/**
 * A mutable set of Unicode characters and multicharacter strings.  Objects of this class
 * represent <em>character classes</em> used in regular expressions.
 * A character specifies a subset of Unicode code points.  Legal
 * code points are U+0000 to U+10FFFF, inclusive.
 *
 * <p>The UnicodeSet class is not designed to be subclassed.
 *
 * <p><code>UnicodeSet</code> supports two APIs. The first is the
 * <em>operand</em> API that allows the caller to modify the value of
 * a <code>UnicodeSet</code> object. It conforms to Java 2's
 * <code>java.util.Set</code> interface, although
 * <code>UnicodeSet</code> does not actually implement that
 * interface. All methods of <code>Set</code> are supported, with the
 * modification that they take a character range or single character
 * instead of an <code>Object</code>, and they take a
 * <code>UnicodeSet</code> instead of a <code>Collection</code>.  The
 * operand API may be thought of in terms of boolean logic: a boolean
 * OR is implemented by <code>add</code>, a boolean AND is implemented
 * by <code>retain</code>, a boolean XOR is implemented by
 * <code>complement</code> taking an argument, and a boolean NOT is
 * implemented by <code>complement</code> with no argument.  In terms
 * of traditional set theory function names, <code>add</code> is a
 * union, <code>retain</code> is an intersection, <code>remove</code>
 * is an asymmetric difference, and <code>complement</code> with no
 * argument is a set complement with respect to the superset range
 * <code>MIN_VALUE-MAX_VALUE</code>
 *
 * <p>The second API is the
 * <code>applyPattern()</code>/<code>toPattern()</code> API from the
 * <code>java.text.Format</code>-derived classes.  Unlike the
 * methods that add characters, add categories, and control the logic
 * of the set, the method <code>applyPattern()</code> sets all
 * attributes of a <code>UnicodeSet</code> at once, based on a
 * string pattern.
 *
 * <p><b>Pattern syntax</b></p>
 *
 * Patterns are accepted by the constructors and the
 * <code>applyPattern()</code> methods and returned by the
 * <code>toPattern()</code> method.  These patterns follow a syntax
 * similar to that employed by version 8 regular expression character
 * classes.  Here are some simple examples:
 *
 * \htmlonly<blockquote>\endhtmlonly
 *   <table>
 *     <tr align="top">
 *       <td nowrap valign="top" align="left"><code>[]</code></td>
 *       <td valign="top">No characters</td>
 *     </tr><tr align="top">
 *       <td nowrap valign="top" align="left"><code>[a]</code></td>
 *       <td valign="top">The character 'a'</td>
 *     </tr><tr align="top">
 *       <td nowrap valign="top" align="left"><code>[ae]</code></td>
 *       <td valign="top">The characters 'a' and 'e'</td>
 *     </tr>
 *     <tr>
 *       <td nowrap valign="top" align="left"><code>[a-e]</code></td>
 *       <td valign="top">The characters 'a' through 'e' inclusive, in Unicode code
 *       point order</td>
 *     </tr>
 *     <tr>
 *       <td nowrap valign="top" align="left"><code>[\\u4E01]</code></td>
 *       <td valign="top">The character U+4E01</td>
 *     </tr>
 *     <tr>
 *       <td nowrap valign="top" align="left"><code>[a{ab}{ac}]</code></td>
 *       <td valign="top">The character 'a' and the multicharacter strings &quot;ab&quot; and
 *       &quot;ac&quot;</td>
 *     </tr>
 *     <tr>
 *       <td nowrap valign="top" align="left"><code>[\\p{Lu}]</code></td>
 *       <td valign="top">All characters in the general category Uppercase Letter</td>
 *     </tr>
 *   </table>
 * \htmlonly</blockquote>\endhtmlonly
 *
 * Any character may be preceded by a backslash in order to remove any special
 * meaning.  White space characters, as defined by UCharacter.isWhitespace(), are
 * ignored, unless they are escaped.
 *
 * <p>Property patterns specify a set of characters having a certain
 * property as defined by the Unicode standard.  Both the POSIX-like
 * "[:Lu:]" and the Perl-like syntax "\\p{Lu}" are recognized.  For a
 * complete list of supported property patterns, see the User's Guide
 * for UnicodeSet at
 * <a href="https://unicode-org.github.io/icu/userguide/strings/unicodeset">
 * https://unicode-org.github.io/icu/userguide/strings/unicodeset</a>.
 * Actual determination of property data is defined by the underlying
 * Unicode database as implemented by UCharacter.
 *
 * <p>Patterns specify individual characters, ranges of characters, and
 * Unicode property sets.  When elements are concatenated, they
 * specify their union.  To complement a set, place a '^' immediately
 * after the opening '['.  Property patterns are inverted by modifying
 * their delimiters; "[:^foo]" and "\\P{foo}".  In any other location,
 * '^' has no special meaning.
 *
 * <p>Since ICU 70, "[^...]", "[:^foo]", "\\P{foo}", and "[:binaryProperty=No:]"
 * perform a “code point complement” (all code points minus the original set),
 * removing all multicharacter strings,
 * equivalent to <code>.complement().removeAllStrings()</code>.
 * The complement() API function continues to perform a
 * symmetric difference with all code points and thus retains all multicharacter strings.
 *
 * <p>Ranges are indicated by placing two a '-' between two
 * characters, as in "a-z".  This specifies the range of all
 * characters from the left to the right, in Unicode order.  If the
 * left character is greater than or equal to the
 * right character it is a syntax error.  If a '-' occurs as the first
 * character after the opening '[' or '[^', or if it occurs as the
 * last character before the closing ']', then it is taken as a
 * literal.  Thus "[a\-b]", "[-ab]", and "[ab-]" all indicate the same
 * set of three characters, 'a', 'b', and '-'.
 *
 * <p>Sets may be intersected using the '&' operator or the asymmetric
 * set difference may be taken using the '-' operator, for example,
 * "[[:L:]&[\\u0000-\\u0FFF]]" indicates the set of all Unicode letters
 * with values less than 4096.  Operators ('&' and '|') have equal
 * precedence and bind left-to-right.  Thus
 * "[[:L:]-[a-z]-[\\u0100-\\u01FF]]" is equivalent to
 * "[[[:L:]-[a-z]]-[\\u0100-\\u01FF]]".  This only really matters for
 * difference; intersection is commutative.
 *
 * <table>
 * <tr valign=top><td nowrap><code>[a]</code><td>The set containing 'a'
 * <tr valign=top><td nowrap><code>[a-z]</code><td>The set containing 'a'
 * through 'z' and all letters in between, in Unicode order
 * <tr valign=top><td nowrap><code>[^a-z]</code><td>The set containing
 * all characters but 'a' through 'z',
 * that is, U+0000 through 'a'-1 and 'z'+1 through U+10FFFF
 * <tr valign=top><td nowrap><code>[[<em>pat1</em>][<em>pat2</em>]]</code>
 * <td>The union of sets specified by <em>pat1</em> and <em>pat2</em>
 * <tr valign=top><td nowrap><code>[[<em>pat1</em>]&[<em>pat2</em>]]</code>
 * <td>The intersection of sets specified by <em>pat1</em> and <em>pat2</em>
 * <tr valign=top><td nowrap><code>[[<em>pat1</em>]-[<em>pat2</em>]]</code>
 * <td>The asymmetric difference of sets specified by <em>pat1</em> and
 * <em>pat2</em>
 * <tr valign=top><td nowrap><code>[:Lu:] or \\p{Lu}</code>
 * <td>The set of characters having the specified
 * Unicode property; in
 * this case, Unicode uppercase letters
 * <tr valign=top><td nowrap><code>[:^Lu:] or \\P{Lu}</code>
 * <td>The set of characters <em>not</em> having the given
 * Unicode property
 * </table>
 *
 * <p><b>Formal syntax</b></p>
 *
 * \htmlonly<blockquote>\endhtmlonly
 *   <table>
 *     <tr align="top">
 *       <td nowrap valign="top" align="right"><code>pattern :=&nbsp; </code></td>
 *       <td valign="top"><code>('[' '^'? item* ']') |
 *       property</code></td>
 *     </tr>
 *     <tr align="top">
 *       <td nowrap valign="top" align="right"><code>item :=&nbsp; </code></td>
 *       <td valign="top"><code>char | (char '-' char) | pattern-expr<br>
 *       </code></td>
 *     </tr>
 *     <tr align="top">
 *       <td nowrap valign="top" align="right"><code>pattern-expr :=&nbsp; </code></td>
 *       <td valign="top"><code>pattern | pattern-expr pattern |
 *       pattern-expr op pattern<br>
 *       </code></td>
 *     </tr>
 *     <tr align="top">
 *       <td nowrap valign="top" align="right"><code>op :=&nbsp; </code></td>
 *       <td valign="top"><code>'&amp;' | '-'<br>
 *       </code></td>
 *     </tr>
 *     <tr align="top">
 *       <td nowrap valign="top" align="right"><code>special :=&nbsp; </code></td>
 *       <td valign="top"><code>'[' | ']' | '-'<br>
 *       </code></td>
 *     </tr>
 *     <tr align="top">
 *       <td nowrap valign="top" align="right"><code>char :=&nbsp; </code></td>
 *       <td valign="top"><em>any character that is not</em><code> special<br>
 *       | ('\' </code><em>any character</em><code>)<br>
 *       | ('\\u' hex hex hex hex)<br>
 *       </code></td>
 *     </tr>
 *     <tr align="top">
 *       <td nowrap valign="top" align="right"><code>hex :=&nbsp; </code></td>
 *       <td valign="top"><code>'0' | '1' | '2' | '3' | '4' | '5' | '6' | '7' | '8' | '9' |<br>
 *       &nbsp;&nbsp;&nbsp;&nbsp;'A' | 'B' | 'C' | 'D' | 'E' | 'F' | 'a' | 'b' | 'c' | 'd' | 'e' | 'f'</code></td>
 *     </tr>
 *     <tr>
 *       <td nowrap valign="top" align="right"><code>property :=&nbsp; </code></td>
 *       <td valign="top"><em>a Unicode property set pattern</em></td>
 *     </tr>
 *   </table>
 *   <br>
 *   <table border="1">
 *     <tr>
 *       <td>Legend: <table>
 *         <tr>
 *           <td nowrap valign="top"><code>a := b</code></td>
 *           <td width="20" valign="top">&nbsp; </td>
 *           <td valign="top"><code>a</code> may be replaced by <code>b</code> </td>
 *         </tr>
 *         <tr>
 *           <td nowrap valign="top"><code>a?</code></td>
 *           <td valign="top"></td>
 *           <td valign="top">zero or one instance of <code>a</code><br>
 *           </td>
 *         </tr>
 *         <tr>
 *           <td nowrap valign="top"><code>a*</code></td>
 *           <td valign="top"></td>
 *           <td valign="top">one or more instances of <code>a</code><br>
 *           </td>
 *         </tr>
 *         <tr>
 *           <td nowrap valign="top"><code>a | b</code></td>
 *           <td valign="top"></td>
 *           <td valign="top">either <code>a</code> or <code>b</code><br>
 *           </td>
 *         </tr>
 *         <tr>
 *           <td nowrap valign="top"><code>'a'</code></td>
 *           <td valign="top"></td>
 *           <td valign="top">the literal string between the quotes </td>
 *         </tr>
 *       </table>
 *       </td>
 *     </tr>
 *   </table>
 * \htmlonly</blockquote>\endhtmlonly
 * 
 * <p>Note:
 *  - Most UnicodeSet methods do not take a UErrorCode parameter because
 *   there are usually very few opportunities for failure other than a shortage
 *   of memory, error codes in low-level C++ string methods would be inconvenient,
 *   and the error code as the last parameter (ICU convention) would prevent
 *   the use of default parameter values.
 *   Instead, such methods set the UnicodeSet into a "bogus" state
 *   (see isBogus()) if an error occurs.
 *
 * @author Alan Liu
 * @stable ICU 2.0
 */
class U_COMMON_API UnicodeSet final : public UnicodeFilter {
private:
    /**
     * Enough for sets with few ranges.
     * For example, White_Space has 10 ranges, list length 21.
     */
    static constexpr int32_t INITIAL_CAPACITY = 25;
    // fFlags constant
    static constexpr uint8_t kIsBogus = 1;  // This set is bogus (i.e. not valid)

    UChar32* list = stackList; // MUST be terminated with HIGH
    int32_t capacity = INITIAL_CAPACITY; // capacity of list
    int32_t len = 1; // length of list used; 1 <= len <= capacity
    uint8_t fFlags = 0;         // Bit flag (see constants above)

    BMPSet *bmpSet = nullptr; // The set is frozen iff either bmpSet or stringSpan is not nullptr.
    UChar32* buffer = nullptr; // internal buffer, may be nullptr
    int32_t bufferCapacity = 0; // capacity of buffer

    /**
     * The pattern representation of this set.  This may not be the
     * most economical pattern.  It is the pattern supplied to
     * applyPattern(), with variables substituted and whitespace
     * removed.  For sets constructed without applyPattern(), or
     * modified using the non-pattern API, this string will be empty,
     * indicating that toPattern() must generate a pattern
     * representation from the inversion list.
     */
    char16_t *pat = nullptr;
    int32_t patLen = 0;

    UVector* strings = nullptr; // maintained in sorted order
    UnicodeSetStringSpan *stringSpan = nullptr;

    /**
     * Initial list array.
     * Avoids some heap allocations, and list is never nullptr.
     * Increases the object size a bit.
     */
    UChar32 stackList[INITIAL_CAPACITY];

public:
    /**
     * Determine if this object contains a valid set.
     * A bogus set has no value. It is different from an empty set.
     * It can be used to indicate that no set value is available.
     *
     * @return true if the set is bogus/invalid, false otherwise
     * @see setToBogus()
     * @stable ICU 4.0
     */
    inline UBool isBogus(void) const;

    /**
     * Make this UnicodeSet object invalid.
     * The string will test true with isBogus().
     *
     * A bogus set has no value. It is different from an empty set.
     * It can be used to indicate that no set value is available.
     *
     * This utility function is used throughout the UnicodeSet
     * implementation to indicate that a UnicodeSet operation failed,
     * and may be used in other functions,
     * especially but not exclusively when such functions do not
     * take a UErrorCode for simplicity.
     *
     * @see isBogus()
     * @stable ICU 4.0
     */
    void setToBogus();

public:

    enum {
        /**
         * Minimum value that can be stored in a UnicodeSet.
         * @stable ICU 2.4
         */
        MIN_VALUE = 0,

        /**
         * Maximum value that can be stored in a UnicodeSet.
         * @stable ICU 2.4
         */
        MAX_VALUE = 0x10ffff
    };

    //----------------------------------------------------------------
    // Constructors &c
    //----------------------------------------------------------------

public:

    /**
     * Constructs an empty set.
     * @stable ICU 2.0
     */
    UnicodeSet();

    /**
     * Constructs a set containing the given range. If <code>end <
     * start</code> then an empty set is created.
     *
     * @param start first character, inclusive, of range
     * @param end last character, inclusive, of range
     * @stable ICU 2.4
     */
    UnicodeSet(UChar32 start, UChar32 end);

#ifndef U_HIDE_INTERNAL_API
    /**
     * @internal
     */
    enum ESerialization {
      kSerialized  /* result of serialize() */
    };

    /**
     * Constructs a set from the output of serialize().
     *
     * @param buffer the 16 bit array
     * @param bufferLen the original length returned from serialize()
     * @param serialization the value 'kSerialized'
     * @param status error code
     *
     * @internal
     */
    UnicodeSet(const uint16_t buffer[], int32_t bufferLen,
               ESerialization serialization, UErrorCode &status);
#endif  /* U_HIDE_INTERNAL_API */

    /**
     * Constructs a set from the given pattern.  See the class
     * description for the syntax of the pattern language.
     * @param pattern a string specifying what characters are in the set
     * @param status returns <code>U_ILLEGAL_ARGUMENT_ERROR</code> if the pattern
     * contains a syntax error.
     * @stable ICU 2.0
     */
    UnicodeSet(const UnicodeString& pattern,
               UErrorCode& status);

#ifndef U_HIDE_INTERNAL_API
    /**
     * Constructs a set from the given pattern.  See the class
     * description for the syntax of the pattern language.
     * @param pattern a string specifying what characters are in the set
     * @param options bitmask for options to apply to the pattern.
     * Valid options are USET_IGNORE_SPACE and
     * at most one of USET_CASE_INSENSITIVE, USET_ADD_CASE_MAPPINGS, USET_SIMPLE_CASE_INSENSITIVE.
     * These case options are mutually exclusive.
     * @param symbols a symbol table mapping variable names to values
     * and stand-in characters to UnicodeSets; may be nullptr
     * @param status returns <code>U_ILLEGAL_ARGUMENT_ERROR</code> if the pattern
     * contains a syntax error.
     * @internal
     */
    UnicodeSet(const UnicodeString& pattern,
               uint32_t options,
               const SymbolTable* symbols,
               UErrorCode& status);
#endif  /* U_HIDE_INTERNAL_API */

    /**
     * Constructs a set from the given pattern.  See the class description
     * for the syntax of the pattern language.
     * @param pattern a string specifying what characters are in the set
     * @param pos on input, the position in pattern at which to start parsing.
     * On output, the position after the last character parsed.
     * @param options bitmask for options to apply to the pattern.
     * Valid options are USET_IGNORE_SPACE and
     * at most one of USET_CASE_INSENSITIVE, USET_ADD_CASE_MAPPINGS, USET_SIMPLE_CASE_INSENSITIVE.
     * These case options are mutually exclusive.
     * @param symbols a symbol table mapping variable names to values
     * and stand-in characters to UnicodeSets; may be nullptr
     * @param status input-output error code
     * @stable ICU 2.8
     */
    UnicodeSet(const UnicodeString& pattern, ParsePosition& pos,
               uint32_t options,
               const SymbolTable* symbols,
               UErrorCode& status);

    /**
     * Constructs a set that is identical to the given UnicodeSet.
     * @stable ICU 2.0
     */
    UnicodeSet(const UnicodeSet& o);

    /**
     * Destructs the set.
     * @stable ICU 2.0
     */
    virtual ~UnicodeSet();

    /**
     * Assigns this object to be a copy of another.
     * A frozen set will not be modified.
     * @stable ICU 2.0
     */
    UnicodeSet& operator=(const UnicodeSet& o);

    /**
     * Compares the specified object with this set for equality.  Returns
     * <tt>true</tt> if the two sets
     * have the same size, and every member of the specified set is
     * contained in this set (or equivalently, every member of this set is
     * contained in the specified set).
     *
     * @param o set to be compared for equality with this set.
     * @return <tt>true</tt> if the specified set is equal to this set.
     * @stable ICU 2.0
     */
    virtual bool operator==(const UnicodeSet& o) const;

    /**
     * Compares the specified object with this set for equality.  Returns
     * <tt>true</tt> if the specified set is not equal to this set.
     * @stable ICU 2.0
     */
    inline bool operator!=(const UnicodeSet& o) const;

    /**
     * Returns a copy of this object.  All UnicodeFunctor objects have
     * to support cloning in order to allow classes using
     * UnicodeFunctors, such as Transliterator, to implement cloning.
     * If this set is frozen, then the clone will be frozen as well.
     * Use cloneAsThawed() for a mutable clone of a frozen set.
     * @see cloneAsThawed
     * @stable ICU 2.0
     */
    virtual UnicodeSet* clone() const override;

    /**
     * Returns the hash code value for this set.
     *
     * @return the hash code value for this set.
     * @see Object#hashCode()
     * @stable ICU 2.0
     */
    virtual int32_t hashCode(void) const;

    /**
     * Get a UnicodeSet pointer from a USet
     *
     * @param uset a USet (the ICU plain C type for UnicodeSet)
     * @return the corresponding UnicodeSet pointer.
     *
     * @stable ICU 4.2
     */
    inline static UnicodeSet *fromUSet(USet *uset);

    /**
     * Get a UnicodeSet pointer from a const USet
     *
     * @param uset a const USet (the ICU plain C type for UnicodeSet)
     * @return the corresponding UnicodeSet pointer.
     *
     * @stable ICU 4.2
     */
    inline static const UnicodeSet *fromUSet(const USet *uset);
    
    /**
     * Produce a USet * pointer for this UnicodeSet.
     * USet is the plain C type for UnicodeSet
     *
     * @return a USet pointer for this UnicodeSet
     * @stable ICU 4.2
     */
    inline USet *toUSet();


    /**
     * Produce a const USet * pointer for this UnicodeSet.
     * USet is the plain C type for UnicodeSet
     *
     * @return a const USet pointer for this UnicodeSet
     * @stable ICU 4.2
     */
    inline const USet * toUSet() const;


    //----------------------------------------------------------------
    // Freezable API
    //----------------------------------------------------------------

    /**
     * Determines whether the set has been frozen (made immutable) or not.
     * See the ICU4J Freezable interface for details.
     * @return true/false for whether the set has been frozen
     * @see freeze
     * @see cloneAsThawed
     * @stable ICU 3.8
     */
    inline UBool isFrozen() const;

    /**
     * Freeze the set (make it immutable).
     * Once frozen, it cannot be unfrozen and is therefore thread-safe
     * until it is deleted.
     * See the ICU4J Freezable interface for details.
     * Freezing the set may also make some operations faster, for example
     * contains() and span().
     * A frozen set will not be modified. (It remains frozen.)
     * @return this set.
     * @see isFrozen
     * @see cloneAsThawed
     * @stable ICU 3.8
     */
    UnicodeSet *freeze();

    /**
     * Clone the set and make the clone mutable.
     * See the ICU4J Freezable interface for details.
     * @return the mutable clone
     * @see freeze
     * @see isFrozen
     * @stable ICU 3.8
     */
    UnicodeSet *cloneAsThawed() const;

    //----------------------------------------------------------------
    // Public API
    //----------------------------------------------------------------

    /**
     * Make this object represent the range `start - end`.
     * If `start > end` then this object is set to an empty range.
     * A frozen set will not be modified.
     *
     * @param start first character in the set, inclusive
     * @param end last character in the set, inclusive
     * @stable ICU 2.4
     */
    UnicodeSet& set(UChar32 start, UChar32 end);

    /**
     * Return true if the given position, in the given pattern, appears
     * to be the start of a UnicodeSet pattern.
     * @stable ICU 2.4
     */
    static UBool resemblesPattern(const UnicodeString& pattern,
                                  int32_t pos);

    /**
     * Modifies this set to represent the set specified by the given
     * pattern, ignoring Unicode Pattern_White_Space characters.
     * See the class description for the syntax of the pattern language.
     * A frozen set will not be modified.
     * @param pattern a string specifying what characters are in the set
     * @param status returns <code>U_ILLEGAL_ARGUMENT_ERROR</code> if the pattern
     * contains a syntax error.
     * <em> Empties the set passed before applying the pattern.</em>
     * @return a reference to this
     * @stable ICU 2.0
     */
    UnicodeSet& applyPattern(const UnicodeString& pattern,
                             UErrorCode& status);

#ifndef U_HIDE_INTERNAL_API
    /**
     * Modifies this set to represent the set specified by the given
     * pattern, optionally ignoring Unicode Pattern_White_Space characters.
     * See the class description for the syntax of the pattern language.
     * A frozen set will not be modified.
     * @param pattern a string specifying what characters are in the set
     * @param options bitmask for options to apply to the pattern.
     * Valid options are USET_IGNORE_SPACE and
     * at most one of USET_CASE_INSENSITIVE, USET_ADD_CASE_MAPPINGS, USET_SIMPLE_CASE_INSENSITIVE.
     * These case options are mutually exclusive.
     * @param symbols a symbol table mapping variable names to
     * values and stand-ins to UnicodeSets; may be nullptr
     * @param status returns <code>U_ILLEGAL_ARGUMENT_ERROR</code> if the pattern
     * contains a syntax error.
     *<em> Empties the set passed before applying the pattern.</em>
     * @return a reference to this
     * @internal
     */
    UnicodeSet& applyPattern(const UnicodeString& pattern,
                             uint32_t options,
                             const SymbolTable* symbols,
                             UErrorCode& status);
#endif  /* U_HIDE_INTERNAL_API */

    /**
     * Parses the given pattern, starting at the given position.  The
     * character at pattern.charAt(pos.getIndex()) must be '[', or the
     * parse fails.  Parsing continues until the corresponding closing
     * ']'.  If a syntax error is encountered between the opening and
     * closing brace, the parse fails.  Upon return from a successful
     * parse, the ParsePosition is updated to point to the character
     * following the closing ']', and a StringBuffer containing a
     * pairs list for the parsed pattern is returned.  This method calls
     * itself recursively to parse embedded subpatterns.
     *<em> Empties the set passed before applying the pattern.</em>
     * A frozen set will not be modified.
     *
     * @param pattern the string containing the pattern to be parsed.
     * The portion of the string from pos.getIndex(), which must be a
     * '[', to the corresponding closing ']', is parsed.
     * @param pos upon entry, the position at which to being parsing.
     * The character at pattern.charAt(pos.getIndex()) must be a '['.
     * Upon return from a successful parse, pos.getIndex() is either
     * the character after the closing ']' of the parsed pattern, or
     * pattern.length() if the closing ']' is the last character of
     * the pattern string.
     * @param options bitmask for options to apply to the pattern.
     * Valid options are USET_IGNORE_SPACE and
     * at most one of USET_CASE_INSENSITIVE, USET_ADD_CASE_MAPPINGS, USET_SIMPLE_CASE_INSENSITIVE.
     * These case options are mutually exclusive.
     * @param symbols a symbol table mapping variable names to
     * values and stand-ins to UnicodeSets; may be nullptr
     * @param status returns <code>U_ILLEGAL_ARGUMENT_ERROR</code> if the pattern
     * contains a syntax error.
     * @return a reference to this
     * @stable ICU 2.8
     */
    UnicodeSet& applyPattern(const UnicodeString& pattern,
                             ParsePosition& pos,
                             uint32_t options,
                             const SymbolTable* symbols,
                             UErrorCode& status);

    /**
     * Returns a string representation of this set.  If the result of
     * calling this function is passed to a UnicodeSet constructor, it
     * will produce another set that is equal to this one.
     * A frozen set will not be modified.
     * @param result the string to receive the rules.  Previous
     * contents will be deleted.
     * @param escapeUnprintable if true then convert unprintable
     * character to their hex escape representations, \\uxxxx or
     * \\Uxxxxxxxx.  Unprintable characters are those other than
     * U+000A, U+0020..U+007E.
     * @stable ICU 2.0
     */
    virtual UnicodeString& toPattern(UnicodeString& result,
                                     UBool escapeUnprintable = false) const override;

    /**
     * Modifies this set to contain those code points which have the given value
     * for the given binary or enumerated property, as returned by
     * u_getIntPropertyValue.  Prior contents of this set are lost.
     * A frozen set will not be modified.
     *
     * @param prop a property in the range UCHAR_BIN_START..UCHAR_BIN_LIMIT-1
     * or UCHAR_INT_START..UCHAR_INT_LIMIT-1
     * or UCHAR_MASK_START..UCHAR_MASK_LIMIT-1.
     *
     * @param value a value in the range u_getIntPropertyMinValue(prop)..
     * u_getIntPropertyMaxValue(prop), with one exception.  If prop is
     * UCHAR_GENERAL_CATEGORY_MASK, then value should not be a UCharCategory, but
     * rather a mask value produced by U_GET_GC_MASK().  This allows grouped
     * categories such as [:L:] to be represented.
     *
     * @param ec error code input/output parameter
     *
     * @return a reference to this set
     *
     * @stable ICU 2.4
     */
    UnicodeSet& applyIntPropertyValue(UProperty prop,
                                      int32_t value,
                                      UErrorCode& ec);

    /**
     * Modifies this set to contain those code points which have the
     * given value for the given property.  Prior contents of this
     * set are lost.
     * A frozen set will not be modified.
     *
     * @param prop a property alias, either short or long.  The name is matched
     * loosely.  See PropertyAliases.txt for names and a description of loose
     * matching.  If the value string is empty, then this string is interpreted
     * as either a General_Category value alias, a Script value alias, a binary
     * property alias, or a special ID.  Special IDs are matched loosely and
     * correspond to the following sets:
     *
     * "ANY" = [\\u0000-\\U0010FFFF],
     * "ASCII" = [\\u0000-\\u007F],
     * "Assigned" = [:^Cn:].
     *
     * @param value a value alias, either short or long.  The name is matched
     * loosely.  See PropertyValueAliases.txt for names and a description of
     * loose matching.  In addition to aliases listed, numeric values and
     * canonical combining classes may be expressed numerically, e.g., ("nv",
     * "0.5") or ("ccc", "220").  The value string may also be empty.
     *
     * @param ec error code input/output parameter
     *
     * @return a reference to this set
     *
     * @stable ICU 2.4
     */
    UnicodeSet& applyPropertyAlias(const UnicodeString& prop,
                                   const UnicodeString& value,
                                   UErrorCode& ec);

    /**
     * Returns the number of elements in this set (its cardinality).
     * Note than the elements of a set may include both individual
     * codepoints and strings.
     *
     * This is slower than getRangeCount() because
     * it counts the code points of all ranges.
     *
     * @return the number of elements in this set (its cardinality).
     * @stable ICU 2.0
     * @see getRangeCount
     */
    virtual int32_t size(void) const;

    /**
     * Returns <tt>true</tt> if this set contains no elements.
     *
     * @return <tt>true</tt> if this set contains no elements.
     * @stable ICU 2.0
     */
    virtual UBool isEmpty(void) const;

    /**
     * @return true if this set contains multi-character strings or the empty string.
     * @stable ICU 70
     */
    UBool hasStrings() const;

    /**
     * Returns true if this set contains the given character.
     * This function works faster with a frozen set.
     * @param c character to be checked for containment
     * @return true if the test condition is met
     * @stable ICU 2.0
     */
    virtual UBool contains(UChar32 c) const override;

    /**
     * Returns true if this set contains every character
     * of the given range.
     * @param start first character, inclusive, of the range
     * @param end last character, inclusive, of the range
     * @return true if the test condition is met
     * @stable ICU 2.0
     */
    virtual UBool contains(UChar32 start, UChar32 end) const;

    /**
     * Returns <tt>true</tt> if this set contains the given
     * multicharacter string.
     * @param s string to be checked for containment
     * @return <tt>true</tt> if this set contains the specified string
     * @stable ICU 2.4
     */
    UBool contains(const UnicodeString& s) const;

    /**
     * Returns true if this set contains all the characters and strings
     * of the given set.
     * @param c set to be checked for containment
     * @return true if the test condition is met
     * @stable ICU 2.4
     */
    virtual UBool containsAll(const UnicodeSet& c) const;

    /**
     * Returns true if this set contains all the characters
     * of the given string.
     * @param s string containing characters to be checked for containment
     * @return true if the test condition is met
     * @stable ICU 2.4
     */
    UBool containsAll(const UnicodeString& s) const;

    /**
     * Returns true if this set contains none of the characters
     * of the given range.
     * @param start first character, inclusive, of the range
     * @param end last character, inclusive, of the range
     * @return true if the test condition is met
     * @stable ICU 2.4
     */
    UBool containsNone(UChar32 start, UChar32 end) const;

    /**
     * Returns true if this set contains none of the characters and strings
     * of the given set.
     * @param c set to be checked for containment
     * @return true if the test condition is met
     * @stable ICU 2.4
     */
    UBool containsNone(const UnicodeSet& c) const;

    /**
     * Returns true if this set contains none of the characters
     * of the given string.
     * @param s string containing characters to be checked for containment
     * @return true if the test condition is met
     * @stable ICU 2.4
     */
    UBool containsNone(const UnicodeString& s) const;

    /**
     * Returns true if this set contains one or more of the characters
     * in the given range.
     * @param start first character, inclusive, of the range
     * @param end last character, inclusive, of the range
     * @return true if the condition is met
     * @stable ICU 2.4
     */
    inline UBool containsSome(UChar32 start, UChar32 end) const;

    /**
     * Returns true if this set contains one or more of the characters
     * and strings of the given set.
     * @param s The set to be checked for containment
     * @return true if the condition is met
     * @stable ICU 2.4
     */
    inline UBool containsSome(const UnicodeSet& s) const;

    /**
     * Returns true if this set contains one or more of the characters
     * of the given string.
     * @param s string containing characters to be checked for containment
     * @return true if the condition is met
     * @stable ICU 2.4
     */
    inline UBool containsSome(const UnicodeString& s) const;

    /**
     * Returns the length of the initial substring of the input string which
     * consists only of characters and strings that are contained in this set
     * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE),
     * or only of characters and strings that are not contained
     * in this set (USET_SPAN_NOT_CONTAINED).
     * See USetSpanCondition for details.
     * Similar to the strspn() C library function.
     * Unpaired surrogates are treated according to contains() of their surrogate code points.
     * This function works faster with a frozen set and with a non-negative string length argument.
     * @param s start of the string
     * @param length of the string; can be -1 for NUL-terminated
     * @param spanCondition specifies the containment condition
     * @return the length of the initial substring according to the spanCondition;
     *         0 if the start of the string does not fit the spanCondition
     * @stable ICU 3.8
     * @see USetSpanCondition
     */
    int32_t span(const char16_t *s, int32_t length, USetSpanCondition spanCondition) const;

    /**
     * Returns the end of the substring of the input string according to the USetSpanCondition.
     * Same as <code>start+span(s.getBuffer()+start, s.length()-start, spanCondition)</code>
     * after pinning start to 0<=start<=s.length().
     * @param s the string
     * @param start the start index in the string for the span operation
     * @param spanCondition specifies the containment condition
     * @return the exclusive end of the substring according to the spanCondition;
     *         the substring s.tempSubStringBetween(start, end) fulfills the spanCondition
     * @stable ICU 4.4
     * @see USetSpanCondition
     */
    inline int32_t span(const UnicodeString &s, int32_t start, USetSpanCondition spanCondition) const;

    /**
     * Returns the start of the trailing substring of the input string which
     * consists only of characters and strings that are contained in this set
     * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE),
     * or only of characters and strings that are not contained
     * in this set (USET_SPAN_NOT_CONTAINED).
     * See USetSpanCondition for details.
     * Unpaired surrogates are treated according to contains() of their surrogate code points.
     * This function works faster with a frozen set and with a non-negative string length argument.
     * @param s start of the string
     * @param length of the string; can be -1 for NUL-terminated
     * @param spanCondition specifies the containment condition
     * @return the start of the trailing substring according to the spanCondition;
     *         the string length if the end of the string does not fit the spanCondition
     * @stable ICU 3.8
     * @see USetSpanCondition
     */
    int32_t spanBack(const char16_t *s, int32_t length, USetSpanCondition spanCondition) const;

    /**
     * Returns the start of the substring of the input string according to the USetSpanCondition.
     * Same as <code>spanBack(s.getBuffer(), limit, spanCondition)</code>
     * after pinning limit to 0<=end<=s.length().
     * @param s the string
     * @param limit the exclusive-end index in the string for the span operation
     *              (use s.length() or INT32_MAX for spanning back from the end of the string)
     * @param spanCondition specifies the containment condition
     * @return the start of the substring according to the spanCondition;
     *         the substring s.tempSubStringBetween(start, limit) fulfills the spanCondition
     * @stable ICU 4.4
     * @see USetSpanCondition
     */
    inline int32_t spanBack(const UnicodeString &s, int32_t limit, USetSpanCondition spanCondition) const;

    /**
     * Returns the length of the initial substring of the input string which
     * consists only of characters and strings that are contained in this set
     * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE),
     * or only of characters and strings that are not contained
     * in this set (USET_SPAN_NOT_CONTAINED).
     * See USetSpanCondition for details.
     * Similar to the strspn() C library function.
     * Malformed byte sequences are treated according to contains(0xfffd).
     * This function works faster with a frozen set and with a non-negative string length argument.
     * @param s start of the string (UTF-8)
     * @param length of the string; can be -1 for NUL-terminated
     * @param spanCondition specifies the containment condition
     * @return the length of the initial substring according to the spanCondition;
     *         0 if the start of the string does not fit the spanCondition
     * @stable ICU 3.8
     * @see USetSpanCondition
     */
    int32_t spanUTF8(const char *s, int32_t length, USetSpanCondition spanCondition) const;

    /**
     * Returns the start of the trailing substring of the input string which
     * consists only of characters and strings that are contained in this set
     * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE),
     * or only of characters and strings that are not contained
     * in this set (USET_SPAN_NOT_CONTAINED).
     * See USetSpanCondition for details.
     * Malformed byte sequences are treated according to contains(0xfffd).
     * This function works faster with a frozen set and with a non-negative string length argument.
     * @param s start of the string (UTF-8)
     * @param length of the string; can be -1 for NUL-terminated
     * @param spanCondition specifies the containment condition
     * @return the start of the trailing substring according to the spanCondition;
     *         the string length if the end of the string does not fit the spanCondition
     * @stable ICU 3.8
     * @see USetSpanCondition
     */
    int32_t spanBackUTF8(const char *s, int32_t length, USetSpanCondition spanCondition) const;

    /**
     * Implement UnicodeMatcher::matches()
     * @stable ICU 2.4
     */
    virtual UMatchDegree matches(const Replaceable& text,
                         int32_t& offset,
                         int32_t limit,
                         UBool incremental) override;

private:
    /**
     * Returns the longest match for s in text at the given position.
     * If limit > start then match forward from start+1 to limit
     * matching all characters except s.charAt(0).  If limit < start,
     * go backward starting from start-1 matching all characters
     * except s.charAt(s.length()-1).  This method assumes that the
     * first character, text.charAt(start), matches s, so it does not
     * check it.
     * @param text the text to match
     * @param start the first character to match.  In the forward
     * direction, text.charAt(start) is matched against s.charAt(0).
     * In the reverse direction, it is matched against
     * s.charAt(s.length()-1).
     * @param limit the limit offset for matching, either last+1 in
     * the forward direction, or last-1 in the reverse direction,
     * where last is the index of the last character to match.
     * @param s
     * @return If part of s matches up to the limit, return |limit -
     * start|.  If all of s matches before reaching the limit, return
     * s.length().  If there is a mismatch between s and text, return
     * 0
     */
    static int32_t matchRest(const Replaceable& text,
                             int32_t start, int32_t limit,
                             const UnicodeString& s);

    /**
     * Returns the smallest value i such that c < list[i].  Caller
     * must ensure that c is a legal value or this method will enter
     * an infinite loop.  This method performs a binary search.
     * @param c a character in the range MIN_VALUE..MAX_VALUE
     * inclusive
     * @return the smallest integer i in the range 0..len-1,
     * inclusive, such that c < list[i]
     */
    int32_t findCodePoint(UChar32 c) const;

public:

    /**
     * Implementation of UnicodeMatcher API.  Union the set of all
     * characters that may be matched by this object into the given
     * set.
     * @param toUnionTo the set into which to union the source characters
     * @stable ICU 2.4
     */
    virtual void addMatchSetTo(UnicodeSet& toUnionTo) const override;

    /**
     * Returns the index of the given character within this set, where
     * the set is ordered by ascending code point.  If the character
     * is not in this set, return -1.  The inverse of this method is
     * <code>charAt()</code>.
     * @return an index from 0..size()-1, or -1
     * @stable ICU 2.4
     */
    int32_t indexOf(UChar32 c) const;

    /**
     * Returns the character at the given index within this set, where
     * the set is ordered by ascending code point.  If the index is
     * out of range for characters, returns (UChar32)-1.
     * The inverse of this method is <code>indexOf()</code>.
     *
     * For iteration, this is slower than UnicodeSetIterator or
     * getRangeCount()/getRangeStart()/getRangeEnd(),
     * because for each call it skips linearly over <code>index</code>
     * characters in the ranges.
     *
     * @param index an index from 0..size()-1
     * @return the character at the given index, or (UChar32)-1.
     * @stable ICU 2.4
     */
    UChar32 charAt(int32_t index) const;

    /**
     * Adds the specified range to this set if it is not already
     * present.  If this set already contains the specified range,
     * the call leaves this set unchanged.  If <code>start > end</code>
     * then an empty range is added, leaving the set unchanged.
     * This is equivalent to a boolean logic OR, or a set UNION.
     * A frozen set will not be modified.
     *
     * @param start first character, inclusive, of range to be added
     * to this set.
     * @param end last character, inclusive, of range to be added
     * to this set.
     * @stable ICU 2.0
     */
    virtual UnicodeSet& add(UChar32 start, UChar32 end);

    /**
     * Adds the specified character to this set if it is not already
     * present.  If this set already contains the specified character,
     * the call leaves this set unchanged.
     * A frozen set will not be modified.
     *
     * @param c the character (code point)
     * @return this object, for chaining
     * @stable ICU 2.0
     */
    UnicodeSet& add(UChar32 c);

    /**
     * Adds the specified multicharacter to this set if it is not already
     * present.  If this set already contains the multicharacter,
     * the call leaves this set unchanged.
     * Thus "ch" => {"ch"}
     * A frozen set will not be modified.
     *
     * @param s the source string
     * @return this object, for chaining
     * @stable ICU 2.4
     */
    UnicodeSet& add(const UnicodeString& s);

 private:
    /**
     * @return a code point IF the string consists of a single one.
     * otherwise returns -1.
     * @param s string to test
     */
    static int32_t getSingleCP(const UnicodeString& s);

    void _add(const UnicodeString& s);

 public:
    /**
     * Adds each of the characters in this string to the set. Note: "ch" => {"c", "h"}
     * If this set already contains any particular character, it has no effect on that character.
     * A frozen set will not be modified.
     * @param s the source string
     * @return this object, for chaining
     * @stable ICU 2.4
     */
    UnicodeSet& addAll(const UnicodeString& s);

    /**
     * Retains EACH of the characters in this string. Note: "ch" == {"c", "h"}
     * A frozen set will not be modified.
     * @param s the source string
     * @return this object, for chaining
     * @stable ICU 2.4
     */
    UnicodeSet& retainAll(const UnicodeString& s);

    /**
     * Complement EACH of the characters in this string. Note: "ch" == {"c", "h"}
     * A frozen set will not be modified.
     * @param s the source string
     * @return this object, for chaining
     * @stable ICU 2.4
     */
    UnicodeSet& complementAll(const UnicodeString& s);

    /**
     * Remove EACH of the characters in this string. Note: "ch" == {"c", "h"}
     * A frozen set will not be modified.
     * @param s the source string
     * @return this object, for chaining
     * @stable ICU 2.4
     */
    UnicodeSet& removeAll(const UnicodeString& s);

    /**
     * Makes a set from a multicharacter string. Thus "ch" => {"ch"}
     *
     * @param s the source string
     * @return a newly created set containing the given string.
     * The caller owns the return object and is responsible for deleting it.
     * @stable ICU 2.4
     */
    static UnicodeSet* U_EXPORT2 createFrom(const UnicodeString& s);


    /**
     * Makes a set from each of the characters in the string. Thus "ch" => {"c", "h"}
     * @param s the source string
     * @return a newly created set containing the given characters
     * The caller owns the return object and is responsible for deleting it.
     * @stable ICU 2.4
     */
    static UnicodeSet* U_EXPORT2 createFromAll(const UnicodeString& s);

    /**
     * Retain only the elements in this set that are contained in the
     * specified range.  If <code>start > end</code> then an empty range is
     * retained, leaving the set empty.  This is equivalent to
     * a boolean logic AND, or a set INTERSECTION.
     * A frozen set will not be modified.
     *
     * @param start first character, inclusive, of range
     * @param end last character, inclusive, of range
     * @stable ICU 2.0
     */
    virtual UnicodeSet& retain(UChar32 start, UChar32 end);


    /**
     * Retain the specified character from this set if it is present.
     * A frozen set will not be modified.
     *
     * @param c the character (code point)
     * @return this object, for chaining
     * @stable ICU 2.0
     */
    UnicodeSet& retain(UChar32 c);

    /**
     * Retains only the specified string from this set if it is present.
     * Upon return this set will be empty if it did not contain s, or
     * will only contain s if it did contain s.
     * A frozen set will not be modified.
     *
     * @param s the source string
     * @return this object, for chaining
     * @stable ICU 69
     */
    UnicodeSet& retain(const UnicodeString &s);

    /**
     * Removes the specified range from this set if it is present.
     * The set will not contain the specified range once the call
     * returns.  If <code>start > end</code> then an empty range is
     * removed, leaving the set unchanged.
     * A frozen set will not be modified.
     *
     * @param start first character, inclusive, of range to be removed
     * from this set.
     * @param end last character, inclusive, of range to be removed
     * from this set.
     * @stable ICU 2.0
     */
    virtual UnicodeSet& remove(UChar32 start, UChar32 end);

    /**
     * Removes the specified character from this set if it is present.
     * The set will not contain the specified range once the call
     * returns.
     * A frozen set will not be modified.
     *
     * @param c the character (code point)
     * @return this object, for chaining
     * @stable ICU 2.0
     */
    UnicodeSet& remove(UChar32 c);

    /**
     * Removes the specified string from this set if it is present.
     * The set will not contain the specified character once the call
     * returns.
     * A frozen set will not be modified.
     * @param s the source string
     * @return this object, for chaining
     * @stable ICU 2.4
     */
    UnicodeSet& remove(const UnicodeString& s);

    /**
     * This is equivalent to
     * <code>complement(MIN_VALUE, MAX_VALUE)</code>.
     *
     * <strong>Note:</strong> This performs a symmetric difference with all code points
     * <em>and thus retains all multicharacter strings</em>.
     * In order to achieve a “code point complement” (all code points minus this set),
     * the easiest is to <code>.complement().removeAllStrings()</code>.
     *
     * A frozen set will not be modified.
     * @stable ICU 2.0
     */
    virtual UnicodeSet& complement();

    /**
     * Complements the specified range in this set.  Any character in
     * the range will be removed if it is in this set, or will be
     * added if it is not in this set.  If <code>start > end</code>
     * then an empty range is complemented, leaving the set unchanged.
     * This is equivalent to a boolean logic XOR.
     * A frozen set will not be modified.
     *
     * @param start first character, inclusive, of range
     * @param end last character, inclusive, of range
     * @stable ICU 2.0
     */
    virtual UnicodeSet& complement(UChar32 start, UChar32 end);

    /**
     * Complements the specified character in this set.  The character
     * will be removed if it is in this set, or will be added if it is
     * not in this set.
     * A frozen set will not be modified.
     *
     * @param c the character (code point)
     * @return this object, for chaining
     * @stable ICU 2.0
     */
    UnicodeSet& complement(UChar32 c);

    /**
     * Complement the specified string in this set.
     * The string will be removed if it is in this set, or will be added if it is not in this set.
     * A frozen set will not be modified.
     *
     * @param s the string to complement
     * @return this object, for chaining
     * @stable ICU 2.4
     */
    UnicodeSet& complement(const UnicodeString& s);

    /**
     * Adds all of the elements in the specified set to this set if
     * they're not already present.  This operation effectively
     * modifies this set so that its value is the <i>union</i> of the two
     * sets.  The behavior of this operation is unspecified if the specified
     * collection is modified while the operation is in progress.
     * A frozen set will not be modified.
     *
     * @param c set whose elements are to be added to this set.
     * @see #add(UChar32, UChar32)
     * @stable ICU 2.0
     */
    virtual UnicodeSet& addAll(const UnicodeSet& c);

    /**
     * Retains only the elements in this set that are contained in the
     * specified set.  In other words, removes from this set all of
     * its elements that are not contained in the specified set.  This
     * operation effectively modifies this set so that its value is
     * the <i>intersection</i> of the two sets.
     * A frozen set will not be modified.
     *
     * @param c set that defines which elements this set will retain.
     * @stable ICU 2.0
     */
    virtual UnicodeSet& retainAll(const UnicodeSet& c);

    /**
     * Removes from this set all of its elements that are contained in the
     * specified set.  This operation effectively modifies this
     * set so that its value is the <i>asymmetric set difference</i> of
     * the two sets.
     * A frozen set will not be modified.
     *
     * @param c set that defines which elements will be removed from
     *          this set.
     * @stable ICU 2.0
     */
    virtual UnicodeSet& removeAll(const UnicodeSet& c);

    /**
     * Complements in this set all elements contained in the specified
     * set.  Any character in the other set will be removed if it is
     * in this set, or will be added if it is not in this set.
     * A frozen set will not be modified.
     *
     * @param c set that defines which elements will be xor'ed from
     *          this set.
     * @stable ICU 2.4
     */
    virtual UnicodeSet& complementAll(const UnicodeSet& c);

    /**
     * Removes all of the elements from this set.  This set will be
     * empty after this call returns.
     * A frozen set will not be modified.
     * @stable ICU 2.0
     */
    virtual UnicodeSet& clear(void);

    /**
     * Close this set over the given attribute.  For the attribute
     * USET_CASE_INSENSITIVE, the result is to modify this set so that:
     *
     * 1. For each character or string 'a' in this set, all strings or
     * characters 'b' such that foldCase(a) == foldCase(b) are added
     * to this set.
     *
     * 2. For each string 'e' in the resulting set, if e !=
     * foldCase(e), 'e' will be removed.
     *
     * Example: [aq\\u00DF{Bc}{bC}{Fi}] => [aAqQ\\u00DF\\uFB01{ss}{bc}{fi}]
     *
     * (Here foldCase(x) refers to the operation u_strFoldCase, and a
     * == b denotes that the contents are the same, not pointer
     * comparison.)
     *
     * A frozen set will not be modified.
     *
     * @param attribute bitmask for attributes to close over.
     * Valid options:
     * At most one of USET_CASE_INSENSITIVE, USET_ADD_CASE_MAPPINGS, USET_SIMPLE_CASE_INSENSITIVE.
     * These case options are mutually exclusive.
     * Unrelated options bits are ignored.
     * @return a reference to this set.
     * @stable ICU 4.2
     */
    UnicodeSet& closeOver(int32_t attribute);

    /**
     * Remove all strings from this set.
     *
     * @return a reference to this set.
     * @stable ICU 4.2
     */
    virtual UnicodeSet &removeAllStrings();

    /**
     * Iteration method that returns the number of ranges contained in
     * this set.
     * @see #getRangeStart
     * @see #getRangeEnd
     * @stable ICU 2.4
     */
    virtual int32_t getRangeCount(void) const;

    /**
     * Iteration method that returns the first character in the
     * specified range of this set.
     * @see #getRangeCount
     * @see #getRangeEnd
     * @stable ICU 2.4
     */
    virtual UChar32 getRangeStart(int32_t index) const;

    /**
     * Iteration method that returns the last character in the
     * specified range of this set.
     * @see #getRangeStart
     * @see #getRangeEnd
     * @stable ICU 2.4
     */
    virtual UChar32 getRangeEnd(int32_t index) const;

    /**
     * Serializes this set into an array of 16-bit integers.  Serialization
     * (currently) only records the characters in the set; multicharacter
     * strings are ignored.
     *
     * The array has following format (each line is one 16-bit
     * integer):
     *
     *  length     = (n+2*m) | (m!=0?0x8000:0)
     *  bmpLength  = n; present if m!=0
     *  bmp[0]
     *  bmp[1]
     *  ...
     *  bmp[n-1]
     *  supp-high[0]
     *  supp-low[0]
     *  supp-high[1]
     *  supp-low[1]
     *  ...
     *  supp-high[m-1]
     *  supp-low[m-1]
     *
     * The array starts with a header.  After the header are n bmp
     * code points, then m supplementary code points.  Either n or m
     * or both may be zero.  n+2*m is always <= 0x7FFF.
     *
     * If there are no supplementary characters (if m==0) then the
     * header is one 16-bit integer, 'length', with value n.
     *
     * If there are supplementary characters (if m!=0) then the header
     * is two 16-bit integers.  The first, 'length', has value
     * (n+2*m)|0x8000.  The second, 'bmpLength', has value n.
     *
     * After the header the code points are stored in ascending order.
     * Supplementary code points are stored as most significant 16
     * bits followed by least significant 16 bits.
     *
     * @param dest pointer to buffer of destCapacity 16-bit integers.
     * May be nullptr only if destCapacity is zero.
     * @param destCapacity size of dest, or zero.  Must not be negative.
     * @param ec error code.  Will be set to U_INDEX_OUTOFBOUNDS_ERROR
     * if n+2*m > 0x7FFF.  Will be set to U_BUFFER_OVERFLOW_ERROR if
     * n+2*m+(m!=0?2:1) > destCapacity.
     * @return the total length of the serialized format, including
     * the header, that is, n+2*m+(m!=0?2:1), or 0 on error other
     * than U_BUFFER_OVERFLOW_ERROR.
     * @stable ICU 2.4
     */
    int32_t serialize(uint16_t *dest, int32_t destCapacity, UErrorCode& ec) const;

    /**
     * Reallocate this objects internal structures to take up the least
     * possible space, without changing this object's value.
     * A frozen set will not be modified.
     * @stable ICU 2.4
     */
    virtual UnicodeSet& compact();

    /**
     * Return the class ID for this class.  This is useful only for
     * comparing to a return value from getDynamicClassID().  For example:
     * <pre>
     * .      Base* polymorphic_pointer = createPolymorphicObject();
     * .      if (polymorphic_pointer->getDynamicClassID() ==
     * .          Derived::getStaticClassID()) ...
     * </pre>
     * @return          The class ID for all objects of this class.
     * @stable ICU 2.0
     */
    static UClassID U_EXPORT2 getStaticClassID(void);

    /**
     * Implement UnicodeFunctor API.
     *
     * @return The class ID for this object. All objects of a given
     * class have the same class ID.  Objects of other classes have
     * different class IDs.
     * @stable ICU 2.4
     */
    virtual UClassID getDynamicClassID(void) const override;

private:

    // Private API for the USet API

    friend class USetAccess;

    const UnicodeString* getString(int32_t index) const;

    //----------------------------------------------------------------
    // RuleBasedTransliterator support
    //----------------------------------------------------------------

private:

    /**
     * Returns <tt>true</tt> if this set contains any character whose low byte
     * is the given value.  This is used by <tt>RuleBasedTransliterator</tt> for
     * indexing.
     */
    virtual UBool matchesIndexValue(uint8_t v) const override;

private:
    friend class RBBIRuleScanner;

    //----------------------------------------------------------------
    // Implementation: Clone as thawed (see ICU4J Freezable)
    //----------------------------------------------------------------

    UnicodeSet(const UnicodeSet& o, UBool /* asThawed */);
    UnicodeSet& copyFrom(const UnicodeSet& o, UBool asThawed);

    //----------------------------------------------------------------
    // Implementation: Pattern parsing
    //----------------------------------------------------------------

    void applyPatternIgnoreSpace(const UnicodeString& pattern,
                                 ParsePosition& pos,
                                 const SymbolTable* symbols,
                                 UErrorCode& status);

    void applyPattern(RuleCharacterIterator& chars,
                      const SymbolTable* symbols,
                      UnicodeString& rebuiltPat,
                      uint32_t options,
                      UnicodeSet& (UnicodeSet::*caseClosure)(int32_t attribute),
                      int32_t depth,
                      UErrorCode& ec);

    void closeOverCaseInsensitive(bool simple);
    void closeOverAddCaseMappings();

    //----------------------------------------------------------------
    // Implementation: Utility methods
    //----------------------------------------------------------------

    static int32_t nextCapacity(int32_t minCapacity);

    bool ensureCapacity(int32_t newLen);

    bool ensureBufferCapacity(int32_t newLen);

    void swapBuffers(void);

    UBool allocateStrings(UErrorCode &status);
    int32_t stringsSize() const;
    UBool stringsContains(const UnicodeString &s) const;

    UnicodeString& _toPattern(UnicodeString& result,
                              UBool escapeUnprintable) const;

    UnicodeString& _generatePattern(UnicodeString& result,
                                    UBool escapeUnprintable) const;

    static void _appendToPat(UnicodeString& buf, const UnicodeString& s, UBool escapeUnprintable);

    static void _appendToPat(UnicodeString& buf, UChar32 c, UBool escapeUnprintable);

    static void _appendToPat(UnicodeString &result, UChar32 start, UChar32 end,
                             UBool escapeUnprintable);

    //----------------------------------------------------------------
    // Implementation: Fundamental operators
    //----------------------------------------------------------------

    void exclusiveOr(const UChar32* other, int32_t otherLen, int8_t polarity);

    void add(const UChar32* other, int32_t otherLen, int8_t polarity);

    void retain(const UChar32* other, int32_t otherLen, int8_t polarity);

    /**
     * Return true if the given position, in the given pattern, appears
     * to be the start of a property set pattern [:foo:], \\p{foo}, or
     * \\P{foo}, or \\N{name}.
     */
    static UBool resemblesPropertyPattern(const UnicodeString& pattern,
                                          int32_t pos);

    static UBool resemblesPropertyPattern(RuleCharacterIterator& chars,
                                          int32_t iterOpts);

    /**
     * Parse the given property pattern at the given parse position
     * and set this UnicodeSet to the result.
     *
     * The original design document is out of date, but still useful.
     * Ignore the property and value names:
     * https://htmlpreview.github.io/?https://github.com/unicode-org/icu-docs/blob/main/design/unicodeset_properties.html
     *
     * Recognized syntax:
     *
     * [:foo:] [:^foo:] - white space not allowed within "[:" or ":]"
     * \\p{foo} \\P{foo}  - white space not allowed within "\\p" or "\\P"
     * \\N{name}         - white space not allowed within "\\N"
     *
     * Other than the above restrictions, Unicode Pattern_White_Space characters are ignored.
     * Case is ignored except in "\\p" and "\\P" and "\\N".  In 'name' leading
     * and trailing space is deleted, and internal runs of whitespace
     * are collapsed to a single space.
     *
     * We support binary properties, enumerated properties, and the
     * following non-enumerated properties:
     *
     *  Numeric_Value
     *  Name
     *  Unicode_1_Name
     *
     * @param pattern the pattern string
     * @param ppos on entry, the position at which to begin parsing.
     * This should be one of the locations marked '^':
     *
     *   [:blah:]     \\p{blah}     \\P{blah}     \\N{name}
     *   ^       %    ^       %    ^       %    ^       %
     *
     * On return, the position after the last character parsed, that is,
     * the locations marked '%'.  If the parse fails, ppos is returned
     * unchanged.
     * @param ec status
     * @return a reference to this.
     */
    UnicodeSet& applyPropertyPattern(const UnicodeString& pattern,
                                     ParsePosition& ppos,
                                     UErrorCode &ec);

    void applyPropertyPattern(RuleCharacterIterator& chars,
                              UnicodeString& rebuiltPat,
                              UErrorCode& ec);

    /**
     * A filter that returns true if the given code point should be
     * included in the UnicodeSet being constructed.
     */
    typedef UBool (*Filter)(UChar32 codePoint, void* context);

    /**
     * Given a filter, set this UnicodeSet to the code points
     * contained by that filter.  The filter MUST be
     * property-conformant.  That is, if it returns value v for one
     * code point, then it must return v for all affiliated code
     * points, as defined by the inclusions list.  See
     * getInclusions().
     * src is a UPropertySource value.
     */
    void applyFilter(Filter filter,
                     void* context,
                     const UnicodeSet* inclusions,
                     UErrorCode &status);

    /**
     * Set the new pattern to cache.
     */
    void setPattern(const UnicodeString& newPat) {
        setPattern(newPat.getBuffer(), newPat.length());
    }
    void setPattern(const char16_t *newPat, int32_t newPatLen);
    /**
     * Release existing cached pattern.
     */
    void releasePattern();

    friend class UnicodeSetIterator;
};



inline bool UnicodeSet::operator!=(const UnicodeSet& o) const {
    return !operator==(o);
}

inline UBool UnicodeSet::isFrozen() const {
    return (UBool)(bmpSet!=nullptr || stringSpan!=nullptr);
}

inline UBool UnicodeSet::containsSome(UChar32 start, UChar32 end) const {
    return !containsNone(start, end);
}

inline UBool UnicodeSet::containsSome(const UnicodeSet& s) const {
    return !containsNone(s);
}

inline UBool UnicodeSet::containsSome(const UnicodeString& s) const {
    return !containsNone(s);
}

inline UBool UnicodeSet::isBogus() const {
    return (UBool)(fFlags & kIsBogus);
}

inline UnicodeSet *UnicodeSet::fromUSet(USet *uset) {
    return reinterpret_cast<UnicodeSet *>(uset);
}

inline const UnicodeSet *UnicodeSet::fromUSet(const USet *uset) {
    return reinterpret_cast<const UnicodeSet *>(uset);
}

inline USet *UnicodeSet::toUSet() {
    return reinterpret_cast<USet *>(this);
}

inline const USet *UnicodeSet::toUSet() const {
    return reinterpret_cast<const USet *>(this);
}

inline int32_t UnicodeSet::span(const UnicodeString &s, int32_t start, USetSpanCondition spanCondition) const {
    int32_t sLength=s.length();
    if(start<0) {
        start=0;
    } else if(start>sLength) {
        start=sLength;
    }
    return start+span(s.getBuffer()+start, sLength-start, spanCondition);
}

inline int32_t UnicodeSet::spanBack(const UnicodeString &s, int32_t limit, USetSpanCondition spanCondition) const {
    int32_t sLength=s.length();
    if(limit<0) {
        limit=0;
    } else if(limit>sLength) {
        limit=sLength;
    }
    return spanBack(s.getBuffer(), limit, spanCondition);
}

U_NAMESPACE_END

#endif /* U_SHOW_CPLUSPLUS_API */

#endif