summaryrefslogtreecommitdiffstats
path: root/intl/icu/source/common/ustring.cpp
blob: 4f7443cc886b7e3fc5256fa6e70d9dcd39ac522e (plain)
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
// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
******************************************************************************
*
*   Copyright (C) 1998-2016, International Business Machines
*   Corporation and others.  All Rights Reserved.
*
******************************************************************************
*
* File ustring.cpp
*
* Modification History:
*
*   Date        Name        Description
*   12/07/98    bertrand    Creation.
******************************************************************************
*/

#include "unicode/utypes.h"
#include "unicode/putil.h"
#include "unicode/uchar.h"
#include "unicode/ustring.h"
#include "unicode/utf16.h"
#include "cstring.h"
#include "cwchar.h"
#include "cmemory.h"
#include "ustr_imp.h"

/* ANSI string.h - style functions ------------------------------------------ */

/* U+ffff is the highest BMP code point, the highest one that fits into a 16-bit char16_t */
#define U_BMP_MAX 0xffff

/* Forward binary string search functions ----------------------------------- */

/*
 * Test if a substring match inside a string is at code point boundaries.
 * All pointers refer to the same buffer.
 * The limit pointer may be nullptr, all others must be real pointers.
 */
static inline UBool
isMatchAtCPBoundary(const char16_t *start, const char16_t *match, const char16_t *matchLimit, const char16_t *limit) {
    if(U16_IS_TRAIL(*match) && start!=match && U16_IS_LEAD(*(match-1))) {
        /* the leading edge of the match is in the middle of a surrogate pair */
        return false;
    }
    if(U16_IS_LEAD(*(matchLimit-1)) && matchLimit!=limit && U16_IS_TRAIL(*matchLimit)) {
        /* the trailing edge of the match is in the middle of a surrogate pair */
        return false;
    }
    return true;
}

U_CAPI char16_t * U_EXPORT2
u_strFindFirst(const char16_t *s, int32_t length,
               const char16_t *sub, int32_t subLength) {
    const char16_t *start, *p, *q, *subLimit;
    char16_t c, cs, cq;

    if(sub==nullptr || subLength<-1) {
        return (char16_t *)s;
    }
    if(s==nullptr || length<-1) {
        return nullptr;
    }

    start=s;

    if(length<0 && subLength<0) {
        /* both strings are NUL-terminated */
        if((cs=*sub++)==0) {
            return (char16_t *)s;
        }
        if(*sub==0 && !U16_IS_SURROGATE(cs)) {
            /* the substring consists of a single, non-surrogate BMP code point */
            return u_strchr(s, cs);
        }

        while((c=*s++)!=0) {
            if(c==cs) {
                /* found first substring char16_t, compare rest */
                p=s;
                q=sub;
                for(;;) {
                    if((cq=*q)==0) {
                        if(isMatchAtCPBoundary(start, s-1, p, nullptr)) {
                            return (char16_t *)(s-1); /* well-formed match */
                        } else {
                            break; /* no match because surrogate pair is split */
                        }
                    }
                    if((c=*p)==0) {
                        return nullptr; /* no match, and none possible after s */
                    }
                    if(c!=cq) {
                        break; /* no match */
                    }
                    ++p;
                    ++q;
                }
            }
        }

        /* not found */
        return nullptr;
    }

    if(subLength<0) {
        subLength=u_strlen(sub);
    }
    if(subLength==0) {
        return (char16_t *)s;
    }

    /* get sub[0] to search for it fast */
    cs=*sub++;
    --subLength;
    subLimit=sub+subLength;

    if(subLength==0 && !U16_IS_SURROGATE(cs)) {
        /* the substring consists of a single, non-surrogate BMP code point */
        return length<0 ? u_strchr(s, cs) : u_memchr(s, cs, length);
    }

    if(length<0) {
        /* s is NUL-terminated */
        while((c=*s++)!=0) {
            if(c==cs) {
                /* found first substring char16_t, compare rest */
                p=s;
                q=sub;
                for(;;) {
                    if(q==subLimit) {
                        if(isMatchAtCPBoundary(start, s-1, p, nullptr)) {
                            return (char16_t *)(s-1); /* well-formed match */
                        } else {
                            break; /* no match because surrogate pair is split */
                        }
                    }
                    if((c=*p)==0) {
                        return nullptr; /* no match, and none possible after s */
                    }
                    if(c!=*q) {
                        break; /* no match */
                    }
                    ++p;
                    ++q;
                }
            }
        }
    } else {
        const char16_t *limit, *preLimit;

        /* subLength was decremented above */
        if(length<=subLength) {
            return nullptr; /* s is shorter than sub */
        }

        limit=s+length;

        /* the substring must start before preLimit */
        preLimit=limit-subLength;

        while(s!=preLimit) {
            c=*s++;
            if(c==cs) {
                /* found first substring char16_t, compare rest */
                p=s;
                q=sub;
                for(;;) {
                    if(q==subLimit) {
                        if(isMatchAtCPBoundary(start, s-1, p, limit)) {
                            return (char16_t *)(s-1); /* well-formed match */
                        } else {
                            break; /* no match because surrogate pair is split */
                        }
                    }
                    if(*p!=*q) {
                        break; /* no match */
                    }
                    ++p;
                    ++q;
                }
            }
        }
    }

    /* not found */
    return nullptr;
}

U_CAPI char16_t * U_EXPORT2
u_strstr(const char16_t *s, const char16_t *substring) {
    return u_strFindFirst(s, -1, substring, -1);
}

U_CAPI char16_t * U_EXPORT2
u_strchr(const char16_t *s, char16_t c) {
    if(U16_IS_SURROGATE(c)) {
        /* make sure to not find half of a surrogate pair */
        return u_strFindFirst(s, -1, &c, 1);
    } else {
        char16_t cs;

        /* trivial search for a BMP code point */
        for(;;) {
            if((cs=*s)==c) {
                return (char16_t *)s;
            }
            if(cs==0) {
                return nullptr;
            }
            ++s;
        }
    }
}

U_CAPI char16_t * U_EXPORT2
u_strchr32(const char16_t *s, UChar32 c) {
    if((uint32_t)c<=U_BMP_MAX) {
        /* find BMP code point */
        return u_strchr(s, (char16_t)c);
    } else if((uint32_t)c<=UCHAR_MAX_VALUE) {
        /* find supplementary code point as surrogate pair */
        char16_t cs, lead=U16_LEAD(c), trail=U16_TRAIL(c);

        while((cs=*s++)!=0) {
            if(cs==lead && *s==trail) {
                return (char16_t *)(s-1);
            }
        }
        return nullptr;
    } else {
        /* not a Unicode code point, not findable */
        return nullptr;
    }
}

U_CAPI char16_t * U_EXPORT2
u_memchr(const char16_t *s, char16_t c, int32_t count) {
    if(count<=0) {
        return nullptr; /* no string */
    } else if(U16_IS_SURROGATE(c)) {
        /* make sure to not find half of a surrogate pair */
        return u_strFindFirst(s, count, &c, 1);
    } else {
        /* trivial search for a BMP code point */
        const char16_t *limit=s+count;
        do {
            if(*s==c) {
                return (char16_t *)s;
            }
        } while(++s!=limit);
        return nullptr;
    }
}

U_CAPI char16_t * U_EXPORT2
u_memchr32(const char16_t *s, UChar32 c, int32_t count) {
    if((uint32_t)c<=U_BMP_MAX) {
        /* find BMP code point */
        return u_memchr(s, (char16_t)c, count);
    } else if(count<2) {
        /* too short for a surrogate pair */
        return nullptr;
    } else if((uint32_t)c<=UCHAR_MAX_VALUE) {
        /* find supplementary code point as surrogate pair */
        const char16_t *limit=s+count-1; /* -1 so that we do not need a separate check for the trail unit */
        char16_t lead=U16_LEAD(c), trail=U16_TRAIL(c);

        do {
            if(*s==lead && *(s+1)==trail) {
                return (char16_t *)s;
            }
        } while(++s!=limit);
        return nullptr;
    } else {
        /* not a Unicode code point, not findable */
        return nullptr;
    }
}

/* Backward binary string search functions ---------------------------------- */

U_CAPI char16_t * U_EXPORT2
u_strFindLast(const char16_t *s, int32_t length,
              const char16_t *sub, int32_t subLength) {
    const char16_t *start, *limit, *p, *q, *subLimit;
    char16_t c, cs;

    if(sub==nullptr || subLength<-1) {
        return (char16_t *)s;
    }
    if(s==nullptr || length<-1) {
        return nullptr;
    }

    /*
     * This implementation is more lazy than the one for u_strFindFirst():
     * There is no special search code for NUL-terminated strings.
     * It does not seem to be worth it for searching substrings to
     * search forward and find all matches like in u_strrchr() and similar.
     * Therefore, we simply get both string lengths and search backward.
     *
     * markus 2002oct23
     */

    if(subLength<0) {
        subLength=u_strlen(sub);
    }
    if(subLength==0) {
        return (char16_t *)s;
    }

    /* get sub[subLength-1] to search for it fast */
    subLimit=sub+subLength;
    cs=*(--subLimit);
    --subLength;

    if(subLength==0 && !U16_IS_SURROGATE(cs)) {
        /* the substring consists of a single, non-surrogate BMP code point */
        return length<0 ? u_strrchr(s, cs) : u_memrchr(s, cs, length);
    }

    if(length<0) {
        length=u_strlen(s);
    }

    /* subLength was decremented above */
    if(length<=subLength) {
        return nullptr; /* s is shorter than sub */
    }

    start=s;
    limit=s+length;

    /* the substring must start no later than s+subLength */
    s+=subLength;

    while(s!=limit) {
        c=*(--limit);
        if(c==cs) {
            /* found last substring char16_t, compare rest */
            p=limit;
            q=subLimit;
            for(;;) {
                if(q==sub) {
                    if(isMatchAtCPBoundary(start, p, limit+1, start+length)) {
                        return (char16_t *)p; /* well-formed match */
                    } else {
                        break; /* no match because surrogate pair is split */
                    }
                }
                if(*(--p)!=*(--q)) {
                    break; /* no match */
                }
            }
        }
    }

    /* not found */
    return nullptr;
}

U_CAPI char16_t * U_EXPORT2
u_strrstr(const char16_t *s, const char16_t *substring) {
    return u_strFindLast(s, -1, substring, -1);
}

U_CAPI char16_t * U_EXPORT2
u_strrchr(const char16_t *s, char16_t c) {
    if(U16_IS_SURROGATE(c)) {
        /* make sure to not find half of a surrogate pair */
        return u_strFindLast(s, -1, &c, 1);
    } else {
        const char16_t *result=nullptr;
        char16_t cs;

        /* trivial search for a BMP code point */
        for(;;) {
            if((cs=*s)==c) {
                result=s;
            }
            if(cs==0) {
                return (char16_t *)result;
            }
            ++s;
        }
    }
}

U_CAPI char16_t * U_EXPORT2
u_strrchr32(const char16_t *s, UChar32 c) {
    if((uint32_t)c<=U_BMP_MAX) {
        /* find BMP code point */
        return u_strrchr(s, (char16_t)c);
    } else if((uint32_t)c<=UCHAR_MAX_VALUE) {
        /* find supplementary code point as surrogate pair */
        const char16_t *result=nullptr;
        char16_t cs, lead=U16_LEAD(c), trail=U16_TRAIL(c);

        while((cs=*s++)!=0) {
            if(cs==lead && *s==trail) {
                result=s-1;
            }
        }
        return (char16_t *)result;
    } else {
        /* not a Unicode code point, not findable */
        return nullptr;
    }
}

U_CAPI char16_t * U_EXPORT2
u_memrchr(const char16_t *s, char16_t c, int32_t count) {
    if(count<=0) {
        return nullptr; /* no string */
    } else if(U16_IS_SURROGATE(c)) {
        /* make sure to not find half of a surrogate pair */
        return u_strFindLast(s, count, &c, 1);
    } else {
        /* trivial search for a BMP code point */
        const char16_t *limit=s+count;
        do {
            if(*(--limit)==c) {
                return (char16_t *)limit;
            }
        } while(s!=limit);
        return nullptr;
    }
}

U_CAPI char16_t * U_EXPORT2
u_memrchr32(const char16_t *s, UChar32 c, int32_t count) {
    if((uint32_t)c<=U_BMP_MAX) {
        /* find BMP code point */
        return u_memrchr(s, (char16_t)c, count);
    } else if(count<2) {
        /* too short for a surrogate pair */
        return nullptr;
    } else if((uint32_t)c<=UCHAR_MAX_VALUE) {
        /* find supplementary code point as surrogate pair */
        const char16_t *limit=s+count-1;
        char16_t lead=U16_LEAD(c), trail=U16_TRAIL(c);

        do {
            if(*limit==trail && *(limit-1)==lead) {
                return (char16_t *)(limit-1);
            }
        } while(s!=--limit);
        return nullptr;
    } else {
        /* not a Unicode code point, not findable */
        return nullptr;
    }
}

/* Tokenization functions --------------------------------------------------- */

/*
 * Match each code point in a string against each code point in the matchSet.
 * Return the index of the first string code point that
 * is (polarity==true) or is not (false) contained in the matchSet.
 * Return -(string length)-1 if there is no such code point.
 */
static int32_t
_matchFromSet(const char16_t *string, const char16_t *matchSet, UBool polarity) {
    int32_t matchLen, matchBMPLen, strItr, matchItr;
    UChar32 stringCh, matchCh;
    char16_t c, c2;

    /* first part of matchSet contains only BMP code points */
    matchBMPLen = 0;
    while((c = matchSet[matchBMPLen]) != 0 && U16_IS_SINGLE(c)) {
        ++matchBMPLen;
    }

    /* second part of matchSet contains BMP and supplementary code points */
    matchLen = matchBMPLen;
    while(matchSet[matchLen] != 0) {
        ++matchLen;
    }

    for(strItr = 0; (c = string[strItr]) != 0;) {
        ++strItr;
        if(U16_IS_SINGLE(c)) {
            if(polarity) {
                for(matchItr = 0; matchItr < matchLen; ++matchItr) {
                    if(c == matchSet[matchItr]) {
                        return strItr - 1; /* one matches */
                    }
                }
            } else {
                for(matchItr = 0; matchItr < matchLen; ++matchItr) {
                    if(c == matchSet[matchItr]) {
                        goto endloop;
                    }
                }
                return strItr - 1; /* none matches */
            }
        } else {
            /*
             * No need to check for string length before U16_IS_TRAIL
             * because c2 could at worst be the terminating NUL.
             */
            if(U16_IS_SURROGATE_LEAD(c) && U16_IS_TRAIL(c2 = string[strItr])) {
                ++strItr;
                stringCh = U16_GET_SUPPLEMENTARY(c, c2);
            } else {
                stringCh = c; /* unpaired trail surrogate */
            }

            if(polarity) {
                for(matchItr = matchBMPLen; matchItr < matchLen;) {
                    U16_NEXT(matchSet, matchItr, matchLen, matchCh);
                    if(stringCh == matchCh) {
                        return strItr - U16_LENGTH(stringCh); /* one matches */
                    }
                }
            } else {
                for(matchItr = matchBMPLen; matchItr < matchLen;) {
                    U16_NEXT(matchSet, matchItr, matchLen, matchCh);
                    if(stringCh == matchCh) {
                        goto endloop;
                    }
                }
                return strItr - U16_LENGTH(stringCh); /* none matches */
            }
        }
endloop:
        /* wish C had continue with labels like Java... */;
    }

    /* Didn't find it. */
    return -strItr-1;
}

/* Search for a codepoint in a string that matches one of the matchSet codepoints. */
U_CAPI char16_t * U_EXPORT2
u_strpbrk(const char16_t *string, const char16_t *matchSet)
{
    int32_t idx = _matchFromSet(string, matchSet, true);
    if(idx >= 0) {
        return (char16_t *)string + idx;
    } else {
        return nullptr;
    }
}

/* Search for a codepoint in a string that matches one of the matchSet codepoints. */
U_CAPI int32_t U_EXPORT2
u_strcspn(const char16_t *string, const char16_t *matchSet)
{
    int32_t idx = _matchFromSet(string, matchSet, true);
    if(idx >= 0) {
        return idx;
    } else {
        return -idx - 1; /* == u_strlen(string) */
    }
}

/* Search for a codepoint in a string that does not match one of the matchSet codepoints. */
U_CAPI int32_t U_EXPORT2
u_strspn(const char16_t *string, const char16_t *matchSet)
{
    int32_t idx = _matchFromSet(string, matchSet, false);
    if(idx >= 0) {
        return idx;
    } else {
        return -idx - 1; /* == u_strlen(string) */
    }
}

/* ----- Text manipulation functions --- */

U_CAPI char16_t* U_EXPORT2
u_strtok_r(char16_t *src,
     const char16_t *delim,
           char16_t   **saveState)
{
    char16_t *tokSource;
    char16_t *nextToken;
    uint32_t nonDelimIdx;

    /* If saveState is nullptr, the user messed up. */
    if (src != nullptr) {
        tokSource = src;
        *saveState = src; /* Set to "src" in case there are no delimiters */
    }
    else if (*saveState) {
        tokSource = *saveState;
    }
    else {
        /* src == nullptr && *saveState == nullptr */
        /* This shouldn't happen. We already finished tokenizing. */
        return nullptr;
    }

    /* Skip initial delimiters */
    nonDelimIdx = u_strspn(tokSource, delim);
    tokSource = &tokSource[nonDelimIdx];

    if (*tokSource) {
        nextToken = u_strpbrk(tokSource, delim);
        if (nextToken != nullptr) {
            /* Create a token */
            *(nextToken++) = 0;
            *saveState = nextToken;
            return tokSource;
        }
        else if (*saveState) {
            /* Return the last token */
            *saveState = nullptr;
            return tokSource;
        }
    }
    else {
        /* No tokens were found. Only delimiters were left. */
        *saveState = nullptr;
    }
    return nullptr;
}

/* Miscellaneous functions -------------------------------------------------- */

U_CAPI char16_t* U_EXPORT2
u_strcat(char16_t  *dst,
    const char16_t  *src)
{
    char16_t *anchor = dst;            /* save a pointer to start of dst */

    while(*dst != 0) {              /* To end of first string          */
        ++dst;
    }
    while((*(dst++) = *(src++)) != 0) {     /* copy string 2 over              */
    }

    return anchor;
}

U_CAPI char16_t*  U_EXPORT2
u_strncat(char16_t  *dst,
     const char16_t  *src,
     int32_t     n ) 
{
    if(n > 0) {
        char16_t *anchor = dst;            /* save a pointer to start of dst */

        while(*dst != 0) {              /* To end of first string          */
            ++dst;
        }
        while((*dst = *src) != 0) {     /* copy string 2 over              */
            ++dst;
            if(--n == 0) {
                *dst = 0;
                break;
            }
            ++src;
        }

        return anchor;
    } else {
        return dst;
    }
}

/* ----- Text property functions --- */

U_CAPI int32_t   U_EXPORT2
u_strcmp(const char16_t *s1,
    const char16_t *s2)
{
    char16_t  c1, c2;

    for(;;) {
        c1=*s1++;
        c2=*s2++;
        if (c1 != c2 || c1 == 0) {
            break;
        }
    }
    return (int32_t)c1 - (int32_t)c2;
}

U_CFUNC int32_t U_EXPORT2
uprv_strCompare(const char16_t *s1, int32_t length1,
                const char16_t *s2, int32_t length2,
                UBool strncmpStyle, UBool codePointOrder) {
    const char16_t *start1, *start2, *limit1, *limit2;
    char16_t c1, c2;

    /* setup for fix-up */
    start1=s1;
    start2=s2;

    /* compare identical prefixes - they do not need to be fixed up */
    if(length1<0 && length2<0) {
        /* strcmp style, both NUL-terminated */
        if(s1==s2) {
            return 0;
        }

        for(;;) {
            c1=*s1;
            c2=*s2;
            if(c1!=c2) {
                break;
            }
            if(c1==0) {
                return 0;
            }
            ++s1;
            ++s2;
        }

        /* setup for fix-up */
        limit1=limit2=nullptr;
    } else if(strncmpStyle) {
        /* special handling for strncmp, assume length1==length2>=0 but also check for NUL */
        if(s1==s2) {
            return 0;
        }

        limit1=start1+length1;

        for(;;) {
            /* both lengths are same, check only one limit */
            if(s1==limit1) {
                return 0;
            }

            c1=*s1;
            c2=*s2;
            if(c1!=c2) {
                break;
            }
            if(c1==0) {
                return 0;
            }
            ++s1;
            ++s2;
        }

        /* setup for fix-up */
        limit2=start2+length1; /* use length1 here, too, to enforce assumption */
    } else {
        /* memcmp/UnicodeString style, both length-specified */
        int32_t lengthResult;

        if(length1<0) {
            length1=u_strlen(s1);
        }
        if(length2<0) {
            length2=u_strlen(s2);
        }

        /* limit1=start1+min(length1, length2) */
        if(length1<length2) {
            lengthResult=-1;
            limit1=start1+length1;
        } else if(length1==length2) {
            lengthResult=0;
            limit1=start1+length1;
        } else /* length1>length2 */ {
            lengthResult=1;
            limit1=start1+length2;
        }

        if(s1==s2) {
            return lengthResult;
        }

        for(;;) {
            /* check pseudo-limit */
            if(s1==limit1) {
                return lengthResult;
            }

            c1=*s1;
            c2=*s2;
            if(c1!=c2) {
                break;
            }
            ++s1;
            ++s2;
        }

        /* setup for fix-up */
        limit1=start1+length1;
        limit2=start2+length2;
    }

    /* if both values are in or above the surrogate range, fix them up */
    if(c1>=0xd800 && c2>=0xd800 && codePointOrder) {
        /* subtract 0x2800 from BMP code points to make them smaller than supplementary ones */
        if(
            (c1<=0xdbff && (s1+1)!=limit1 && U16_IS_TRAIL(*(s1+1))) ||
            (U16_IS_TRAIL(c1) && start1!=s1 && U16_IS_LEAD(*(s1-1)))
        ) {
            /* part of a surrogate pair, leave >=d800 */
        } else {
            /* BMP code point - may be surrogate code point - make <d800 */
            c1-=0x2800;
        }

        if(
            (c2<=0xdbff && (s2+1)!=limit2 && U16_IS_TRAIL(*(s2+1))) ||
            (U16_IS_TRAIL(c2) && start2!=s2 && U16_IS_LEAD(*(s2-1)))
        ) {
            /* part of a surrogate pair, leave >=d800 */
        } else {
            /* BMP code point - may be surrogate code point - make <d800 */
            c2-=0x2800;
        }
    }

    /* now c1 and c2 are in the requested (code unit or code point) order */
    return (int32_t)c1-(int32_t)c2;
}

/*
 * Compare two strings as presented by UCharIterators.
 * Use code unit or code point order.
 * When the function returns, it is undefined where the iterators
 * have stopped.
 */
U_CAPI int32_t U_EXPORT2
u_strCompareIter(UCharIterator *iter1, UCharIterator *iter2, UBool codePointOrder) {
    UChar32 c1, c2;

    /* argument checking */
    if(iter1==nullptr || iter2==nullptr) {
        return 0; /* bad arguments */
    }
    if(iter1==iter2) {
        return 0; /* identical iterators */
    }

    /* reset iterators to start? */
    iter1->move(iter1, 0, UITER_START);
    iter2->move(iter2, 0, UITER_START);

    /* compare identical prefixes - they do not need to be fixed up */
    for(;;) {
        c1=iter1->next(iter1);
        c2=iter2->next(iter2);
        if(c1!=c2) {
            break;
        }
        if(c1==-1) {
            return 0;
        }
    }

    /* if both values are in or above the surrogate range, fix them up */
    if(c1>=0xd800 && c2>=0xd800 && codePointOrder) {
        /* subtract 0x2800 from BMP code points to make them smaller than supplementary ones */
        if(
            (c1<=0xdbff && U16_IS_TRAIL(iter1->current(iter1))) ||
            (U16_IS_TRAIL(c1) && (iter1->previous(iter1), U16_IS_LEAD(iter1->previous(iter1))))
        ) {
            /* part of a surrogate pair, leave >=d800 */
        } else {
            /* BMP code point - may be surrogate code point - make <d800 */
            c1-=0x2800;
        }

        if(
            (c2<=0xdbff && U16_IS_TRAIL(iter2->current(iter2))) ||
            (U16_IS_TRAIL(c2) && (iter2->previous(iter2), U16_IS_LEAD(iter2->previous(iter2))))
        ) {
            /* part of a surrogate pair, leave >=d800 */
        } else {
            /* BMP code point - may be surrogate code point - make <d800 */
            c2-=0x2800;
        }
    }

    /* now c1 and c2 are in the requested (code unit or code point) order */
    return (int32_t)c1-(int32_t)c2;
}

#if 0
/*
 * u_strCompareIter() does not leave the iterators _on_ the different units.
 * This is possible but would cost a few extra indirect function calls to back
 * up if the last unit (c1 or c2 respectively) was >=0.
 *
 * Consistently leaving them _behind_ the different units is not an option
 * because the current "unit" is the end of the string if that is reached,
 * and in such a case the iterator does not move.
 * For example, when comparing "ab" with "abc", both iterators rest _on_ the end
 * of their strings. Calling previous() on each does not move them to where
 * the comparison fails.
 *
 * So the simplest semantics is to not define where the iterators end up.
 *
 * The following fragment is part of what would need to be done for backing up.
 */
void fragment {
        /* iff a surrogate is part of a surrogate pair, leave >=d800 */
        if(c1<=0xdbff) {
            if(!U16_IS_TRAIL(iter1->current(iter1))) {
                /* lead surrogate code point - make <d800 */
                c1-=0x2800;
            }
        } else if(c1<=0xdfff) {
            int32_t idx=iter1->getIndex(iter1, UITER_CURRENT);
            iter1->previous(iter1); /* ==c1 */
            if(!U16_IS_LEAD(iter1->previous(iter1))) {
                /* trail surrogate code point - make <d800 */
                c1-=0x2800;
            }
            /* go back to behind where the difference is */
            iter1->move(iter1, idx, UITER_ZERO);
        } else /* 0xe000<=c1<=0xffff */ {
            /* BMP code point - make <d800 */
            c1-=0x2800;
        }
}
#endif

U_CAPI int32_t U_EXPORT2
u_strCompare(const char16_t *s1, int32_t length1,
             const char16_t *s2, int32_t length2,
             UBool codePointOrder) {
    /* argument checking */
    if(s1==nullptr || length1<-1 || s2==nullptr || length2<-1) {
        return 0;
    }
    return uprv_strCompare(s1, length1, s2, length2, false, codePointOrder);
}

/* String compare in code point order - u_strcmp() compares in code unit order. */
U_CAPI int32_t U_EXPORT2
u_strcmpCodePointOrder(const char16_t *s1, const char16_t *s2) {
    return uprv_strCompare(s1, -1, s2, -1, false, true);
}

U_CAPI int32_t   U_EXPORT2
u_strncmp(const char16_t  *s1,
     const char16_t  *s2,
     int32_t     n) 
{
    if(n > 0) {
        int32_t rc;
        for(;;) {
            rc = (int32_t)*s1 - (int32_t)*s2;
            if(rc != 0 || *s1 == 0 || --n == 0) {
                return rc;
            }
            ++s1;
            ++s2;
        }
    } else {
        return 0;
    }
}

U_CAPI int32_t U_EXPORT2
u_strncmpCodePointOrder(const char16_t *s1, const char16_t *s2, int32_t n) {
    return uprv_strCompare(s1, n, s2, n, true, true);
}

U_CAPI char16_t* U_EXPORT2
u_strcpy(char16_t  *dst,
    const char16_t  *src)
{
    char16_t *anchor = dst;            /* save a pointer to start of dst */

    while((*(dst++) = *(src++)) != 0) {     /* copy string 2 over              */
    }

    return anchor;
}

U_CAPI char16_t*  U_EXPORT2
u_strncpy(char16_t  *dst,
     const char16_t  *src,
     int32_t     n) 
{
    char16_t *anchor = dst;            /* save a pointer to start of dst */

    /* copy string 2 over */
    while(n > 0 && (*(dst++) = *(src++)) != 0) {
        --n;
    }

    return anchor;
}

U_CAPI int32_t   U_EXPORT2
u_strlen(const char16_t *s)
{
#if U_SIZEOF_WCHAR_T == U_SIZEOF_UCHAR
    return (int32_t)uprv_wcslen((const wchar_t *)s);
#else
    const char16_t *t = s;
    while(*t != 0) {
      ++t;
    }
    return t - s;
#endif
}

U_CAPI int32_t U_EXPORT2
u_countChar32(const char16_t *s, int32_t length) {
    int32_t count;

    if(s==nullptr || length<-1) {
        return 0;
    }

    count=0;
    if(length>=0) {
        while(length>0) {
            ++count;
            if(U16_IS_LEAD(*s) && length>=2 && U16_IS_TRAIL(*(s+1))) {
                s+=2;
                length-=2;
            } else {
                ++s;
                --length;
            }
        }
    } else /* length==-1 */ {
        char16_t c;

        for(;;) {
            if((c=*s++)==0) {
                break;
            }
            ++count;

            /*
             * sufficient to look ahead one because of UTF-16;
             * safe to look ahead one because at worst that would be the terminating NUL
             */
            if(U16_IS_LEAD(c) && U16_IS_TRAIL(*s)) {
                ++s;
            }
        }
    }
    return count;
}

U_CAPI UBool U_EXPORT2
u_strHasMoreChar32Than(const char16_t *s, int32_t length, int32_t number) {

    if(number<0) {
        return true;
    }
    if(s==nullptr || length<-1) {
        return false;
    }

    if(length==-1) {
        /* s is NUL-terminated */
        char16_t c;

        /* count code points until they exceed */
        for(;;) {
            if((c=*s++)==0) {
                return false;
            }
            if(number==0) {
                return true;
            }
            if(U16_IS_LEAD(c) && U16_IS_TRAIL(*s)) {
                ++s;
            }
            --number;
        }
    } else {
        /* length>=0 known */
        const char16_t *limit;
        int32_t maxSupplementary;

        /* s contains at least (length+1)/2 code points: <=2 UChars per cp */
        if(((length+1)/2)>number) {
            return true;
        }

        /* check if s does not even contain enough UChars */
        maxSupplementary=length-number;
        if(maxSupplementary<=0) {
            return false;
        }
        /* there are maxSupplementary=length-number more UChars than asked-for code points */

        /*
         * count code points until they exceed and also check that there are
         * no more than maxSupplementary supplementary code points (char16_t pairs)
         */
        limit=s+length;
        for(;;) {
            if(s==limit) {
                return false;
            }
            if(number==0) {
                return true;
            }
            if(U16_IS_LEAD(*s++) && s!=limit && U16_IS_TRAIL(*s)) {
                ++s;
                if(--maxSupplementary<=0) {
                    /* too many pairs - too few code points */
                    return false;
                }
            }
            --number;
        }
    }
}

U_CAPI char16_t * U_EXPORT2
u_memcpy(char16_t *dest, const char16_t *src, int32_t count) {
    if(count > 0) {
        uprv_memcpy(dest, src, (size_t)count*U_SIZEOF_UCHAR);
    }
    return dest;
}

U_CAPI char16_t * U_EXPORT2
u_memmove(char16_t *dest, const char16_t *src, int32_t count) {
    if(count > 0) {
        uprv_memmove(dest, src, (size_t)count*U_SIZEOF_UCHAR);
    }
    return dest;
}

U_CAPI char16_t * U_EXPORT2
u_memset(char16_t *dest, char16_t c, int32_t count) {
    if(count > 0) {
        char16_t *ptr = dest;
        char16_t *limit = dest + count;

        while (ptr < limit) {
            *(ptr++) = c;
        }
    }
    return dest;
}

U_CAPI int32_t U_EXPORT2
u_memcmp(const char16_t *buf1, const char16_t *buf2, int32_t count) {
    if(count > 0) {
        const char16_t *limit = buf1 + count;
        int32_t result;

        while (buf1 < limit) {
            result = (int32_t)(uint16_t)*buf1 - (int32_t)(uint16_t)*buf2;
            if (result != 0) {
                return result;
            }
            buf1++;
            buf2++;
        }
    }
    return 0;
}

U_CAPI int32_t U_EXPORT2
u_memcmpCodePointOrder(const char16_t *s1, const char16_t *s2, int32_t count) {
    return uprv_strCompare(s1, count, s2, count, false, true);
}

/* u_unescape & support fns ------------------------------------------------- */

/* This map must be in ASCENDING ORDER OF THE ESCAPE CODE */
static const char16_t UNESCAPE_MAP[] = {
    /*"   0x22, 0x22 */
    /*'   0x27, 0x27 */
    /*?   0x3F, 0x3F */
    /*\   0x5C, 0x5C */
    /*a*/ 0x61, 0x07,
    /*b*/ 0x62, 0x08,
    /*e*/ 0x65, 0x1b,
    /*f*/ 0x66, 0x0c,
    /*n*/ 0x6E, 0x0a,
    /*r*/ 0x72, 0x0d,
    /*t*/ 0x74, 0x09,
    /*v*/ 0x76, 0x0b
};
enum { UNESCAPE_MAP_LENGTH = UPRV_LENGTHOF(UNESCAPE_MAP) };

/* Convert one octal digit to a numeric value 0..7, or -1 on failure */
static int32_t _digit8(char16_t c) {
    if (c >= u'0' && c <= u'7') {
        return c - u'0';
    }
    return -1;
}

/* Convert one hex digit to a numeric value 0..F, or -1 on failure */
static int32_t _digit16(char16_t c) {
    if (c >= u'0' && c <= u'9') {
        return c - u'0';
    }
    if (c >= u'A' && c <= u'F') {
        return c - (u'A' - 10);
    }
    if (c >= u'a' && c <= u'f') {
        return c - (u'a' - 10);
    }
    return -1;
}

/* Parse a single escape sequence.  Although this method deals in
 * UChars, it does not use C++ or UnicodeString.  This allows it to
 * be used from C contexts. */
U_CAPI UChar32 U_EXPORT2
u_unescapeAt(UNESCAPE_CHAR_AT charAt,
             int32_t *offset,
             int32_t length,
             void *context) {

    int32_t start = *offset;
    UChar32 c;
    UChar32 result = 0;
    int8_t n = 0;
    int8_t minDig = 0;
    int8_t maxDig = 0;
    int8_t bitsPerDigit = 4; 
    int32_t dig;
    UBool braces = false;

    /* Check that offset is in range */
    if (*offset < 0 || *offset >= length) {
        goto err;
    }

    /* Fetch first char16_t after '\\' */
    c = charAt((*offset)++, context);

    /* Convert hexadecimal and octal escapes */
    switch (c) {
    case u'u':
        minDig = maxDig = 4;
        break;
    case u'U':
        minDig = maxDig = 8;
        break;
    case u'x':
        minDig = 1;
        if (*offset < length && charAt(*offset, context) == u'{') {
            ++(*offset);
            braces = true;
            maxDig = 8;
        } else {
            maxDig = 2;
        }
        break;
    default:
        dig = _digit8(c);
        if (dig >= 0) {
            minDig = 1;
            maxDig = 3;
            n = 1; /* Already have first octal digit */
            bitsPerDigit = 3;
            result = dig;
        }
        break;
    }
    if (minDig != 0) {
        while (*offset < length && n < maxDig) {
            c = charAt(*offset, context);
            dig = (bitsPerDigit == 3) ? _digit8(c) : _digit16(c);
            if (dig < 0) {
                break;
            }
            result = (result << bitsPerDigit) | dig;
            ++(*offset);
            ++n;
        }
        if (n < minDig) {
            goto err;
        }
        if (braces) {
            if (c != u'}') {
                goto err;
            }
            ++(*offset);
        }
        if (result < 0 || result >= 0x110000) {
            goto err;
        }
        /* If an escape sequence specifies a lead surrogate, see if
         * there is a trail surrogate after it, either as an escape or
         * as a literal.  If so, join them up into a supplementary.
         */
        if (*offset < length && U16_IS_LEAD(result)) {
            int32_t ahead = *offset + 1;
            c = charAt(*offset, context);
            if (c == u'\\' && ahead < length) {
                // Calling ourselves recursively may cause a stack overflow if
                // we have repeated escaped lead surrogates.
                // Limit the length to 11 ("x{0000DFFF}") after ahead.
                int32_t tailLimit = ahead + 11;
                if (tailLimit > length) {
                    tailLimit = length;
                }
                c = u_unescapeAt(charAt, &ahead, tailLimit, context);
            }
            if (U16_IS_TRAIL(c)) {
                *offset = ahead;
                result = U16_GET_SUPPLEMENTARY(result, c);
            }
        }
        return result;
    }

    /* Convert C-style escapes in table */
    for (int32_t i=0; i<UNESCAPE_MAP_LENGTH; i+=2) {
        if (c == UNESCAPE_MAP[i]) {
            return UNESCAPE_MAP[i+1];
        } else if (c < UNESCAPE_MAP[i]) {
            break;
        }
    }

    /* Map \cX to control-X: X & 0x1F */
    if (c == u'c' && *offset < length) {
        c = charAt((*offset)++, context);
        if (U16_IS_LEAD(c) && *offset < length) {
            char16_t c2 = charAt(*offset, context);
            if (U16_IS_TRAIL(c2)) {
                ++(*offset);
                c = U16_GET_SUPPLEMENTARY(c, c2);
            }
        }
        return 0x1F & c;
    }

    /* If no special forms are recognized, then consider
     * the backslash to generically escape the next character.
     * Deal with surrogate pairs. */
    if (U16_IS_LEAD(c) && *offset < length) {
        char16_t c2 = charAt(*offset, context);
        if (U16_IS_TRAIL(c2)) {
            ++(*offset);
            return U16_GET_SUPPLEMENTARY(c, c2);
        }
    }
    return c;

 err:
    /* Invalid escape sequence */
    *offset = start; /* Reset to initial value */
    return (UChar32)0xFFFFFFFF;
}

/* u_unescapeAt() callback to return a char16_t from a char* */
static char16_t U_CALLCONV
_charPtr_charAt(int32_t offset, void *context) {
    char16_t c16;
    /* It would be more efficient to access the invariant tables
     * directly but there is no API for that. */
    u_charsToUChars(((char*) context) + offset, &c16, 1);
    return c16;
}

/* Append an escape-free segment of the text; used by u_unescape() */
static void _appendUChars(char16_t *dest, int32_t destCapacity,
                          const char *src, int32_t srcLen) {
    if (destCapacity < 0) {
        destCapacity = 0;
    }
    if (srcLen > destCapacity) {
        srcLen = destCapacity;
    }
    u_charsToUChars(src, dest, srcLen);
}

/* Do an invariant conversion of char* -> char16_t*, with escape parsing */
U_CAPI int32_t U_EXPORT2
u_unescape(const char *src, char16_t *dest, int32_t destCapacity) {
    const char *segment = src;
    int32_t i = 0;
    char c;

    while ((c=*src) != 0) {
        /* '\\' intentionally written as compiler-specific
         * character constant to correspond to compiler-specific
         * char* constants. */
        if (c == '\\') {
            int32_t lenParsed = 0;
            UChar32 c32;
            if (src != segment) {
                if (dest != nullptr) {
                    _appendUChars(dest + i, destCapacity - i,
                                  segment, (int32_t)(src - segment));
                }
                i += (int32_t)(src - segment);
            }
            ++src; /* advance past '\\' */
            c32 = (UChar32)u_unescapeAt(_charPtr_charAt, &lenParsed, (int32_t)uprv_strlen(src), (void*)src);
            if (lenParsed == 0) {
                goto err;
            }
            src += lenParsed; /* advance past escape seq. */
            if (dest != nullptr && U16_LENGTH(c32) <= (destCapacity - i)) {
                U16_APPEND_UNSAFE(dest, i, c32);
            } else {
                i += U16_LENGTH(c32);
            }
            segment = src;
        } else {
            ++src;
        }
    }
    if (src != segment) {
        if (dest != nullptr) {
            _appendUChars(dest + i, destCapacity - i,
                          segment, (int32_t)(src - segment));
        }
        i += (int32_t)(src - segment);
    }
    if (dest != nullptr && i < destCapacity) {
        dest[i] = 0;
    }
    return i;

 err:
    if (dest != nullptr && destCapacity > 0) {
        *dest = 0;
    }
    return 0;
}

/* NUL-termination of strings ----------------------------------------------- */

/**
 * NUL-terminate a string no matter what its type.
 * Set warning and error codes accordingly.
 */
#define __TERMINATE_STRING(dest, destCapacity, length, pErrorCode) UPRV_BLOCK_MACRO_BEGIN { \
    if(pErrorCode!=nullptr && U_SUCCESS(*pErrorCode)) {                    \
        /* not a public function, so no complete argument checking */   \
                                                                        \
        if(length<0) {                                                  \
            /* assume that the caller handles this */                   \
        } else if(length<destCapacity) {                                \
            /* NUL-terminate the string, the NUL fits */                \
            dest[length]=0;                                             \
            /* unset the not-terminated warning but leave all others */ \
            if(*pErrorCode==U_STRING_NOT_TERMINATED_WARNING) {          \
                *pErrorCode=U_ZERO_ERROR;                               \
            }                                                           \
        } else if(length==destCapacity) {                               \
            /* unable to NUL-terminate, but the string itself fit - set a warning code */ \
            *pErrorCode=U_STRING_NOT_TERMINATED_WARNING;                \
        } else /* length>destCapacity */ {                              \
            /* even the string itself did not fit - set an error code */ \
            *pErrorCode=U_BUFFER_OVERFLOW_ERROR;                        \
        }                                                               \
    } \
} UPRV_BLOCK_MACRO_END

U_CAPI char16_t U_EXPORT2
u_asciiToUpper(char16_t c) {
    if (u'a' <= c && c <= u'z') {
        c = c + u'A' - u'a';
    }
    return c;
}

U_CAPI int32_t U_EXPORT2
u_terminateUChars(char16_t *dest, int32_t destCapacity, int32_t length, UErrorCode *pErrorCode) {
    __TERMINATE_STRING(dest, destCapacity, length, pErrorCode);
    return length;
}

U_CAPI int32_t U_EXPORT2
u_terminateChars(char *dest, int32_t destCapacity, int32_t length, UErrorCode *pErrorCode) {
    __TERMINATE_STRING(dest, destCapacity, length, pErrorCode);
    return length;
}

U_CAPI int32_t U_EXPORT2
u_terminateUChar32s(UChar32 *dest, int32_t destCapacity, int32_t length, UErrorCode *pErrorCode) {
    __TERMINATE_STRING(dest, destCapacity, length, pErrorCode);
    return length;
}

U_CAPI int32_t U_EXPORT2
u_terminateWChars(wchar_t *dest, int32_t destCapacity, int32_t length, UErrorCode *pErrorCode) {
    __TERMINATE_STRING(dest, destCapacity, length, pErrorCode);
    return length;
}

// Compute the hash code for a string -------------------------------------- ***

// Moved here from uhash.c so that UnicodeString::hashCode() does not depend
// on UHashtable code.

/*
  Compute the hash by iterating sparsely over about 32 (up to 63)
  characters spaced evenly through the string.  For each character,
  multiply the previous hash value by a prime number and add the new
  character in, like a linear congruential random number generator,
  producing a pseudorandom deterministic value well distributed over
  the output range. [LIU]
*/

#define STRING_HASH(TYPE, STR, STRLEN, DEREF) UPRV_BLOCK_MACRO_BEGIN { \
    uint32_t hash = 0;                        \
    const TYPE *p = (const TYPE*) STR;        \
    if (p != nullptr) {                          \
        int32_t len = (int32_t)(STRLEN);      \
        int32_t inc = ((len - 32) / 32) + 1;  \
        const TYPE *limit = p + len;          \
        while (p<limit) {                     \
            hash = (hash * 37) + DEREF;       \
            p += inc;                         \
        }                                     \
    }                                         \
    return static_cast<int32_t>(hash);        \
} UPRV_BLOCK_MACRO_END

/* Used by UnicodeString to compute its hashcode - Not public API. */
U_CAPI int32_t U_EXPORT2
ustr_hashUCharsN(const char16_t *str, int32_t length) {
    STRING_HASH(char16_t, str, length, *p);
}

U_CAPI int32_t U_EXPORT2
ustr_hashCharsN(const char *str, int32_t length) {
    STRING_HASH(uint8_t, str, length, *p);
}

U_CAPI int32_t U_EXPORT2
ustr_hashICharsN(const char *str, int32_t length) {
    STRING_HASH(char, str, length, (uint8_t)uprv_tolower(*p));
}