summaryrefslogtreecommitdiff
path: root/Build/source/libs/icu/icu-src/source/test/intltest/ssearch.cpp
blob: bea90a1d12431dcc176d05c3d965dcf84d2d43af (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
/*
 **********************************************************************
 *   Copyright (C) 2005-2016, International Business Machines
 *   Corporation and others.  All Rights Reserved.
 **********************************************************************
 */

#include "unicode/utypes.h"

#if !UCONFIG_NO_COLLATION

#include "cmemory.h"
#include "cstring.h"
#include "usrchimp.h"

#include "unicode/coll.h"
#include "unicode/tblcoll.h"
#include "unicode/usearch.h"
#include "unicode/uset.h"
#include "unicode/ustring.h"

#include "unicode/coleitr.h"
#include "unicode/regex.h"        // TODO: make conditional on regexp being built.

#include "colldata.h"
#include "ssearch.h"
#include "xmlparser.h"

#include <stdio.h>  // for sprintf

char testId[100];

#define TEST_ASSERT(x) {if (!(x)) { \
    errln("Failure in file %s, line %d, test ID = \"%s\"", __FILE__, __LINE__, testId);}}

#define TEST_ASSERT_M(x, m) {if (!(x)) { \
    dataerrln("Failure in file %s, line %d.   \"%s\"", __FILE__, __LINE__, m);return;}}

#define TEST_ASSERT_SUCCESS(errcode) {if (U_FAILURE(errcode)) { \
    dataerrln("Failure in file %s, line %d, test ID \"%s\", status = \"%s\"", \
          __FILE__, __LINE__, testId, u_errorName(errcode));}}

#define NEW_ARRAY(type, count) (type *) uprv_malloc((count) * sizeof(type))
#define DELETE_ARRAY(array) uprv_free((void *) (array))

//---------------------------------------------------------------------------
//
//  Test class boilerplate
//
//---------------------------------------------------------------------------
SSearchTest::SSearchTest()
{
}

SSearchTest::~SSearchTest()
{
}

void SSearchTest::runIndexedTest( int32_t index, UBool exec, const char* &name, char *params )
{
    if (exec) logln("TestSuite SSearchTest: ");
    switch (index) {
#if !UCONFIG_NO_BREAK_ITERATION
       case 0: name = "searchTest";
            if (exec) searchTest();
            break;

        case 1: name = "offsetTest";
            if (exec) offsetTest();
            break;

        case 2: name = "monkeyTest";
            if (exec) monkeyTest(params);
            break;

        case 3: name = "sharpSTest";
            if (exec) sharpSTest();
            break;

        case 4: name = "goodSuffixTest";
            if (exec) goodSuffixTest();
            break;

        case 5: name = "searchTime";
            if (exec) searchTime();
            break;
#endif
        default: name = "";
            break; //needed to end loop
    }
}


#if !UCONFIG_NO_BREAK_ITERATION

#define PATH_BUFFER_SIZE 2048
const char *SSearchTest::getPath(char buffer[2048], const char *filename) {
    UErrorCode status = U_ZERO_ERROR;
    const char *testDataDirectory = IntlTest::getSourceTestData(status);

    if (U_FAILURE(status) || strlen(testDataDirectory) + strlen(filename) + 1 >= PATH_BUFFER_SIZE) {
        errln("ERROR: getPath() failed - %s", u_errorName(status));
        return NULL;
    }

    strcpy(buffer, testDataDirectory);
    strcat(buffer, filename);
    return buffer;
}


void SSearchTest::searchTest()
{
#if !UCONFIG_NO_REGULAR_EXPRESSIONS && !UCONFIG_NO_FILE_IO
    UErrorCode status = U_ZERO_ERROR;
    char path[PATH_BUFFER_SIZE];
    const char *testFilePath = getPath(path, "ssearch.xml");

    if (testFilePath == NULL) {
        return; /* Couldn't get path: error message already output. */
    }

    LocalPointer<UXMLParser> parser(UXMLParser::createParser(status));
    TEST_ASSERT_SUCCESS(status);
    LocalPointer<UXMLElement> root(parser->parseFile(testFilePath, status));
    TEST_ASSERT_SUCCESS(status);
    if (U_FAILURE(status)) {
        return;
    }

    const UnicodeString *debugTestCase = root->getAttribute("debug");
    if (debugTestCase != NULL) {
//       setenv("USEARCH_DEBUG", "1", 1);
    }


    const UXMLElement *testCase;
    int32_t tc = 0;

    while((testCase = root->nextChildElement(tc)) != NULL) {

        if (testCase->getTagName().compare("test-case") != 0) {
            errln("ssearch, unrecognized XML Element in test file");
            continue;
        }
        const UnicodeString *id       = testCase->getAttribute("id");
        *testId = 0;
        if (id != NULL) {
            id->extract(0, id->length(), testId,  sizeof(testId), US_INV);
        }

        // If debugging test case has been specified and this is not it, skip to next.
        if (id!=NULL && debugTestCase!=NULL && *id != *debugTestCase) {
            continue;
        }
        //
        //  Get the requested collation strength.
        //    Default is tertiary if the XML attribute is missing from the test case.
        //
        const UnicodeString *strength = testCase->getAttribute("strength");
        UColAttributeValue collatorStrength = UCOL_PRIMARY;
        if      (strength==NULL)          { collatorStrength = UCOL_TERTIARY;}
        else if (*strength=="PRIMARY")    { collatorStrength = UCOL_PRIMARY;}
        else if (*strength=="SECONDARY")  { collatorStrength = UCOL_SECONDARY;}
        else if (*strength=="TERTIARY")   { collatorStrength = UCOL_TERTIARY;}
        else if (*strength=="QUATERNARY") { collatorStrength = UCOL_QUATERNARY;}
        else if (*strength=="IDENTICAL")  { collatorStrength = UCOL_IDENTICAL;}
        else {
            // Bogus value supplied for strength.  Shouldn't happen, even from
            //  typos, if the  XML source has been validated.
            //  This assert is a little deceiving in that strength can be
            //   any of the allowed values, not just TERTIARY, but it will
            //   do the job of getting the error output.
            TEST_ASSERT(*strength=="TERTIARY")
        }

        //
        // Get the collator normalization flag.  Default is UCOL_OFF.
        //
        UColAttributeValue normalize = UCOL_OFF;
        const UnicodeString *norm = testCase->getAttribute("norm");
        TEST_ASSERT (norm==NULL || *norm=="ON" || *norm=="OFF");
        if (norm!=NULL && *norm=="ON") {
            normalize = UCOL_ON;
        }

        //
        // Get the alternate_handling flag. Default is UCOL_NON_IGNORABLE.
        //
        UColAttributeValue alternateHandling = UCOL_NON_IGNORABLE;
        const UnicodeString *alt = testCase->getAttribute("alternate_handling");
        TEST_ASSERT (alt == NULL || *alt == "SHIFTED" || *alt == "NON_IGNORABLE");
        if (alt != NULL && *alt == "SHIFTED") {
            alternateHandling = UCOL_SHIFTED;
        }

        const UnicodeString defLocale("en");
        char  clocale[100];
        const UnicodeString *locale   = testCase->getAttribute("locale");
        if (locale == NULL || locale->length()==0) {
            locale = &defLocale;
        };
        locale->extract(0, locale->length(), clocale, sizeof(clocale), NULL);


        UnicodeString  text;
        UnicodeString  target;
        UnicodeString  pattern;
        int32_t        expectedMatchStart = -1;
        int32_t        expectedMatchLimit = -1;
        const UXMLElement  *n;
        int32_t                nodeCount = 0;

        n = testCase->getChildElement("pattern");
        TEST_ASSERT(n != NULL);
        if (n==NULL) {
            continue;
        }
        text = n->getText(FALSE);
        text = text.unescape();
        pattern.append(text);
        nodeCount++;

        n = testCase->getChildElement("pre");
        if (n!=NULL) {
            text = n->getText(FALSE);
            text = text.unescape();
            target.append(text);
            nodeCount++;
        }

        n = testCase->getChildElement("m");
        if (n!=NULL) {
            expectedMatchStart = target.length();
            text = n->getText(FALSE);
            text = text.unescape();
            target.append(text);
            expectedMatchLimit = target.length();
            nodeCount++;
        }

        n = testCase->getChildElement("post");
        if (n!=NULL) {
            text = n->getText(FALSE);
            text = text.unescape();
            target.append(text);
            nodeCount++;
        }

        //  Check that there weren't extra things in the XML
        TEST_ASSERT(nodeCount == testCase->countChildren());

        // Open a collator and StringSearch based on the parameters
        //   obtained from the XML.
        //
        status = U_ZERO_ERROR;
        LocalUCollatorPointer collator(ucol_open(clocale, &status));
        ucol_setStrength(collator.getAlias(), collatorStrength);
        ucol_setAttribute(collator.getAlias(), UCOL_NORMALIZATION_MODE, normalize, &status);
        ucol_setAttribute(collator.getAlias(), UCOL_ALTERNATE_HANDLING, alternateHandling, &status);
        LocalUStringSearchPointer uss(usearch_openFromCollator(pattern.getBuffer(), pattern.length(),
                                                               target.getBuffer(), target.length(),
                                                               collator.getAlias(),
                                                               NULL,     // the break iterator
                                                               &status));

        TEST_ASSERT_SUCCESS(status);
        if (U_FAILURE(status)) {
            continue;
        }

        int32_t foundStart = 0;
        int32_t foundLimit = 0;
        UBool   foundMatch;

        //
        // Do the search, check the match result against the expected results.
        //
        foundMatch= usearch_search(uss.getAlias(), 0, &foundStart, &foundLimit, &status);
        TEST_ASSERT_SUCCESS(status);
        if ((foundMatch && expectedMatchStart<0) ||
            (foundStart != expectedMatchStart)   ||
            (foundLimit != expectedMatchLimit)) {
                TEST_ASSERT(FALSE);   //  ouput generic error position
                infoln("Found, expected match start = %d, %d \n"
                       "Found, expected match limit = %d, %d",
                foundStart, expectedMatchStart, foundLimit, expectedMatchLimit);
        }

        // In case there are other matches...
        // (should we only do this if the test case passed?)
        while (foundMatch) {
            expectedMatchStart = foundStart;
            expectedMatchLimit = foundLimit;

            foundMatch = usearch_search(uss.getAlias(), foundLimit, &foundStart, &foundLimit, &status);
        }

        uss.adoptInstead(usearch_openFromCollator(pattern.getBuffer(), pattern.length(),
            target.getBuffer(), target.length(),
            collator.getAlias(),
            NULL,
            &status));

        //
        // Do the backwards search, check the match result against the expected results.
        //
        foundMatch= usearch_searchBackwards(uss.getAlias(), target.length(), &foundStart, &foundLimit, &status);
        TEST_ASSERT_SUCCESS(status);
        if ((foundMatch && expectedMatchStart<0) ||
            (foundStart != expectedMatchStart)   ||
            (foundLimit != expectedMatchLimit)) {
                TEST_ASSERT(FALSE);   //  ouput generic error position
                infoln("Found, expected backwards match start = %d, %d \n"
                       "Found, expected backwards match limit = %d, %d",
                foundStart, expectedMatchStart, foundLimit, expectedMatchLimit);
        }
    }
#endif
}

struct Order
{
    int32_t order;
    int32_t lowOffset;
    int32_t highOffset;
};

class OrderList
{
public:
    OrderList();
    OrderList(UCollator *coll, const UnicodeString &string, int32_t stringOffset = 0);
    ~OrderList();

    int32_t size(void) const;
    void add(int32_t order, int32_t low, int32_t high);
    const Order *get(int32_t index) const;
    int32_t getLowOffset(int32_t index) const;
    int32_t getHighOffset(int32_t index) const;
    int32_t getOrder(int32_t index) const;
    void reverse(void);
    UBool compare(const OrderList &other) const;
    UBool matchesAt(int32_t offset, const OrderList &other) const;

private:
    Order *list;
    int32_t listMax;
    int32_t listSize;
};

OrderList::OrderList()
  : list(NULL),  listMax(16), listSize(0)
{
    list = new Order[listMax];
}

OrderList::OrderList(UCollator *coll, const UnicodeString &string, int32_t stringOffset)
    : list(NULL), listMax(16), listSize(0)
{
    UErrorCode status = U_ZERO_ERROR;
    UCollationElements *elems = ucol_openElements(coll, string.getBuffer(), string.length(), &status);
    uint32_t strengthMask = 0;
    int32_t order, low, high;

    switch (ucol_getStrength(coll))
    {
    default:
        strengthMask |= UCOL_TERTIARYORDERMASK;
        U_FALLTHROUGH;
    case UCOL_SECONDARY:
        strengthMask |= UCOL_SECONDARYORDERMASK;
        U_FALLTHROUGH;
    case UCOL_PRIMARY:
        strengthMask |= UCOL_PRIMARYORDERMASK;
    }

    list = new Order[listMax];

    ucol_setOffset(elems, stringOffset, &status);

    do {
        low   = ucol_getOffset(elems);
        order = ucol_next(elems, &status);
        high  = ucol_getOffset(elems);

        if (order != UCOL_NULLORDER) {
            order &= strengthMask;
        }

        if (order != UCOL_IGNORABLE) {
            add(order, low, high);
        }
    } while (order != UCOL_NULLORDER);

    ucol_closeElements(elems);
}

OrderList::~OrderList()
{
    delete[] list;
}

void OrderList::add(int32_t order, int32_t low, int32_t high)
{
    if (listSize >= listMax) {
        listMax *= 2;

        Order *newList = new Order[listMax];

        uprv_memcpy(newList, list, listSize * sizeof(Order));
        delete[] list;
        list = newList;
    }

    list[listSize].order      = order;
    list[listSize].lowOffset  = low;
    list[listSize].highOffset = high;

    listSize += 1;
}

const Order *OrderList::get(int32_t index) const
{
    if (index >= listSize) {
        return NULL;
    }

    return &list[index];
}

int32_t OrderList::getLowOffset(int32_t index) const
{
    const Order *order = get(index);

    if (order != NULL) {
        return order->lowOffset;
    }

    return -1;
}

int32_t OrderList::getHighOffset(int32_t index) const
{
    const Order *order = get(index);

    if (order != NULL) {
        return order->highOffset;
    }

    return -1;
}

int32_t OrderList::getOrder(int32_t index) const
{
    const Order *order = get(index);

    if (order != NULL) {
        return order->order;
    }

    return UCOL_NULLORDER;
}

int32_t OrderList::size() const
{
    return listSize;
}

void OrderList::reverse()
{
    for(int32_t f = 0, b = listSize - 1; f < b; f += 1, b -= 1) {
        Order swap = list[b];

        list[b] = list[f];
        list[f] = swap;
    }
}

UBool OrderList::compare(const OrderList &other) const
{
    if (listSize != other.listSize) {
        return FALSE;
    }

    for(int32_t i = 0; i < listSize; i += 1) {
        if (list[i].order  != other.list[i].order ||
            list[i].lowOffset != other.list[i].lowOffset ||
            list[i].highOffset != other.list[i].highOffset) {
                return FALSE;
        }
    }

    return TRUE;
}

UBool OrderList::matchesAt(int32_t offset, const OrderList &other) const
{
    // NOTE: sizes include the NULLORDER, which we don't want to compare.
    int32_t otherSize = other.size() - 1;

    if (listSize - 1 - offset < otherSize) {
        return FALSE;
    }

    for (int32_t i = offset, j = 0; j < otherSize; i += 1, j += 1) {
        if (getOrder(i) != other.getOrder(j)) {
            return FALSE;
        }
    }

    return TRUE;
}

static char *printOffsets(char *buffer, OrderList &list)
{
    int32_t size = list.size();
    char *s = buffer;

    for(int32_t i = 0; i < size; i += 1) {
        const Order *order = list.get(i);

        if (i != 0) {
            s += sprintf(s, ", ");
        }

        s += sprintf(s, "(%d, %d)", order->lowOffset, order->highOffset);
    }

    return buffer;
}

static char *printOrders(char *buffer, OrderList &list)
{
    int32_t size = list.size();
    char *s = buffer;

    for(int32_t i = 0; i < size; i += 1) {
        const Order *order = list.get(i);

        if (i != 0) {
            s += sprintf(s, ", ");
        }

        s += sprintf(s, "%8.8X", order->order);
    }

    return buffer;
}

void SSearchTest::offsetTest()
{
    const char *test[] = {
        // The sequence \u0FB3\u0F71\u0F71\u0F80 contains a discontiguous
        // contraction (\u0FB3\u0F71\u0F80) logically followed by \u0F71.
        "\\u1E33\\u0FB3\\u0F71\\u0F71\\u0F80\\uD835\\uDF6C\\u01B0",

        "\\ua191\\u16ef\\u2036\\u017a",

#if 0
        // This results in a complex interaction between contraction,
        // expansion and normalization that confuses the backwards offset fixups.
        "\\u0F7F\\u0F80\\u0F81\\u0F82\\u0F83\\u0F84\\u0F85",
#endif

        "\\u0F80\\u0F81\\u0F82\\u0F83\\u0F84\\u0F85",
        "\\u07E9\\u07EA\\u07F1\\u07F2\\u07F3",

        "\\u02FE\\u02FF"
        "\\u0300\\u0301\\u0302\\u0303\\u0304\\u0305\\u0306\\u0307\\u0308\\u0309\\u030A\\u030B\\u030C\\u030D\\u030E\\u030F"
        "\\u0310\\u0311\\u0312\\u0313\\u0314\\u0315\\u0316\\u0317\\u0318\\u0319\\u031A\\u031B\\u031C\\u031D\\u031E\\u031F"
        "\\u0320\\u0321\\u0322\\u0323\\u0324\\u0325\\u0326\\u0327\\u0328\\u0329\\u032A\\u032B\\u032C\\u032D\\u032E\\u032F"
        "\\u0330\\u0331\\u0332\\u0333\\u0334\\u0335\\u0336\\u0337\\u0338\\u0339\\u033A\\u033B\\u033C\\u033D\\u033E\\u033F"
        "\\u0340\\u0341\\u0342\\u0343\\u0344\\u0345\\u0346\\u0347\\u0348\\u0349\\u034A\\u034B\\u034C\\u034D\\u034E", // currently not working, see #8081

        "\\u02FE\\u02FF\\u0300\\u0301\\u0302\\u0303\\u0316\\u0317\\u0318", // currently not working, see #8081
        "a\\u02FF\\u0301\\u0316", // currently not working, see #8081
        "a\\u02FF\\u0316\\u0301",
        "a\\u0430\\u0301\\u0316",
        "a\\u0430\\u0316\\u0301",
        "abc\\u0E41\\u0301\\u0316",
        "abc\\u0E41\\u0316\\u0301",
        "\\u0E41\\u0301\\u0316",
        "\\u0E41\\u0316\\u0301",
        "a\\u0301\\u0316",
        "a\\u0316\\u0301",
        "\\uAC52\\uAC53",
        "\\u34CA\\u34CB",
        "\\u11ED\\u11EE",
        "\\u30C3\\u30D0",
        "p\\u00E9ch\\u00E9",
        "a\\u0301\\u0325",
        "a\\u0300\\u0325",
        "a\\u0325\\u0300",
        "A\\u0323\\u0300B",
        "A\\u0300\\u0323B",
        "A\\u0301\\u0323B",
        "A\\u0302\\u0301\\u0323B",
        "abc",
        "ab\\u0300c",
        "ab\\u0300\\u0323c",
        " \\uD800\\uDC00\\uDC00",
        "a\\uD800\\uDC00\\uDC00",
        "A\\u0301\\u0301",
        "A\\u0301\\u0323",
        "A\\u0301\\u0323B",
        "B\\u0301\\u0323C",
        "A\\u0300\\u0323B",
        "\\u0301A\\u0301\\u0301",
        "abcd\\r\\u0301",
        "p\\u00EAche",
        "pe\\u0302che",
    };

    int32_t testCount = UPRV_LENGTHOF(test);
    UErrorCode status = U_ZERO_ERROR;
    RuleBasedCollator *col = (RuleBasedCollator *) Collator::createInstance(Locale::getEnglish(), status);
    if (U_FAILURE(status)) {
        errcheckln(status, "Failed to create collator in offsetTest! - %s", u_errorName(status));
        return;
    }
    char buffer[4096];  // A bit of a hack... just happens to be long enough for all the test cases...
                        // We could allocate one that's the right size by (CE_count * 10) + 2
                        // 10 chars is enough room for 8 hex digits plus ", ". 2 extra chars for "[" and "]"

    col->setAttribute(UCOL_NORMALIZATION_MODE, UCOL_ON, status);

    for(int32_t i = 0; i < testCount; i += 1) {
        UnicodeString ts = CharsToUnicodeString(test[i]);
        CollationElementIterator *iter = col->createCollationElementIterator(ts);
        OrderList forwardList;
        OrderList backwardList;
        int32_t order, low, high;

        do {
            low   = iter->getOffset();
            order = iter->next(status);
            high  = iter->getOffset();

            forwardList.add(order, low, high);
        } while (order != CollationElementIterator::NULLORDER);

        iter->reset();
        iter->setOffset(ts.length(), status);

        backwardList.add(CollationElementIterator::NULLORDER, iter->getOffset(), iter->getOffset());

        do {
            high  = iter->getOffset();
            order = iter->previous(status);
            low   = iter->getOffset();

            if (order == CollationElementIterator::NULLORDER) {
                break;
            }

            backwardList.add(order, low, high);
        } while (TRUE);

        backwardList.reverse();

        if (forwardList.compare(backwardList)) {
            logln("Works with \"%s\"", test[i]);
            logln("Forward offsets:  [%s]", printOffsets(buffer, forwardList));
//          logln("Backward offsets: [%s]", printOffsets(buffer, backwardList));

            logln("Forward CEs:  [%s]", printOrders(buffer, forwardList));
//          logln("Backward CEs: [%s]", printOrders(buffer, backwardList));

            logln();
        } else {
            errln("Fails with \"%s\"", test[i]);
            infoln("Forward offsets:  [%s]", printOffsets(buffer, forwardList));
            infoln("Backward offsets: [%s]", printOffsets(buffer, backwardList));

            infoln("Forward CEs:  [%s]", printOrders(buffer, forwardList));
            infoln("Backward CEs: [%s]", printOrders(buffer, backwardList));

            infoln();
        }
        delete iter;
    }
    delete col;
}

#if 0
static UnicodeString &escape(const UnicodeString &string, UnicodeString &buffer)
{
    for(int32_t i = 0; i < string.length(); i += 1) {
        UChar32 ch = string.char32At(i);

        if (ch >= 0x0020 && ch <= 0x007F) {
            if (ch == 0x005C) {
                buffer.append("\\\\");
            } else {
                buffer.append(ch);
            }
        } else {
            char cbuffer[12];

            if (ch <= 0xFFFFL) {
                sprintf(cbuffer, "\\u%4.4X", ch);
            } else {
                sprintf(cbuffer, "\\U%8.8X", ch);
            }

            buffer.append(cbuffer);
        }

        if (ch >= 0x10000L) {
            i += 1;
        }
    }

    return buffer;
}
#endif

void SSearchTest::sharpSTest()
{
    UErrorCode status = U_ZERO_ERROR;
    UCollator *coll = NULL;
    UnicodeString lp  = "fuss";
    UnicodeString sp = "fu\\u00DF";
    UnicodeString targets[]  = {"fu\\u00DF", "fu\\u00DFball", "1fu\\u00DFball", "12fu\\u00DFball", "123fu\\u00DFball", "1234fu\\u00DFball",
                                "ffu\\u00DF", "fufu\\u00DF", "fusfu\\u00DF",
                                "fuss", "ffuss", "fufuss", "fusfuss", "1fuss", "12fuss", "123fuss", "1234fuss", "fu\\u00DF", "1fu\\u00DF", "12fu\\u00DF", "123fu\\u00DF", "1234fu\\u00DF"};
    int32_t start = -1, end = -1;

    coll = ucol_openFromShortString("LEN_S1", FALSE, NULL, &status);
    TEST_ASSERT_SUCCESS(status);

    UnicodeString lpUnescaped = lp.unescape();
    UnicodeString spUnescaped = sp.unescape();

    LocalUStringSearchPointer ussLong(usearch_openFromCollator(lpUnescaped.getBuffer(), lpUnescaped.length(),
                                                           lpUnescaped.getBuffer(), lpUnescaped.length(),   // actual test data will be set later
                                                           coll,
                                                           NULL,     // the break iterator
                                                           &status));

    LocalUStringSearchPointer ussShort(usearch_openFromCollator(spUnescaped.getBuffer(), spUnescaped.length(),
                                                           spUnescaped.getBuffer(), spUnescaped.length(),   // actual test data will be set later
                                                           coll,
                                                           NULL,     // the break iterator
                                                           &status));
    TEST_ASSERT_SUCCESS(status);

    for (uint32_t t = 0; t < UPRV_LENGTHOF(targets); t += 1) {
        UBool bFound;
        UnicodeString target = targets[t].unescape();

        start = end = -1;
        usearch_setText(ussLong.getAlias(), target.getBuffer(), target.length(), &status);
        bFound = usearch_search(ussLong.getAlias(), 0, &start, &end, &status);
        TEST_ASSERT_SUCCESS(status);
        if (bFound) {
            logln("Test %d: found long pattern at [%d, %d].", t, start, end);
        } else {
            dataerrln("Test %d: did not find long pattern.", t);
        }

        usearch_setText(ussShort.getAlias(), target.getBuffer(), target.length(), &status);
        bFound = usearch_search(ussShort.getAlias(), 0, &start, &end, &status);
        TEST_ASSERT_SUCCESS(status);
        if (bFound) {
            logln("Test %d: found long pattern at [%d, %d].", t, start, end);
        } else {
            dataerrln("Test %d: did not find long pattern.", t);
        }
    }

    ucol_close(coll);
}

void SSearchTest::goodSuffixTest()
{
    UErrorCode status = U_ZERO_ERROR;
    UCollator *coll = NULL;
    UnicodeString pat = /*"gcagagag"*/ "fxeld";
    UnicodeString target = /*"gcatcgcagagagtatacagtacg"*/ "cloveldfxeld";
    int32_t start = -1, end = -1;
    UBool bFound;

    coll = ucol_open(NULL, &status);
    TEST_ASSERT_SUCCESS(status);

    LocalUStringSearchPointer ss(usearch_openFromCollator(pat.getBuffer(), pat.length(),
                                                          target.getBuffer(), target.length(),
                                                          coll,
                                                          NULL,     // the break iterator
                                                          &status));
    TEST_ASSERT_SUCCESS(status);

    bFound = usearch_search(ss.getAlias(), 0, &start, &end, &status);
    TEST_ASSERT_SUCCESS(status);
    if (bFound) {
        logln("Found pattern at [%d, %d].", start, end);
    } else {
        dataerrln("Did not find pattern.");
    }

    ucol_close(coll);
}

//
//  searchTime()    A quick and dirty performance test for string search.
//                  Probably  doesn't really belong as part of intltest, but it
//                  does check that the search succeeds, and gets the right result,
//                  so it serves as a functionality test also.
//
//                  To run as a perf test, up the loop count, select by commenting
//                  and uncommenting in the code the operation to be measured,
//                  rebuild, and measure the running time of this test alone.
//
//                     time LD_LIBRARY_PATH=whatever  ./intltest  collate/SSearchTest/searchTime
//
void SSearchTest::searchTime() {
    static const char *longishText =
"Whylom, as olde stories tellen us,\n"
"Ther was a duk that highte Theseus:\n"
"Of Athenes he was lord and governour,\n"
"And in his tyme swich a conquerour,\n"
"That gretter was ther noon under the sonne.\n"
"Ful many a riche contree hadde he wonne;\n"
"What with his wisdom and his chivalrye,\n"
"He conquered al the regne of Femenye,\n"
"That whylom was y-cleped Scithia;\n"
"And weddede the quene Ipolita,\n"
"And broghte hir hoom with him in his contree\n"
"With muchel glorie and greet solempnitee,\n"
"And eek hir yonge suster Emelye.\n"
"And thus with victorie and with melodye\n"
"Lete I this noble duk to Athenes ryde,\n"
"And al his hoost, in armes, him bisyde.\n"
"And certes, if it nere to long to here,\n"
"I wolde han told yow fully the manere,\n"
"How wonnen was the regne of Femenye\n"
"By Theseus, and by his chivalrye;\n"
"And of the grete bataille for the nones\n"
"Bitwixen Athen's and Amazones;\n"
"And how asseged was Ipolita,\n"
"The faire hardy quene of Scithia;\n"
"And of the feste that was at hir weddinge,\n"
"And of the tempest at hir hoom-cominge;\n"
"But al that thing I moot as now forbere.\n"
"I have, God woot, a large feeld to ere,\n"
"And wayke been the oxen in my plough.\n"
"The remenant of the tale is long y-nough.\n"
"I wol nat letten eek noon of this route;\n"
"Lat every felawe telle his tale aboute,\n"
"And lat see now who shal the soper winne;\n"
"And ther I lefte, I wol ageyn biginne.\n"
"This duk, of whom I make mencioun,\n"
"When he was come almost unto the toun,\n"
"In al his wele and in his moste pryde,\n"
"He was war, as he caste his eye asyde,\n"
"Wher that ther kneled in the hye weye\n"
"A companye of ladies, tweye and tweye,\n"
"Ech after other, clad in clothes blake; \n"
"But swich a cry and swich a wo they make,\n"
"That in this world nis creature livinge,\n"
"That herde swich another weymentinge;\n"
"And of this cry they nolde never stenten,\n"
"Til they the reynes of his brydel henten.\n"
"'What folk ben ye, that at myn hoomcominge\n"
"Perturben so my feste with cryinge'?\n"
"Quod Theseus, 'have ye so greet envye\n"
"Of myn honour, that thus compleyne and crye? \n"
"Or who hath yow misboden, or offended?\n"
"And telleth me if it may been amended;\n"
"And why that ye ben clothed thus in blak'?\n"
"The eldest lady of hem alle spak,\n"
"When she hadde swowned with a deedly chere,\n"
"That it was routhe for to seen and here,\n"
"And seyde: 'Lord, to whom Fortune hath yiven\n"
"Victorie, and as a conquerour to liven,\n"
"Noght greveth us your glorie and your honour;\n"
"But we biseken mercy and socour.\n"
"Have mercy on our wo and our distresse.\n"
"Som drope of pitee, thurgh thy gentilesse,\n"
"Up-on us wrecched wommen lat thou falle.\n"
"For certes, lord, ther nis noon of us alle,\n"
"That she nath been a duchesse or a quene;\n"
"Now be we caitifs, as it is wel sene:\n"
"Thanked be Fortune, and hir false wheel,\n"
"That noon estat assureth to be weel.\n"
"And certes, lord, t'abyden your presence,\n"
"Here in the temple of the goddesse Clemence\n"
"We han ben waytinge al this fourtenight;\n"
"Now help us, lord, sith it is in thy might.\n"
"I wrecche, which that wepe and waille thus,\n"
"Was whylom wyf to king Capaneus,\n"
"That starf at Thebes, cursed be that day!\n"
"And alle we, that been in this array,\n"
"And maken al this lamentacioun,\n"
"We losten alle our housbondes at that toun,\n"
"Whyl that the sege ther-aboute lay.\n"
"And yet now th'olde Creon, weylaway!\n"
"The lord is now of Thebes the citee, \n"
"Fulfild of ire and of iniquitee,\n"
"He, for despyt, and for his tirannye,\n"
"To do the dede bodyes vileinye,\n"
"Of alle our lordes, whiche that ben slawe,\n"
"Hath alle the bodyes on an heep y-drawe,\n"
"And wol nat suffren hem, by noon assent,\n"
"Neither to been y-buried nor y-brent,\n"
"But maketh houndes ete hem in despyt. zet'\n";

const char *cPattern = "maketh houndes ete hem";
//const char *cPattern = "Whylom";
//const char *cPattern = "zet";
    const char *testId = "searchTime()";   // for error macros.
    UnicodeString target = longishText;
    UErrorCode status = U_ZERO_ERROR;


    LocalUCollatorPointer collator(ucol_open("en", &status));
    //ucol_setStrength(collator.getAlias(), collatorStrength);
    //ucol_setAttribute(collator.getAlias(), UCOL_NORMALIZATION_MODE, normalize, &status);
    UnicodeString uPattern = cPattern;
    LocalUStringSearchPointer uss(usearch_openFromCollator(uPattern.getBuffer(), uPattern.length(),
                                                           target.getBuffer(), target.length(),
                                                           collator.getAlias(),
                                                           NULL,     // the break iterator
                                                           &status));
    TEST_ASSERT_SUCCESS(status);

//  int32_t foundStart;
//  int32_t foundEnd;
    UBool   found;

    // Find the match position usgin strstr
    const char *pm = strstr(longishText, cPattern);
    TEST_ASSERT_M(pm!=NULL, "No pattern match with strstr");
    int32_t  refMatchPos = (int32_t)(pm - longishText);
    int32_t  icuMatchPos;
    int32_t  icuMatchEnd;
    usearch_search(uss.getAlias(), 0, &icuMatchPos, &icuMatchEnd, &status);
    TEST_ASSERT_SUCCESS(status);
    TEST_ASSERT_M(refMatchPos == icuMatchPos, "strstr and icu give different match positions.");

    int32_t i;
    // int32_t j=0;

    // Try loopcounts around 100000 to some millions, depending on the operation,
    //   to get runtimes of at least several seconds.
    for (i=0; i<10000; i++) {
        found = usearch_search(uss.getAlias(), 0, &icuMatchPos, &icuMatchEnd, &status);
        (void)found;   // Suppress set but not used warning.
        //TEST_ASSERT_SUCCESS(status);
        //TEST_ASSERT(found);

        // usearch_setOffset(uss.getAlias(), 0, &status);
        // icuMatchPos = usearch_next(uss.getAlias(), &status);

         // The i+j stuff is to confuse the optimizer and get it to actually leave the
         //   call to strstr in place.
         //pm = strstr(longishText+j, cPattern);
         //j = (j + i)%5;
    }

    //printf("%ld, %d\n", pm-longishText, j);
}

//----------------------------------------------------------------------------------------
//
//   Random Numbers.  Similar to standard lib rand() and srand()
//                    Not using library to
//                      1.  Get same results on all platforms.
//                      2.  Get access to current seed, to more easily reproduce failures.
//
//---------------------------------------------------------------------------------------
static uint32_t m_seed = 1;

static uint32_t m_rand()
{
    m_seed = m_seed * 1103515245 + 12345;
    return (uint32_t)(m_seed/65536) % 32768;
}

class Monkey
{
public:
    virtual void append(UnicodeString &test, UnicodeString &alternate) = 0;

protected:
    Monkey();
    virtual ~Monkey();
};

Monkey::Monkey()
{
    // ook?
}

Monkey::~Monkey()
{
    // ook?
}

class SetMonkey : public Monkey
{
public:
    SetMonkey(const USet *theSet);
    ~SetMonkey();

    virtual void append(UnicodeString &test, UnicodeString &alternate);

private:
    const USet *set;
};

SetMonkey::SetMonkey(const USet *theSet)
    : Monkey(), set(theSet)
{
    // ook?
}

SetMonkey::~SetMonkey()
{
    //ook...
}

void SetMonkey::append(UnicodeString &test, UnicodeString &alternate)
{
    int32_t size = uset_size(set);
    int32_t index = m_rand() % size;
    UChar32 ch = uset_charAt(set, index);
    UnicodeString str(ch);

    test.append(str);
    alternate.append(str); // flip case, or some junk?
}

class StringSetMonkey : public Monkey
{
public:
    StringSetMonkey(const USet *theSet, UCollator *theCollator, CollData *theCollData);
    ~StringSetMonkey();

    void append(UnicodeString &testCase, UnicodeString &alternate);

private:
    UnicodeString &generateAlternative(const UnicodeString &testCase, UnicodeString &alternate);

    const USet *set;
    UCollator  *coll;
    CollData   *collData;
};

StringSetMonkey::StringSetMonkey(const USet *theSet, UCollator *theCollator, CollData *theCollData)
: Monkey(), set(theSet), coll(theCollator), collData(theCollData)
{
    // ook.
}

StringSetMonkey::~StringSetMonkey()
{
    // ook?
}

void StringSetMonkey::append(UnicodeString &testCase, UnicodeString &alternate)
{
    int32_t itemCount = uset_getItemCount(set), len = 0;
    int32_t index = m_rand() % itemCount;
    UChar32 rangeStart = 0, rangeEnd = 0;
    UChar buffer[16];
    UErrorCode err = U_ZERO_ERROR;

    len = uset_getItem(set, index, &rangeStart, &rangeEnd, buffer, 16, &err);

    if (len == 0) {
        int32_t offset = m_rand() % (rangeEnd - rangeStart + 1);
        UChar32 ch = rangeStart + offset;
        UnicodeString str(ch);

        testCase.append(str);
        generateAlternative(str, alternate);
    } else if (len > 0) {
        // should check that len < 16...
        UnicodeString str(buffer, len);

        testCase.append(str);
        generateAlternative(str, alternate);
    } else {
        // shouldn't happen...
    }
}

UnicodeString &StringSetMonkey::generateAlternative(const UnicodeString &testCase, UnicodeString &alternate)
{
    // find out shortest string for the longest sequence of ces.
    // needs to be refined to use dynamic programming, but will be roughly right
    UErrorCode status = U_ZERO_ERROR;
    CEList ceList(coll, testCase, status);
    UnicodeString alt;
    int32_t offset = 0;

    if (ceList.size() == 0) {
        return alternate.append(testCase);
    }

    while (offset < ceList.size()) {
        int32_t ce = ceList.get(offset);
        const StringList *strings = collData->getStringList(ce);

        if (strings == NULL) {
            return alternate.append(testCase);
        }

        int32_t stringCount = strings->size();
        int32_t tries = 0;

        // find random string that generates the same CEList
        const CEList *ceList2 = NULL;
        const UnicodeString *string = NULL;
              UBool matches = FALSE;

        do {
            int32_t s = m_rand() % stringCount;

            if (tries++ > stringCount) {
                alternate.append(testCase);
                return alternate;
            }

            string = strings->get(s);
            ceList2 = collData->getCEList(string);
            matches = ceList.matchesAt(offset, ceList2);

            if (! matches) {
                collData->freeCEList((CEList *) ceList2);
            }
        } while (! matches);

        alt.append(*string);
        offset += ceList2->size();
        collData->freeCEList(ceList2);
    }

    const CEList altCEs(coll, alt, status);

    if (ceList.matchesAt(0, &altCEs)) {
        return alternate.append(alt);
    }

    return alternate.append(testCase);
}

static void generateTestCase(UCollator *coll, Monkey *monkeys[], int32_t monkeyCount, UnicodeString &testCase, UnicodeString &alternate)
{
    int32_t pieces = (m_rand() % 4) + 1;
    UErrorCode status = U_ZERO_ERROR;
    UBool matches;

    do {
        testCase.remove();
        alternate.remove();
        monkeys[0]->append(testCase, alternate);

        for(int32_t piece = 0; piece < pieces; piece += 1) {
            int32_t monkey = m_rand() % monkeyCount;

            monkeys[monkey]->append(testCase, alternate);
        }

        const CEList ceTest(coll, testCase, status);
        const CEList ceAlt(coll, alternate, status);

        matches = ceTest.matchesAt(0, &ceAlt);
    } while (! matches);
}

static UBool simpleSearch(UCollator *coll, const UnicodeString &target, int32_t offset, const UnicodeString &pattern, int32_t &matchStart, int32_t &matchEnd)
{
    UErrorCode      status = U_ZERO_ERROR;
    OrderList       targetOrders(coll, target, offset);
    OrderList       patternOrders(coll, pattern);
    int32_t         targetSize  = targetOrders.size() - 1;
    int32_t         patternSize = patternOrders.size() - 1;
    UBreakIterator *charBreakIterator = ubrk_open(UBRK_CHARACTER, ucol_getLocaleByType(coll, ULOC_VALID_LOCALE, &status),
                                                  target.getBuffer(), target.length(), &status);

    if (patternSize == 0) {
        // Searching for an empty pattern always fails
        matchStart = matchEnd = -1;
        ubrk_close(charBreakIterator);
        return FALSE;
    }

    matchStart = matchEnd = -1;

    for(int32_t i = 0; i < targetSize; i += 1) {
        if (targetOrders.matchesAt(i, patternOrders)) {
            int32_t start    = targetOrders.getLowOffset(i);
            int32_t maxLimit = targetOrders.getLowOffset(i + patternSize);
            int32_t minLimit = targetOrders.getLowOffset(i + patternSize - 1);

            // if the low and high offsets of the first CE in
            // the match are the same, it means that the match
            // starts in the middle of an expansion - all but
            // the first CE of the expansion will have the offset
            // of the following character.
            if (start == targetOrders.getHighOffset(i)) {
                continue;
            }

            // Make sure match starts on a grapheme boundary
            if (! ubrk_isBoundary(charBreakIterator, start)) {
                continue;
            }

            // If the low and high offsets of the CE after the match
            // are the same, it means that the match ends in the middle
            // of an expansion sequence.
            if (maxLimit == targetOrders.getHighOffset(i + patternSize) &&
                targetOrders.getOrder(i + patternSize) != UCOL_NULLORDER) {
                continue;
            }

            int32_t mend = maxLimit;

            // Find the first grapheme break after the character index
            // of the last CE in the match. If it's after character index
            // that's after the last CE in the match, use that index
            // as the end of the match.
            if (minLimit < maxLimit) {
                // When the last CE's low index is same with its high index, the CE is likely
                // a part of expansion. In this case, the index is located just after the
                // character corresponding to the CEs compared above. If the index is right
                // at the break boundary, move the position to the next boundary will result
                // incorrect match length when there are ignorable characters exist between
                // the position and the next character produces CE(s). See ticket#8482.
                if (minLimit == targetOrders.getHighOffset(i + patternSize - 1) && ubrk_isBoundary(charBreakIterator, minLimit)) {
                    mend = minLimit;
                } else {
                    int32_t nba = ubrk_following(charBreakIterator, minLimit);

                    if (nba >= targetOrders.getHighOffset(i + patternSize - 1)) {
                        mend = nba;
                    }
                }
            }

            if (mend > maxLimit) {
                continue;
            }

            if (! ubrk_isBoundary(charBreakIterator, mend)) {
                continue;
            }

            matchStart = start;
            matchEnd   = mend;

            ubrk_close(charBreakIterator);
            return TRUE;
        }
    }

    ubrk_close(charBreakIterator);
    return FALSE;
}

#if !UCONFIG_NO_REGULAR_EXPRESSIONS
static int32_t  getIntParam(UnicodeString name, UnicodeString &params, int32_t defaultVal) {
    int32_t val = defaultVal;

    name.append(" *= *(-?\\d+)");

    UErrorCode status = U_ZERO_ERROR;
    RegexMatcher m(name, params, 0, status);

    if (m.find()) {
        // The param exists.  Convert the string to an int.
        char valString[100];
        int32_t paramLength = m.end(1, status) - m.start(1, status);

        if (paramLength >= (int32_t)(sizeof(valString)-1)) {
            paramLength = (int32_t)(sizeof(valString)-2);
        }

        params.extract(m.start(1, status), paramLength, valString, sizeof(valString));
        val = uprv_strtol(valString,  NULL, 10);

        // Delete this parameter from the params string.
        m.reset();
        params = m.replaceFirst("", status);
    }

  //U_ASSERT(U_SUCCESS(status));
    if (! U_SUCCESS(status)) {
        val = defaultVal;
    }

    return val;
}
#endif

#if !UCONFIG_NO_COLLATION
int32_t SSearchTest::monkeyTestCase(UCollator *coll, const UnicodeString &testCase, const UnicodeString &pattern, const UnicodeString &altPattern,
                                    const char *name, const char *strength, uint32_t seed)
{
    UErrorCode status = U_ZERO_ERROR;
    int32_t actualStart = -1, actualEnd = -1;
  //int32_t expectedStart = prefix.length(), expectedEnd = prefix.length() + altPattern.length();
    int32_t expectedStart = -1, expectedEnd = -1;
    int32_t notFoundCount = 0;
    LocalUStringSearchPointer uss(usearch_openFromCollator(pattern.getBuffer(), pattern.length(),
                                                           testCase.getBuffer(), testCase.length(),
                                                           coll,
                                                           NULL,     // the break iterator
                                                           &status));

    // **** TODO: find *all* matches, not just first one ****
    simpleSearch(coll, testCase, 0, pattern, expectedStart, expectedEnd);

    usearch_search(uss.getAlias(), 0, &actualStart, &actualEnd, &status);

    if (expectedStart >= 0 && (actualStart != expectedStart || actualEnd != expectedEnd)) {
        errln("Search for <pattern> in <%s> failed: expected [%d, %d], got [%d, %d]\n"
              "    strength=%s seed=%d",
              name, expectedStart, expectedEnd, actualStart, actualEnd, strength, seed);
    }

    if (expectedStart == -1 && actualStart == -1) {
        notFoundCount += 1;
    }

    // **** TODO: find *all* matches, not just first one ****
    simpleSearch(coll, testCase, 0, altPattern, expectedStart, expectedEnd);

    usearch_setPattern(uss.getAlias(), altPattern.getBuffer(), altPattern.length(), &status);

    usearch_search(uss.getAlias(), 0, &actualStart, &actualEnd, &status);

    if (expectedStart >= 0 && (actualStart != expectedStart || actualEnd != expectedEnd)) {
        errln("Search for <alt_pattern> in <%s> failed: expected [%d, %d], got [%d, %d]\n"
              "    strength=%s seed=%d",
              name, expectedStart, expectedEnd, actualStart, actualEnd, strength, seed);
    }

    if (expectedStart == -1 && actualStart == -1) {
        notFoundCount += 1;
    }

    return notFoundCount;
}
#endif

void SSearchTest::monkeyTest(char *params)
{
    // ook!
    UErrorCode status = U_ZERO_ERROR;
  //UCollator *coll = ucol_open(NULL, &status);
    UCollator *coll = ucol_openFromShortString("S1", FALSE, NULL, &status);

    if (U_FAILURE(status)) {
        errcheckln(status, "Failed to create collator in MonkeyTest! - %s", u_errorName(status));
        return;
    }

    CollData  *monkeyData = new CollData(coll, status);

    USet *expansions   = uset_openEmpty();
    USet *contractions = uset_openEmpty();

    ucol_getContractionsAndExpansions(coll, contractions, expansions, FALSE, &status);

    U_STRING_DECL(letter_pattern, "[[:letter:]-[:ideographic:]-[:hangul:]]", 39);
    U_STRING_INIT(letter_pattern, "[[:letter:]-[:ideographic:]-[:hangul:]]", 39);
    USet *letters = uset_openPattern(letter_pattern, 39, &status);
    SetMonkey letterMonkey(letters);
    StringSetMonkey contractionMonkey(contractions, coll, monkeyData);
    StringSetMonkey expansionMonkey(expansions, coll, monkeyData);
    UnicodeString testCase;
    UnicodeString alternate;
    UnicodeString pattern, altPattern;
    UnicodeString prefix, altPrefix;
    UnicodeString suffix, altSuffix;

    Monkey *monkeys[] = {
        &letterMonkey,
        &contractionMonkey,
        &expansionMonkey,
        &contractionMonkey,
        &expansionMonkey,
        &contractionMonkey,
        &expansionMonkey,
        &contractionMonkey,
        &expansionMonkey};
    int32_t monkeyCount = UPRV_LENGTHOF(monkeys);
    // int32_t nonMatchCount = 0;

    UCollationStrength strengths[] = {UCOL_PRIMARY, UCOL_SECONDARY, UCOL_TERTIARY};
    const char *strengthNames[] = {"primary", "secondary", "tertiary"};
    int32_t strengthCount = UPRV_LENGTHOF(strengths);
    int32_t loopCount = quick? 1000 : 10000;
    int32_t firstStrength = 0;
    int32_t lastStrength  = strengthCount - 1; //*/ 0;

    if (params != NULL) {
#if !UCONFIG_NO_REGULAR_EXPRESSIONS
        UnicodeString p(params);

        loopCount = getIntParam("loop", p, loopCount);
        m_seed    = getIntParam("seed", p, m_seed);

        RegexMatcher m(" *strength *= *(primary|secondary|tertiary) *", p, 0, status);
        if (m.find()) {
            UnicodeString breakType = m.group(1, status);

            for (int32_t s = 0; s < strengthCount; s += 1) {
                if (breakType == strengthNames[s]) {
                    firstStrength = lastStrength = s;
                    break;
                }
            }

            m.reset();
            p = m.replaceFirst("", status);
        }

        if (RegexMatcher("\\S", p, 0, status).find()) {
            // Each option is stripped out of the option string as it is processed.
            // All options have been checked.  The option string should have been completely emptied..
            char buf[100];
            p.extract(buf, sizeof(buf), NULL, status);
            buf[sizeof(buf)-1] = 0;
            errln("Unrecognized or extra parameter:  %s\n", buf);
            return;
        }
#else
        infoln("SSearchTest built with UCONFIG_NO_REGULAR_EXPRESSIONS: ignoring parameters.");
#endif
    }

    for(int32_t s = firstStrength; s <= lastStrength; s += 1) {
        int32_t notFoundCount = 0;

        logln("Setting strength to %s.", strengthNames[s]);
        ucol_setStrength(coll, strengths[s]);

        // TODO: try alternate prefix and suffix too?
        // TODO: alternates are only equal at primary strength. Is this OK?
        for(int32_t t = 0; t < loopCount; t += 1) {
            uint32_t seed = m_seed;
            // int32_t  nmc = 0;

            generateTestCase(coll, monkeys, monkeyCount, pattern, altPattern);
            generateTestCase(coll, monkeys, monkeyCount, prefix,  altPrefix);
            generateTestCase(coll, monkeys, monkeyCount, suffix,  altSuffix);

            // pattern
            notFoundCount += monkeyTestCase(coll, pattern, pattern, altPattern, "pattern", strengthNames[s], seed);

            testCase.remove();
            testCase.append(prefix);
            testCase.append(/*alt*/pattern);

            // prefix + pattern
            notFoundCount += monkeyTestCase(coll, testCase, pattern, altPattern, "prefix + pattern", strengthNames[s], seed);

            testCase.append(suffix);

            // prefix + pattern + suffix
            notFoundCount += monkeyTestCase(coll, testCase, pattern, altPattern, "prefix + pattern + suffix", strengthNames[s], seed);

            testCase.remove();
            testCase.append(pattern);
            testCase.append(suffix);

            // pattern + suffix
            notFoundCount += monkeyTestCase(coll, testCase, pattern, altPattern, "pattern + suffix", strengthNames[s], seed);
        }

       logln("For strength %s the not found count is %d.", strengthNames[s], notFoundCount);
    }

    uset_close(contractions);
    uset_close(expansions);
    uset_close(letters);
    delete monkeyData;

    ucol_close(coll);
}

#endif

#endif