summaryrefslogtreecommitdiff
path: root/Build/source/libs/icu-xetex/test/intltest/convtest.cpp
blob: 909c951464f8828421d0eb232e48c980d10cd196 (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
/*
*******************************************************************************
*
*   Copyright (C) 2003-2007, International Business Machines
*   Corporation and others.  All Rights Reserved.
*
*******************************************************************************
*   file name:  convtest.cpp
*   encoding:   US-ASCII
*   tab size:   8 (not used)
*   indentation:4
*
*   created on: 2003jul15
*   created by: Markus W. Scherer
*
*   Test file for data-driven conversion tests.
*/

#include "unicode/utypes.h"

#if !UCONFIG_NO_LEGACY_CONVERSION
/*
 * Note: Turning off all of convtest.cpp if !UCONFIG_NO_LEGACY_CONVERSION
 * is slightly unnecessary - it removes tests for Unicode charsets
 * like UTF-8 that should work.
 * However, there is no easy way for the test to detect whether a test case
 * is for a Unicode charset, so it would be difficult to only exclude those.
 * Also, regular testing of ICU is done with all modules on, therefore
 * not testing conversion for a custom configuration like this should be ok.
 */

#include "unicode/ucnv.h"
#include "unicode/unistr.h"
#include "unicode/parsepos.h"
#include "unicode/uniset.h"
#include "unicode/ustring.h"
#include "unicode/ures.h"
#include "convtest.h"
#include "unicode/tstdtmod.h"
#include <string.h>
#include <stdlib.h>

#define LENGTHOF(array) (int32_t)(sizeof(array)/sizeof((array)[0]))

enum {
    // characters used in test data for callbacks
    SUB_CB='?',
    SKIP_CB='0',
    STOP_CB='.',
    ESC_CB='&'
};

ConversionTest::ConversionTest() {
    UErrorCode errorCode=U_ZERO_ERROR;
    utf8Cnv=ucnv_open("UTF-8", &errorCode);
    ucnv_setToUCallBack(utf8Cnv, UCNV_TO_U_CALLBACK_STOP, NULL, NULL, NULL, &errorCode);
    if(U_FAILURE(errorCode)) {
        errln("unable to open UTF-8 converter");
    }
}

ConversionTest::~ConversionTest() {
    ucnv_close(utf8Cnv);
}

void
ConversionTest::runIndexedTest(int32_t index, UBool exec, const char *&name, char * /*par*/) {
    if (exec) logln("TestSuite ConversionTest: ");
    switch (index) {
        case 0: name="TestToUnicode"; if (exec) TestToUnicode(); break;
        case 1: name="TestFromUnicode"; if (exec) TestFromUnicode(); break;
        case 2: name="TestGetUnicodeSet"; if (exec) TestGetUnicodeSet(); break;
        default: name=""; break; //needed to end loop
    }
}

// test data interface ----------------------------------------------------- ***

void
ConversionTest::TestToUnicode() {
    ConversionCase cc;
    char charset[100], cbopt[4];
    const char *option;
    UnicodeString s, unicode;
    int32_t offsetsLength;
    UConverterToUCallback callback;

    TestDataModule *dataModule;
    TestData *testData;
    const DataMap *testCase;
    UErrorCode errorCode;
    int32_t i;

    errorCode=U_ZERO_ERROR;
    dataModule=TestDataModule::getTestDataModule("conversion", *this, errorCode);
    if(U_SUCCESS(errorCode)) {
        testData=dataModule->createTestData("toUnicode", errorCode);
        if(U_SUCCESS(errorCode)) {
            for(i=0; testData->nextCase(testCase, errorCode); ++i) {
                if(U_FAILURE(errorCode)) {
                    errln("error retrieving conversion/toUnicode test case %d - %s",
                            i, u_errorName(errorCode));
                    errorCode=U_ZERO_ERROR;
                    continue;
                }

                cc.caseNr=i;

                s=testCase->getString("charset", errorCode);
                s.extract(0, 0x7fffffff, charset, sizeof(charset), "");
                cc.charset=charset;

                cc.bytes=testCase->getBinary(cc.bytesLength, "bytes", errorCode);
                unicode=testCase->getString("unicode", errorCode);
                cc.unicode=unicode.getBuffer();
                cc.unicodeLength=unicode.length();

                offsetsLength=0;
                cc.offsets=testCase->getIntVector(offsetsLength, "offsets", errorCode);
                if(offsetsLength==0) {
                    cc.offsets=NULL;
                } else if(offsetsLength!=unicode.length()) {
                    errln("toUnicode[%d] unicode[%d] and offsets[%d] must have the same length",
                            i, unicode.length(), offsetsLength);
                    errorCode=U_ILLEGAL_ARGUMENT_ERROR;
                }

                cc.finalFlush= 0!=testCase->getInt28("flush", errorCode);
                cc.fallbacks= 0!=testCase->getInt28("fallbacks", errorCode);

                s=testCase->getString("errorCode", errorCode);
                if(s==UNICODE_STRING("invalid", 7)) {
                    cc.outErrorCode=U_INVALID_CHAR_FOUND;
                } else if(s==UNICODE_STRING("illegal", 7)) {
                    cc.outErrorCode=U_ILLEGAL_CHAR_FOUND;
                } else if(s==UNICODE_STRING("truncated", 9)) {
                    cc.outErrorCode=U_TRUNCATED_CHAR_FOUND;
                } else if(s==UNICODE_STRING("illesc", 6)) {
                    cc.outErrorCode=U_ILLEGAL_ESCAPE_SEQUENCE;
                } else if(s==UNICODE_STRING("unsuppesc", 9)) {
                    cc.outErrorCode=U_UNSUPPORTED_ESCAPE_SEQUENCE;
                } else {
                    cc.outErrorCode=U_ZERO_ERROR;
                }

                s=testCase->getString("callback", errorCode);
                s.extract(0, 0x7fffffff, cbopt, sizeof(cbopt), "");
                cc.cbopt=cbopt;
                switch(cbopt[0]) {
                case SUB_CB:
                    callback=UCNV_TO_U_CALLBACK_SUBSTITUTE;
                    break;
                case SKIP_CB:
                    callback=UCNV_TO_U_CALLBACK_SKIP;
                    break;
                case STOP_CB:
                    callback=UCNV_TO_U_CALLBACK_STOP;
                    break;
                case ESC_CB:
                    callback=UCNV_TO_U_CALLBACK_ESCAPE;
                    break;
                default:
                    callback=NULL;
                    break;
                }
                option=callback==NULL ? cbopt : cbopt+1;
                if(*option==0) {
                    option=NULL;
                }

                cc.invalidChars=testCase->getBinary(cc.invalidLength, "invalidChars", errorCode);

                if(U_FAILURE(errorCode)) {
                    errln("error parsing conversion/toUnicode test case %d - %s",
                            i, u_errorName(errorCode));
                    errorCode=U_ZERO_ERROR;
                } else {
                    logln("TestToUnicode[%d] %s", i, charset);
                    ToUnicodeCase(cc, callback, option);
                }
            }
            delete testData;
        }
        delete dataModule;
    }
    else {
        errln("Failed: could not load test conversion data");
    }
}

void
ConversionTest::TestFromUnicode() {
    ConversionCase cc;
    char charset[100], cbopt[4];
    const char *option;
    UnicodeString s, unicode, invalidUChars;
    int32_t offsetsLength, index;
    UConverterFromUCallback callback;

    TestDataModule *dataModule;
    TestData *testData;
    const DataMap *testCase;
    const UChar *p;
    UErrorCode errorCode;
    int32_t i, length;

    errorCode=U_ZERO_ERROR;
    dataModule=TestDataModule::getTestDataModule("conversion", *this, errorCode);
    if(U_SUCCESS(errorCode)) {
        testData=dataModule->createTestData("fromUnicode", errorCode);
        if(U_SUCCESS(errorCode)) {
            for(i=0; testData->nextCase(testCase, errorCode); ++i) {
                if(U_FAILURE(errorCode)) {
                    errln("error retrieving conversion/fromUnicode test case %d - %s",
                            i, u_errorName(errorCode));
                    errorCode=U_ZERO_ERROR;
                    continue;
                }

                cc.caseNr=i;

                s=testCase->getString("charset", errorCode);
                s.extract(0, 0x7fffffff, charset, sizeof(charset), "");
                cc.charset=charset;

                unicode=testCase->getString("unicode", errorCode);
                cc.unicode=unicode.getBuffer();
                cc.unicodeLength=unicode.length();
                cc.bytes=testCase->getBinary(cc.bytesLength, "bytes", errorCode);

                offsetsLength=0;
                cc.offsets=testCase->getIntVector(offsetsLength, "offsets", errorCode);
                if(offsetsLength==0) {
                    cc.offsets=NULL;
                } else if(offsetsLength!=cc.bytesLength) {
                    errln("fromUnicode[%d] bytes[%d] and offsets[%d] must have the same length",
                            i, cc.bytesLength, offsetsLength);
                    errorCode=U_ILLEGAL_ARGUMENT_ERROR;
                }

                cc.finalFlush= 0!=testCase->getInt28("flush", errorCode);
                cc.fallbacks= 0!=testCase->getInt28("fallbacks", errorCode);

                s=testCase->getString("errorCode", errorCode);
                if(s==UNICODE_STRING("invalid", 7)) {
                    cc.outErrorCode=U_INVALID_CHAR_FOUND;
                } else if(s==UNICODE_STRING("illegal", 7)) {
                    cc.outErrorCode=U_ILLEGAL_CHAR_FOUND;
                } else if(s==UNICODE_STRING("truncated", 9)) {
                    cc.outErrorCode=U_TRUNCATED_CHAR_FOUND;
                } else {
                    cc.outErrorCode=U_ZERO_ERROR;
                }

                s=testCase->getString("callback", errorCode);
                cc.setSub=0; // default: no subchar

                if((index=s.indexOf((UChar)0))>0) {
                    // read NUL-separated subchar first, if any
                    // copy the subchar from Latin-1 characters
                    // start after the NUL
                    p=s.getTerminatedBuffer();
                    length=index+1;
                    p+=length;
                    length=s.length()-length;
                    if(length<=0 || length>=(int32_t)sizeof(cc.subchar)) {
                        errorCode=U_ILLEGAL_ARGUMENT_ERROR;
                    } else {
                        int32_t j;

                        for(j=0; j<length; ++j) {
                            cc.subchar[j]=(char)p[j];
                        }
                        // NUL-terminate the subchar
                        cc.subchar[j]=0;
                        cc.setSub=1;
                    }

                    // remove the NUL and subchar from s
                    s.truncate(index);
                } else if((index=s.indexOf((UChar)0x3d))>0) /* '=' */ {
                    // read a substitution string, separated by an equal sign
                    p=s.getBuffer()+index+1;
                    length=s.length()-(index+1);
                    if(length<0 || length>=LENGTHOF(cc.subString)) {
                        errorCode=U_ILLEGAL_ARGUMENT_ERROR;
                    } else {
                        u_memcpy(cc.subString, p, length);
                        // NUL-terminate the subString
                        cc.subString[length]=0;
                        cc.setSub=-1;
                    }

                    // remove the equal sign and subString from s
                    s.truncate(index);
                }

                s.extract(0, 0x7fffffff, cbopt, sizeof(cbopt), "");
                cc.cbopt=cbopt;
                switch(cbopt[0]) {
                case SUB_CB:
                    callback=UCNV_FROM_U_CALLBACK_SUBSTITUTE;
                    break;
                case SKIP_CB:
                    callback=UCNV_FROM_U_CALLBACK_SKIP;
                    break;
                case STOP_CB:
                    callback=UCNV_FROM_U_CALLBACK_STOP;
                    break;
                case ESC_CB:
                    callback=UCNV_FROM_U_CALLBACK_ESCAPE;
                    break;
                default:
                    callback=NULL;
                    break;
                }
                option=callback==NULL ? cbopt : cbopt+1;
                if(*option==0) {
                    option=NULL;
                }

                invalidUChars=testCase->getString("invalidUChars", errorCode);
                cc.invalidUChars=invalidUChars.getBuffer();
                cc.invalidLength=invalidUChars.length();

                if(U_FAILURE(errorCode)) {
                    errln("error parsing conversion/fromUnicode test case %d - %s",
                            i, u_errorName(errorCode));
                    errorCode=U_ZERO_ERROR;
                } else {
                    logln("TestFromUnicode[%d] %s", i, charset);
                    FromUnicodeCase(cc, callback, option);
                }
            }
            delete testData;
        }
        delete dataModule;
    }
    else {
        errln("Failed: could not load test conversion data");
    }
}

static const UChar ellipsis[]={ 0x2e, 0x2e, 0x2e };

void
ConversionTest::TestGetUnicodeSet() {
    char charset[100];
    UnicodeString s, map, mapnot;
    int32_t which;

    ParsePosition pos;
    UnicodeSet cnvSet, mapSet, mapnotSet, diffSet;
    UnicodeSet *cnvSetPtr = &cnvSet;
    UConverter *cnv;

    TestDataModule *dataModule;
    TestData *testData;
    const DataMap *testCase;
    UErrorCode errorCode;
    int32_t i;

    errorCode=U_ZERO_ERROR;
    dataModule=TestDataModule::getTestDataModule("conversion", *this, errorCode);
    if(U_SUCCESS(errorCode)) {
        testData=dataModule->createTestData("getUnicodeSet", errorCode);
        if(U_SUCCESS(errorCode)) {
            for(i=0; testData->nextCase(testCase, errorCode); ++i) {
                if(U_FAILURE(errorCode)) {
                    errln("error retrieving conversion/getUnicodeSet test case %d - %s",
                            i, u_errorName(errorCode));
                    errorCode=U_ZERO_ERROR;
                    continue;
                }

                s=testCase->getString("charset", errorCode);
                s.extract(0, 0x7fffffff, charset, sizeof(charset), "");

                map=testCase->getString("map", errorCode);
                mapnot=testCase->getString("mapnot", errorCode);

                which=testCase->getInt28("which", errorCode);

                if(U_FAILURE(errorCode)) {
                    errln("error parsing conversion/getUnicodeSet test case %d - %s",
                            i, u_errorName(errorCode));
                    errorCode=U_ZERO_ERROR;
                    continue;
                }

                // test this test case
                mapSet.clear();
                mapnotSet.clear();

                pos.setIndex(0);
                mapSet.applyPattern(map, pos, 0, NULL, errorCode);
                if(U_FAILURE(errorCode) || pos.getIndex()!=map.length()) {
                    errln("error creating the map set for conversion/getUnicodeSet test case %d - %s\n"
                          "    error index %d  index %d  U+%04x",
                            i, u_errorName(errorCode), pos.getErrorIndex(), pos.getIndex(), map.char32At(pos.getIndex()));
                    errorCode=U_ZERO_ERROR;
                    continue;
                }

                pos.setIndex(0);
                mapnotSet.applyPattern(mapnot, pos, 0, NULL, errorCode);
                if(U_FAILURE(errorCode) || pos.getIndex()!=mapnot.length()) {
                    errln("error creating the mapnot set for conversion/getUnicodeSet test case %d - %s\n"
                          "    error index %d  index %d  U+%04x",
                            i, u_errorName(errorCode), pos.getErrorIndex(), pos.getIndex(), mapnot.char32At(pos.getIndex()));
                    errorCode=U_ZERO_ERROR;
                    continue;
                }

                logln("TestGetUnicodeSet[%d] %s", i, charset);

                cnv=cnv_open(charset, errorCode);
                if(U_FAILURE(errorCode)) {
                    errln("error opening \"%s\" for conversion/getUnicodeSet test case %d - %s",
                            charset, i, u_errorName(errorCode));
                    errorCode=U_ZERO_ERROR;
                    continue;
                }

                ucnv_getUnicodeSet(cnv, (USet *)cnvSetPtr, (UConverterUnicodeSet)which, &errorCode);
                ucnv_close(cnv);

                if(U_FAILURE(errorCode)) {
                    errln("error in ucnv_getUnicodeSet(\"%s\") for conversion/getUnicodeSet test case %d - %s",
                            charset, i, u_errorName(errorCode));
                    errorCode=U_ZERO_ERROR;
                    continue;
                }

                // are there items that must be in cnvSet but are not?
                (diffSet=mapSet).removeAll(cnvSet);
                if(!diffSet.isEmpty()) {
                    diffSet.toPattern(s, TRUE);
                    if(s.length()>100) {
                        s.replace(100, 0x7fffffff, ellipsis, LENGTHOF(ellipsis));
                    }
                    errln("error: ucnv_getUnicodeSet(\"%s\") is missing items - conversion/getUnicodeSet test case %d",
                            charset, i);
                    errln(s);
                }

                // are there items that must not be in cnvSet but are?
                (diffSet=mapnotSet).retainAll(cnvSet);
                if(!diffSet.isEmpty()) {
                    diffSet.toPattern(s, TRUE);
                    if(s.length()>100) {
                        s.replace(100, 0x7fffffff, ellipsis, LENGTHOF(ellipsis));
                    }
                    errln("error: ucnv_getUnicodeSet(\"%s\") contains unexpected items - conversion/getUnicodeSet test case %d",
                            charset, i);
                    errln(s);
                }
            }
            delete testData;
        }
        delete dataModule;
    }
    else {
        errln("Failed: could not load test conversion data");
    }
}

// open testdata or ICU data converter ------------------------------------- ***

UConverter *
ConversionTest::cnv_open(const char *name, UErrorCode &errorCode) {
    if(name!=NULL && *name=='*') {
        /* loadTestData(): set the data directory */
        return ucnv_openPackage(loadTestData(errorCode), name+1, &errorCode);
    } else {
        return ucnv_open(name, &errorCode);
    }
}

// output helpers ---------------------------------------------------------- ***

static inline char
hexDigit(uint8_t digit) {
    return digit<=9 ? (char)('0'+digit) : (char)('a'-10+digit);
}

static char *
printBytes(const uint8_t *bytes, int32_t length, char *out) {
    uint8_t b;

    if(length>0) {
        b=*bytes++;
        --length;
        *out++=hexDigit((uint8_t)(b>>4));
        *out++=hexDigit((uint8_t)(b&0xf));
    }

    while(length>0) {
        b=*bytes++;
        --length;
        *out++=' ';
        *out++=hexDigit((uint8_t)(b>>4));
        *out++=hexDigit((uint8_t)(b&0xf));
    }
    *out++=0;
    return out;
}

static char *
printUnicode(const UChar *unicode, int32_t length, char *out) {
    UChar32 c;
    int32_t i;

    for(i=0; i<length;) {
        if(i>0) {
            *out++=' ';
        }
        U16_NEXT(unicode, i, length, c);
        // write 4..6 digits
        if(c>=0x100000) {
            *out++='1';
        }
        if(c>=0x10000) {
            *out++=hexDigit((uint8_t)((c>>16)&0xf));
        }
        *out++=hexDigit((uint8_t)((c>>12)&0xf));
        *out++=hexDigit((uint8_t)((c>>8)&0xf));
        *out++=hexDigit((uint8_t)((c>>4)&0xf));
        *out++=hexDigit((uint8_t)(c&0xf));
    }
    *out++=0;
    return out;
}

static char *
printOffsets(const int32_t *offsets, int32_t length, char *out) {
    int32_t i, o, d;

    if(offsets==NULL) {
        length=0;
    }

    for(i=0; i<length; ++i) {
        if(i>0) {
            *out++=' ';
        }
        o=offsets[i];

        // print all offsets with 2 characters each (-x, -9..99, xx)
        if(o<-9) {
            *out++='-';
            *out++='x';
        } else if(o<0) {
            *out++='-';
            *out++=(char)('0'-o);
        } else if(o<=99) {
            *out++=(d=o/10)==0 ? ' ' : (char)('0'+d);
            *out++=(char)('0'+o%10);
        } else /* o>99 */ {
            *out++='x';
            *out++='x';
        }
    }
    *out++=0;
    return out;
}

// toUnicode test worker functions ----------------------------------------- ***

static int32_t
stepToUnicode(ConversionCase &cc, UConverter *cnv,
              UChar *result, int32_t resultCapacity,
              int32_t *resultOffsets, /* also resultCapacity */
              int32_t step,
              UErrorCode *pErrorCode) {
    const char *source, *sourceLimit, *bytesLimit;
    UChar *target, *targetLimit, *resultLimit;
    UBool flush;

    source=(const char *)cc.bytes;
    target=result;
    bytesLimit=source+cc.bytesLength;
    resultLimit=result+resultCapacity;

    if(step>=0) {
        // call ucnv_toUnicode() with in/out buffers no larger than (step) at a time
        // move only one buffer (in vs. out) at a time to be extra mean
        // step==0 performs bulk conversion and generates offsets

        // initialize the partial limits for the loop
        if(step==0) {
            // use the entire buffers
            sourceLimit=bytesLimit;
            targetLimit=resultLimit;
            flush=cc.finalFlush;
        } else {
            // start with empty partial buffers
            sourceLimit=source;
            targetLimit=target;
            flush=FALSE;

            // output offsets only for bulk conversion
            resultOffsets=NULL;
        }

        for(;;) {
            // resetting the opposite conversion direction must not affect this one
            ucnv_resetFromUnicode(cnv);

            // convert
            ucnv_toUnicode(cnv,
                &target, targetLimit,
                &source, sourceLimit,
                resultOffsets,
                flush, pErrorCode);

            // check pointers and errors
            if(source>sourceLimit || target>targetLimit) {
                *pErrorCode=U_INTERNAL_PROGRAM_ERROR;
                break;
            } else if(*pErrorCode==U_BUFFER_OVERFLOW_ERROR) {
                if(target!=targetLimit) {
                    // buffer overflow must only be set when the target is filled
                    *pErrorCode=U_INTERNAL_PROGRAM_ERROR;
                    break;
                } else if(targetLimit==resultLimit) {
                    // not just a partial overflow
                    break;
                }

                // the partial target is filled, set a new limit, reset the error and continue
                targetLimit=(resultLimit-target)>=step ? target+step : resultLimit;
                *pErrorCode=U_ZERO_ERROR;
            } else if(U_FAILURE(*pErrorCode)) {
                // some other error occurred, done
                break;
            } else {
                if(source!=sourceLimit) {
                    // when no error occurs, then the input must be consumed
                    *pErrorCode=U_INTERNAL_PROGRAM_ERROR;
                    break;
                }

                if(sourceLimit==bytesLimit) {
                    // we are done
                    break;
                }

                // the partial conversion succeeded, set a new limit and continue
                sourceLimit=(bytesLimit-source)>=step ? source+step : bytesLimit;
                flush=(UBool)(cc.finalFlush && sourceLimit==bytesLimit);
            }
        }
    } else /* step<0 */ {
        /*
         * step==-1: call only ucnv_getNextUChar()
         * otherwise alternate between ucnv_toUnicode() and ucnv_getNextUChar()
         *   if step==-2 or -3, then give ucnv_toUnicode() the whole remaining input,
         *   else give it at most (-step-2)/2 bytes
         */
        UChar32 c;

        // end the loop by getting an index out of bounds error
        for(;;) {
            // resetting the opposite conversion direction must not affect this one
            ucnv_resetFromUnicode(cnv);

            // convert
            if((step&1)!=0 /* odd: -1, -3, -5, ... */) {
                sourceLimit=source; // use sourceLimit not as a real limit
                                    // but to remember the pre-getNextUChar source pointer
                c=ucnv_getNextUChar(cnv, &source, bytesLimit, pErrorCode);

                // check pointers and errors
                if(*pErrorCode==U_INDEX_OUTOFBOUNDS_ERROR) {
                    if(source!=bytesLimit) {
                        *pErrorCode=U_INTERNAL_PROGRAM_ERROR;
                    } else {
                        *pErrorCode=U_ZERO_ERROR;
                    }
                    break;
                } else if(U_FAILURE(*pErrorCode)) {
                    break;
                }
                // source may not move if c is from previous overflow

                if(target==resultLimit) {
                    *pErrorCode=U_BUFFER_OVERFLOW_ERROR;
                    break;
                }
                if(c<=0xffff) {
                    *target++=(UChar)c;
                } else {
                    *target++=U16_LEAD(c);
                    if(target==resultLimit) {
                        *pErrorCode=U_BUFFER_OVERFLOW_ERROR;
                        break;
                    }
                    *target++=U16_TRAIL(c);
                }

                // alternate between -n-1 and -n but leave -1 alone
                if(step<-1) {
                    ++step;
                }
            } else /* step is even */ {
                // allow only one UChar output
                targetLimit=target<resultLimit ? target+1 : resultLimit;

                // as with ucnv_getNextUChar(), we always flush (if we go to bytesLimit)
                // and never output offsets
                if(step==-2) {
                    sourceLimit=bytesLimit;
                } else {
                    sourceLimit=source+(-step-2)/2;
                    if(sourceLimit>bytesLimit) {
                        sourceLimit=bytesLimit;
                    }
                }

                ucnv_toUnicode(cnv,
                    &target, targetLimit,
                    &source, sourceLimit,
                    NULL, (UBool)(sourceLimit==bytesLimit), pErrorCode);

                // check pointers and errors
                if(*pErrorCode==U_BUFFER_OVERFLOW_ERROR) {
                    if(target!=targetLimit) {
                        // buffer overflow must only be set when the target is filled
                        *pErrorCode=U_INTERNAL_PROGRAM_ERROR;
                        break;
                    } else if(targetLimit==resultLimit) {
                        // not just a partial overflow
                        break;
                    }

                    // the partial target is filled, set a new limit and continue
                    *pErrorCode=U_ZERO_ERROR;
                } else if(U_FAILURE(*pErrorCode)) {
                    // some other error occurred, done
                    break;
                } else {
                    if(source!=sourceLimit) {
                        // when no error occurs, then the input must be consumed
                        *pErrorCode=U_INTERNAL_PROGRAM_ERROR;
                        break;
                    }

                    // we are done (flush==TRUE) but we continue, to get the index out of bounds error above
                }

                --step;
            }
        }
    }

    return (int32_t)(target-result);
}

UBool
ConversionTest::ToUnicodeCase(ConversionCase &cc, UConverterToUCallback callback, const char *option) {
    UConverter *cnv;
    UErrorCode errorCode;

    // open the converter
    errorCode=U_ZERO_ERROR;
    cnv=cnv_open(cc.charset, errorCode);
    if(U_FAILURE(errorCode)) {
        errln("toUnicode[%d](%s cb=\"%s\" fb=%d flush=%d) ucnv_open() failed - %s",
                cc.caseNr, cc.charset, cc.cbopt, cc.fallbacks, cc.finalFlush, u_errorName(errorCode));
        return FALSE;
    }

    // set the callback
    if(callback!=NULL) {
        ucnv_setToUCallBack(cnv, callback, option, NULL, NULL, &errorCode);
        if(U_FAILURE(errorCode)) {
            errln("toUnicode[%d](%s cb=\"%s\" fb=%d flush=%d) ucnv_setToUCallBack() failed - %s",
                    cc.caseNr, cc.charset, cc.cbopt, cc.fallbacks, cc.finalFlush, u_errorName(errorCode));
            ucnv_close(cnv);
            return FALSE;
        }
    }

    int32_t resultOffsets[256];
    UChar result[256];
    int32_t resultLength;
    UBool ok;

    static const struct {
        int32_t step;
        const char *name;
    } steps[]={
        { 0, "bulk" }, // must be first for offsets to be checked
        { 1, "step=1" },
        { 3, "step=3" },
        { 7, "step=7" },
        { -1, "getNext" },
        { -2, "toU(bulk)+getNext" },
        { -3, "getNext+toU(bulk)" },
        { -4, "toU(1)+getNext" },
        { -5, "getNext+toU(1)" },
        { -12, "toU(5)+getNext" },
        { -13, "getNext+toU(5)" },
    };
    int32_t i, step;

    ok=TRUE;
    for(i=0; i<LENGTHOF(steps) && ok; ++i) {
        step=steps[i].step;
        if(step<0 && !cc.finalFlush) {
            // skip ucnv_getNextUChar() if !finalFlush because
            // ucnv_getNextUChar() always implies flush
            continue;
        }
        if(step!=0) {
            // bulk test is first, then offsets are not checked any more
            cc.offsets=NULL;
        }
        else {
            memset(resultOffsets, -1, LENGTHOF(resultOffsets));
        }
        memset(result, -1, LENGTHOF(result));
        errorCode=U_ZERO_ERROR;
        resultLength=stepToUnicode(cc, cnv,
                                result, LENGTHOF(result),
                                step==0 ? resultOffsets : NULL,
                                step, &errorCode);
        ok=checkToUnicode(
                cc, cnv, steps[i].name,
                result, resultLength,
                cc.offsets!=NULL ? resultOffsets : NULL,
                errorCode);
        if(U_FAILURE(errorCode) || !cc.finalFlush) {
            // reset if an error occurred or we did not flush
            // otherwise do nothing to make sure that flushing resets
            ucnv_resetToUnicode(cnv);
        }
        if (resultOffsets[resultLength] != -1) {
            errln("toUnicode[%d](%s) Conversion wrote too much to offsets at index %d",
                cc.caseNr, cc.charset, resultLength);
        }
        if (result[resultLength] != (UChar)-1) {
            errln("toUnicode[%d](%s) Conversion wrote too much to result at index %d",
                cc.caseNr, cc.charset, resultLength);
        }
    }

    // not a real loop, just a convenience for breaking out of the block
    while(ok && cc.finalFlush) {
        // test ucnv_toUChars()
        memset(result, 0, sizeof(result));

        errorCode=U_ZERO_ERROR;
        resultLength=ucnv_toUChars(cnv,
                        result, LENGTHOF(result),
                        (const char *)cc.bytes, cc.bytesLength,
                        &errorCode);
        ok=checkToUnicode(
                cc, cnv, "toUChars",
                result, resultLength,
                NULL,
                errorCode);
        if(!ok) {
            break;
        }

        // test preflighting
        // keep the correct result for simple checking
        errorCode=U_ZERO_ERROR;
        resultLength=ucnv_toUChars(cnv,
                        NULL, 0,
                        (const char *)cc.bytes, cc.bytesLength,
                        &errorCode);
        if(errorCode==U_STRING_NOT_TERMINATED_WARNING || errorCode==U_BUFFER_OVERFLOW_ERROR) {
            errorCode=U_ZERO_ERROR;
        }
        ok=checkToUnicode(
                cc, cnv, "preflight toUChars",
                result, resultLength,
                NULL,
                errorCode);
        break;
    }

    ucnv_close(cnv);
    return ok;
}

UBool
ConversionTest::checkToUnicode(ConversionCase &cc, UConverter *cnv, const char *name,
                               const UChar *result, int32_t resultLength,
                               const int32_t *resultOffsets,
                               UErrorCode resultErrorCode) {
    char resultInvalidChars[8];
    int8_t resultInvalidLength;
    UErrorCode errorCode;

    const char *msg;

    // reset the message; NULL will mean "ok"
    msg=NULL;

    errorCode=U_ZERO_ERROR;
    resultInvalidLength=sizeof(resultInvalidChars);
    ucnv_getInvalidChars(cnv, resultInvalidChars, &resultInvalidLength, &errorCode);
    if(U_FAILURE(errorCode)) {
        errln("toUnicode[%d](%s cb=\"%s\" fb=%d flush=%d %s) ucnv_getInvalidChars() failed - %s",
                cc.caseNr, cc.charset, cc.cbopt, cc.fallbacks, cc.finalFlush, name, u_errorName(errorCode));
        return FALSE;
    }

    // check everything that might have gone wrong
    if(cc.unicodeLength!=resultLength) {
        msg="wrong result length";
    } else if(0!=u_memcmp(cc.unicode, result, cc.unicodeLength)) {
        msg="wrong result string";
    } else if(cc.offsets!=NULL && 0!=memcmp(cc.offsets, resultOffsets, cc.unicodeLength*sizeof(*cc.offsets))) {
        msg="wrong offsets";
    } else if(cc.outErrorCode!=resultErrorCode) {
        msg="wrong error code";
    } else if(cc.invalidLength!=resultInvalidLength) {
        msg="wrong length of last invalid input";
    } else if(0!=memcmp(cc.invalidChars, resultInvalidChars, cc.invalidLength)) {
        msg="wrong last invalid input";
    }

    if(msg==NULL) {
        return TRUE;
    } else {
        char buffer[2000]; // one buffer for all strings
        char *s, *bytesString, *unicodeString, *resultString,
            *offsetsString, *resultOffsetsString,
            *invalidCharsString, *resultInvalidCharsString;

        bytesString=s=buffer;
        s=printBytes(cc.bytes, cc.bytesLength, bytesString);
        s=printUnicode(cc.unicode, cc.unicodeLength, unicodeString=s);
        s=printUnicode(result, resultLength, resultString=s);
        s=printOffsets(cc.offsets, cc.unicodeLength, offsetsString=s);
        s=printOffsets(resultOffsets, resultLength, resultOffsetsString=s);
        s=printBytes(cc.invalidChars, cc.invalidLength, invalidCharsString=s);
        s=printBytes((uint8_t *)resultInvalidChars, resultInvalidLength, resultInvalidCharsString=s);

        if((s-buffer)>(int32_t)sizeof(buffer)) {
            errln("toUnicode[%d](%s cb=\"%s\" fb=%d flush=%d %s) fatal error: checkToUnicode() test output buffer overflow writing %d chars\n",
                    cc.caseNr, cc.charset, cc.cbopt, cc.fallbacks, cc.finalFlush, name, (int)(s-buffer));
            exit(1);
        }

        errln("toUnicode[%d](%s cb=\"%s\" fb=%d flush=%d %s) failed: %s\n"
              "  bytes <%s>[%d]\n"
              " expected <%s>[%d]\n"
              "  result  <%s>[%d]\n"
              " offsets         <%s>\n"
              "  result offsets <%s>\n"
              " error code expected %s got %s\n"
              "  invalidChars expected <%s> got <%s>\n",
              cc.caseNr, cc.charset, cc.cbopt, cc.fallbacks, cc.finalFlush, name, msg,
              bytesString, cc.bytesLength,
              unicodeString, cc.unicodeLength,
              resultString, resultLength,
              offsetsString,
              resultOffsetsString,
              u_errorName(cc.outErrorCode), u_errorName(resultErrorCode),
              invalidCharsString, resultInvalidCharsString);

        return FALSE;
    }
}

// fromUnicode test worker functions --------------------------------------- ***

static int32_t
stepFromUTF8(ConversionCase &cc,
             UConverter *utf8Cnv, UConverter *cnv,
             char *result, int32_t resultCapacity,
             int32_t step,
             UErrorCode *pErrorCode) {
    const char *source, *sourceLimit, *utf8Limit;
    UChar pivotBuffer[32];
    UChar *pivotSource, *pivotTarget, *pivotLimit;
    char *target, *targetLimit, *resultLimit;
    UBool flush;

    source=cc.utf8;
    pivotSource=pivotTarget=pivotBuffer;
    target=result;
    utf8Limit=source+cc.utf8Length;
    resultLimit=result+resultCapacity;

    // call ucnv_convertEx() with in/out buffers no larger than (step) at a time
    // move only one buffer (in vs. out) at a time to be extra mean
    // step==0 performs bulk conversion

    // initialize the partial limits for the loop
    if(step==0) {
        // use the entire buffers
        sourceLimit=utf8Limit;
        targetLimit=resultLimit;
        flush=cc.finalFlush;

        pivotLimit=pivotBuffer+LENGTHOF(pivotBuffer);
    } else {
        // start with empty partial buffers
        sourceLimit=source;
        targetLimit=target;
        flush=FALSE;

        // empty pivot is not allowed, make it of length step
        pivotLimit=pivotBuffer+step;
    }

    for(;;) {
        // resetting the opposite conversion direction must not affect this one
        ucnv_resetFromUnicode(utf8Cnv);
        ucnv_resetToUnicode(cnv);

        // convert
        ucnv_convertEx(cnv, utf8Cnv,
            &target, targetLimit,
            &source, sourceLimit,
            pivotBuffer, &pivotSource, &pivotTarget, pivotLimit,
            FALSE, flush, pErrorCode);

        // check pointers and errors
        if(source>sourceLimit || target>targetLimit) {
            *pErrorCode=U_INTERNAL_PROGRAM_ERROR;
            break;
        } else if(*pErrorCode==U_BUFFER_OVERFLOW_ERROR) {
            if(target!=targetLimit) {
                // buffer overflow must only be set when the target is filled
                *pErrorCode=U_INTERNAL_PROGRAM_ERROR;
                break;
            } else if(targetLimit==resultLimit) {
                // not just a partial overflow
                break;
            }

            // the partial target is filled, set a new limit, reset the error and continue
            targetLimit=(resultLimit-target)>=step ? target+step : resultLimit;
            *pErrorCode=U_ZERO_ERROR;
        } else if(U_FAILURE(*pErrorCode)) {
            if(pivotSource==pivotBuffer) {
                // toUnicode error, should not occur
                // toUnicode errors are tested in cintltst TestConvertExFromUTF8()
                break;
            } else {
                // fromUnicode error
                // some other error occurred, done
                break;
            }
        } else {
            if(source!=sourceLimit) {
                // when no error occurs, then the input must be consumed
                *pErrorCode=U_INTERNAL_PROGRAM_ERROR;
                break;
            }

            if(sourceLimit==utf8Limit) {
                // we are done
                if(*pErrorCode==U_STRING_NOT_TERMINATED_WARNING) {
                    // ucnv_convertEx() warns about not terminating the output
                    // but ucnv_fromUnicode() does not and so
                    // checkFromUnicode() does not expect it
                    *pErrorCode=U_ZERO_ERROR;
                }
                break;
            }

            // the partial conversion succeeded, set a new limit and continue
            sourceLimit=(utf8Limit-source)>=step ? source+step : utf8Limit;
            flush=(UBool)(cc.finalFlush && sourceLimit==utf8Limit);
        }
    }

    return (int32_t)(target-result);
}

static int32_t
stepFromUnicode(ConversionCase &cc, UConverter *cnv,
                char *result, int32_t resultCapacity,
                int32_t *resultOffsets, /* also resultCapacity */
                int32_t step,
                UErrorCode *pErrorCode) {
    const UChar *source, *sourceLimit, *unicodeLimit;
    char *target, *targetLimit, *resultLimit;
    UBool flush;

    source=cc.unicode;
    target=result;
    unicodeLimit=source+cc.unicodeLength;
    resultLimit=result+resultCapacity;

    // call ucnv_fromUnicode() with in/out buffers no larger than (step) at a time
    // move only one buffer (in vs. out) at a time to be extra mean
    // step==0 performs bulk conversion and generates offsets

    // initialize the partial limits for the loop
    if(step==0) {
        // use the entire buffers
        sourceLimit=unicodeLimit;
        targetLimit=resultLimit;
        flush=cc.finalFlush;
    } else {
        // start with empty partial buffers
        sourceLimit=source;
        targetLimit=target;
        flush=FALSE;

        // output offsets only for bulk conversion
        resultOffsets=NULL;
    }

    for(;;) {
        // resetting the opposite conversion direction must not affect this one
        ucnv_resetToUnicode(cnv);

        // convert
        ucnv_fromUnicode(cnv,
            &target, targetLimit,
            &source, sourceLimit,
            resultOffsets,
            flush, pErrorCode);

        // check pointers and errors
        if(source>sourceLimit || target>targetLimit) {
            *pErrorCode=U_INTERNAL_PROGRAM_ERROR;
            break;
        } else if(*pErrorCode==U_BUFFER_OVERFLOW_ERROR) {
            if(target!=targetLimit) {
                // buffer overflow must only be set when the target is filled
                *pErrorCode=U_INTERNAL_PROGRAM_ERROR;
                break;
            } else if(targetLimit==resultLimit) {
                // not just a partial overflow
                break;
            }

            // the partial target is filled, set a new limit, reset the error and continue
            targetLimit=(resultLimit-target)>=step ? target+step : resultLimit;
            *pErrorCode=U_ZERO_ERROR;
        } else if(U_FAILURE(*pErrorCode)) {
            // some other error occurred, done
            break;
        } else {
            if(source!=sourceLimit) {
                // when no error occurs, then the input must be consumed
                *pErrorCode=U_INTERNAL_PROGRAM_ERROR;
                break;
            }

            if(sourceLimit==unicodeLimit) {
                // we are done
                break;
            }

            // the partial conversion succeeded, set a new limit and continue
            sourceLimit=(unicodeLimit-source)>=step ? source+step : unicodeLimit;
            flush=(UBool)(cc.finalFlush && sourceLimit==unicodeLimit);
        }
    }

    return (int32_t)(target-result);
}

UBool
ConversionTest::FromUnicodeCase(ConversionCase &cc, UConverterFromUCallback callback, const char *option) {
    UConverter *cnv;
    UErrorCode errorCode;

    // open the converter
    errorCode=U_ZERO_ERROR;
    cnv=cnv_open(cc.charset, errorCode);
    if(U_FAILURE(errorCode)) {
        errln("fromUnicode[%d](%s cb=\"%s\" fb=%d flush=%d) ucnv_open() failed - %s",
                cc.caseNr, cc.charset, cc.cbopt, cc.fallbacks, cc.finalFlush, u_errorName(errorCode));
        return FALSE;
    }
    ucnv_resetToUnicode(utf8Cnv);

    // set the callback
    if(callback!=NULL) {
        ucnv_setFromUCallBack(cnv, callback, option, NULL, NULL, &errorCode);
        if(U_FAILURE(errorCode)) {
            errln("fromUnicode[%d](%s cb=\"%s\" fb=%d flush=%d) ucnv_setFromUCallBack() failed - %s",
                    cc.caseNr, cc.charset, cc.cbopt, cc.fallbacks, cc.finalFlush, u_errorName(errorCode));
            ucnv_close(cnv);
            return FALSE;
        }
    }

    // set the fallbacks flag
    // TODO change with Jitterbug 2401, then add a similar call for toUnicode too
    ucnv_setFallback(cnv, cc.fallbacks);

    // set the subchar
    int32_t length;

    if(cc.setSub>0) {
        length=(int32_t)strlen(cc.subchar);
        ucnv_setSubstChars(cnv, cc.subchar, (int8_t)length, &errorCode);
        if(U_FAILURE(errorCode)) {
            errln("fromUnicode[%d](%s cb=\"%s\" fb=%d flush=%d) ucnv_setSubstChars() failed - %s",
                    cc.caseNr, cc.charset, cc.cbopt, cc.fallbacks, cc.finalFlush, u_errorName(errorCode));
            ucnv_close(cnv);
            return FALSE;
        }
    } else if(cc.setSub<0) {
        ucnv_setSubstString(cnv, cc.subString, -1, &errorCode);
        if(U_FAILURE(errorCode)) {
            errln("fromUnicode[%d](%s cb=\"%s\" fb=%d flush=%d) ucnv_setSubstString() failed - %s",
                    cc.caseNr, cc.charset, cc.cbopt, cc.fallbacks, cc.finalFlush, u_errorName(errorCode));
            ucnv_close(cnv);
            return FALSE;
        }
    }

    // convert unicode to utf8
    char utf8[256];
    cc.utf8=utf8;
    u_strToUTF8(utf8, LENGTHOF(utf8), &cc.utf8Length,
                cc.unicode, cc.unicodeLength,
                &errorCode);
    if(U_FAILURE(errorCode)) {
        // skip UTF-8 testing of a string with an unpaired surrogate,
        // or of one that's too long
        // toUnicode errors are tested in cintltst TestConvertExFromUTF8()
        cc.utf8Length=-1;
    }

    int32_t resultOffsets[256];
    char result[256];
    int32_t resultLength;
    UBool ok;

    static const struct {
        int32_t step;
        const char *name, *utf8Name;
    } steps[]={
        { 0, "bulk",   "utf8" }, // must be first for offsets to be checked
        { 1, "step=1", "utf8 step=1" },
        { 3, "step=3", "utf8 step=3" },
        { 7, "step=7", "utf8 step=7" }
    };
    int32_t i, step;

    ok=TRUE;
    for(i=0; i<LENGTHOF(steps) && ok; ++i) {
        step=steps[i].step;
        memset(resultOffsets, -1, LENGTHOF(resultOffsets));
        memset(result, -1, LENGTHOF(result));
        errorCode=U_ZERO_ERROR;
        resultLength=stepFromUnicode(cc, cnv,
                                result, LENGTHOF(result),
                                step==0 ? resultOffsets : NULL,
                                step, &errorCode);
        ok=checkFromUnicode(
                cc, cnv, steps[i].name,
                (uint8_t *)result, resultLength,
                cc.offsets!=NULL ? resultOffsets : NULL,
                errorCode);
        if(U_FAILURE(errorCode) || !cc.finalFlush) {
            // reset if an error occurred or we did not flush
            // otherwise do nothing to make sure that flushing resets
            ucnv_resetFromUnicode(cnv);
        }
        if (resultOffsets[resultLength] != -1) {
            errln("fromUnicode[%d](%s) Conversion wrote too much to offsets at index %d",
                cc.caseNr, cc.charset, resultLength);
        }
        if (result[resultLength] != (char)-1) {
            errln("fromUnicode[%d](%s) Conversion wrote too much to result at index %d",
                cc.caseNr, cc.charset, resultLength);
        }

        // bulk test is first, then offsets are not checked any more
        cc.offsets=NULL;

        // test direct conversion from UTF-8
        if(cc.utf8Length>=0) {
            errorCode=U_ZERO_ERROR;
            resultLength=stepFromUTF8(cc, utf8Cnv, cnv,
                                    result, LENGTHOF(result),
                                    step, &errorCode);
            ok=checkFromUnicode(
                    cc, cnv, steps[i].utf8Name,
                    (uint8_t *)result, resultLength,
                    NULL,
                    errorCode);
            if(U_FAILURE(errorCode) || !cc.finalFlush) {
                // reset if an error occurred or we did not flush
                // otherwise do nothing to make sure that flushing resets
                ucnv_resetToUnicode(utf8Cnv);
                ucnv_resetFromUnicode(cnv);
            }
        }
    }

    // not a real loop, just a convenience for breaking out of the block
    while(ok && cc.finalFlush) {
        // test ucnv_fromUChars()
        memset(result, 0, sizeof(result));

        errorCode=U_ZERO_ERROR;
        resultLength=ucnv_fromUChars(cnv,
                        result, LENGTHOF(result),
                        cc.unicode, cc.unicodeLength,
                        &errorCode);
        ok=checkFromUnicode(
                cc, cnv, "fromUChars",
                (uint8_t *)result, resultLength,
                NULL,
                errorCode);
        if(!ok) {
            break;
        }

        // test preflighting
        // keep the correct result for simple checking
        errorCode=U_ZERO_ERROR;
        resultLength=ucnv_fromUChars(cnv,
                        NULL, 0,
                        cc.unicode, cc.unicodeLength,
                        &errorCode);
        if(errorCode==U_STRING_NOT_TERMINATED_WARNING || errorCode==U_BUFFER_OVERFLOW_ERROR) {
            errorCode=U_ZERO_ERROR;
        }
        ok=checkFromUnicode(
                cc, cnv, "preflight fromUChars",
                (uint8_t *)result, resultLength,
                NULL,
                errorCode);
        break;
    }

    ucnv_close(cnv);
    return ok;
}

UBool
ConversionTest::checkFromUnicode(ConversionCase &cc, UConverter *cnv, const char *name,
                                 const uint8_t *result, int32_t resultLength,
                                 const int32_t *resultOffsets,
                                 UErrorCode resultErrorCode) {
    UChar resultInvalidUChars[8];
    int8_t resultInvalidLength;
    UErrorCode errorCode;

    const char *msg;

    // reset the message; NULL will mean "ok"
    msg=NULL;

    errorCode=U_ZERO_ERROR;
    resultInvalidLength=LENGTHOF(resultInvalidUChars);
    ucnv_getInvalidUChars(cnv, resultInvalidUChars, &resultInvalidLength, &errorCode);
    if(U_FAILURE(errorCode)) {
        errln("fromUnicode[%d](%s cb=\"%s\" fb=%d flush=%d %s) ucnv_getInvalidUChars() failed - %s",
                cc.caseNr, cc.charset, cc.cbopt, cc.fallbacks, cc.finalFlush, name, u_errorName(errorCode));
        return FALSE;
    }

    // check everything that might have gone wrong
    if(cc.bytesLength!=resultLength) {
        msg="wrong result length";
    } else if(0!=memcmp(cc.bytes, result, cc.bytesLength)) {
        msg="wrong result string";
    } else if(cc.offsets!=NULL && 0!=memcmp(cc.offsets, resultOffsets, cc.bytesLength*sizeof(*cc.offsets))) {
        msg="wrong offsets";
    } else if(cc.outErrorCode!=resultErrorCode) {
        msg="wrong error code";
    } else if(cc.invalidLength!=resultInvalidLength) {
        msg="wrong length of last invalid input";
    } else if(0!=u_memcmp(cc.invalidUChars, resultInvalidUChars, cc.invalidLength)) {
        msg="wrong last invalid input";
    }

    if(msg==NULL) {
        return TRUE;
    } else {
        char buffer[2000]; // one buffer for all strings
        char *s, *unicodeString, *bytesString, *resultString,
            *offsetsString, *resultOffsetsString,
            *invalidCharsString, *resultInvalidUCharsString;

        unicodeString=s=buffer;
        s=printUnicode(cc.unicode, cc.unicodeLength, unicodeString);
        s=printBytes(cc.bytes, cc.bytesLength, bytesString=s);
        s=printBytes(result, resultLength, resultString=s);
        s=printOffsets(cc.offsets, cc.bytesLength, offsetsString=s);
        s=printOffsets(resultOffsets, resultLength, resultOffsetsString=s);
        s=printUnicode(cc.invalidUChars, cc.invalidLength, invalidCharsString=s);
        s=printUnicode(resultInvalidUChars, resultInvalidLength, resultInvalidUCharsString=s);

        if((s-buffer)>(int32_t)sizeof(buffer)) {
            errln("fromUnicode[%d](%s cb=\"%s\" fb=%d flush=%d %s) fatal error: checkFromUnicode() test output buffer overflow writing %d chars\n",
                    cc.caseNr, cc.charset, cc.cbopt, cc.fallbacks, cc.finalFlush, name, (int)(s-buffer));
            exit(1);
        }

        errln("fromUnicode[%d](%s cb=\"%s\" fb=%d flush=%d %s) failed: %s\n"
              "  unicode <%s>[%d]\n"
              " expected <%s>[%d]\n"
              "  result  <%s>[%d]\n"
              " offsets         <%s>\n"
              "  result offsets <%s>\n"
              " error code expected %s got %s\n"
              "  invalidChars expected <%s> got <%s>\n",
              cc.caseNr, cc.charset, cc.cbopt, cc.fallbacks, cc.finalFlush, name, msg,
              unicodeString, cc.unicodeLength,
              bytesString, cc.bytesLength,
              resultString, resultLength,
              offsetsString,
              resultOffsetsString,
              u_errorName(cc.outErrorCode), u_errorName(resultErrorCode),
              invalidCharsString, resultInvalidUCharsString);

        return FALSE;
    }
}

#endif /* #if !UCONFIG_NO_LEGACY_CONVERSION */