summaryrefslogtreecommitdiff
path: root/Build/source/libs/icu/icu-src/source/test/intltest/rbbitst.cpp
blob: 73d4393843be9227cb05573d1a74fe13cf7f10db (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
3972
3973
3974
3975
3976
3977
3978
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
4114
4115
4116
4117
4118
4119
4120
4121
4122
4123
4124
4125
4126
4127
4128
4129
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4148
4149
4150
4151
4152
4153
4154
4155
4156
4157
4158
4159
4160
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
4176
4177
4178
4179
4180
4181
4182
4183
4184
4185
4186
4187
4188
4189
4190
4191
4192
4193
4194
4195
4196
4197
4198
4199
4200
4201
4202
4203
4204
4205
4206
4207
4208
4209
4210
4211
4212
4213
4214
4215
4216
4217
4218
4219
4220
4221
4222
4223
4224
4225
4226
4227
4228
4229
4230
4231
4232
4233
4234
4235
4236
4237
4238
4239
4240
4241
4242
4243
4244
4245
4246
4247
4248
4249
4250
4251
4252
4253
4254
4255
4256
4257
4258
4259
4260
4261
4262
4263
4264
4265
4266
4267
4268
4269
4270
4271
4272
4273
4274
4275
4276
4277
4278
4279
4280
4281
4282
4283
4284
4285
4286
4287
4288
4289
4290
4291
4292
4293
4294
4295
4296
4297
4298
4299
4300
4301
4302
4303
4304
4305
4306
4307
4308
4309
4310
4311
4312
4313
4314
4315
4316
4317
4318
4319
4320
4321
4322
4323
4324
4325
4326
4327
4328
4329
4330
4331
4332
4333
4334
4335
4336
4337
4338
4339
4340
4341
4342
4343
4344
4345
4346
4347
4348
4349
4350
4351
4352
4353
4354
4355
4356
4357
4358
4359
4360
4361
4362
4363
4364
4365
4366
4367
4368
4369
4370
4371
4372
4373
4374
4375
4376
4377
4378
4379
4380
4381
4382
4383
4384
4385
4386
4387
4388
4389
4390
4391
4392
4393
4394
4395
4396
4397
4398
4399
4400
4401
4402
4403
4404
4405
4406
4407
4408
4409
4410
4411
4412
4413
4414
4415
4416
4417
4418
4419
4420
4421
4422
4423
4424
4425
4426
4427
4428
4429
4430
4431
4432
4433
4434
4435
4436
4437
4438
4439
4440
4441
4442
4443
4444
4445
4446
4447
4448
4449
4450
4451
4452
4453
4454
4455
4456
4457
4458
4459
4460
4461
4462
4463
4464
4465
4466
4467
4468
4469
4470
4471
4472
4473
4474
4475
4476
4477
4478
4479
4480
4481
4482
4483
4484
4485
4486
4487
4488
4489
4490
4491
4492
4493
4494
4495
4496
4497
4498
4499
4500
4501
4502
4503
4504
4505
4506
4507
4508
4509
4510
4511
4512
4513
4514
4515
4516
4517
4518
4519
4520
4521
4522
4523
4524
4525
4526
4527
4528
4529
4530
4531
4532
4533
4534
4535
4536
4537
4538
4539
4540
4541
4542
4543
4544
4545
4546
4547
4548
4549
4550
4551
4552
4553
4554
4555
4556
4557
4558
4559
4560
4561
4562
4563
4564
4565
4566
4567
4568
4569
4570
4571
4572
4573
4574
4575
4576
4577
4578
4579
4580
4581
4582
4583
4584
4585
4586
4587
4588
4589
4590
4591
4592
4593
4594
4595
4596
4597
4598
4599
4600
4601
4602
4603
4604
4605
4606
4607
4608
4609
4610
4611
4612
4613
4614
4615
4616
4617
4618
4619
4620
4621
4622
4623
4624
4625
4626
4627
4628
4629
4630
4631
4632
4633
4634
4635
4636
4637
4638
4639
4640
4641
4642
4643
4644
4645
4646
4647
4648
4649
4650
4651
4652
4653
4654
4655
4656
4657
4658
4659
4660
4661
4662
4663
4664
4665
4666
4667
4668
4669
4670
4671
4672
4673
4674
4675
4676
4677
4678
4679
4680
4681
4682
4683
4684
4685
4686
4687
4688
4689
4690
4691
4692
4693
4694
4695
4696
4697
4698
4699
4700
4701
4702
4703
4704
4705
4706
4707
4708
4709
4710
4711
4712
4713
4714
4715
4716
4717
4718
4719
4720
4721
4722
4723
4724
4725
4726
4727
4728
4729
4730
4731
4732
4733
4734
4735
4736
4737
4738
4739
4740
4741
4742
4743
4744
4745
4746
4747
4748
4749
4750
4751
4752
4753
4754
4755
4756
4757
4758
4759
4760
4761
4762
4763
4764
4765
4766
4767
4768
4769
4770
4771
4772
4773
4774
4775
4776
4777
4778
4779
4780
4781
4782
4783
4784
4785
4786
4787
4788
4789
4790
4791
4792
4793
4794
4795
4796
4797
4798
4799
4800
4801
4802
4803
4804
4805
4806
4807
4808
4809
4810
4811
4812
4813
4814
4815
4816
4817
4818
4819
4820
4821
4822
4823
4824
4825
4826
4827
4828
4829
4830
4831
4832
4833
4834
4835
4836
4837
4838
4839
4840
4841
4842
4843
4844
4845
4846
4847
4848
4849
4850
4851
4852
4853
4854
4855
4856
4857
4858
4859
4860
4861
4862
4863
4864
4865
4866
4867
4868
4869
4870
4871
4872
4873
4874
4875
4876
4877
4878
4879
4880
4881
4882
4883
4884
4885
4886
4887
4888
4889
4890
4891
4892
4893
4894
4895
4896
4897
4898
4899
4900
4901
4902
4903
4904
4905
4906
4907
4908
4909
4910
4911
4912
4913
4914
4915
4916
4917
4918
4919
4920
4921
4922
4923
4924
4925
4926
4927
4928
4929
4930
4931
4932
4933
4934
4935
4936
4937
4938
4939
4940
4941
4942
4943
4944
4945
4946
4947
4948
4949
4950
4951
4952
4953
4954
4955
4956
4957
4958
4959
4960
4961
4962
4963
4964
4965
4966
4967
4968
4969
4970
4971
4972
4973
4974
4975
4976
4977
4978
4979
4980
4981
4982
4983
4984
4985
4986
4987
4988
4989
4990
4991
4992
4993
4994
4995
4996
4997
4998
4999
5000
5001
5002
5003
5004
5005
5006
5007
5008
5009
5010
5011
5012
5013
5014
5015
5016
5017
5018
5019
5020
5021
5022
5023
5024
5025
5026
5027
5028
5029
5030
5031
5032
5033
5034
5035
5036
5037
5038
5039
5040
5041
5042
5043
5044
5045
5046
5047
5048
5049
5050
5051
5052
5053
5054
5055
5056
5057
5058
5059
5060
5061
5062
5063
5064
5065
5066
5067
5068
5069
5070
5071
5072
5073
5074
5075
5076
5077
5078
5079
5080
5081
5082
5083
5084
5085
5086
5087
5088
5089
5090
5091
5092
5093
5094
5095
5096
5097
5098
5099
5100
5101
5102
5103
5104
5105
5106
5107
5108
5109
5110
5111
5112
5113
5114
5115
5116
5117
5118
5119
5120
5121
5122
5123
5124
5125
5126
5127
5128
5129
5130
5131
5132
5133
5134
5135
5136
5137
5138
5139
5140
5141
5142
5143
5144
5145
5146
5147
5148
5149
5150
5151
5152
5153
5154
5155
5156
5157
5158
5159
5160
5161
5162
5163
5164
5165
5166
5167
5168
5169
5170
5171
5172
5173
5174
5175
5176
5177
5178
5179
5180
5181
5182
5183
5184
5185
5186
5187
5188
5189
5190
5191
5192
5193
5194
5195
5196
5197
5198
5199
5200
5201
5202
5203
5204
5205
5206
5207
5208
5209
5210
5211
5212
5213
5214
5215
5216
5217
5218
5219
5220
5221
5222
5223
5224
5225
5226
5227
5228
5229
5230
5231
5232
5233
5234
5235
5236
5237
5238
5239
5240
5241
5242
5243
5244
5245
5246
5247
5248
5249
5250
5251
5252
5253
5254
5255
5256
5257
5258
5259
5260
5261
5262
5263
5264
5265
5266
5267
5268
5269
5270
5271
5272
5273
5274
5275
5276
5277
5278
5279
5280
5281
5282
5283
5284
5285
5286
5287
5288
5289
5290
5291
5292
5293
5294
5295
5296
5297
5298
5299
5300
5301
5302
5303
5304
5305
5306
5307
5308
5309
5310
5311
5312
5313
5314
5315
5316
5317
5318
5319
5320
5321
5322
5323
5324
5325
5326
5327
5328
5329
5330
5331
5332
5333
5334
5335
5336
5337
5338
5339
5340
5341
5342
5343
5344
5345
5346
5347
5348
5349
5350
5351
5352
5353
5354
5355
5356
5357
5358
5359
5360
5361
5362
5363
5364
5365
5366
5367
5368
5369
5370
5371
5372
5373
5374
5375
5376
5377
5378
5379
5380
5381
5382
5383
5384
5385
5386
5387
5388
5389
5390
5391
5392
5393
5394
5395
5396
5397
5398
5399
5400
5401
5402
5403
5404
5405
5406
5407
5408
5409
5410
5411
5412
5413
5414
5415
5416
// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/********************************************************************
 * COPYRIGHT:
 * Copyright (c) 1999-2016, International Business Machines Corporation and
 * others. All Rights Reserved.
 ********************************************************************/
/************************************************************************
*   Date        Name        Description
*   12/15/99    Madhu        Creation.
*   01/12/2000  Madhu        Updated for changed API and added new tests
************************************************************************/

#include "unicode/utypes.h"
#if !UCONFIG_NO_BREAK_ITERATION

#include <algorithm>
#include <sstream>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <utility>
#include <vector>

#include "unicode/brkiter.h"
#include "unicode/localpointer.h"
#include "unicode/numfmt.h"
#include "unicode/rbbi.h"
#if !UCONFIG_NO_REGULAR_EXPRESSIONS
#include "unicode/regex.h"
#endif
#include "unicode/schriter.h"
#include "unicode/uchar.h"
#include "unicode/utf16.h"
#include "unicode/ucnv.h"
#include "unicode/uniset.h"
#include "unicode/uscript.h"
#include "unicode/ustring.h"
#include "unicode/utext.h"
#include "unicode/utrace.h"

#include "charstr.h"
#include "cmemory.h"
#include "cstr.h"
#include "intltest.h"
#include "lstmbe.h"
#include "rbbitst.h"
#include "rbbidata.h"
#include "utypeinfo.h"  // for 'typeid' to work
#include "uvector.h"
#include "uvectr32.h"


#if !UCONFIG_NO_FILTERED_BREAK_ITERATION
#include "unicode/filteredbrk.h"
#endif // !UCONFIG_NO_FILTERED_BREAK_ITERATION

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

#define TEST_ASSERT_SUCCESS(errcode) UPRV_BLOCK_MACRO_BEGIN { \
    if (U_FAILURE(errcode)) { \
        errcheckln(errcode, "Failure in file %s, line %d, status = \"%s\"", __FILE__, __LINE__, u_errorName(errcode)); \
    } \
} UPRV_BLOCK_MACRO_END

#define MONKEY_ERROR(msg, fRuleFileName, index, seed) { \
    IntlTest::gTest->errln("\n%s:%d %s at index %d. Parameters to reproduce: @\"type=%s seed=%u loop=1\"", \
                    __FILE__, __LINE__, msg, index, fRuleFileName, seed); \
}

//---------------------------------------------
// runIndexedTest
//---------------------------------------------


//  Note:  Before adding new tests to this file, check whether the desired test data can
//         simply be added to the file testdata/rbbitest.txt.  In most cases it can,
//         it's much less work than writing a new test, diagnostic output in the event of failures
//         is good, and the test data file will is shared with ICU4J, so eventually the test
//         will run there as well, without additional effort.

void RBBITest::runIndexedTest( int32_t index, UBool exec, const char* &name, char* params )
{
    if (exec) logln("TestSuite RuleBasedBreakIterator: ");
    fTestParams = params;

    TESTCASE_AUTO_BEGIN;
#if !UCONFIG_NO_FILE_IO
    TESTCASE_AUTO(TestBug4153072);
#endif
#if !UCONFIG_NO_FILE_IO
    TESTCASE_AUTO(TestUnicodeFiles);
#endif
    TESTCASE_AUTO(TestGetAvailableLocales);
    TESTCASE_AUTO(TestGetDisplayName);
#if !UCONFIG_NO_FILE_IO
    TESTCASE_AUTO(TestEndBehaviour);
    TESTCASE_AUTO(TestWordBreaks);
    TESTCASE_AUTO(TestWordBoundary);
    TESTCASE_AUTO(TestLineBreaks);
    TESTCASE_AUTO(TestSentBreaks);
    TESTCASE_AUTO(TestExtended);
#endif
#if !UCONFIG_NO_REGULAR_EXPRESSIONS && !UCONFIG_NO_FILE_IO
    TESTCASE_AUTO(TestMonkey);
#endif
#if !UCONFIG_NO_FILE_IO
    TESTCASE_AUTO(TestBug3818);
#endif
    TESTCASE_AUTO(TestDebug);
#if !UCONFIG_NO_FILE_IO
    TESTCASE_AUTO(TestBug5775);
#endif
    TESTCASE_AUTO(TestBug9983);
    TESTCASE_AUTO(TestDictRules);
    TESTCASE_AUTO(TestBug5532);
    TESTCASE_AUTO(TestBug7547);
    TESTCASE_AUTO(TestBug12797);
    TESTCASE_AUTO(TestBug12918);
    TESTCASE_AUTO(TestBug12932);
    TESTCASE_AUTO(TestEmoji);
    TESTCASE_AUTO(TestBug12519);
    TESTCASE_AUTO(TestBug12677);
    TESTCASE_AUTO(TestTableRedundancies);
    TESTCASE_AUTO(TestBug13447);
    TESTCASE_AUTO(TestReverse);
    TESTCASE_AUTO(TestBug13692);
    TESTCASE_AUTO(TestDebugRules);
    TESTCASE_AUTO(Test8BitsTrieWith8BitStateTable);
    TESTCASE_AUTO(Test8BitsTrieWith16BitStateTable);
    TESTCASE_AUTO(Test16BitsTrieWith8BitStateTable);
    TESTCASE_AUTO(Test16BitsTrieWith16BitStateTable);
    TESTCASE_AUTO(TestTable_8_16_Bits);
    TESTCASE_AUTO(TestBug13590);
    TESTCASE_AUTO(TestUnpairedSurrogate);
    TESTCASE_AUTO(TestLSTMThai);
    TESTCASE_AUTO(TestLSTMBurmese);

#if U_ENABLE_TRACING
    TESTCASE_AUTO(TestTraceCreateCharacter);
    TESTCASE_AUTO(TestTraceCreateWord);
    TESTCASE_AUTO(TestTraceCreateSentence);
    TESTCASE_AUTO(TestTraceCreateTitle);
    TESTCASE_AUTO(TestTraceCreateLine);
    TESTCASE_AUTO(TestTraceCreateLineNormal);
    TESTCASE_AUTO(TestTraceCreateLineLoose);
    TESTCASE_AUTO(TestTraceCreateLineStrict);
    TESTCASE_AUTO(TestTraceCreateBreakEngine);
#endif

    TESTCASE_AUTO_END;
}


//--------------------------------------------------------------------------------------
//
//    RBBITest    constructor and destructor
//
//--------------------------------------------------------------------------------------

RBBITest::RBBITest() {
    fTestParams = NULL;
}


RBBITest::~RBBITest() {
}


static void printStringBreaks(UText *tstr, int expected[], int expectedCount) {
    UErrorCode status = U_ZERO_ERROR;
    char name[100];
    printf("code    alpha extend alphanum type word sent line name\n");
    int nextExpectedIndex = 0;
    utext_setNativeIndex(tstr, 0);
    for (int j = 0; j < static_cast<int>(utext_nativeLength(tstr)); j=static_cast<int>(utext_getNativeIndex(tstr))) {
        if (nextExpectedIndex < expectedCount && j >= expected[nextExpectedIndex] ) {
            printf("------------------------------------------------ %d\n", j);
            ++nextExpectedIndex;
        }

        UChar32 c = utext_next32(tstr);
        u_charName(c, U_UNICODE_CHAR_NAME, name, 100, &status);
        printf("%7x %5d %6d %8d %4s %4s %4s %4s %s\n", (int)c,
                           u_isUAlphabetic(c),
                           u_hasBinaryProperty(c, UCHAR_GRAPHEME_EXTEND),
                           u_isalnum(c),
                           u_getPropertyValueName(UCHAR_GENERAL_CATEGORY,
                                                  u_charType(c),
                                                  U_SHORT_PROPERTY_NAME),
                           u_getPropertyValueName(UCHAR_WORD_BREAK,
                                                  u_getIntPropertyValue(c,
                                                          UCHAR_WORD_BREAK),
                                                  U_SHORT_PROPERTY_NAME),
                           u_getPropertyValueName(UCHAR_SENTENCE_BREAK,
                                   u_getIntPropertyValue(c,
                                           UCHAR_SENTENCE_BREAK),
                                   U_SHORT_PROPERTY_NAME),
                           u_getPropertyValueName(UCHAR_LINE_BREAK,
                                   u_getIntPropertyValue(c,
                                           UCHAR_LINE_BREAK),
                                   U_SHORT_PROPERTY_NAME),
                           name);
    }
}


static void printStringBreaks(const UnicodeString &ustr, int expected[], int expectedCount) {
   UErrorCode status = U_ZERO_ERROR;
   UText *tstr = NULL;
   tstr = utext_openConstUnicodeString(NULL, &ustr, &status);
   if (U_FAILURE(status)) {
       printf("printStringBreaks, utext_openConstUnicodeString() returns %s\n", u_errorName(status));
       return;
    }
   printStringBreaks(tstr, expected, expectedCount);
   utext_close(tstr);
}


void RBBITest::TestBug3818() {
    UErrorCode  status = U_ZERO_ERROR;

    // Four Thai words...
    static const UChar thaiWordData[] = {  0x0E43,0x0E2B,0x0E0D,0x0E48, 0x0E43,0x0E2B,0x0E0D,0x0E48,
                                           0x0E43,0x0E2B,0x0E0D,0x0E48, 0x0E43,0x0E2B,0x0E0D,0x0E48, 0 };
    UnicodeString  thaiStr(thaiWordData);

    BreakIterator* bi = BreakIterator::createWordInstance(Locale("th"), status);
    if (U_FAILURE(status) || bi == NULL) {
        errcheckln(status, "Fail at file %s, line %d, status = %s", __FILE__, __LINE__, u_errorName(status));
        return;
    }
    bi->setText(thaiStr);

    int32_t  startOfSecondWord = bi->following(1);
    if (startOfSecondWord != 4) {
        errln("Fail at file %s, line %d expected start of word at 4, got %d",
            __FILE__, __LINE__, startOfSecondWord);
    }
    startOfSecondWord = bi->following(0);
    if (startOfSecondWord != 4) {
        errln("Fail at file %s, line %d expected start of word at 4, got %d",
            __FILE__, __LINE__, startOfSecondWord);
    }
    delete bi;
}


//---------------------------------------------
//
//     other tests
//
//---------------------------------------------

void RBBITest::TestGetAvailableLocales()
{
    int32_t locCount = 0;
    const Locale* locList = BreakIterator::getAvailableLocales(locCount);

    if (locCount == 0)
        dataerrln("getAvailableLocales() returned an empty list!");
    // Just make sure that it's returning good memory.
    int32_t i;
    for (i = 0; i < locCount; ++i) {
        logln(locList[i].getName());
    }
}

//Testing the BreakIterator::getDisplayName() function
void RBBITest::TestGetDisplayName()
{
    UnicodeString   result;

    BreakIterator::getDisplayName(Locale::getUS(), result);
    if (Locale::getDefault() == Locale::getUS() && result != "English (United States)")
        dataerrln("BreakIterator::getDisplayName() failed: expected \"English (United States)\", got \""
                + result);

    BreakIterator::getDisplayName(Locale::getFrance(), Locale::getUS(), result);
    if (result != "French (France)")
        dataerrln("BreakIterator::getDisplayName() failed: expected \"French (France)\", got \""
                + result);
}
/**
 * Test End Behaviour
 * @bug 4068137
 */
void RBBITest::TestEndBehaviour()
{
    UErrorCode status = U_ZERO_ERROR;
    UnicodeString testString("boo.");
    BreakIterator *wb = BreakIterator::createWordInstance(Locale::getDefault(), status);
    if (U_FAILURE(status))
    {
        errcheckln(status, "Failed to create the BreakIterator for default locale in TestEndBehaviour. - %s", u_errorName(status));
        return;
    }
    wb->setText(testString);

    if (wb->first() != 0)
        errln("Didn't get break at beginning of string.");
    if (wb->next() != 3)
        errln("Didn't get break before period in \"boo.\"");
    if (wb->current() != 4 && wb->next() != 4)
        errln("Didn't get break at end of string.");
    delete wb;
}
/*
 * @bug 4153072
 */
void RBBITest::TestBug4153072() {
    UErrorCode status = U_ZERO_ERROR;
    BreakIterator *iter = BreakIterator::createWordInstance(Locale::getDefault(), status);
    if (U_FAILURE(status))
    {
        errcheckln(status, "Failed to create the BreakIterator for default locale in TestBug4153072 - %s", u_errorName(status));
        return;
    }
    UnicodeString str("...Hello, World!...");
    int32_t begin = 3;
    int32_t end = str.length() - 3;
    UBool onBoundary;

    StringCharacterIterator* textIterator = new StringCharacterIterator(str, begin, end, begin);
    iter->adoptText(textIterator);
    int index;
    // Note: with the switch to UText, there is no way to restrict the
    //       iteration range to begin at an index other than zero.
    //       String character iterators created with a non-zero bound are
    //         treated by RBBI as being empty.
    for (index = -1; index < begin + 1; ++index) {
        onBoundary = iter->isBoundary(index);
        if (index == 0?  !onBoundary : onBoundary) {
            errln((UnicodeString)"Didn't handle isBoundary correctly with offset = " + index +
                            " and begin index = " + begin);
        }
    }
    delete iter;
}


//
// Test for problem reported by Ashok Matoria on 9 July 2007
//    One.<kSoftHyphen><kSpace>Two.
//
//    Sentence break at start (0) and then on calling next() it breaks at
//   'T' of "Two". Now, at this point if I do next() and
//    then previous(), it breaks at <kSOftHyphen> instead of 'T' of "Two".
//
void RBBITest::TestBug5775() {
    UErrorCode status = U_ZERO_ERROR;
    BreakIterator *bi = BreakIterator::createSentenceInstance(Locale::getEnglish(), status);
    TEST_ASSERT_SUCCESS(status);
    if (U_FAILURE(status)) {
        return;
    }
// Check for status first for better handling of no data errors.
    TEST_ASSERT(bi != NULL);
    if (bi == NULL) {
        return;
    }

    UnicodeString s("One.\\u00ad Two.", -1, US_INV);
    //               01234      56789
    s = s.unescape();
    bi->setText(s);
    int pos = bi->next();
    TEST_ASSERT(pos == 6);
    pos = bi->next();
    TEST_ASSERT(pos == 10);
    pos = bi->previous();
    TEST_ASSERT(pos == 6);
    delete bi;
}



//------------------------------------------------------------------------------
//
//   RBBITest::Extended    Run  RBBI Tests from an external test data file
//
//------------------------------------------------------------------------------

struct TestParams {
    BreakIterator   *bi;                   // Break iterator is set while parsing test source.
                                           //   Changed out whenever test data changes break type.

    UnicodeString    dataToBreak;          // Data that is built up while parsing the test.
    UVector32       *expectedBreaks;       // Expected break positions, matches dataToBreak UnicodeString.
    UVector32       *srcLine;              // Positions in source file, indexed same as dataToBreak.
    UVector32       *srcCol;

    UText           *textToBreak;          // UText, could be UTF8 or UTF16.
    UVector32       *textMap;              // Map from UTF-16 dataToBreak offsets to UText offsets.
    CharString       utf8String;           // UTF-8 form of text to break.

    TestParams(UErrorCode &status) : dataToBreak() {
        bi               = NULL;
        expectedBreaks   = new UVector32(status);
        srcLine          = new UVector32(status);
        srcCol           = new UVector32(status);
        textToBreak      = NULL;
        textMap          = new UVector32(status);
    }

    ~TestParams() {
        delete bi;
        delete expectedBreaks;
        delete srcLine;
        delete srcCol;
        utext_close(textToBreak);
        delete textMap;
    }

    int32_t getSrcLine(int32_t bp);
    int32_t getExpectedBreak(int32_t bp);
    int32_t getSrcCol(int32_t bp);

    void setUTF16(UErrorCode &status);
    void setUTF8(UErrorCode &status);
};

// Append a UnicodeString to a CharString with UTF-8 encoding.
// Substitute any invalid chars.
//   Note: this is used with test data that includes a few unpaired surrogates in the UTF-16 that will be substituted.
static void CharStringAppend(CharString &dest, const UnicodeString &src, UErrorCode &status) {
    if (U_FAILURE(status)) {
        return;
    }
    int32_t utf8Length;
    u_strToUTF8WithSub(NULL, 0, &utf8Length,            // Output Buffer, NULL for preflight.
                       src.getBuffer(), src.length(),   // UTF-16 data
                       0xfffd, NULL,                    // Substitution char, number of subs.
                       &status);
    if (U_FAILURE(status) && status != U_BUFFER_OVERFLOW_ERROR) {
        return;
    }
    status = U_ZERO_ERROR;
    int32_t capacity;
    char *buffer = dest.getAppendBuffer(utf8Length, utf8Length, capacity, status);
    u_strToUTF8WithSub(buffer, utf8Length, NULL,
                       src.getBuffer(), src.length(),
                       0xfffd, NULL, &status);
    dest.append(buffer, utf8Length, status);
}


void TestParams::setUTF16(UErrorCode &status) {
    textToBreak = utext_openUnicodeString(textToBreak, &dataToBreak, &status);
    textMap->removeAllElements();
    for (int32_t i=0; i<dataToBreak.length(); i++) {
        if (i == dataToBreak.getChar32Start(i)) {
            textMap->addElement(i, status);
        } else {
            textMap->addElement(-1, status);
        }
    }
    textMap->addElement(dataToBreak.length(), status);
    U_ASSERT(dataToBreak.length() + 1 == textMap->size());
}


void TestParams::setUTF8(UErrorCode &status) {
    if (U_FAILURE(status)) {
        return;
    }
    utf8String.clear();
    CharStringAppend(utf8String, dataToBreak, status);
    textToBreak = utext_openUTF8(textToBreak, utf8String.data(), utf8String.length(), &status);
    if (U_FAILURE(status)) {
        return;
    }

    textMap->removeAllElements();
    int32_t utf16Index = 0;
    for (;;) {
        textMap->addElement(utf16Index, status);
        UChar32 c32 = utext_current32(textToBreak);
        if (c32 < 0) {
            break;
        }
        utf16Index += U16_LENGTH(c32);
        utext_next32(textToBreak);
        while (textMap->size() < utext_getNativeIndex(textToBreak)) {
            textMap->addElement(-1, status);
        }
    }
    U_ASSERT(utext_nativeLength(textToBreak) + 1 == textMap->size());
}


int32_t TestParams::getSrcLine(int32_t bp) {
    if (bp >= textMap->size()) {
        bp = textMap->size() - 1;
    }
    int32_t i = 0;
    for(; bp >= 0 ; --bp) {
        // Move to a character boundary if we are not on one already.
        i = textMap->elementAti(bp);
        if (i >= 0) {
            break;
        }
    }
    return srcLine->elementAti(i);
}


int32_t TestParams::getExpectedBreak(int32_t bp) {
    if (bp >= textMap->size()) {
        return 0;
    }
    int32_t i = textMap->elementAti(bp);
    int32_t retVal = 0;
    if (i >= 0) {
        retVal = expectedBreaks->elementAti(i);
    }
    return retVal;
}


int32_t TestParams::getSrcCol(int32_t bp) {
    if (bp >= textMap->size()) {
        bp = textMap->size() - 1;
    }
    int32_t i = 0;
    for(; bp >= 0; --bp) {
        // Move bp to a character boundary if we are not on one already.
        i = textMap->elementAti(bp);
        if (i >= 0) {
            break;
        }
    }
    return srcCol->elementAti(i);
}


void RBBITest::executeTest(TestParams *t, UErrorCode &status) {
    int32_t    bp;
    int32_t    prevBP;
    int32_t    i;

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

    if (t->bi == NULL) {
        return;
    }

    t->bi->setText(t->textToBreak, status);
    //
    //  Run the iterator forward
    //
    prevBP = -1;
    for (bp = t->bi->first(); bp != BreakIterator::DONE; bp = t->bi->next()) {
        if (prevBP ==  bp) {
            // Fail for lack of forward progress.
            errln("Forward Iteration, no forward progress.  Break Pos=%4d  File line,col=%4d,%4d",
                bp, t->getSrcLine(bp), t->getSrcCol(bp));
            break;
        }

        // Check that there we didn't miss an expected break between the last one
        //  and this one.
        for (i=prevBP+1; i<bp; i++) {
            if (t->getExpectedBreak(i) != 0) {
                int expected[] = {0, i};
                printStringBreaks(t->dataToBreak, expected, 2);
                errln("Forward Iteration, break expected, but not found.  Pos=%4d  File line,col= %4d,%4d",
                      i, t->getSrcLine(i), t->getSrcCol(i));
            }
        }

        // Check that the break we did find was expected
        if (t->getExpectedBreak(bp) == 0) {
            int expected[] = {0, bp};
            printStringBreaks(t->textToBreak, expected, 2);
            errln("Forward Iteration, break found, but not expected.  Pos=%4d  File line,col= %4d,%4d",
                bp, t->getSrcLine(bp), t->getSrcCol(bp));
        } else {
            // The break was expected.
            //   Check that the {nnn} tag value is correct.
            int32_t expectedTagVal = t->getExpectedBreak(bp);
            if (expectedTagVal == -1) {
                expectedTagVal = 0;
            }
            int32_t line = t->getSrcLine(bp);
            int32_t rs = t->bi->getRuleStatus();
            if (rs != expectedTagVal) {
                errln("Incorrect status for forward break.  Pos=%4d  File line,col= %4d,%4d.\n"
                      "          Actual, Expected status = %4d, %4d",
                    bp, line, t->getSrcCol(bp), rs, expectedTagVal);
            }
        }

        prevBP = bp;
    }

    // Verify that there were no missed expected breaks after the last one found
    for (i=prevBP+1; i<utext_nativeLength(t->textToBreak); i++) {
        if (t->getExpectedBreak(i) != 0) {
            errln("Forward Iteration, break expected, but not found.  Pos=%4d  File line,col= %4d,%4d",
                      i, t->getSrcLine(i), t->getSrcCol(i));
        }
    }

    //
    //  Run the iterator backwards, verify that the same breaks are found.
    //
    prevBP = static_cast<int32_t>(utext_nativeLength(t->textToBreak) + 2); // start with a phony value for the last break pos seen.
    bp = t->bi->last();
    while (bp != BreakIterator::DONE) {
        if (prevBP ==  bp) {
            // Fail for lack of progress.
            errln("Reverse Iteration, no progress.  Break Pos=%4d  File line,col=%4d,%4d",
                bp, t->getSrcLine(bp), t->getSrcCol(bp));
            break;
        }

        // Check that we didn't miss an expected break between the last one
        //  and this one.  (UVector returns zeros for index out of bounds.)
        for (i=prevBP-1; i>bp; i--) {
            if (t->getExpectedBreak(i) != 0) {
                errln("Reverse Iteration, break expected, but not found.  Pos=%4d  File line,col= %4d,%4d",
                      i, t->getSrcLine(i), t->getSrcCol(i));
            }
        }

        // Check that the break we did find was expected
        if (t->getExpectedBreak(bp) == 0) {
            errln("Reverse Itertion, break found, but not expected.  Pos=%4d  File line,col= %4d,%4d",
                   bp, t->getSrcLine(bp), t->getSrcCol(bp));
        } else {
            // The break was expected.
            //   Check that the {nnn} tag value is correct.
            int32_t expectedTagVal = t->getExpectedBreak(bp);
            if (expectedTagVal == -1) {
                expectedTagVal = 0;
            }
            int line = t->getSrcLine(bp);
            int32_t rs = t->bi->getRuleStatus();
            if (rs != expectedTagVal) {
                errln("Incorrect status for reverse break.  Pos=%4d  File line,col= %4d,%4d.\n"
                      "          Actual, Expected status = %4d, %4d",
                    bp, line, t->getSrcCol(bp), rs, expectedTagVal);
            }
        }

        prevBP = bp;
        bp = t->bi->previous();
    }

    // Verify that there were no missed breaks prior to the last one found
    for (i=prevBP-1; i>=0; i--) {
        if (t->getExpectedBreak(i) != 0) {
            errln("Forward Itertion, break expected, but not found.  Pos=%4d  File line,col= %4d,%4d",
                      i, t->getSrcLine(i), t->getSrcCol(i));
        }
    }

    // Check isBoundary()
    for (i=0; i < utext_nativeLength(t->textToBreak); i++) {
        UBool boundaryExpected = (t->getExpectedBreak(i) != 0);
        UBool boundaryFound    = t->bi->isBoundary(i);
        if (boundaryExpected != boundaryFound) {
            errln("isBoundary(%d) incorrect. File line,col= %4d,%4d\n"
                  "        Expected, Actual= %s, %s",
                  i, t->getSrcLine(i), t->getSrcCol(i),
                  boundaryExpected ? "true":"false", boundaryFound? "true" : "false");
        }
    }

    // Check following()
    for (i=0; i < static_cast<int32_t>(utext_nativeLength(t->textToBreak)); i++) {
        int32_t actualBreak = t->bi->following(i);
        int32_t expectedBreak = BreakIterator::DONE;
        for (int32_t j=i+1; j <= static_cast<int32_t>(utext_nativeLength(t->textToBreak)); j++) {
            if (t->getExpectedBreak(j) != 0) {
                expectedBreak = j;
                break;
            }
        }
        if (expectedBreak != actualBreak) {
            errln("following(%d) incorrect. File line,col= %4d,%4d\n"
                  "        Expected, Actual= %d, %d",
                  i, t->getSrcLine(i), t->getSrcCol(i), expectedBreak, actualBreak);
        }
    }

    // Check preceding()
    for (i=static_cast<int32_t>(utext_nativeLength(t->textToBreak)); i>=0; i--) {
        int32_t actualBreak = t->bi->preceding(i);
        int32_t expectedBreak = BreakIterator::DONE;

        // For UTF-8 & UTF-16 supplementals, all code units of a character are equivalent.
        // preceding(trailing byte) will return the index of some preceding code point,
        // not the lead byte of the current code point, even though that has a smaller index.
        // Therefore, start looking at the expected break data not at i-1, but at
        // the start of code point index - 1.
        utext_setNativeIndex(t->textToBreak, i);
        int32_t j = static_cast<int32_t>(utext_getNativeIndex(t->textToBreak) - 1);
        for (; j >= 0; j--) {
            if (t->getExpectedBreak(j) != 0) {
                expectedBreak = j;
                break;
            }
        }
        if (expectedBreak != actualBreak) {
            errln("preceding(%d) incorrect. File line,col= %4d,%4d\n"
                  "        Expected, Actual= %d, %d",
                  i, t->getSrcLine(i), t->getSrcCol(i), expectedBreak, actualBreak);
        }
    }
}

void RBBITest::TestExtended() {
     // The expectations in this test heavily depends on the Thai dictionary.
     // Therefore, we skip this test under the LSTM configuration.
     if (skipDictionaryTest()) {
         return;
     }
  // Skip test for now when UCONFIG_NO_FILTERED_BREAK_ITERATION is set. This
  // data driven test closely entangles filtered and regular data.
#if !UCONFIG_NO_REGULAR_EXPRESSIONS && !UCONFIG_NO_FILTERED_BREAK_ITERATION
    UErrorCode      status  = U_ZERO_ERROR;
    Locale          locale("");

    TestParams          tp(status);

    RegexMatcher      localeMatcher(UnicodeString(u"<locale *([\\p{L}\\p{Nd}_@&=-]*) *>"), 0, status);
    if (U_FAILURE(status)) {
        dataerrln("Failure in file %s, line %d, status = \"%s\"", __FILE__, __LINE__, u_errorName(status));
    }

    //
    //  Open and read the test data file.
    //
    const char *testDataDirectory = IntlTest::getSourceTestData(status);
    CharString testFileName(testDataDirectory, -1, status);
    testFileName.append("rbbitst.txt", -1, status);

    int    len;
    UChar *testFile = ReadAndConvertFile(testFileName.data(), len, "UTF-8", status);
    if (U_FAILURE(status)) {
        errln("%s:%d Error %s opening file rbbitst.txt", __FILE__, __LINE__, u_errorName(status));
        return;
    }

    bool skipTest = false; // Skip this test?

    //
    //  Put the test data into a UnicodeString
    //
    UnicodeString testString(FALSE, testFile, len);

    enum EParseState{
        PARSE_COMMENT,
        PARSE_TAG,
        PARSE_DATA,
        PARSE_NUM,
        PARSE_RULES
    }
    parseState = PARSE_TAG;

    EParseState savedState = PARSE_TAG;

    int32_t    lineNum  = 1;
    int32_t    colStart = 0;
    int32_t    column   = 0;
    int32_t    charIdx  = 0;

    int32_t    tagValue = 0;             // The numeric value of a <nnn> tag.

    UnicodeString       rules;           // Holds rules from a <rules> ... </rules> block
    int32_t             rulesFirstLine = 0;  // Line number of the start of current <rules> block

    for (charIdx = 0; charIdx < len; ) {
        status = U_ZERO_ERROR;
        UChar  c = testString.charAt(charIdx);
        charIdx++;
        if (c == u'\r' && charIdx<len && testString.charAt(charIdx) == u'\n') {
            // treat CRLF as a unit
            c = u'\n';
            charIdx++;
        }
        if (c == u'\n' || c == u'\r') {
            lineNum++;
            colStart = charIdx;
        }
        column = charIdx - colStart + 1;

        switch (parseState) {
        case PARSE_COMMENT:
            if (c == u'\n' || c == u'\r') {
                parseState = savedState;
            }
            break;

        case PARSE_TAG:
            {
            if (c == u'#') {
                parseState = PARSE_COMMENT;
                savedState = PARSE_TAG;
                break;
            }
            if (u_isUWhiteSpace(c)) {
                break;
            }
            if (testString.compare(charIdx-1, 6, u"<word>") == 0) {
                delete tp.bi;
                tp.bi = BreakIterator::createWordInstance(locale,  status);
                skipTest = false;
                charIdx += 5;
                break;
            }
            if (testString.compare(charIdx-1, 6, u"<char>") == 0) {
                delete tp.bi;
                tp.bi = BreakIterator::createCharacterInstance(locale,  status);
                skipTest = false;
                charIdx += 5;
                break;
            }
            if (testString.compare(charIdx-1, 6, u"<line>") == 0) {
                delete tp.bi;
                tp.bi = BreakIterator::createLineInstance(locale,  status);
                skipTest = false;
                charIdx += 5;
                break;
            }
            if (testString.compare(charIdx-1, 6, u"<sent>") == 0) {
                delete tp.bi;
                tp.bi = BreakIterator::createSentenceInstance(locale,  status);
                skipTest = false;
                charIdx += 5;
                break;
            }
            if (testString.compare(charIdx-1, 7, u"<title>") == 0) {
                delete tp.bi;
                tp.bi = BreakIterator::createTitleInstance(locale,  status);
                charIdx += 6;
                break;
            }

            if (testString.compare(charIdx-1, 7, u"<rules>") == 0 ||
                testString.compare(charIdx-1, 10, u"<badrules>") == 0) {
                charIdx = testString.indexOf(u'>', charIdx) + 1;
                parseState = PARSE_RULES;
                rules.remove();
                rulesFirstLine = lineNum;
                break;
            }

            // <locale  loc_name>
            localeMatcher.reset(testString);
            if (localeMatcher.lookingAt(charIdx-1, status)) {
                UnicodeString localeName = localeMatcher.group(1, status);
                char localeName8[100];
                localeName.extract(0, localeName.length(), localeName8, sizeof(localeName8), 0);
                locale = Locale::createFromName(localeName8);
                charIdx += localeMatcher.group(0, status).length() - 1;
                TEST_ASSERT_SUCCESS(status);
                break;
            }
            if (testString.compare(charIdx-1, 6, u"<data>") == 0) {
                parseState = PARSE_DATA;
                charIdx += 5;
                tp.dataToBreak = "";
                tp.expectedBreaks->removeAllElements();
                tp.srcCol ->removeAllElements();
                tp.srcLine->removeAllElements();
                break;
            }

            errln("line %d: Tag expected in test file.", lineNum);
            parseState = PARSE_COMMENT;
            savedState = PARSE_DATA;
            goto end_test; // Stop the test.
            }
            break;

        case PARSE_RULES:
            if (testString.compare(charIdx-1, 8, u"</rules>") == 0) {
                charIdx += 7;
                parseState = PARSE_TAG;
                delete tp.bi;
                UParseError pe;
                tp.bi = new RuleBasedBreakIterator(rules, pe, status);
                skipTest = U_FAILURE(status);
                if (U_FAILURE(status)) {
                    errln("file rbbitst.txt: %d - Error %s creating break iterator from rules.",
                        rulesFirstLine + pe.line - 1, u_errorName(status));
                }
            } else if (testString.compare(charIdx-1, 11, u"</badrules>") == 0) {
                charIdx += 10;
                parseState = PARSE_TAG;
                UErrorCode ec = U_ZERO_ERROR;
                UParseError pe;
                RuleBasedBreakIterator bi(rules, pe, ec);
                if (U_SUCCESS(ec)) {
                    errln("file rbbitst.txt: %d - Expected, but did not get, a failure creating break iterator from rules.",
                        rulesFirstLine + pe.line - 1);
                }
            } else {
                rules.append(c);
            }
            break;

        case PARSE_DATA:
            if (c == u'•') {
                int32_t  breakIdx = tp.dataToBreak.length();
                if (tp.expectedBreaks->size() > breakIdx) {
                    errln("rbbitst.txt:%d:%d adjacent expected breaks with no intervening test text",
                          lineNum, column);
                }
                tp.expectedBreaks->setSize(breakIdx+1);
                tp.expectedBreaks->setElementAt(-1, breakIdx);
                tp.srcLine->setSize(breakIdx+1);
                tp.srcLine->setElementAt(lineNum, breakIdx);
                tp.srcCol ->setSize(breakIdx+1);
                tp.srcCol ->setElementAt(column, breakIdx);
                break;
            }

            if (testString.compare(charIdx-1, 7, u"</data>") == 0) {
                // Add final entry to mappings from break location to source file position.
                //  Need one extra because last break position returned is after the
                //    last char in the data, not at the last char.
                tp.srcLine->addElement(lineNum, status);
                tp.srcCol ->addElement(column, status);

                parseState = PARSE_TAG;
                charIdx += 6;

                if (!skipTest) {
                    // RUN THE TEST!
                    status = U_ZERO_ERROR;
                    tp.setUTF16(status);
                    executeTest(&tp, status);
                    TEST_ASSERT_SUCCESS(status);

                    // Run again, this time with UTF-8 text wrapped in a UText.
                    status = U_ZERO_ERROR;
                    tp.setUTF8(status);
                    TEST_ASSERT_SUCCESS(status);
                    executeTest(&tp, status);
                }
                break;
            }

            if (testString.compare(charIdx-1, 3, u"\\N{") == 0) {
                // Named character, e.g. \N{COMBINING GRAVE ACCENT}
                // Get the code point from the name and insert it into the test data.
                //   (Damn, no API takes names in Unicode  !!!
                //    we've got to take it back to char *)
                int32_t nameEndIdx = testString.indexOf(u'}', charIdx);
                int32_t nameLength = nameEndIdx - (charIdx+2);
                char charNameBuf[200];
                UChar32 theChar = -1;
                if (nameEndIdx != -1) {
                    UErrorCode status = U_ZERO_ERROR;
                    testString.extract(charIdx+2, nameLength, charNameBuf, sizeof(charNameBuf));
                    charNameBuf[sizeof(charNameBuf)-1] = 0;
                    theChar = u_charFromName(U_UNICODE_CHAR_NAME, charNameBuf, &status);
                    if (U_FAILURE(status)) {
                        theChar = -1;
                    }
                }
                if (theChar == -1) {
                    errln("Error in named character in test file at line %d, col %d",
                        lineNum, column);
                } else {
                    // Named code point was recognized.  Insert it
                    //   into the test data.
                    tp.dataToBreak.append(theChar);
                    while (tp.dataToBreak.length() > tp.srcLine->size()) {
                        tp.srcLine->addElement(lineNum, status);
                        tp.srcCol ->addElement(column, status);
                    }
                }
                if (nameEndIdx > charIdx) {
                    charIdx = nameEndIdx+1;

                }
                break;
            }



            if (testString.compare(charIdx-1, 2, u"<>") == 0) {
                charIdx++;
                int32_t  breakIdx = tp.dataToBreak.length();
                tp.expectedBreaks->setSize(breakIdx+1);
                tp.expectedBreaks->setElementAt(-1, breakIdx);
                tp.srcLine->setSize(breakIdx+1);
                tp.srcLine->setElementAt(lineNum, breakIdx);
                tp.srcCol ->setSize(breakIdx+1);
                tp.srcCol ->setElementAt(column, breakIdx);
                break;
            }

            if (c == u'<') {
                tagValue   = 0;
                parseState = PARSE_NUM;
                break;
            }

            if (c == u'#' && column==3) {   // TODO:  why is column off so far?
                parseState = PARSE_COMMENT;
                savedState = PARSE_DATA;
                break;
            }

            if (c == u'\\') {
                // Check for \ at end of line, a line continuation.
                //     Advance over (discard) the newline
                UChar32 cp = testString.char32At(charIdx);
                if (cp == u'\r' && charIdx<len && testString.charAt(charIdx+1) == u'\n') {
                    // We have a CR LF
                    //  Need an extra increment of the input ptr to move over both of them
                    charIdx++;
                }
                if (cp == u'\n' || cp == u'\r') {
                    lineNum++;
                    colStart = charIdx;
                    charIdx++;
                    break;
                }

                // Let unescape handle the back slash.
                cp = testString.unescapeAt(charIdx);
                if (cp != -1) {
                    // Escape sequence was recognized.  Insert the char
                    //   into the test data.
                    tp.dataToBreak.append(cp);
                    while (tp.dataToBreak.length() > tp.srcLine->size()) {
                        tp.srcLine->addElement(lineNum, status);
                        tp.srcCol ->addElement(column, status);
                    }
                    break;
                }


                // Not a recognized backslash escape sequence.
                // Take the next char as a literal.
                //  TODO:  Should this be an error?
                c = testString.charAt(charIdx);
                charIdx = testString.moveIndex32(charIdx, 1);
            }

            // Normal, non-escaped data char.
            tp.dataToBreak.append(c);

            // Save the mapping from offset in the data to line/column numbers in
            //   the original input file.  Will be used for better error messages only.
            //   If there's an expected break before this char, the slot in the mapping
            //     vector will already be set for this char; don't overwrite it.
            if (tp.dataToBreak.length() > tp.srcLine->size()) {
                tp.srcLine->addElement(lineNum, status);
                tp.srcCol ->addElement(column, status);
            }
            break;


        case PARSE_NUM:
            // We are parsing an expected numeric tag value, like <1234>,
            //   within a chunk of data.
            if (u_isUWhiteSpace(c)) {
                break;
            }

            if (c == u'>') {
                // Finished the number.  Add the info to the expected break data,
                //   and switch parse state back to doing plain data.
                parseState = PARSE_DATA;
                if (tagValue == 0) {
                    tagValue = -1;
                }
                int32_t  breakIdx = tp.dataToBreak.length();
                if (tp.expectedBreaks->size() > breakIdx) {
                    errln("rbbitst.txt:%d:%d adjacent expected breaks with no intervening test text",
                          lineNum, column);
                }
                tp.expectedBreaks->setSize(breakIdx+1);
                tp.expectedBreaks->setElementAt(tagValue, breakIdx);
                tp.srcLine->setSize(breakIdx+1);
                tp.srcLine->setElementAt(lineNum, breakIdx);
                tp.srcCol ->setSize(breakIdx+1);
                tp.srcCol ->setElementAt(column, breakIdx);
                break;
            }

            if (u_isdigit(c)) {
                tagValue = tagValue*10 + u_charDigitValue(c);
                break;
            }

            errln("Syntax Error in test file at line %d, col %d",
                lineNum, column);
            parseState = PARSE_COMMENT;
            goto end_test; // Stop the test
            break;
        }


        if (U_FAILURE(status)) {
            dataerrln("ICU Error %s while parsing test file at line %d.",
                u_errorName(status), lineNum);
            status = U_ZERO_ERROR;
            goto end_test; // Stop the test
        }

    }

    // Reached end of test file. Raise an error if parseState indicates that we are
    //   within a block that should have been terminated.

    if (parseState == PARSE_RULES) {
        errln("rbbitst.txt:%d <rules> block beginning at line %d is not closed.",
            lineNum, rulesFirstLine);
    }
    if (parseState == PARSE_DATA) {
        errln("rbbitst.txt:%d <data> block not closed.", lineNum);
    }


end_test:
    delete [] testFile;
#endif
}

//-------------------------------------------------------------------------------
//
//  TestDictRules   create a break iterator from source rules that includes a
//                  dictionary range.   Regression for bug #7130.  Source rules
//                  do not declare a break iterator type (word, line, sentence, etc.
//                  but the dictionary code, without a type, would loop.
//
//-------------------------------------------------------------------------------
void RBBITest::TestDictRules() {
    const char *rules =  "$dictionary = [a-z]; \n"
                         "!!forward; \n"
                         "$dictionary $dictionary; \n"
                         "!!reverse; \n"
                         "$dictionary $dictionary; \n";
    const char *text = "aa";
    UErrorCode status = U_ZERO_ERROR;
    UParseError parseError;

    RuleBasedBreakIterator bi(rules, parseError, status);
    if (U_SUCCESS(status)) {
        UnicodeString utext = text;
        bi.setText(utext);
        int32_t position;
        int32_t loops;
        for (loops = 0; loops<10; loops++) {
            position = bi.next();
            if (position == RuleBasedBreakIterator::DONE) {
                break;
            }
        }
        TEST_ASSERT(loops == 1);
    } else {
        dataerrln("Error creating RuleBasedBreakIterator: %s", u_errorName(status));
    }
}



//--------------------------------------------------------------------------------------------
//
//   Run tests from each of the boundary test data files distributed by the Unicode Consortium
//
//-------------------------------------------------------------------------------------------
void RBBITest::TestUnicodeFiles() {
    RuleBasedBreakIterator  *bi;
    UErrorCode               status = U_ZERO_ERROR;

    bi =  (RuleBasedBreakIterator *)BreakIterator::createCharacterInstance(Locale::getEnglish(), status);
    TEST_ASSERT_SUCCESS(status);
    if (U_SUCCESS(status)) {
        runUnicodeTestData("GraphemeBreakTest.txt", bi);
    }
    delete bi;

    bi =  (RuleBasedBreakIterator *)BreakIterator::createWordInstance(Locale::getEnglish(), status);
    TEST_ASSERT_SUCCESS(status);
    if (U_SUCCESS(status)) {
        runUnicodeTestData("WordBreakTest.txt", bi);
    }
    delete bi;

    bi =  (RuleBasedBreakIterator *)BreakIterator::createSentenceInstance(Locale::getEnglish(), status);
    TEST_ASSERT_SUCCESS(status);
    if (U_SUCCESS(status)) {
        runUnicodeTestData("SentenceBreakTest.txt", bi);
    }
    delete bi;

    bi =  (RuleBasedBreakIterator *)BreakIterator::createLineInstance(Locale::getEnglish(), status);
    TEST_ASSERT_SUCCESS(status);
    if (U_SUCCESS(status)) {
        runUnicodeTestData("LineBreakTest.txt", bi);
    }
    delete bi;
}


// Check for test cases from the Unicode test data files that are known to fail
// and should be skipped as known issues because ICU does not fully implement
// the Unicode specifications, or because ICU includes tailorings that differ from
// the Unicode standard.
//
// Test cases are identified by the test data sequence, which tends to be more stable
// across Unicode versions than the test file line numbers.
//
// The test case with ticket "10666" is a dummy, included as an example.

UBool RBBITest::testCaseIsKnownIssue(const UnicodeString &testCase, const char *fileName) {
    static struct TestCase {
        const char *fTicketNum;
        const char *fFileName;
        const UChar *fString;
    } badTestCases[] = {
        {"10666", "GraphemeBreakTest.txt", u"\u0020\u0020\u0033"},    // Fake example, for illustration.
        // The following tests were originally for
        // Issue 8151, move the Finnish tailoring of the line break of hyphens to root.
        // However, that ticket has been closed as fixed but these tests still fail, so
        // ICU-21097 has been created to investigate and address these remaining issues.
        {"21097",  "LineBreakTest.txt", u"-#"},
        {"21097",  "LineBreakTest.txt", u"\u002d\u0308\u0023"},
        {"21097",  "LineBreakTest.txt", u"\u002d\u00a7"},
        {"21097",  "LineBreakTest.txt", u"\u002d\u0308\u00a7"},
        {"21097",  "LineBreakTest.txt", u"\u002d\U00050005"},
        {"21097",  "LineBreakTest.txt", u"\u002d\u0308\U00050005"},
        {"21097",  "LineBreakTest.txt", u"\u002d\u0e01"},
        {"21097",  "LineBreakTest.txt", u"\u002d\u0308\u0e01"},

        // The following tests were originally for
        // Issue ICU-12017 Improve line break around numbers.
        // However, that ticket has been closed as fixed but these tests still fail, so
        // ICU-21097 has been created to investigate and address these remaining issues.
        {"21097", "LineBreakTest.txt", u"\u002C\u0030"},   // ",0"
        {"21097", "LineBreakTest.txt", u"\u002C\u0308\u0030"},
        {"21097", "LineBreakTest.txt", u"equals .35 cents"},
        {"21097", "LineBreakTest.txt", u"a.2 "},
        {"21097", "LineBreakTest.txt", u"a.2 \u0915"},
        {"21097", "LineBreakTest.txt", u"a.2 \u672C"},
        {"21097", "LineBreakTest.txt", u"a.2\u3000\u672C"},
        {"21097", "LineBreakTest.txt", u"a.2\u3000\u307E"},
        {"21097", "LineBreakTest.txt", u"a.2\u3000\u0033"},
        {"21097", "LineBreakTest.txt", u"A.1 \uBABB"},
        {"21097", "LineBreakTest.txt", u"\uBD24\uC5B4\u002E\u0020\u0041\u002E\u0032\u0020\uBCFC"},
        {"21097", "LineBreakTest.txt", u"\uBD10\uC694\u002E\u0020\u0041\u002E\u0033\u0020\uBABB"},
        {"21097", "LineBreakTest.txt", u"\uC694\u002E\u0020\u0041\u002E\u0034\u0020\uBABB"},
        {"21097", "LineBreakTest.txt", u"a.2\u3000\u300C"},
    };

    for (int n=0; n<UPRV_LENGTHOF(badTestCases); n++) {
        const TestCase &badCase = badTestCases[n];
        if (!strcmp(fileName, badCase.fFileName) &&
                testCase == UnicodeString(badCase.fString)) {
            return logKnownIssue(badCase.fTicketNum);
        }
    }
    return FALSE;
}


//--------------------------------------------------------------------------------------------
//
//   Run tests from one of the boundary test data files distributed by the Unicode Consortium
//
//-------------------------------------------------------------------------------------------
void RBBITest::runUnicodeTestData(const char *fileName, RuleBasedBreakIterator *bi) {
#if !UCONFIG_NO_REGULAR_EXPRESSIONS
    UErrorCode  status = U_ZERO_ERROR;

    //
    //  Open and read the test data file, put it into a UnicodeString.
    //
    const char *testDataDirectory = IntlTest::getSourceTestData(status);
    char testFileName[1000];
    if (testDataDirectory == NULL || strlen(testDataDirectory) >= sizeof(testFileName)) {
        dataerrln("Can't open test data.  Path too long.");
        return;
    }
    strcpy(testFileName, testDataDirectory);
    strcat(testFileName, fileName);

    logln("Opening data file %s\n", fileName);

    int    len;
    UChar *testFile = ReadAndConvertFile(testFileName, len, "UTF-8", status);
    if (status != U_FILE_ACCESS_ERROR) {
        TEST_ASSERT_SUCCESS(status);
        TEST_ASSERT(testFile != NULL);
    }
    if (U_FAILURE(status) || testFile == NULL) {
        return; /* something went wrong, error already output */
    }
    UnicodeString testFileAsString(TRUE, testFile, len);

    //
    //  Parse the test data file using a regular expression.
    //  Each kind of token is recognized in its own capture group; what type of item was scanned
    //     is identified by which group had a match.
    //
    //    Capture Group  #                  1          2            3            4           5
    //    Parses this item:               divide       x      hex digits   comment \n  unrecognized \n
    //
    UnicodeString tokenExpr("[ \t]*(?:(\\u00F7)|(\\u00D7)|([0-9a-fA-F]+)|((?:#.*?)?$.)|(.*?$.))", -1, US_INV);
    RegexMatcher    tokenMatcher(tokenExpr, testFileAsString, UREGEX_MULTILINE | UREGEX_DOTALL, status);
    UnicodeString   testString;
    UVector32       breakPositions(status);
    int             lineNumber = 1;
    TEST_ASSERT_SUCCESS(status);
    if (U_FAILURE(status)) {
        return;
    }

    //
    //  Scan through each test case, building up the string to be broken in testString,
    //   and the positions that should be boundaries in the breakPositions vector.
    //
    int spin = 0;
    while (tokenMatcher.find()) {
        if(tokenMatcher.hitEnd()) {
          /* Shouldn't Happen(TM).  This means we didn't find the symbols we were looking for.
             This occurred when the text file was corrupt (wasn't marked as UTF-8)
             and caused an infinite loop here on EBCDIC systems!
          */
          fprintf(stderr,"FAIL: hit end of file %s for the %8dth time- corrupt data file?\r", fileName, ++spin);
          //       return;
        }
        if (tokenMatcher.start(1, status) >= 0) {
            // Scanned a divide sign, indicating a break position in the test data.
            if (testString.length()>0) {
                breakPositions.addElement(testString.length(), status);
            }
        }
        else if (tokenMatcher.start(2, status) >= 0) {
            // Scanned an 'x', meaning no break at this position in the test data
            //   Nothing to be done here.
            }
        else if (tokenMatcher.start(3, status) >= 0) {
            // Scanned Hex digits.  Convert them to binary, append to the character data string.
            const UnicodeString &hexNumber = tokenMatcher.group(3, status);
            int length = hexNumber.length();
            if (length<=8) {
                char buf[10];
                hexNumber.extract (0, length, buf, sizeof(buf), US_INV);
                UChar32 c = (UChar32)strtol(buf, NULL, 16);
                if (c<=0x10ffff) {
                    testString.append(c);
                } else {
                    errln("Error: Unicode Character value out of range. \'%s\', line %d.\n",
                       fileName, lineNumber);
                }
            } else {
                errln("Syntax Error: Hex Unicode Character value must have no more than 8 digits at \'%s\', line %d.\n",
                       fileName, lineNumber);
             }
        }
        else if (tokenMatcher.start(4, status) >= 0) {
            // Scanned to end of a line, possibly skipping over a comment in the process.
            //   If the line from the file contained test data, run the test now.
            if (testString.length() > 0 && !testCaseIsKnownIssue(testString, fileName)) {
                checkUnicodeTestCase(fileName, lineNumber, testString, &breakPositions, bi);
            }

            // Clear out this test case.
            //    The string and breakPositions vector will be refilled as the next
            //       test case is parsed.
            testString.remove();
            breakPositions.removeAllElements();
            lineNumber++;
        } else {
            // Scanner catchall.  Something unrecognized appeared on the line.
            char token[16];
            UnicodeString uToken = tokenMatcher.group(0, status);
            uToken.extract(0, uToken.length(), token, (uint32_t)sizeof(token));
            token[sizeof(token)-1] = 0;
            errln("Syntax error in test data file \'%s\', line %d.  Scanning \"%s\"\n", fileName, lineNumber, token);

            // Clean up, in preparation for continuing with the next line.
            testString.remove();
            breakPositions.removeAllElements();
            lineNumber++;
        }
        TEST_ASSERT_SUCCESS(status);
        if (U_FAILURE(status)) {
            break;
        }
    }

    delete [] testFile;
 #endif   // !UCONFIG_NO_REGULAR_EXPRESSIONS
}

//--------------------------------------------------------------------------------------------
//
//   checkUnicodeTestCase()   Run one test case from one of the Unicode Consortium
//                            test data files.  Do only a simple, forward-only check -
//                            this test is mostly to check that ICU and the Unicode
//                            data agree with each other.
//
//--------------------------------------------------------------------------------------------
void RBBITest::checkUnicodeTestCase(const char *testFileName, int lineNumber,
                         const UnicodeString &testString,   // Text data to be broken
                         UVector32 *breakPositions,         // Positions where breaks should be found.
                         RuleBasedBreakIterator *bi) {
    int32_t pos;                 // Break Position in the test string
    int32_t expectedI = 0;       // Index of expected break position in the vector of expected results.
    int32_t expectedPos;         // Expected break position (index into test string)

    bi->setText(testString);
    pos = bi->first();
    pos = bi->next();

    while (pos != BreakIterator::DONE) {
        if (expectedI >= breakPositions->size()) {
            errln("Test file \"%s\", line %d, unexpected break found at position %d",
                testFileName, lineNumber, pos);
            break;
        }
        expectedPos = breakPositions->elementAti(expectedI);
        if (pos < expectedPos) {
            errln("Test file \"%s\", line %d, unexpected break found at position %d",
                testFileName, lineNumber, pos);
            break;
        }
        if (pos > expectedPos) {
            errln("Test file \"%s\", line %d, failed to find expected break at position %d",
                testFileName, lineNumber, expectedPos);
            break;
        }
        pos = bi->next();
        expectedI++;
    }

    if (pos==BreakIterator::DONE && expectedI<breakPositions->size()) {
        errln("Test file \"%s\", line %d, failed to find expected break at position %d",
            testFileName, lineNumber, breakPositions->elementAti(expectedI));
    }
}



#if !UCONFIG_NO_REGULAR_EXPRESSIONS
//---------------------------------------------------------------------------------------
//
//   class RBBIMonkeyKind
//
//      Monkey Test for Break Iteration
//      Abstract interface class.   Concrete derived classes independently
//      implement the break rules for different iterator types.
//
//      The Monkey Test itself uses doesn't know which type of break iterator it is
//      testing, but works purely in terms of the interface defined here.
//
//---------------------------------------------------------------------------------------
class RBBIMonkeyKind {
public:
    // Return a UVector of UnicodeSets, representing the character classes used
    //   for this type of iterator.
    virtual  UVector  *charClasses() = 0;

    // Set the test text on which subsequent calls to next() will operate
    virtual  void      setText(const UnicodeString &s) = 0;

    // Find the next break position, starting from the prev break position, or from zero.
    // Return -1 after reaching end of string.
    virtual  int32_t   next(int32_t i) = 0;

    // Name of each character class, parallel with charClasses. Used for debugging output
    // of characters.
    virtual  std::vector<std::string>&     characterClassNames();

    void setAppliedRule(int32_t position, const char* value);

    std::string getAppliedRule(int32_t position);

    virtual ~RBBIMonkeyKind();
    UErrorCode deferredStatus;

    std::string classNameFromCodepoint(const UChar32 c);
    unsigned int maxClassNameSize();

 protected:
     RBBIMonkeyKind();
     std::vector<std::string> classNames;
     std::vector<std::string> appliedRules;

    // Clear `appliedRules` and fill it with empty strings in the size of test text.
    void prepareAppliedRules(int32_t size );

 private:

};

RBBIMonkeyKind::RBBIMonkeyKind() {
    deferredStatus = U_ZERO_ERROR;
}

RBBIMonkeyKind::~RBBIMonkeyKind() {
}

std::vector<std::string>& RBBIMonkeyKind::characterClassNames() {
    return classNames;
}

void RBBIMonkeyKind::prepareAppliedRules(int32_t size) {
    // Remove all the information in the `appliedRules`.
    appliedRules.clear();
    appliedRules.resize(size + 1);
}

void RBBIMonkeyKind::setAppliedRule(int32_t position, const char* value) {
    appliedRules[position] = value;
}

std::string RBBIMonkeyKind::getAppliedRule(int32_t position){
    return appliedRules[position];
}

std::string RBBIMonkeyKind::classNameFromCodepoint(const UChar32 c) {
    // Simply iterate through charClasses to find character's class
    for (int aClassNum = 0; aClassNum < charClasses()->size(); aClassNum++) {
        UnicodeSet *classSet = (UnicodeSet *)charClasses()->elementAt(aClassNum);
        if (classSet->contains(c)) {
            return classNames[aClassNum];
        }
    }
    U_ASSERT(FALSE);  // This should not happen.
    return "bad class name";
}

unsigned int RBBIMonkeyKind::maxClassNameSize() {
    unsigned int maxSize = 0;
    for (int aClassNum = 0; aClassNum < charClasses()->size(); aClassNum++) {
        auto aClassNumSize = static_cast<unsigned int>(classNames[aClassNum].size());
        if (aClassNumSize > maxSize) {
            maxSize = aClassNumSize;
        }
    }
    return maxSize;
}

//----------------------------------------------------------------------------------------
//
//   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 RBBICharMonkey      Character (Grapheme Cluster) specific implementation
//                             of RBBIMonkeyKind.
//
//------------------------------------------------------------------------------------------
class RBBICharMonkey: public RBBIMonkeyKind {
public:
    RBBICharMonkey();
    virtual          ~RBBICharMonkey();
    virtual  UVector *charClasses() override;
    virtual  void     setText(const UnicodeString &s) override;
    virtual  int32_t  next(int32_t i) override;
private:
    UVector   *fSets;

    UnicodeSet  *fCRLFSet;
    UnicodeSet  *fControlSet;
    UnicodeSet  *fExtendSet;
    UnicodeSet  *fZWJSet;
    UnicodeSet  *fRegionalIndicatorSet;
    UnicodeSet  *fPrependSet;
    UnicodeSet  *fSpacingSet;
    UnicodeSet  *fLSet;
    UnicodeSet  *fVSet;
    UnicodeSet  *fTSet;
    UnicodeSet  *fLVSet;
    UnicodeSet  *fLVTSet;
    UnicodeSet  *fHangulSet;
    UnicodeSet  *fExtendedPictSet;
    UnicodeSet  *fViramaSet;
    UnicodeSet  *fLinkingConsonantSet;
    UnicodeSet  *fExtCccZwjSet;
    UnicodeSet  *fAnySet;

    const UnicodeString *fText;
};


RBBICharMonkey::RBBICharMonkey() {
    UErrorCode  status = U_ZERO_ERROR;

    fText = NULL;

    fCRLFSet    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\r\\n]"), status);
    fControlSet = new UnicodeSet(UNICODE_STRING_SIMPLE("[[\\p{Grapheme_Cluster_Break = Control}]]"), status);
    fExtendSet  = new UnicodeSet(UNICODE_STRING_SIMPLE("[[\\p{Grapheme_Cluster_Break = Extend}]]"), status);
    fZWJSet     = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Grapheme_Cluster_Break = ZWJ}]"), status);
    fRegionalIndicatorSet =
                  new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Grapheme_Cluster_Break = Regional_Indicator}]"), status);
    fPrependSet = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Grapheme_Cluster_Break = Prepend}]"), status);
    fSpacingSet = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Grapheme_Cluster_Break = SpacingMark}]"), status);
    fLSet       = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Grapheme_Cluster_Break = L}]"), status);
    fVSet       = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Grapheme_Cluster_Break = V}]"), status);
    fTSet       = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Grapheme_Cluster_Break = T}]"), status);
    fLVSet      = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Grapheme_Cluster_Break = LV}]"), status);
    fLVTSet     = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Grapheme_Cluster_Break = LVT}]"), status);
    fHangulSet  = new UnicodeSet();
    fHangulSet->addAll(*fLSet);
    fHangulSet->addAll(*fVSet);
    fHangulSet->addAll(*fTSet);
    fHangulSet->addAll(*fLVSet);
    fHangulSet->addAll(*fLVTSet);

    fExtendedPictSet  = new UnicodeSet(u"[:Extended_Pictographic:]", status);
    fViramaSet        = new UnicodeSet(u"[\\p{Gujr}\\p{sc=Telu}\\p{sc=Mlym}\\p{sc=Orya}\\p{sc=Beng}\\p{sc=Deva}&"
                                        "\\p{Indic_Syllabic_Category=Virama}]", status);
    fLinkingConsonantSet = new UnicodeSet(u"[\\p{Gujr}\\p{sc=Telu}\\p{sc=Mlym}\\p{sc=Orya}\\p{sc=Beng}\\p{sc=Deva}&"
                                        "\\p{Indic_Syllabic_Category=Consonant}]", status);
    fExtCccZwjSet     = new UnicodeSet(u"[[\\p{gcb=Extend}-\\p{ccc=0}] \\p{gcb=ZWJ}]", status);
    fAnySet           = new UnicodeSet(0, 0x10ffff);

    // Create sets of characters, and add the names of the above character sets.
    // In each new ICU release, add new names corresponding to the sets above.
    fSets             = new UVector(status);

    // Important: Keep class names the same as the class contents.
    fSets->addElement(fCRLFSet, status); classNames.push_back("CRLF");
    fSets->addElement(fControlSet, status); classNames.push_back("Control");
    fSets->addElement(fExtendSet, status); classNames.push_back("Extended");
    fSets->addElement(fRegionalIndicatorSet, status); classNames.push_back("RegionalIndicator");
    if (!fPrependSet->isEmpty()) {
        fSets->addElement(fPrependSet, status); classNames.push_back("Prepend");
    }
    fSets->addElement(fSpacingSet, status); classNames.push_back("Spacing");
    fSets->addElement(fHangulSet, status); classNames.push_back("Hangul");
    fSets->addElement(fZWJSet, status); classNames.push_back("ZWJ");
    fSets->addElement(fExtendedPictSet, status); classNames.push_back("ExtendedPict");
    fSets->addElement(fViramaSet, status); classNames.push_back("Virama");
    fSets->addElement(fLinkingConsonantSet, status); classNames.push_back("LinkingConsonant");
    fSets->addElement(fExtCccZwjSet, status); classNames.push_back("ExtCcccZwj");
    fSets->addElement(fAnySet, status); classNames.push_back("Any");

    if (U_FAILURE(status)) {
        deferredStatus = status;
    }
}


void RBBICharMonkey::setText(const UnicodeString &s) {
    fText = &s;
    prepareAppliedRules(s.length());
}



int32_t RBBICharMonkey::next(int32_t prevPos) {
    int    p0, p1, p2, p3;    // Indices of the significant code points around the
                              //   break position being tested.  The candidate break
                              //   location is before p2.

    int     breakPos = -1;

    UChar32 c0, c1, c2, c3;   // The code points at p0, p1, p2 & p3.
    UChar32 cBase;            // for (X Extend*) patterns, the X character.

    if (U_FAILURE(deferredStatus)) {
        return -1;
    }

    // Previous break at end of string.  return DONE.
    if (prevPos >= fText->length()) {
        return -1;
    }

    p0 = p1 = p2 = p3 = prevPos;
    c3 =  fText->char32At(prevPos);
    c0 = c1 = c2 = cBase = 0;
    (void)p0;   // suppress set but not used warning.
    (void)c0;

    // Loop runs once per "significant" character position in the input text.
    for (;;) {
        // Move all of the positions forward in the input string.
        p0 = p1;  c0 = c1;
        p1 = p2;  c1 = c2;
        p2 = p3;  c2 = c3;

        // Advance p3 by one codepoint
        p3 = fText->moveIndex32(p3, 1);
        c3 = fText->char32At(p3);

        if (p1 == p2) {
            // Still warming up the loop.  (won't work with zero length strings, but we don't care)
            continue;
        }

        if (p2 == fText->length()) {
            setAppliedRule(p2, "End of String");
            break;
        }

        //     No Extend or Format characters may appear between the CR and LF,
        //     which requires the additional check for p2 immediately following p1.
        //
        if (c1==0x0D && c2==0x0A && p1==(p2-1)) {
          setAppliedRule(p2, "GB3   CR x LF");
          continue;
        }

        if (fControlSet->contains(c1) ||
            c1 == 0x0D ||
            c1 == 0x0A)  {
          setAppliedRule(p2, "GB4   ( Control | CR | LF ) <break>");
          break;
        }

        if (fControlSet->contains(c2) ||
            c2 == 0x0D ||
            c2 == 0x0A)  {
            setAppliedRule(p2, "GB5   <break>  ( Control | CR | LF )");
            break;
        }

        if (fLSet->contains(c1) &&
               (fLSet->contains(c2)  ||
                fVSet->contains(c2)  ||
                fLVSet->contains(c2) ||
                fLVTSet->contains(c2))) {
            setAppliedRule(p2, "GB6   L x ( L | V | LV | LVT )");
            continue;
        }

        if ((fLVSet->contains(c1) || fVSet->contains(c1)) &&
            (fVSet->contains(c2) || fTSet->contains(c2)))  {
            setAppliedRule(p2, "GB7    ( LV | V )  x  ( V | T )");
            continue;
        }

        if ((fLVTSet->contains(c1) || fTSet->contains(c1)) &&
            fTSet->contains(c2))  {
            setAppliedRule(p2, "GB8   ( LVT | T)  x T");
            continue;
        }

        if (fExtendSet->contains(c2) || fZWJSet->contains(c2))  {
            if (!fExtendSet->contains(c1)) {
                cBase = c1;
            }
            setAppliedRule(p2, "GB9   x (Extend | ZWJ)");
            continue;
        }

        if (fSpacingSet->contains(c2)) {
            setAppliedRule(p2, "GB9a  x  SpacingMark");
            continue;
        }

        if (fPrependSet->contains(c1)) {
            setAppliedRule(p2, "GB9b  Prepend x");
            continue;
        }

        //   Note: Viramas are also included in the ExtCccZwj class.
        if (fLinkingConsonantSet->contains(c2)) {
            int pi = p1;
            bool sawVirama = false;
            while (pi > 0 && fExtCccZwjSet->contains(fText->char32At(pi))) {
                if (fViramaSet->contains(fText->char32At(pi))) {
                    sawVirama = true;
                }
                pi = fText->moveIndex32(pi, -1);
            }
            if (sawVirama && fLinkingConsonantSet->contains(fText->char32At(pi))) {
              setAppliedRule(p2, "GB9.3  LinkingConsonant ExtCccZwj* Virama ExtCccZwj* × LinkingConsonant");
              continue;
            }
        }

        if (fExtendedPictSet->contains(cBase) && fZWJSet->contains(c1) && fExtendedPictSet->contains(c2)) {
          setAppliedRule(p2, "GB11  Extended_Pictographic Extend * ZWJ x Extended_Pictographic");
          continue;
        }

        //                   Note: The first if condition is a little tricky. We only need to force
        //                      a break if there are three or more contiguous RIs. If there are
        //                      only two, a break following will occur via other rules, and will include
        //                      any trailing extend characters, which is needed behavior.
        if (fRegionalIndicatorSet->contains(c0) && fRegionalIndicatorSet->contains(c1)
                && fRegionalIndicatorSet->contains(c2)) {
          setAppliedRule(p2, "GB12-13  Regional_Indicator x Regional_Indicator");
          break;
        }
        if (fRegionalIndicatorSet->contains(c1) && fRegionalIndicatorSet->contains(c2)) {
          setAppliedRule(p2, "GB12-13  Regional_Indicator x Regional_Indicator");
          continue;
        }

        setAppliedRule(p2, "GB999 Any <break> Any");
        break;
    }

    breakPos = p2;
    return breakPos;
}



UVector  *RBBICharMonkey::charClasses() {
    return fSets;
}

RBBICharMonkey::~RBBICharMonkey() {
    delete fSets;
    delete fCRLFSet;
    delete fControlSet;
    delete fExtendSet;
    delete fRegionalIndicatorSet;
    delete fPrependSet;
    delete fSpacingSet;
    delete fLSet;
    delete fVSet;
    delete fTSet;
    delete fLVSet;
    delete fLVTSet;
    delete fHangulSet;
    delete fAnySet;
    delete fZWJSet;
    delete fExtendedPictSet;
    delete fViramaSet;
    delete fLinkingConsonantSet;
    delete fExtCccZwjSet;
}

//------------------------------------------------------------------------------------------
//
//   class RBBIWordMonkey      Word Break specific implementation
//                             of RBBIMonkeyKind.
//
//------------------------------------------------------------------------------------------
class RBBIWordMonkey: public RBBIMonkeyKind {
public:
    RBBIWordMonkey();
    virtual          ~RBBIWordMonkey();
    virtual  UVector *charClasses() override;
    virtual  void     setText(const UnicodeString &s) override;
    virtual int32_t   next(int32_t i) override;
private:
    UVector      *fSets;

    UnicodeSet  *fCRSet;
    UnicodeSet  *fLFSet;
    UnicodeSet  *fNewlineSet;
    UnicodeSet  *fRegionalIndicatorSet;
    UnicodeSet  *fKatakanaSet;
    UnicodeSet  *fHebrew_LetterSet;
    UnicodeSet  *fALetterSet;
    UnicodeSet  *fSingle_QuoteSet;
    UnicodeSet  *fDouble_QuoteSet;
    UnicodeSet  *fMidNumLetSet;
    UnicodeSet  *fMidLetterSet;
    UnicodeSet  *fMidNumSet;
    UnicodeSet  *fNumericSet;
    UnicodeSet  *fFormatSet;
    UnicodeSet  *fOtherSet = nullptr;
    UnicodeSet  *fExtendSet;
    UnicodeSet  *fExtendNumLetSet;
    UnicodeSet  *fWSegSpaceSet;
    UnicodeSet  *fDictionarySet = nullptr;
    UnicodeSet  *fZWJSet;
    UnicodeSet  *fExtendedPictSet;

    const UnicodeString  *fText;
};


RBBIWordMonkey::RBBIWordMonkey()
{
    UErrorCode  status = U_ZERO_ERROR;

    fSets            = new UVector(status);

    fCRSet            = new UnicodeSet(u"[\\p{Word_Break = CR}]",           status);
    fLFSet            = new UnicodeSet(u"[\\p{Word_Break = LF}]",           status);
    fNewlineSet       = new UnicodeSet(u"[\\p{Word_Break = Newline}]",      status);
    fKatakanaSet      = new UnicodeSet(u"[\\p{Word_Break = Katakana}]",     status);
    fRegionalIndicatorSet =  new UnicodeSet(u"[\\p{Word_Break = Regional_Indicator}]", status);
    fHebrew_LetterSet = new UnicodeSet(u"[\\p{Word_Break = Hebrew_Letter}]", status);
    fALetterSet       = new UnicodeSet(u"[\\p{Word_Break = ALetter}]", status);
    fSingle_QuoteSet  = new UnicodeSet(u"[\\p{Word_Break = Single_Quote}]",    status);
    fDouble_QuoteSet  = new UnicodeSet(u"[\\p{Word_Break = Double_Quote}]",    status);
    fMidNumLetSet     = new UnicodeSet(u"[\\p{Word_Break = MidNumLet}]",    status);
    fMidLetterSet     = new UnicodeSet(u"[\\p{Word_Break = MidLetter}]",    status);
    fMidNumSet        = new UnicodeSet(u"[\\p{Word_Break = MidNum}]",       status);
    fNumericSet       = new UnicodeSet(u"[\\p{Word_Break = Numeric}]", status);
    fFormatSet        = new UnicodeSet(u"[\\p{Word_Break = Format}]",       status);
    fExtendNumLetSet  = new UnicodeSet(u"[\\p{Word_Break = ExtendNumLet}]", status);
    // There are some sc=Hani characters with WB=Extend.
    // The break rules need to pick one or the other because
    // Extend overlapping with something else is messy.
    // For Unicode 13, we chose to keep U+16FF0 & U+16FF1
    // in $Han (for $dictionary) and out of $Extend.
    fExtendSet        = new UnicodeSet(u"[\\p{Word_Break = Extend}-[:Hani:]]", status);
    fWSegSpaceSet     = new UnicodeSet(u"[\\p{Word_Break = WSegSpace}]",    status);

    fZWJSet           = new UnicodeSet(u"[\\p{Word_Break = ZWJ}]",          status);
    fExtendedPictSet  = new UnicodeSet(u"[:Extended_Pictographic:]", status);
    if(U_FAILURE(status)) {
        IntlTest::gTest->errln("%s:%d %s", __FILE__, __LINE__, u_errorName(status));
        deferredStatus = status;
        return;
    }

    fDictionarySet = new UnicodeSet(u"[[\\uac00-\\ud7a3][:Han:][:Hiragana:]]", status);
    fDictionarySet->addAll(*fKatakanaSet);
    fDictionarySet->addAll(UnicodeSet(u"[\\p{LineBreak = Complex_Context}]", status));

    fALetterSet->removeAll(*fDictionarySet);

    fOtherSet        = new UnicodeSet();
    if(U_FAILURE(status)) {
        IntlTest::gTest->errln("%s:%d %s", __FILE__, __LINE__, u_errorName(status));
        deferredStatus = status;
        return;
    }

    fOtherSet->complement();
    fOtherSet->removeAll(*fCRSet);
    fOtherSet->removeAll(*fLFSet);
    fOtherSet->removeAll(*fNewlineSet);
    fOtherSet->removeAll(*fKatakanaSet);
    fOtherSet->removeAll(*fHebrew_LetterSet);
    fOtherSet->removeAll(*fALetterSet);
    fOtherSet->removeAll(*fSingle_QuoteSet);
    fOtherSet->removeAll(*fDouble_QuoteSet);
    fOtherSet->removeAll(*fMidLetterSet);
    fOtherSet->removeAll(*fMidNumSet);
    fOtherSet->removeAll(*fNumericSet);
    fOtherSet->removeAll(*fExtendNumLetSet);
    fOtherSet->removeAll(*fWSegSpaceSet);
    fOtherSet->removeAll(*fFormatSet);
    fOtherSet->removeAll(*fExtendSet);
    fOtherSet->removeAll(*fRegionalIndicatorSet);
    fOtherSet->removeAll(*fZWJSet);
    fOtherSet->removeAll(*fExtendedPictSet);

    // Inhibit dictionary characters from being tested at all.
    fOtherSet->removeAll(*fDictionarySet);

    // Add classes and their names
    fSets->addElement(fCRSet, status); classNames.push_back("CR");
    fSets->addElement(fLFSet, status); classNames.push_back("LF");
    fSets->addElement(fNewlineSet, status); classNames.push_back("Newline");
    fSets->addElement(fRegionalIndicatorSet, status); classNames.push_back("RegionalIndicator");
    fSets->addElement(fHebrew_LetterSet, status); classNames.push_back("Hebrew");
    fSets->addElement(fALetterSet, status); classNames.push_back("ALetter");
    fSets->addElement(fSingle_QuoteSet, status); classNames.push_back("Single Quote");
    fSets->addElement(fDouble_QuoteSet, status); classNames.push_back("Double Quote");
    // Omit Katakana from fSets, which omits Katakana characters
    // from the test data. They are all in the dictionary set,
    // which this (old, to be retired) monkey test cannot handle.
    //fSets->addElement(fKatakanaSet, status);

    fSets->addElement(fMidLetterSet, status); classNames.push_back("MidLetter");
    fSets->addElement(fMidNumLetSet, status); classNames.push_back("MidNumLet");
    fSets->addElement(fMidNumSet, status); classNames.push_back("MidNum");
    fSets->addElement(fNumericSet, status); classNames.push_back("Numeric");
    fSets->addElement(fFormatSet, status); classNames.push_back("Format");
    fSets->addElement(fExtendSet, status); classNames.push_back("Extend");
    fSets->addElement(fOtherSet, status); classNames.push_back("Other");
    fSets->addElement(fExtendNumLetSet, status); classNames.push_back("ExtendNumLet");
    fSets->addElement(fWSegSpaceSet, status); classNames.push_back("WSegSpace");

    fSets->addElement(fZWJSet, status); classNames.push_back("ZWJ");
    fSets->addElement(fExtendedPictSet, status); classNames.push_back("ExtendedPict");

    if (U_FAILURE(status)) {
        deferredStatus = status;
    }
}

void RBBIWordMonkey::setText(const UnicodeString &s) {
    fText       = &s;
    prepareAppliedRules(s.length());
}


int32_t RBBIWordMonkey::next(int32_t prevPos) {
    int    p0, p1, p2, p3;    // Indices of the significant code points around the
                              //   break position being tested.  The candidate break
                              //   location is before p2.

    int     breakPos = -1;

    UChar32 c0, c1, c2, c3;   // The code points at p0, p1, p2 & p3.

    if (U_FAILURE(deferredStatus)) {
        return -1;
    }

    // Prev break at end of string.  return DONE.
    if (prevPos >= fText->length()) {
        return -1;
    }
    p0 = p1 = p2 = p3 = prevPos;
    c3 =  fText->char32At(prevPos);
    c0 = c1 = c2 = 0;
    (void)p0;       // Suppress set but not used warning.

    // Loop runs once per "significant" character position in the input text.
    for (;;) {
        // Move all of the positions forward in the input string.
        p0 = p1;  c0 = c1;
        p1 = p2;  c1 = c2;
        p2 = p3;  c2 = c3;

        // Advance p3 by    X(Extend | Format)*   Rule 4
        //    But do not advance over Extend & Format following a new line. (Unicode 5.1 change)
        do {
            p3 = fText->moveIndex32(p3, 1);
            c3 = fText->char32At(p3);
            if (fCRSet->contains(c2) || fLFSet->contains(c2) || fNewlineSet->contains(c2)) {
               break;
            }
        }
        while (fFormatSet->contains(c3) || fExtendSet->contains(c3) || fZWJSet->contains(c3));


        if (p1 == p2) {
            // Still warming up the loop.  (won't work with zero length strings, but we don't care)
            continue;
        }

        if (p2 == fText->length()) {
            // Reached end of string.  Always a break position.
            break;
        }

        //     No Extend or Format characters may appear between the CR and LF,
        //     which requires the additional check for p2 immediately following p1.
        //
        if (c1==0x0D && c2==0x0A) {
          setAppliedRule(p2, "WB3   CR x LF");
          continue;
        }

        if (fCRSet->contains(c1) || fLFSet->contains(c1) || fNewlineSet->contains(c1)) {
            setAppliedRule(p2, "WB3a  Break before and after newlines (including CR and LF)");
            break;
        }
        if (fCRSet->contains(c2) || fLFSet->contains(c2) || fNewlineSet->contains(c2)) {
            setAppliedRule(p2, "WB3a  Break before and after newlines (including CR and LF)");
            break;
        }

        //              Not ignoring extend chars, so peek into input text to
        //              get the potential ZWJ, the character immediately preceding c2.
        //              Sloppy UChar32 indexing: p2-1 may reference trail half
        //              but char32At will get the full code point.
        if (fZWJSet->contains(fText->char32At(p2 - 1)) && fExtendedPictSet->contains(c2)){
            setAppliedRule(p2, "WB3c  ZWJ x Extended_Pictographic");
            continue;
        }

        if (fWSegSpaceSet->contains(fText->char32At(p2-1)) && fWSegSpaceSet->contains(c2)) {
            setAppliedRule(p2, "WB3d  Keep horizontal whitespace together.");
            continue;
        }

        if ((fALetterSet->contains(c1) || fHebrew_LetterSet->contains(c1)) &&
            (fALetterSet->contains(c2) || fHebrew_LetterSet->contains(c2)))  {
            setAppliedRule(p2, "WB4   (ALetter | Hebrew_Letter) x (ALetter | Hebrew_Letter)");
            continue;
        }

        if ( (fALetterSet->contains(c1) || fHebrew_LetterSet->contains(c1))   &&
             (fMidLetterSet->contains(c2) || fMidNumLetSet->contains(c2) || fSingle_QuoteSet->contains(c2)) &&
             (fALetterSet->contains(c3) || fHebrew_LetterSet->contains(c3))) {
            setAppliedRule(p2,
                           "WB6   (ALetter | Hebrew_Letter)  x  (MidLetter | MidNumLet | Single_Quote) (ALetter _Letter)");
            continue;
        }

        if ((fALetterSet->contains(c0) || fHebrew_LetterSet->contains(c0)) &&
            (fMidLetterSet->contains(c1) || fMidNumLetSet->contains(c1) || fSingle_QuoteSet->contains(c1)) &&
            (fALetterSet->contains(c2) || fHebrew_LetterSet->contains(c2))) {
            setAppliedRule(p2,
                           "WB7   (ALetter | Hebrew_Letter) (MidLetter | MidNumLet | Single_Quote)  x  (ALetter | Hebrew_Letter)");
            continue;
        }

        if (fHebrew_LetterSet->contains(c1) && fSingle_QuoteSet->contains(c2)) {
            setAppliedRule(p2, "WB7a  Hebrew_Letter x Single_Quote");
            continue;
        }

          if (fHebrew_LetterSet->contains(c1) && fDouble_QuoteSet->contains(c2) && fHebrew_LetterSet->contains(c3)) {
            setAppliedRule(p2, "WB7b  Hebrew_Letter x Double_Quote Hebrew_Letter");
            continue;
        }

        if (fHebrew_LetterSet->contains(c0) && fDouble_QuoteSet->contains(c1) && fHebrew_LetterSet->contains(c2)) {
            setAppliedRule(p2, "WB7c  Hebrew_Letter Double_Quote x Hebrew_Letter");
            continue;
        }

        if (fNumericSet->contains(c1) &&
            fNumericSet->contains(c2)) {
            setAppliedRule(p2, "WB8   Numeric x Numeric");
            continue;
        }

        if ((fALetterSet->contains(c1) || fHebrew_LetterSet->contains(c1)) &&
            fNumericSet->contains(c2)) {
            setAppliedRule(p2, "WB9   (ALetter | Hebrew_Letter) x Numeric");
            continue;
        }

        if (fNumericSet->contains(c1) &&
            (fALetterSet->contains(c2) || fHebrew_LetterSet->contains(c2)))  {
            setAppliedRule(p2, "WB10   Numeric x (ALetter | Hebrew_Letter)");
            continue;
        }

          if (fNumericSet->contains(c0) &&
            (fMidNumSet->contains(c1) || fMidNumLetSet->contains(c1) || fSingle_QuoteSet->contains(c1))  &&
            fNumericSet->contains(c2)) {
            setAppliedRule(p2, "WB11  Numeric (MidNum | MidNumLet | Single_Quote)  x  Numeric");
            continue;
        }

        if (fNumericSet->contains(c1) &&
            (fMidNumSet->contains(c2) || fMidNumLetSet->contains(c2) || fSingle_QuoteSet->contains(c2))  &&
            fNumericSet->contains(c3)) {
            setAppliedRule(p2, "WB12  Numeric x (MidNum | MidNumLet | SingleQuote) Numeric");
            continue;
        }

        //            Note: matches UAX 29 rules, but doesn't come into play for ICU because
        //                  all Katakana are handled by the dictionary breaker.
        if (fKatakanaSet->contains(c1) &&
            fKatakanaSet->contains(c2))  {
            setAppliedRule(p2, "WB13  Katakana x Katakana");
            continue;
        }

        if ((fALetterSet->contains(c1) || fHebrew_LetterSet->contains(c1) ||fNumericSet->contains(c1) ||
             fKatakanaSet->contains(c1) || fExtendNumLetSet->contains(c1)) &&
             fExtendNumLetSet->contains(c2)) {
            setAppliedRule(p2,
                           "WB13a (ALetter | Hebrew_Letter | Numeric | KataKana | ExtendNumLet) x ExtendNumLet");
            continue;
        }

        if (fExtendNumLetSet->contains(c1) &&
                (fALetterSet->contains(c2) || fHebrew_LetterSet->contains(c2) ||
                 fNumericSet->contains(c2) || fKatakanaSet->contains(c2)))  {
            setAppliedRule(p2, "WB13b ExtendNumLet x (ALetter | Hebrew_Letter | Numeric | Katakana)");
            continue;
        }

        if (fRegionalIndicatorSet->contains(c0) && fRegionalIndicatorSet->contains(c1)) {
            setAppliedRule(p2, "WB15 - WB17   Group pairs of Regional Indicators.");
            break;
        }
        if (fRegionalIndicatorSet->contains(c1) && fRegionalIndicatorSet->contains(c2)) {
            setAppliedRule(p2, "WB15 - WB17   Group pairs of Regional Indicators.");
            continue;
        }

        setAppliedRule(p2, "WB999");
        break;
    }

    breakPos = p2;
    return breakPos;
}


UVector  *RBBIWordMonkey::charClasses() {
    return fSets;
}

RBBIWordMonkey::~RBBIWordMonkey() {
    delete fSets;
    delete fCRSet;
    delete fLFSet;
    delete fNewlineSet;
    delete fKatakanaSet;
    delete fHebrew_LetterSet;
    delete fALetterSet;
    delete fSingle_QuoteSet;
    delete fDouble_QuoteSet;
    delete fMidNumLetSet;
    delete fMidLetterSet;
    delete fMidNumSet;
    delete fNumericSet;
    delete fFormatSet;
    delete fExtendSet;
    delete fExtendNumLetSet;
    delete fWSegSpaceSet;
    delete fRegionalIndicatorSet;
    delete fDictionarySet;
    delete fOtherSet;
    delete fZWJSet;
    delete fExtendedPictSet;
}




//------------------------------------------------------------------------------------------
//
//   class RBBISentMonkey      Sentence Break specific implementation
//                             of RBBIMonkeyKind.
//
//------------------------------------------------------------------------------------------
class RBBISentMonkey: public RBBIMonkeyKind {
public:
    RBBISentMonkey();
    virtual          ~RBBISentMonkey();
    virtual  UVector *charClasses() override;
    virtual  void     setText(const UnicodeString &s) override;
    virtual int32_t   next(int32_t i) override;
private:
    int               moveBack(int posFrom);
    int               moveForward(int posFrom);
    UChar32           cAt(int pos);

    UVector      *fSets;

    UnicodeSet  *fSepSet;
    UnicodeSet  *fFormatSet;
    UnicodeSet  *fSpSet;
    UnicodeSet  *fLowerSet;
    UnicodeSet  *fUpperSet;
    UnicodeSet  *fOLetterSet;
    UnicodeSet  *fNumericSet;
    UnicodeSet  *fATermSet;
    UnicodeSet  *fSContinueSet;
    UnicodeSet  *fSTermSet;
    UnicodeSet  *fCloseSet;
    UnicodeSet  *fOtherSet;
    UnicodeSet  *fExtendSet;

    const UnicodeString  *fText;
};

RBBISentMonkey::RBBISentMonkey()
{
    UErrorCode  status = U_ZERO_ERROR;

    fSets            = new UVector(status);

    //  Separator Set Note:  Beginning with Unicode 5.1, CR and LF were removed from the separator
    //                       set and made into character classes of their own.  For the monkey impl,
    //                       they remain in SEP, since Sep always appears with CR and LF in the rules.
    fSepSet          = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Sentence_Break = Sep} \\u000a \\u000d]"),     status);
    fFormatSet       = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Sentence_Break = Format}]"),    status);
    fSpSet           = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Sentence_Break = Sp}]"),        status);
    fLowerSet        = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Sentence_Break = Lower}]"),     status);
    fUpperSet        = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Sentence_Break = Upper}]"),     status);
    fOLetterSet      = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Sentence_Break = OLetter}]"),   status);
    fNumericSet      = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Sentence_Break = Numeric}]"),   status);
    fATermSet        = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Sentence_Break = ATerm}]"),     status);
    fSContinueSet    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Sentence_Break = SContinue}]"), status);
    fSTermSet        = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Sentence_Break = STerm}]"),     status);
    fCloseSet        = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Sentence_Break = Close}]"),     status);
    fExtendSet       = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Sentence_Break = Extend}]"),    status);
    fOtherSet        = new UnicodeSet();

    if(U_FAILURE(status)) {
      deferredStatus = status;
      return;
    }

    fOtherSet->complement();
    fOtherSet->removeAll(*fSepSet);
    fOtherSet->removeAll(*fFormatSet);
    fOtherSet->removeAll(*fSpSet);
    fOtherSet->removeAll(*fLowerSet);
    fOtherSet->removeAll(*fUpperSet);
    fOtherSet->removeAll(*fOLetterSet);
    fOtherSet->removeAll(*fNumericSet);
    fOtherSet->removeAll(*fATermSet);
    fOtherSet->removeAll(*fSContinueSet);
    fOtherSet->removeAll(*fSTermSet);
    fOtherSet->removeAll(*fCloseSet);
    fOtherSet->removeAll(*fExtendSet);

    fSets->addElement(fSepSet, status); classNames.push_back("Sep");
    fSets->addElement(fFormatSet, status); classNames.push_back("Format");
    fSets->addElement(fSpSet, status); classNames.push_back("Sp");
    fSets->addElement(fLowerSet, status); classNames.push_back("Lower");
    fSets->addElement(fUpperSet, status); classNames.push_back("Upper");
    fSets->addElement(fOLetterSet, status); classNames.push_back("OLetter");
    fSets->addElement(fNumericSet, status); classNames.push_back("Numeric");
    fSets->addElement(fATermSet, status); classNames.push_back("ATerm");
    fSets->addElement(fSContinueSet, status); classNames.push_back("SContinue");
    fSets->addElement(fSTermSet, status); classNames.push_back("STerm");
    fSets->addElement(fCloseSet, status); classNames.push_back("Close");
    fSets->addElement(fOtherSet, status); classNames.push_back("Other");
    fSets->addElement(fExtendSet, status); classNames.push_back("Extend");

    if (U_FAILURE(status)) {
        deferredStatus = status;
    }
}



void RBBISentMonkey::setText(const UnicodeString &s) {
    fText       = &s;
    prepareAppliedRules(s.length());
}

UVector  *RBBISentMonkey::charClasses() {
    return fSets;
}

//  moveBack()   Find the "significant" code point preceding the index i.
//               Skips over ($Extend | $Format)* .
//
int RBBISentMonkey::moveBack(int i) {
    if (i <= 0) {
        return -1;
    }
    UChar32   c;
    int32_t   j = i;
    do {
        j = fText->moveIndex32(j, -1);
        c = fText->char32At(j);
    }
    while (j>0 &&(fFormatSet->contains(c) || fExtendSet->contains(c)));
    return j;

 }


int RBBISentMonkey::moveForward(int i) {
    if (i>=fText->length()) {
        return fText->length();
    }
    UChar32   c;
    int32_t   j = i;
    do {
        j = fText->moveIndex32(j, 1);
        c = cAt(j);
    }
    while (fFormatSet->contains(c) || fExtendSet->contains(c));
    return j;
}

UChar32 RBBISentMonkey::cAt(int pos) {
    if (pos<0 || pos>=fText->length()) {
        return -1;
    } else {
        return fText->char32At(pos);
    }
}

int32_t RBBISentMonkey::next(int32_t prevPos) {
    int    p0, p1, p2, p3;    // Indices of the significant code points around the
                              //   break position being tested.  The candidate break
                              //   location is before p2.

    int     breakPos = -1;

    UChar32 c0, c1, c2, c3;   // The code points at p0, p1, p2 & p3.
    UChar32 c;

    if (U_FAILURE(deferredStatus)) {
        return -1;
    }

    // Prev break at end of string.  return DONE.
    if (prevPos >= fText->length()) {
        return -1;
    }
    p0 = p1 = p2 = p3 = prevPos;
    c3 =  fText->char32At(prevPos);
    c0 = c1 = c2 = 0;
    (void)p0;     // Suppress set but not used warning.

    // Loop runs once per "significant" character position in the input text.
    for (;;) {
        // Move all of the positions forward in the input string.
        p0 = p1;  c0 = c1;
        p1 = p2;  c1 = c2;
        p2 = p3;  c2 = c3;

        // Advance p3 by    X(Extend | Format)*   Rule 4
        p3 = moveForward(p3);
        c3 = cAt(p3);

        if (c1==0x0d && c2==0x0a && p2==(p1+1)) {
            setAppliedRule(p2, "SB3   CR x LF");
            continue;
        }

        if (fSepSet->contains(c1)) {
            p2 = p1+1;   // Separators don't combine with Extend or Format.

            setAppliedRule(p2, "SB4   Sep  <break>");
            break;
        }

        if (p2 >= fText->length()) {
            // Reached end of string.  Always a break position.
            setAppliedRule(p2, "SB4   Sep  <break>");
            break;
        }

        if (p2 == prevPos) {
            // Still warming up the loop.  (won't work with zero length strings, but we don't care)
            setAppliedRule(p2, "SB4   Sep  <break>");
            continue;
        }

        if (fATermSet->contains(c1) &&  fNumericSet->contains(c2))  {
            setAppliedRule(p2, "SB6   ATerm x Numeric");
            continue;
        }

          if ((fUpperSet->contains(c0) || fLowerSet->contains(c0)) &&
                fATermSet->contains(c1) && fUpperSet->contains(c2)) {
            setAppliedRule(p2, "SB7   (Upper | Lower) ATerm  x  Uppper");
            continue;
        }

        //           Note:  STerm | ATerm are added to the negated part of the expression by a
        //                  note to the Unicode 5.0 documents.
        int p8 = p1;
        while (fSpSet->contains(cAt(p8))) {
            p8 = moveBack(p8);
        }
        while (fCloseSet->contains(cAt(p8))) {
            p8 = moveBack(p8);
        }
        if (fATermSet->contains(cAt(p8))) {
            p8=p2;
            for (;;) {
                c = cAt(p8);
                if (c==-1 || fOLetterSet->contains(c) || fUpperSet->contains(c) ||
                    fLowerSet->contains(c) || fSepSet->contains(c) ||
                    fATermSet->contains(c) || fSTermSet->contains(c))  {

                    setAppliedRule(p2,
                                   "SB8   ATerm Close* Sp*  x  (not (OLettter | Upper | Lower | Sep | STerm | ATerm))* ");
                    break;
                }
                p8 = moveForward(p8);
            }
            if (fLowerSet->contains(cAt(p8))) {

                setAppliedRule(p2,
                               "SB8   ATerm Close* Sp*  x  (not (OLettter | Upper | Lower | Sep | STerm | ATerm))* ");
                continue;
            }
        }

        if (fSContinueSet->contains(c2) || fSTermSet->contains(c2) || fATermSet->contains(c2)) {
            p8 = p1;
            while (fSpSet->contains(cAt(p8))) {
                p8 = moveBack(p8);
            }
            while (fCloseSet->contains(cAt(p8))) {
                p8 = moveBack(p8);
            }
            c = cAt(p8);
            if (fSTermSet->contains(c) || fATermSet->contains(c)) {
                setAppliedRule(p2, "SB8a  (STerm | ATerm) Close* Sp* x (SContinue | STerm | ATerm)");
                continue;
            }
        }

        int p9 = p1;
        while (fCloseSet->contains(cAt(p9))) {
            p9 = moveBack(p9);
        }
        c = cAt(p9);
        if ((fSTermSet->contains(c) || fATermSet->contains(c))) {
            if (fCloseSet->contains(c2) || fSpSet->contains(c2) || fSepSet->contains(c2)) {

                setAppliedRule(p2, "SB9  (STerm | ATerm) Close*  x  (Close | Sp | Sep | CR | LF)");
                continue;
            }
        }

        int p10 = p1;
        while (fSpSet->contains(cAt(p10))) {
            p10 = moveBack(p10);
        }
        while (fCloseSet->contains(cAt(p10))) {
            p10 = moveBack(p10);
        }
        if (fSTermSet->contains(cAt(p10)) || fATermSet->contains(cAt(p10))) {
            if (fSpSet->contains(c2) || fSepSet->contains(c2)) {
                setAppliedRule(p2, "SB10  (Sterm | ATerm) Close* Sp*  x  (Sp | Sep | CR | LF)");
                continue;
            }
        }

        int p11 = p1;
        if (fSepSet->contains(cAt(p11))) {
            p11 = moveBack(p11);
        }
        while (fSpSet->contains(cAt(p11))) {
            p11 = moveBack(p11);
        }
        while (fCloseSet->contains(cAt(p11))) {
            p11 = moveBack(p11);
        }
        if (fSTermSet->contains(cAt(p11)) || fATermSet->contains(cAt(p11))) {
          setAppliedRule(p2, "SB11  (STerm | ATerm) Close* Sp* (Sep | CR | LF)?  <break>");
            break;
        }

        setAppliedRule(p2, "SB12  Any x Any");
        continue;
    }

    breakPos = p2;
    return breakPos;
}

RBBISentMonkey::~RBBISentMonkey() {
    delete fSets;
    delete fSepSet;
    delete fFormatSet;
    delete fSpSet;
    delete fLowerSet;
    delete fUpperSet;
    delete fOLetterSet;
    delete fNumericSet;
    delete fATermSet;
    delete fSContinueSet;
    delete fSTermSet;
    delete fCloseSet;
    delete fOtherSet;
    delete fExtendSet;
}



//-------------------------------------------------------------------------------------------
//
//  RBBILineMonkey
//
//-------------------------------------------------------------------------------------------

class RBBILineMonkey: public RBBIMonkeyKind {
public:
    RBBILineMonkey();
    virtual          ~RBBILineMonkey();
    virtual  UVector *charClasses() override;
    virtual  void     setText(const UnicodeString &s) override;
    virtual  int32_t  next(int32_t i) override;
    virtual  void     rule9Adjust(int32_t pos, UChar32 *posChar, int32_t *nextPos, UChar32 *nextChar);
private:
    UVector      *fSets;

    UnicodeSet  *fBK;
    UnicodeSet  *fCR;
    UnicodeSet  *fLF;
    UnicodeSet  *fCM;
    UnicodeSet  *fNL;
    UnicodeSet  *fSG;
    UnicodeSet  *fWJ;
    UnicodeSet  *fZW;
    UnicodeSet  *fGL;
    UnicodeSet  *fCB;
    UnicodeSet  *fSP;
    UnicodeSet  *fB2;
    UnicodeSet  *fBA;
    UnicodeSet  *fBB;
    UnicodeSet  *fHH;
    UnicodeSet  *fHY;
    UnicodeSet  *fH2;
    UnicodeSet  *fH3;
    UnicodeSet  *fCL;
    UnicodeSet  *fCP;
    UnicodeSet  *fEX;
    UnicodeSet  *fIN;
    UnicodeSet  *fJL;
    UnicodeSet  *fJV;
    UnicodeSet  *fJT;
    UnicodeSet  *fNS;
    UnicodeSet  *fOP;
    UnicodeSet  *fQU;
    UnicodeSet  *fIS;
    UnicodeSet  *fNU;
    UnicodeSet  *fPO;
    UnicodeSet  *fPR;
    UnicodeSet  *fSY;
    UnicodeSet  *fAI;
    UnicodeSet  *fAL;
    UnicodeSet  *fCJ;
    UnicodeSet  *fHL;
    UnicodeSet  *fID;
    UnicodeSet  *fRI;
    UnicodeSet  *fXX;
    UnicodeSet  *fEB;
    UnicodeSet  *fEM;
    UnicodeSet  *fZWJ;
    UnicodeSet  *fOP30;
    UnicodeSet  *fCP30;
    UnicodeSet  *fExtPictUnassigned;

    BreakIterator        *fCharBI;
    const UnicodeString  *fText;
    RegexMatcher         *fNumberMatcher;
};

RBBILineMonkey::RBBILineMonkey() :
    RBBIMonkeyKind(),
    fSets(NULL),

    fCharBI(NULL),
    fText(NULL),
    fNumberMatcher(NULL)

{
    if (U_FAILURE(deferredStatus)) {
        return;
    }

    UErrorCode  status = U_ZERO_ERROR;

    fSets  = new UVector(status);

    fBK    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_Break=BK}]"), status);
    fCR    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=CR}]"), status);
    fLF    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=LF}]"), status);
    fCM    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=CM}]"), status);
    fNL    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=NL}]"), status);
    fWJ    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=WJ}]"), status);
    fZW    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=ZW}]"), status);
    fGL    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=GL}]"), status);
    fCB    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=CB}]"), status);
    fSP    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=SP}]"), status);
    fB2    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=B2}]"), status);
    fBA    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=BA}]"), status);
    fBB    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=BB}]"), status);
    fHH    = new UnicodeSet();
    fHY    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=HY}]"), status);
    fH2    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=H2}]"), status);
    fH3    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=H3}]"), status);
    fCL    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=CL}]"), status);
    fCP    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=CP}]"), status);
    fEX    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=EX}]"), status);
    fIN    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=IN}]"), status);
    fJL    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=JL}]"), status);
    fJV    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=JV}]"), status);
    fJT    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=JT}]"), status);
    fNS    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=NS}]"), status);
    fOP    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=OP}]"), status);
    fQU    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=QU}]"), status);
    fIS    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=IS}]"), status);
    fNU    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=NU}]"), status);
    fPO    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=PO}]"), status);
    fPR    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=PR}]"), status);
    fSY    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=SY}]"), status);
    fAI    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=AI}]"), status);
    fAL    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=AL}]"), status);
    fCJ    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=CJ}]"), status);
    fHL    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=HL}]"), status);
    fID    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=ID}]"), status);
    fRI    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=RI}]"), status);
    fSG    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\ud800-\\udfff]"), status);
    fXX    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=XX}]"), status);
    fEB    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=EB}]"), status);
    fEM    = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=EM}]"), status);
    fZWJ   = new UnicodeSet(UNICODE_STRING_SIMPLE("[\\p{Line_break=ZWJ}]"), status);
    fOP30  = new UnicodeSet(u"[\\p{Line_break=OP}-[\\p{ea=F}\\p{ea=W}\\p{ea=H}]]", status);
    fCP30  = new UnicodeSet(u"[\\p{Line_break=CP}-[\\p{ea=F}\\p{ea=W}\\p{ea=H}]]", status);
    fExtPictUnassigned = new UnicodeSet(u"[\\p{Extended_Pictographic}&\\p{Cn}]", status);

    if (U_FAILURE(status)) {
        deferredStatus = status;
        return;
    }

    fAL->addAll(*fXX);     // Default behavior for XX is identical to AL
    fAL->addAll(*fAI);     // Default behavior for AI is identical to AL
    fAL->addAll(*fSG);     // Default behavior for SG is identical to AL.

    fNS->addAll(*fCJ);     // Default behavior for CJ is identical to NS.
    fCM->addAll(*fZWJ);    // ZWJ behaves as a CM.

    fHH->add(u'\u2010');   // Hyphen, '‐'

    // Sets and names.
    fSets->addElement(fBK, status); classNames.push_back("fBK");
    fSets->addElement(fCR, status); classNames.push_back("fCR");
    fSets->addElement(fLF, status); classNames.push_back("fLF");
    fSets->addElement(fCM, status); classNames.push_back("fCM");
    fSets->addElement(fNL, status); classNames.push_back("fNL");
    fSets->addElement(fWJ, status); classNames.push_back("fWJ");
    fSets->addElement(fZW, status); classNames.push_back("fZW");
    fSets->addElement(fGL, status); classNames.push_back("fGL");
    fSets->addElement(fCB, status); classNames.push_back("fCB");
    fSets->addElement(fSP, status); classNames.push_back("fSP");
    fSets->addElement(fB2, status); classNames.push_back("fB2");
    fSets->addElement(fBA, status); classNames.push_back("fBA");
    fSets->addElement(fBB, status); classNames.push_back("fBB");
    fSets->addElement(fHY, status); classNames.push_back("fHY");
    fSets->addElement(fH2, status); classNames.push_back("fH2");
    fSets->addElement(fH3, status); classNames.push_back("fH3");
    fSets->addElement(fCL, status); classNames.push_back("fCL");
    fSets->addElement(fCP, status); classNames.push_back("fCP");
    fSets->addElement(fEX, status); classNames.push_back("fEX");
    fSets->addElement(fIN, status); classNames.push_back("fIN");
    fSets->addElement(fJL, status); classNames.push_back("fJL");
    fSets->addElement(fJT, status); classNames.push_back("fJT");
    fSets->addElement(fJV, status); classNames.push_back("fJV");
    fSets->addElement(fNS, status); classNames.push_back("fNS");
    fSets->addElement(fOP, status); classNames.push_back("fOP");
    fSets->addElement(fQU, status); classNames.push_back("fQU");
    fSets->addElement(fIS, status); classNames.push_back("fIS");
    fSets->addElement(fNU, status); classNames.push_back("fNU");
    fSets->addElement(fPO, status); classNames.push_back("fPO");
    fSets->addElement(fPR, status); classNames.push_back("fPR");
    fSets->addElement(fSY, status); classNames.push_back("fSY");
    fSets->addElement(fAI, status); classNames.push_back("fAI");
    fSets->addElement(fAL, status); classNames.push_back("fAL");
    fSets->addElement(fHL, status); classNames.push_back("fHL");
    fSets->addElement(fID, status); classNames.push_back("fID");
    fSets->addElement(fRI, status); classNames.push_back("fRI");
    fSets->addElement(fSG, status); classNames.push_back("fSG");
    fSets->addElement(fEB, status); classNames.push_back("fEB");
    fSets->addElement(fEM, status); classNames.push_back("fEM");
    fSets->addElement(fZWJ, status); classNames.push_back("fZWJ");
    // TODO: fOP30 & fCP30 overlap with plain fOP. Probably OK, but fOP/CP chars will be over-represented.
    fSets->addElement(fOP30, status); classNames.push_back("fOP30");
    fSets->addElement(fCP30, status); classNames.push_back("fCP30");
    fSets->addElement(fExtPictUnassigned, status); classNames.push_back("fExtPictUnassigned");

    const char *rules =
            "((\\p{Line_Break=PR}|\\p{Line_Break=PO})(\\p{Line_Break=CM}|\\u200d)*)?"
            "((\\p{Line_Break=OP}|\\p{Line_Break=HY})(\\p{Line_Break=CM}|\\u200d)*)?"
            "((\\p{Line_Break=IS})(\\p{Line_Break=CM}|\\u200d)*)?"
            "\\p{Line_Break=NU}(\\p{Line_Break=CM}|\\u200d)*"
            "((\\p{Line_Break=NU}|\\p{Line_Break=IS}|\\p{Line_Break=SY})(\\p{Line_Break=CM}|\\u200d)*)*"
            "((\\p{Line_Break=CL}|\\p{Line_Break=CP})(\\p{Line_Break=CM}|\\u200d)*)?"
            "((\\p{Line_Break=PR}|\\p{Line_Break=PO})(\\p{Line_Break=CM}|\\u200d)*)?";

    fNumberMatcher = new RegexMatcher(
        UnicodeString(rules, -1, US_INV), 0, status);

    fCharBI = BreakIterator::createCharacterInstance(Locale::getEnglish(), status);

    if (U_FAILURE(status)) {
        deferredStatus = status;
    }

}


void RBBILineMonkey::setText(const UnicodeString &s) {
    fText       = &s;
    fCharBI->setText(s);
    prepareAppliedRules(s.length());
    fNumberMatcher->reset(s);
}

//
//  rule9Adjust
//     Line Break TR rules 9 and 10 implementation.
//     This deals with combining marks and other sequences that
//     that must be treated as if they were something other than what they actually are.
//
//     This is factored out into a separate function because it must be applied twice for
//     each potential break, once to the chars before the position being checked, then
//     again to the text following the possible break.
//
void RBBILineMonkey::rule9Adjust(int32_t pos, UChar32 *posChar, int32_t *nextPos, UChar32 *nextChar) {
    if (pos == -1) {
        // Invalid initial position.  Happens during the warmup iteration of the
        //   main loop in next().
        return;
    }

    int32_t  nPos = *nextPos;

    // LB 9  Keep combining sequences together.
    // advance over any CM class chars.  Note that Line Break CM is different
    // from the normal Grapheme Extend property.
    if (!(fSP->contains(*posChar) || fBK->contains(*posChar) || *posChar==0x0d ||
          *posChar==0x0a ||fNL->contains(*posChar) || fZW->contains(*posChar))) {
        for (;;) {
            *nextChar = fText->char32At(nPos);
            if (!fCM->contains(*nextChar)) {
                break;
            }
            nPos = fText->moveIndex32(nPos, 1);
        }
    }


    // LB 9 Treat X CM* as if it were x.
    //       No explicit action required.

    // LB 10  Treat any remaining combining mark as AL
    if (fCM->contains(*posChar)) {
        *posChar = u'A';
    }

    // Push the updated nextPos and nextChar back to our caller.
    // This only makes a difference if posChar got bigger by consuming a
    // combining sequence.
    *nextPos  = nPos;
    *nextChar = fText->char32At(nPos);
}



int32_t RBBILineMonkey::next(int32_t startPos) {
    UErrorCode status = U_ZERO_ERROR;
    int32_t    pos;       //  Index of the char following a potential break position
    UChar32    thisChar;  //  Character at above position "pos"

    int32_t    prevPos;   //  Index of the char preceding a potential break position
    UChar32    prevChar;  //  Character at above position.  Note that prevChar
                          //   and thisChar may not be adjacent because combining
                          //   characters between them will be ignored.

    int32_t    prevPosX2; //  Second previous character.  Wider context for LB21a.
    UChar32    prevCharX2;

    int32_t    nextPos;   //  Index of the next character following pos.
                          //     Usually skips over combining marks.
    int32_t    nextCPPos; //  Index of the code point following "pos."
                          //     May point to a combining mark.
    int32_t    tPos;      //  temp value.
    UChar32    c;

    if (U_FAILURE(deferredStatus)) {
        return -1;
    }

    if (startPos >= fText->length()) {
        return -1;
    }


    // Initial values for loop.  Loop will run the first time without finding breaks,
    //                           while the invalid values shift out and the "this" and
    //                           "prev" positions are filled in with good values.
    pos      = prevPos   = prevPosX2  = -1;    // Invalid value, serves as flag for initial loop iteration.
    thisChar = prevChar  = prevCharX2 = 0;
    nextPos  = nextCPPos = startPos;


    // Loop runs once per position in the test text, until a break position
    //  is found.
    for (;;) {
        prevPosX2 = prevPos;
        prevCharX2 = prevChar;

        prevPos   = pos;
        prevChar  = thisChar;

        pos       = nextPos;
        thisChar  = fText->char32At(pos);

        nextCPPos = fText->moveIndex32(pos, 1);
        nextPos   = nextCPPos;


        if (pos >= fText->length()) {
            setAppliedRule(pos, "LB2 - Break at end of text.");
            break;
        }


        //             We do this one out-of-order because the adjustment does not change anything
        //             that would match rules LB 3 - LB 6, but after the adjustment, LB 3-6 do need to
        //             be applied.
        rule9Adjust(prevPos, &prevChar, &pos, &thisChar);
        nextCPPos = nextPos = fText->moveIndex32(pos, 1);
        c = fText->char32At(nextPos);
        rule9Adjust(pos, &thisChar, &nextPos, &c);

        // If the loop is still warming up - if we haven't shifted the initial
        //   -1 positions out of prevPos yet - loop back to advance the
        //    position in the input without any further looking for breaks.
        if (prevPos == -1) {
          setAppliedRule(pos, "LB 9 - adjust for combining sequences.");
            continue;
        }


        if (fBK->contains(prevChar)) {
            setAppliedRule(pos, "LB 4  Always break after hard line breaks");
            break;
        }


        if (prevChar == 0x0d && thisChar == 0x0a) {
            setAppliedRule(pos, "LB 5  Break after CR, LF, NL, but not inside CR LF");
            continue;
        }
        if (prevChar == 0x0d ||
            prevChar == 0x0a ||
            prevChar == 0x85)  {
            setAppliedRule(pos, "LB 5  Break after CR, LF, NL, but not inside CR LF");
            break;
        }


        if (thisChar == 0x0d || thisChar == 0x0a || thisChar == 0x85 ||
            fBK->contains(thisChar)) {
            setAppliedRule(pos, "LB 6  Don't break before hard line breaks");
            continue;
        }


        if (fSP->contains(thisChar)) {
            setAppliedRule(pos, "LB 7  Don't break before spaces or zero-width space.");
            continue;
        }

        // !!! ??? Is this the right text for the applied rule?
        if (fZW->contains(thisChar)) {
            setAppliedRule(pos, "LB 7  Don't break before spaces or zero-width space.");
            continue;
        }


        //       ZW SP* ÷
        //       Scan backwards from prevChar for SP* ZW
        tPos = prevPos;
        while (tPos>0 && fSP->contains(fText->char32At(tPos))) {
            tPos = fText->moveIndex32(tPos, -1);
        }
        if (fZW->contains(fText->char32At(tPos))) {
            setAppliedRule(pos, "LB 8  Break after zero width space");
            break;
        }


        //          Move this test up, before LB8a, because numbers can match a longer sequence that would
        //          also match 8a.  e.g. NU ZWJ IS PO     (ZWJ acts like CM)
        if (fNumberMatcher->lookingAt(prevPos, status)) {
            if (U_FAILURE(status)) {
                setAppliedRule(pos, "LB 25 Numbers");
                break;
            }
            // Matched a number.  But could have been just a single digit, which would
            //    not represent a "no break here" between prevChar and thisChar
            int32_t numEndIdx = fNumberMatcher->end(status);  // idx of first char following num
            if (numEndIdx > pos) {
                // Number match includes at least our two chars being checked
                if (numEndIdx > nextPos) {
                    // Number match includes additional chars.  Update pos and nextPos
                    //   so that next loop iteration will continue at the end of the number,
                    //   checking for breaks between last char in number & whatever follows.
                    pos = nextPos = numEndIdx;
                    do {
                        pos = fText->moveIndex32(pos, -1);
                        thisChar = fText->char32At(pos);
                    } while (fCM->contains(thisChar));
                }
                setAppliedRule(pos, "LB 25 Numbers");
                continue;
            }
        }


        //       The monkey test's way of ignoring combining characters doesn't work
        //       for this rule. ZJ is also a CM. Need to get the actual character
        //       preceding "thisChar", not ignoring combining marks, possibly ZJ.
        {
            int32_t prevIdx = fText->moveIndex32(pos, -1);
            UChar32 prevC = fText->char32At(prevIdx);
            if (fZWJ->contains(prevC)) {
                setAppliedRule(pos, "LB 8a ZWJ x");
                continue;
            }
        }


        // appliedRule: "LB 9, 10"; //  Already done, at top of loop.";
        //


        //    x  WJ
        //    WJ  x
        //
        if (fWJ->contains(thisChar) || fWJ->contains(prevChar)) {
            setAppliedRule(pos, "LB 11  Do not break before or after WORD JOINER and related characters.");
            continue;
        }


        if (fGL->contains(prevChar)) {
            setAppliedRule(pos, "LB 12  GL  x");
            continue;
        }


          if (!(fSP->contains(prevChar) ||
              fBA->contains(prevChar) ||
              fHY->contains(prevChar)     ) && fGL->contains(thisChar)) {
              setAppliedRule(pos, "LB 12a  [^SP BA HY] x GL");
              continue;
        }


        if (fCL->contains(thisChar) ||
                fCP->contains(thisChar) ||
                fEX->contains(thisChar) ||
                fSY->contains(thisChar)) {
            setAppliedRule(pos, "LB 13  Don't break before closings.");
            continue;
        }


        //       Scan backwards, checking for this sequence.
        //       The OP char could include combining marks, so we actually check for
        //           OP CM* SP*
        //       Another Twist: The Rule 9 fixes may have changed a SP CM
        //       sequence into a ID char, so before scanning back through spaces,
        //       verify that prevChar is indeed a space.  The prevChar variable
        //       may differ from fText[prevPos]
        tPos = prevPos;
        if (fSP->contains(prevChar)) {
            while (tPos > 0 && fSP->contains(fText->char32At(tPos))) {
                tPos=fText->moveIndex32(tPos, -1);
            }
        }
        while (tPos > 0 && fCM->contains(fText->char32At(tPos))) {
            tPos=fText->moveIndex32(tPos, -1);
        }
        if (fOP->contains(fText->char32At(tPos))) {
            setAppliedRule(pos, "LB 14 Don't break after OP SP*");
            continue;
        }


        if (nextPos < fText->length()) {
            // note: UnicodeString::char32At(length) returns ffff, not distinguishable
            //       from a legit ffff character. So test length separately.
            UChar32 nextChar = fText->char32At(nextPos);
            if (fSP->contains(prevChar) && fIS->contains(thisChar) && fNU->contains(nextChar)) {
                setAppliedRule(pos, "LB 14a Break before an IS that begins a number and follows a space");
                break;
            }
        }


          if (fIS->contains(thisChar)) {
              setAppliedRule(pos, "LB 14b  Do not break before numeric separators, even after spaces.");
              continue;
        }


        if (fOP->contains(thisChar)) {
            // Scan backwards from prevChar to see if it is preceded by QU CM* SP*
            int tPos = prevPos;
            while (tPos>0 && fSP->contains(fText->char32At(tPos))) {
                tPos = fText->moveIndex32(tPos, -1);
            }
            while (tPos>0 && fCM->contains(fText->char32At(tPos))) {
                tPos = fText->moveIndex32(tPos, -1);
            }
            if (fQU->contains(fText->char32At(tPos))) {
                setAppliedRule(pos, "LB 15    QU SP* x OP");
                continue;
            }
        }


        //    Scan backwards for SP* CM* (CL | CP)
        if (fNS->contains(thisChar)) {
            int tPos = prevPos;
            while (tPos>0 && fSP->contains(fText->char32At(tPos))) {
                tPos = fText->moveIndex32(tPos, -1);
            }
            while (tPos>0 && fCM->contains(fText->char32At(tPos))) {
                tPos = fText->moveIndex32(tPos, -1);
            }
            if (fCL->contains(fText->char32At(tPos)) || fCP->contains(fText->char32At(tPos))) {
                setAppliedRule(pos, "LB 16   (CL | CP) SP* x NS");
                continue;
            }
        }


        if (fB2->contains(thisChar)) {
            //  Scan backwards, checking for the B2 CM* SP* sequence.
            tPos = prevPos;
            if (fSP->contains(prevChar)) {
                while (tPos > 0 && fSP->contains(fText->char32At(tPos))) {
                    tPos=fText->moveIndex32(tPos, -1);
                }
            }
            while (tPos > 0 && fCM->contains(fText->char32At(tPos))) {
                tPos=fText->moveIndex32(tPos, -1);
            }
            if (fB2->contains(fText->char32At(tPos))) {
                setAppliedRule(pos, "LB 17   B2 SP* x B2");
                continue;
            }
        }


        if (fSP->contains(prevChar)) {
            setAppliedRule(pos, "LB 18    break after space");
            break;
        }

        //    x   QU
        //    QU  x
        if (fQU->contains(thisChar) || fQU->contains(prevChar)) {
            setAppliedRule(pos, "LB 19");
            continue;
        }

        if (fCB->contains(thisChar) || fCB->contains(prevChar)) {
            setAppliedRule(pos, "LB 20  Break around a CB");
            break;
        }

        //           Don't break between Hyphens and letters if a break precedes the hyphen.
        //           Formerly this was a Finnish tailoring.
        //           Moved to root in ICU 63. This is an ICU customization, not in UAX-14.
        //           ^($HY | $HH) $AL;
        if (fAL->contains(thisChar) && (fHY->contains(prevChar) || fHH->contains(prevChar)) &&
                prevPosX2 == -1) {
            setAppliedRule(pos, "LB 20.09");
            continue;
        }

        if (fBA->contains(thisChar) ||
            fHY->contains(thisChar) ||
            fNS->contains(thisChar) ||
            fBB->contains(prevChar) )   {
            setAppliedRule(pos, "LB 21");
            continue;
        }

        if (fHL->contains(prevCharX2) &&
                (fHY->contains(prevChar) || fBA->contains(prevChar))) {
            setAppliedRule(pos, "LB 21a   HL (HY | BA) x");
            continue;
        }

        if (fSY->contains(prevChar) && fHL->contains(thisChar)) {
            setAppliedRule(pos, "LB 21b SY x HL");
            continue;
        }

        if (fIN->contains(thisChar))   {
            setAppliedRule(pos, "LB 22");
            continue;
        }


        //          (AL | HL) x NU
        //          NU x (AL | HL)
        if ((fAL->contains(prevChar) || fHL->contains(prevChar)) && fNU->contains(thisChar)) {
            setAppliedRule(pos, "LB 23");
            continue;
        }
        if (fNU->contains(prevChar) && (fAL->contains(thisChar) || fHL->contains(thisChar))) {
            setAppliedRule(pos, "LB 23");
            continue;
        }

        // Do not break between numeric prefixes and ideographs, or between ideographs and numeric postfixes.
        //      PR x (ID | EB | EM)
        //     (ID | EB | EM) x PO
        if (fPR->contains(prevChar) &&
                (fID->contains(thisChar) || fEB->contains(thisChar) || fEM->contains(thisChar)))  {
            setAppliedRule(pos, "LB 23a");
            continue;
        }
        if ((fID->contains(prevChar) || fEB->contains(prevChar) || fEM->contains(prevChar)) &&
                fPO->contains(thisChar)) {
            setAppliedRule(pos, "LB 23a");
            continue;
        }

        //   Do not break between prefix and letters or ideographs.
        //         (PR | PO) x (AL | HL)
        //         (AL | HL) x (PR | PO)
        if ((fPR->contains(prevChar) || fPO->contains(prevChar)) &&
                (fAL->contains(thisChar) || fHL->contains(thisChar))) {
            setAppliedRule(pos, "LB 24 no break between prefix and letters or ideographs");
            continue;
        }
        if ((fAL->contains(prevChar) || fHL->contains(prevChar)) &&
                (fPR->contains(thisChar) || fPO->contains(thisChar))) {
            setAppliedRule(pos, "LB 24 no break between prefix and letters or ideographs");
            continue;
        }

        // appliedRule: "LB 25 numbers match"; // moved up, before LB 8a,

        if (fJL->contains(prevChar) && (fJL->contains(thisChar) ||
                                        fJV->contains(thisChar) ||
                                        fH2->contains(thisChar) ||
                                        fH3->contains(thisChar))) {
            setAppliedRule(pos, "LB 26 Do not break a Korean syllable.");
            continue;
                                        }

        if ((fJV->contains(prevChar) || fH2->contains(prevChar))  &&
            (fJV->contains(thisChar) || fJT->contains(thisChar))) {
            setAppliedRule(pos, "LB 26 Do not break a Korean syllable.");
            continue;
        }

        if ((fJT->contains(prevChar) || fH3->contains(prevChar)) &&
            fJT->contains(thisChar)) {
            setAppliedRule(pos, "LB 26 Do not break a Korean syllable.");
            continue;
        }

        if ((fJL->contains(prevChar) || fJV->contains(prevChar) ||
            fJT->contains(prevChar) || fH2->contains(prevChar) || fH3->contains(prevChar)) &&
            fPO->contains(thisChar)) {
            setAppliedRule(pos, "LB 27 Treat a Korean Syllable Block the same as ID.");
            continue;
        }
        if (fPR->contains(prevChar) && (fJL->contains(thisChar) || fJV->contains(thisChar) ||
            fJT->contains(thisChar) || fH2->contains(thisChar) || fH3->contains(thisChar))) {
            setAppliedRule(pos, "LB 27 Treat a Korean Syllable Block the same as ID.");
            continue;
        }



        if ((fAL->contains(prevChar) || fHL->contains(prevChar)) && (fAL->contains(thisChar) || fHL->contains(thisChar))) {
            setAppliedRule(pos, "LB 28  Do not break between alphabetics (\"at\").");
            continue;
        }

          if (fIS->contains(prevChar) && (fAL->contains(thisChar) || fHL->contains(thisChar))) {
              setAppliedRule(pos, "LB 29  Do not break between numeric punctuation and alphabetics (\"e.g.\").");
              continue;
        }

        //          (AL | NU) x OP
        //          CP x (AL | NU)
        if ((fAL->contains(prevChar) || fHL->contains(prevChar) || fNU->contains(prevChar)) && fOP30->contains(thisChar)) {
            setAppliedRule(pos,  "LB 30 No break in letters, numbers, or ordinary symbols, opening/closing punctuation.");
            continue;
        }
        if (fCP30->contains(prevChar) && (fAL->contains(thisChar) || fHL->contains(thisChar) || fNU->contains(thisChar))) {
            setAppliedRule(pos,  "LB 30 No break in letters, numbers, or ordinary symbols, opening/closing punctuation.");
            continue;
        }

        //             RI  x  RI
        if (fRI->contains(prevCharX2) && fRI->contains(prevChar) && fRI->contains(thisChar)) {
            setAppliedRule(pos, "LB30a    RI RI  ÷  RI");
            break;
        }
        if (fRI->contains(prevChar) && fRI->contains(thisChar)) {
            // Two Regional Indicators have been paired.
            // Over-write the trailing one (thisChar) to prevent it from forming another pair with a
            // following RI. This is a hack.
            thisChar = -1;
            setAppliedRule(pos, "LB30a    RI RI  ÷  RI");
            continue;
        }

        // LB30b Do not break between an emoji base (or potential emoji) and an emoji modifier.
        if (fEB->contains(prevChar) && fEM->contains(thisChar)) {
            setAppliedRule(pos, "LB30b    Emoji Base x Emoji Modifier");
            continue;
        }

        if (fExtPictUnassigned->contains(prevChar) && fEM->contains(thisChar)) {
            setAppliedRule(pos, "LB30b    [\\p{Extended_Pictographic}&\\p{Cn}] × EM");
            continue;
        }

        setAppliedRule(pos, "LB 31    Break everywhere else");
        break;
    }

    return pos;
}


UVector  *RBBILineMonkey::charClasses() {
    return fSets;
}


RBBILineMonkey::~RBBILineMonkey() {
    delete fSets;

    delete fBK;
    delete fCR;
    delete fLF;
    delete fCM;
    delete fNL;
    delete fWJ;
    delete fZW;
    delete fGL;
    delete fCB;
    delete fSP;
    delete fB2;
    delete fBA;
    delete fBB;
    delete fHH;
    delete fHY;
    delete fH2;
    delete fH3;
    delete fCL;
    delete fCP;
    delete fEX;
    delete fIN;
    delete fJL;
    delete fJV;
    delete fJT;
    delete fNS;
    delete fOP;
    delete fQU;
    delete fIS;
    delete fNU;
    delete fPO;
    delete fPR;
    delete fSY;
    delete fAI;
    delete fAL;
    delete fCJ;
    delete fHL;
    delete fID;
    delete fRI;
    delete fSG;
    delete fXX;
    delete fEB;
    delete fEM;
    delete fZWJ;
    delete fOP30;
    delete fCP30;
    delete fExtPictUnassigned;

    delete fCharBI;
    delete fNumberMatcher;
}


//-------------------------------------------------------------------------------------------
//
//   TestMonkey
//
//     params
//       seed=nnnnn        Random number starting seed.
//                         Setting the seed allows errors to be reproduced.
//       loop=nnn          Looping count.  Controls running time.
//                         -1:  run forever.
//                          0 or greater:  run length.
//
//       type = char | word | line | sent | title
//
//  Example:
//     intltest  rbbi/RBBITest/TestMonkey@"type=line loop=-1"
//
//-------------------------------------------------------------------------------------------

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 = strtol(valString, NULL, 10);

        // Delete this parameter from the params string.
        m.reset();
        params = m.replaceFirst("", status);
    }
    U_ASSERT(U_SUCCESS(status));
    return val;
}
#endif

#if !UCONFIG_NO_REGULAR_EXPRESSIONS
static void testBreakBoundPreceding(RBBITest *test, UnicodeString ustr,
                                    BreakIterator *bi,
                                    int expected[],
                                    int expectedcount)
{
    int count = 0;
    int i = 0;
    int forward[50];
    bi->setText(ustr);
    for (i = bi->first(); i != BreakIterator::DONE; i = bi->next()) {
        forward[count] = i;
        if (count < expectedcount && expected[count] != i) {
            test->errln("%s:%d break forward test failed: expected %d but got %d",
                        __FILE__, __LINE__, expected[count], i);
            break;
        }
        count ++;
    }
    if (count != expectedcount) {
        printStringBreaks(ustr, expected, expectedcount);
        test->errln("%s:%d break forward test failed: missed %d match",
                    __FILE__, __LINE__, expectedcount - count);
        return;
    }
    // testing boundaries
    for (i = 1; i < expectedcount; i ++) {
        int j = expected[i - 1];
        if (!bi->isBoundary(j)) {
            printStringBreaks(ustr, expected, expectedcount);
            test->errln("%s:%d isBoundary() failed.  Expected boundary at position %d",
                    __FILE__, __LINE__, j);
            return;
        }
        for (j = expected[i - 1] + 1; j < expected[i]; j ++) {
            if (bi->isBoundary(j)) {
                printStringBreaks(ustr, expected, expectedcount);
                test->errln("%s:%d isBoundary() failed.  Not expecting boundary at position %d",
                    __FILE__, __LINE__, j);
                return;
            }
        }
    }

    for (i = bi->last(); i != BreakIterator::DONE; i = bi->previous()) {
        count --;
        if (forward[count] != i) {
            printStringBreaks(ustr, expected, expectedcount);
            test->errln("%s:%d happy break test previous() failed: expected %d but got %d",
                        __FILE__, __LINE__, forward[count], i);
            break;
        }
    }
    if (count != 0) {
        printStringBreaks(ustr, expected, expectedcount);
        test->errln("break test previous() failed: missed a match");
        return;
    }

    // testing preceding
    for (i = 0; i < expectedcount - 1; i ++) {
        // int j = expected[i] + 1;
        int j = ustr.moveIndex32(expected[i], 1);
        for (; j <= expected[i + 1]; j ++) {
            int32_t expectedPreceding = expected[i];
            int32_t actualPreceding = bi->preceding(j);
            if (actualPreceding != expectedPreceding) {
                printStringBreaks(ustr, expected, expectedcount);
                test->errln("%s:%d preceding(%d): expected %d, got %d",
                        __FILE__, __LINE__, j, expectedPreceding, actualPreceding);
                return;
            }
        }
    }
}
#endif

void RBBITest::TestWordBreaks(void)
{
#if !UCONFIG_NO_REGULAR_EXPRESSIONS

    Locale        locale("en");
    UErrorCode    status = U_ZERO_ERROR;
    // BreakIterator  *bi = BreakIterator::createCharacterInstance(locale, status);
    BreakIterator *bi = BreakIterator::createWordInstance(locale, status);
    // Replaced any C+J characters in a row with a random sequence of characters
    // of the same length to make our C+J segmentation not get in the way.
    static const char *strlist[] =
    {
    "\\U000e0032\\u0097\\u0f94\\uc2d8\\u05f4\\U000e0031\\u060d",
    "\\U000e0037\\u2666\\u1202\\u003a\\U000e0031\\u064d\\u0bea\\u091c\\U000e0040\\u003b",
    "\\u0589\\u3e99\\U0001d7f3\\U000e0074\\u1810\\u200e\\U000e004b\\u0027\\U000e0061\\u003a",
    "\\u398c\\U000104a5\\U0001d173\\u102d\\u002e\\uca3b\\u002e\\u002c\\u5622",
    "\\uac00\\u3588\\u009c\\u0953\\u194b",
    "\\u200e\\U000e0072\\u0a4b\\U000e003f\\ufd2b\\u2027\\u002e\\u002e",
    "\\u0602\\u2019\\ua191\\U000e0063\\u0a4c\\u003a\\ub4b5\\u003a\\u827f\\u002e",
    "\\u2f1f\\u1634\\u05f8\\u0944\\u04f2\\u0cdf\\u1f9c\\u05f4\\u002e",
    "\\U000e0042\\u002e\\u0fb8\\u09ef\\u0ed1\\u2044",
    "\\u003b\\u024a\\u102e\\U000e0071\\u0600",
    "\\u2027\\U000e0067\\u0a47\\u00b7",
    "\\u1fcd\\u002c\\u07aa\\u0027\\u11b0",
    "\\u002c\\U000e003c\\U0001d7f4\\u003a\\u0c6f\\u0027",
    "\\u0589\\U000e006e\\u0a42\\U000104a5",
    "\\u0f66\\u2523\\u003a\\u0cae\\U000e0047\\u003a",
    "\\u003a\\u0f21\\u0668\\u0dab\\u003a\\u0655\\u00b7",
    "\\u0027\\u11af\\U000e0057\\u0602",
    "\\U0001d7f2\\U000e007\\u0004\\u0589",
    "\\U000e0022\\u003a\\u10b3\\u003a\\ua21b\\u002e\\U000e0058\\u1732\\U000e002b",
    "\\U0001d7f2\\U000e007d\\u0004\\u0589",
    "\\u82ab\\u17e8\\u0736\\u2019\\U0001d64d",
    "\\ub55c\\u0a68\\U000e0037\\u0cd6\\u002c\\ub959",
    "\\U000e0065\\u302c\\uc986\\u09ee\\U000e0068",
    "\\u0be8\\u002e\\u0c68\\u066e\\u136d\\ufc99\\u59e7",
    "\\u0233\\U000e0020\\u0a69\\u0d6a",
    "\\u206f\\u0741\\ub3ab\\u2019\\ubcac\\u2019",
    "\\u18f4\\U000e0049\\u20e7\\u2027",
    "\\ub315\\U0001d7e5\\U000e0073\\u0c47\\u06f2\\u0c6a\\u0037\\u10fe",
    "\\ua183\\u102d\\u0bec\\u003a",
    "\\u17e8\\u06e7\\u002e\\u096d\\u003b",
    "\\u003a\\u0e57\\u0fad\\u002e",
    "\\u002e\\U000e004c\\U0001d7ea\\u05bb\\ud0fd\\u02de",
    "\\u32e6\\U0001d7f6\\u0fa1\\u206a\\U000e003c\\u0cec\\u003a",
    "\\U000e005d\\u2044\\u0731\\u0650\\u0061",
    "\\u003a\\u0664\\u00b7\\u1fba",
    "\\u003b\\u0027\\u00b7\\u47a3",
    "\\u2027\\U000e0067\\u0a42\\u00b7\\u4edf\\uc26c\\u003a\\u4186\\u041b",
    "\\u0027\\u003a\\U0001d70f\\U0001d7df\\ubf4a\\U0001d7f5\\U0001d177\\u003a\\u0e51\\u1058\\U000e0058\\u00b7\\u0673",
    "\\uc30d\\u002e\\U000e002c\\u0c48\\u003a\\ub5a1\\u0661\\u002c",
    };
    int loop;
    if (U_FAILURE(status)) {
        errcheckln(status, "Creation of break iterator failed %s", u_errorName(status));
        return;
    }
    for (loop = 0; loop < UPRV_LENGTHOF(strlist); loop ++) {
        // printf("looping %d\n", loop);
        UnicodeString ustr = CharsToUnicodeString(strlist[loop]);
        // RBBICharMonkey monkey;
        RBBIWordMonkey monkey;

        int expected[50];
        int expectedcount = 0;

        monkey.setText(ustr);
        int i;
        for (i = 0; i != BreakIterator::DONE; i = monkey.next(i)) {
            expected[expectedcount ++] = i;
        }

        testBreakBoundPreceding(this, ustr, bi, expected, expectedcount);
    }
    delete bi;
#endif
}

void RBBITest::TestWordBoundary(void)
{
    // <data><>\u1d4a\u206e<?>\u0603\U0001d7ff<>\u2019<></data>
    Locale        locale("en");
    UErrorCode    status = U_ZERO_ERROR;
    // BreakIterator  *bi = BreakIterator::createCharacterInstance(locale, status);
    LocalPointer<BreakIterator> bi(BreakIterator::createWordInstance(locale, status), status);
    if (U_FAILURE(status)) {
        errcheckln(status, "%s:%d Creation of break iterator failed %s",
                __FILE__, __LINE__, u_errorName(status));
        return;
    }
    UChar         str[50];
    static const char *strlist[] =
    {
    "\\u200e\\U000e0072\\u0a4b\\U000e003f\\ufd2b\\u2027\\u002e\\u002e",
    "\\U000e0042\\u002e\\u0fb8\\u09ef\\u0ed1\\u2044",
    "\\u003b\\u024a\\u102e\\U000e0071\\u0600",
    "\\u2027\\U000e0067\\u0a47\\u00b7",
    "\\u1fcd\\u002c\\u07aa\\u0027\\u11b0",
    "\\u002c\\U000e003c\\U0001d7f4\\u003a\\u0c6f\\u0027",
    "\\u0589\\U000e006e\\u0a42\\U000104a5",
    "\\u4f66\\ub523\\u003a\\uacae\\U000e0047\\u003a",
    "\\u003a\\u0f21\\u0668\\u0dab\\u003a\\u0655\\u00b7",
    "\\u0027\\u11af\\U000e0057\\u0602",
    "\\U0001d7f2\\U000e007\\u0004\\u0589",
    "\\U000e0022\\u003a\\u10b3\\u003a\\ua21b\\u002e\\U000e0058\\u1732\\U000e002b",
    "\\U0001d7f2\\U000e007d\\u0004\\u0589",
    "\\u82ab\\u17e8\\u0736\\u2019\\U0001d64d",
    "\\u0e01\\ub55c\\u0a68\\U000e0037\\u0cd6\\u002c\\ub959",
    "\\U000e0065\\u302c\\u09ee\\U000e0068",
    "\\u0be8\\u002e\\u0c68\\u066e\\u136d\\ufc99\\u59e7",
    "\\u0233\\U000e0020\\u0a69\\u0d6a",
    "\\u206f\\u0741\\ub3ab\\u2019\\ubcac\\u2019",
    "\\u58f4\\U000e0049\\u20e7\\u2027",
    "\\U0001d7e5\\U000e0073\\u0c47\\u06f2\\u0c6a\\u0037\\u10fe",
    "\\ua183\\u102d\\u0bec\\u003a",
    "\\u17e8\\u06e7\\u002e\\u096d\\u003b",
    "\\u003a\\u0e57\\u0fad\\u002e",
    "\\u002e\\U000e004c\\U0001d7ea\\u05bb\\ud0fd\\u02de",
    "\\u32e6\\U0001d7f6\\u0fa1\\u206a\\U000e003c\\u0cec\\u003a",
    "\\ua2a5\\u0038\\u2044\\u002e\\u0c67\\U000e003c\\u05f4\\u2027\\u05f4\\u2019",
    "\\u003a\\u0664\\u00b7\\u1fba",
    "\\u003b\\u0027\\u00b7\\u47a3",
    };
    int loop;
    for (loop = 0; loop < UPRV_LENGTHOF(strlist); loop ++) {
        u_unescape(strlist[loop], str, UPRV_LENGTHOF(str));
        UnicodeString ustr(str);
        int forward[50];
        int count = 0;

        bi->setText(ustr);
        int prev = -1;
        for (int32_t boundary = bi->first(); boundary != BreakIterator::DONE; boundary = bi->next()) {
            ++count;
            if (count >= UPRV_LENGTHOF(forward)) {
                errln("%s:%d too many breaks found. (loop, count, boundary) = (%d, %d, %d)",
                        __FILE__, __LINE__, loop, count, boundary);
                return;
            }
            forward[count] = boundary;
            if (boundary <= prev) {
                errln("%s:%d bi::next() did not advance. (loop, prev, boundary) = (%d, %d, %d)\n",
                        __FILE__, __LINE__, loop, prev, boundary);
                break;
            }
            for (int32_t nonBoundary = prev + 1; nonBoundary < boundary; nonBoundary ++) {
                if (bi->isBoundary(nonBoundary)) {
                    printStringBreaks(ustr, forward, count);
                    errln("%s:%d isBoundary(nonBoundary) failed. (loop, prev, nonBoundary, boundary) = (%d, %d, %d, %d)",
                           __FILE__, __LINE__, loop, prev, nonBoundary, boundary);
                    return;
                }
            }
            if (!bi->isBoundary(boundary)) {
                printStringBreaks(ustr, forward, count);
                errln("%s:%d happy boundary test failed: expected %d a boundary",
                       __FILE__, __LINE__, boundary);
                return;
            }
            prev = boundary;
        }
    }
}

void RBBITest::TestLineBreaks(void)
{
#if !UCONFIG_NO_REGULAR_EXPRESSIONS
    Locale        locale("en");
    UErrorCode    status = U_ZERO_ERROR;
    BreakIterator *bi = BreakIterator::createLineInstance(locale, status);
    const int32_t  STRSIZE = 50;
    UChar         str[STRSIZE];
    static const char *strlist[] =
    {
     "\\u300f\\ufdfc\\ub798\\u2011\\u2011\\u0020\\u0b43\\u002d\\ubeec\\ufffc",
     "\\u24ba\\u2060\\u3405\\ub290\\u000d\\U000e0032\\ufe35\\u00a0\\u0361\\"
             "U000112ed\\u0f0c\\u000a\\u308e\\ua875\\u0085\\u114d",
     "\\ufffc\\u3063\\u2e08\\u30e3\\u000d\\u002d\\u0ed8\\u002f\\U00011a57\\"
             "u2014\\U000e0105\\u118c\\u000a\\u07f8",
     "\\u0668\\u192b\\u002f\\u2034\\ufe39\\u00b4\\u0cc8\\u2571\\u200b\\u003f",
     "\\ufeff\\ufffc\\u3289\\u0085\\u2772\\u0020\\U000e010a\\u0020\\u2025\\u000a\\U000e0123",
     "\\ufe3c\\u201c\\u000d\\u2025\\u2007\\u201c\\u002d\\u20a0\\u002d\\u30a7\\u17a4",
     "\\u2772\\u0020\\U000e010a\\u0020\\u2025\\u000a\\U000e0123",
     "\\u002d\\uff1b\\u02c8\\u2029\\ufeff\\u0f22\\u2044\\ufe09\\u003a\\u096d\\u2009\\u000a\\u06f7\\u02cc\\u1019\\u2060",
     "\\u2770\\u0020\\U000e010f\\u0020\\u2060\\u000a\\u02cc\\u0bcc\\u060d\\u30e7\\u0f3b\\u002f",
     "\\ufeff\\u0028\\u003b\\U00012fec\\u2010\\u0020\\u0004\\u200b\\u0020\\u275c\\u002f\\u17b1",
     "\\u20a9\\u2014\\u00a2\\u31f1\\u002f\\u0020\\u05b8\\u200b\\u0cc2\\u003b\\u060d\\u02c8\\ua4e8\\u002f\\u17d5",
     "\\u002d\\u136f\\uff63\\u0084\\ua933\\u2028\\u002d\\u431b\\u200b\\u20b0",
     "\\uade3\\u11d6\\u000a\\U0001107d\\u203a\\u201d\\ub070\\u000d\\u2024\\ufffc",
     "\\uff5b\\u101c\\u1806\\u002f\\u2213\\uff5f",
     "\\u2014\\u0a83\\ufdfc\\u003f\\u00a0\\u0020\\u000a\\u2991\\U0001d179\\u0020\\u201d\\U000125f6\\u0a67\\u20a7\\ufeff\\u043f",
     "\\u169b\\U000e0130\\u002d\\u1041\\u0f3d\\u0abf\\u00b0\\u31fb\\u00a0\\u002d\\u02c8\\u003b",
     "\\u2762\\u1680\\u002d\\u2028\\u0027\\u01dc\\ufe56\\u003a\\u000a\\uffe6\\u29fd\\u0020\\u30ee\\u007c\\U0001d178\\u0af1\\u0085",
     "\\u3010\\u200b\\u2029\\ufeff\\ufe6a\\u275b\\U000e013b\\ufe37\\u24d4\\u002d\\u1806\\u256a\\u1806\\u247c\\u0085\\u17ac",
     "\\u99ab\\u0027\\u003b\\u2026\\ueaf0\\u0020\\u0020\\u0313\\u0020\\u3099\\uff09\\u208e\\u2011\\u2007\\u2060\\u000a\\u0020\\u0020\\u300b\\u0bf9",
     "\\u1806\\u060d\\u30f5\\u00b4\\u17e9\\u2544\\u2028\\u2024\\u2011\\u20a3\\u002d\\u09cc\\u1782\\u000d\\uff6f\\u0025",
     "\\u002f\\uf22e\\u1944\\ufe3d\\u0020\\u206f\\u31b3\\u2014\\u002d\\u2025\\u0f0c\\u0085\\u2763",
     "\\u002f\\u2563\\u202f\\u0085\\u17d5\\u200b\\u0020\\U000e0043\\u2014\\u058a\\u3d0a\\ufe57\\u2035\\u2028\\u2029",
     "\\u20ae\\U0001d169\\u9293\\uff1f\\uff1f\\u0021\\u2012\\u2039\\u0085\\u02cc\\u00a2\\u0020\\U000e01ab\\u3085\\u0f3a\\u1806\\u0f0c\\u1945\\u000a\\U0001d7e7",
     "\\u02cc\\ufe6a\\u00a0\\u0021\\u002d\\u7490\\uec2e\\u200b\\u000a",
     "\\uec2e\\u200b\\u000a\\u0020\\u2028\\u2014\\u8945",
     "\\u7490\\uec2e\\u200b\\u000a\\u0020\\u2028\\u2014",
     "\\u0020\\u2028\\u2014\\u8945\\u002c\\u005b",
     "\\u000a\\ufe3c\\u201c\\u000d\\u2025\\u2007\\u201c\\u002d\\u20a0",
     "\\U0001d16e\\ufffc\\u2025\\u0021\\u002d",
     "\\ufffc\\u301b\\u0fa5\\U000e0103\\u2060\\u208e\\u17d5\\u034f\\u1009\\u003a\\u180e\\u2009\\u3111",
     "\\ufffc\\u0020\\u2116\\uff6c\\u200b\\u0ac3\\U0001028f",
     "\\uaeb0\\u0344\\u0085\\ufffc\\u073b\\u2010",
     "\\ufeff\\u0589\\u0085\\u0eb8\\u30fd\\u002f\\u003a\\u2014\\ufe43",
     "\\u09cc\\u256a\\u276d\\u002d\\u3085\\u000d\\u0e05\\u2028\\u0fbb",
     "\\u2034\\u00bb\\u0ae6\\u300c\\u0020\\u31f8\\ufffc",
     "\\u2116\\u0ed2\\uff64\\u02cd\\u2001\\u2060",
     "\\ufe10\\u2060\\u1a5a\\u2060\\u17e4\\ufffc\\ubbe1\\ufe15\\u0020\\u00a0",
         "\\u2060\\u2213\\u200b\\u2019\\uc2dc\\uff6a\\u1736\\u0085\\udb07",
    };
    int loop;
    TEST_ASSERT_SUCCESS(status);
    if (U_FAILURE(status)) {
        return;
    }
    for (loop = 0; loop < UPRV_LENGTHOF(strlist); loop ++) {
        // printf("looping %d\n", loop);
        int32_t t = u_unescape(strlist[loop], str, STRSIZE);
        if (t >= STRSIZE) {
            TEST_ASSERT(FALSE);
            continue;
        }


        UnicodeString ustr(str);
        RBBILineMonkey monkey;
        if (U_FAILURE(monkey.deferredStatus)) {
            continue;
        }

        const int EXPECTEDSIZE = 50;
        int expected[EXPECTEDSIZE];
        int expectedcount = 0;

        monkey.setText(ustr);

        int i;
        for (i = 0; i != BreakIterator::DONE; i = monkey.next(i)) {
            if (expectedcount >= EXPECTEDSIZE) {
                TEST_ASSERT(expectedcount < EXPECTEDSIZE);
                return;
            }
            expected[expectedcount ++] = i;
        }

        testBreakBoundPreceding(this, ustr, bi, expected, expectedcount);
    }
    delete bi;
#endif
}

void RBBITest::TestSentBreaks(void)
{
#if !UCONFIG_NO_REGULAR_EXPRESSIONS
    Locale        locale("en");
    UErrorCode    status = U_ZERO_ERROR;
    BreakIterator *bi = BreakIterator::createSentenceInstance(locale, status);
    UChar         str[200];
    static const char *strlist[] =
    {
     "Now\ris\nthe\r\ntime\n\rfor\r\r",
     "This\n",
     "Hello! how are you? I'am fine. Thankyou. How are you doing? This\n costs $20,00,000.",
     "\"Sentence ending with a quote.\" Bye.",
     "  (This is it).  Testing the sentence iterator. \"This isn't it.\"",
     "Hi! This is a simple sample sentence. (This is it.) This is a simple sample sentence. \"This isn't it.\"",
     "Hi! This is a simple sample sentence. It does not have to make any sense as you can see. ",
     "Nel mezzo del cammin di nostra vita, mi ritrovai in una selva oscura. ",
     "Che la dritta via aveo smarrita. He said, that I said, that you said!! ",
     "Don't rock the boat.\\u2029Because I am the daddy, that is why. Not on my time (el timo.)!",
     "\\U0001040a\\u203a\\u1217\\u2b23\\u000d\\uff3b\\u03dd\\uff57\\u0a69\\u104a\\ufe56\\ufe52"
             "\\u3016\\U000e002f\\U000e0077\\u0662\\u1680\\u2984\\U000e006a\\u002e\\ua6ab\\u104a"
             "\\u002e\\u019b\\u2005\\u002e\\u0477\\u0438\\u0085\\u0441\\u002e\\u5f61\\u202f"
             "\\U0001019f\\uff08\\u27e8\\u055c\\u0352",
     "\\u1f3e\\u004d\\u000a\\ua3e4\\U000e0023\\uff63\\u0c52\\u276d\\U0001d5de\\U0001d171"
             "\\u0e38\\u17e5\\U00012fe6\\u0fa9\\u267f\\u1da3\\u0046\\u03ed\\udc72\\u0030"
             "\\U0001d688\\u0b6d\\u0085\\u0c67\\u1f94\\u0c6c\\u9cb2\\u202a\\u180e\\u000b"
             "\\u002e\\U000e005e\\u035b\\u061f\\u02c1\\U000e0025\\u0357\\u0969\\u202b"
             "\\U000130c5\\u0486\\U000e0123\\u2019\\u01bc\\u2006\\u11ad\\u180e\\u2e05"
             "\\u10b7\\u013e\\u000a\\u002e\\U00013ea4"
    };
    int loop;
    if (U_FAILURE(status)) {
        errcheckln(status, "Creation of break iterator failed %s", u_errorName(status));
        return;
    }
    for (loop = 0; loop < UPRV_LENGTHOF(strlist); loop ++) {
        u_unescape(strlist[loop], str, UPRV_LENGTHOF(str));
        UnicodeString ustr(str);

        RBBISentMonkey monkey;
        if (U_FAILURE(monkey.deferredStatus)) {
            continue;
        }

        const int EXPECTEDSIZE = 50;
        int expected[EXPECTEDSIZE];
        int expectedcount = 0;

        monkey.setText(ustr);

        int i;
        for (i = 0; i != BreakIterator::DONE; i = monkey.next(i)) {
            if (expectedcount >= EXPECTEDSIZE) {
                TEST_ASSERT(expectedcount < EXPECTEDSIZE);
                return;
            }
            expected[expectedcount ++] = i;
        }

        testBreakBoundPreceding(this, ustr, bi, expected, expectedcount);
    }
    delete bi;
#endif
}

void RBBITest::TestMonkey() {
#if !UCONFIG_NO_REGULAR_EXPRESSIONS

    UErrorCode     status    = U_ZERO_ERROR;
    int32_t        loopCount = 500;
    int32_t        seed      = 1;
    UnicodeString  breakType = "all";
    Locale         locale("en");
    UBool          useUText  = FALSE;

    if (quick == FALSE) {
        loopCount = 10000;
    }

    if (fTestParams) {
        UnicodeString p(fTestParams);
        loopCount = getIntParam("loop", p, loopCount);
        seed      = getIntParam("seed", p, seed);

        RegexMatcher m(" *type *= *(char|word|line|sent|title) *", p, 0, status);
        if (m.find()) {
            breakType = m.group(1, status);
            m.reset();
            p = m.replaceFirst("", status);
        }

        RegexMatcher u(" *utext", p, 0, status);
        if (u.find()) {
            useUText = TRUE;
            u.reset();
            p = u.replaceFirst("", status);
        }


        // m.reset(p);
        if (RegexMatcher(UNICODE_STRING_SIMPLE("\\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;
        }

    }

    if (breakType == "char" || breakType == "all") {
        RBBICharMonkey  m;
        BreakIterator  *bi = BreakIterator::createCharacterInstance(locale, status);
        if (U_SUCCESS(status)) {
            RunMonkey(bi, m, "char", seed, loopCount, useUText);
            if (breakType == "all" && useUText==FALSE) {
                // Also run a quick test with UText when "all" is specified
                RunMonkey(bi, m, "char", seed, loopCount, TRUE);
            }
        }
        else {
            errcheckln(status, "Creation of character break iterator failed %s", u_errorName(status));
        }
        delete bi;
    }

    if (breakType == "word" || breakType == "all") {
        logln("Word Break Monkey Test");
        RBBIWordMonkey  m;
        BreakIterator  *bi = BreakIterator::createWordInstance(locale, status);
        if (U_SUCCESS(status)) {
            RunMonkey(bi, m, "word", seed, loopCount, useUText);
        }
        else {
            errcheckln(status, "Creation of word break iterator failed %s", u_errorName(status));
        }
        delete bi;
    }

    if (breakType == "line" || breakType == "all") {
        logln("Line Break Monkey Test");
        RBBILineMonkey  m;
        BreakIterator  *bi = BreakIterator::createLineInstance(locale, status);
        if (loopCount >= 10) {
            loopCount = loopCount / 5;   // Line break runs slower than the others.
        }
        if (U_SUCCESS(status)) {
            RunMonkey(bi, m, "line", seed, loopCount, useUText);
        }
        else {
            errcheckln(status, "Creation of line break iterator failed %s", u_errorName(status));
        }
        delete bi;
    }

    if (breakType == "sent" || breakType == "all"  ) {
        logln("Sentence Break Monkey Test");
        RBBISentMonkey  m;
        BreakIterator  *bi = BreakIterator::createSentenceInstance(locale, status);
        if (loopCount >= 10) {
            loopCount = loopCount / 10;   // Sentence runs slower than the other break types
        }
        if (U_SUCCESS(status)) {
            RunMonkey(bi, m, "sent", seed, loopCount, useUText);
        }
        else {
            errcheckln(status, "Creation of line break iterator failed %s", u_errorName(status));
        }
        delete bi;
    }

#endif
}

//
//  Run a RBBI monkey test.  Common routine, for all break iterator types.
//    Parameters:
//       bi      - the break iterator to use
//       mk      - MonkeyKind, abstraction for obtaining expected results
//       name    - Name of test (char, word, etc.) for use in error messages
//       seed    - Seed for starting random number generator (parameter from user)
//       numIterations
//
void RBBITest::RunMonkey(BreakIterator *bi, RBBIMonkeyKind &mk, const char *name, uint32_t  seed,
                         int32_t numIterations, UBool useUText) {

#if !UCONFIG_NO_REGULAR_EXPRESSIONS

    const int32_t    TESTSTRINGLEN = 500;
    UnicodeString    testText;
    int32_t          numCharClasses;
    UVector          *chClasses;
    int              expectedCount = 0;
    char             expectedBreaks[TESTSTRINGLEN*2 + 1];
    char             forwardBreaks[TESTSTRINGLEN*2 + 1];
    char             reverseBreaks[TESTSTRINGLEN*2+1];
    char             isBoundaryBreaks[TESTSTRINGLEN*2+1];
    char             followingBreaks[TESTSTRINGLEN*2+1];
    char             precedingBreaks[TESTSTRINGLEN*2+1];
    int              i;
    int              loopCount = 0;


    m_seed = seed;

    numCharClasses = mk.charClasses()->size();
    chClasses      = mk.charClasses();

    // Check for errors that occurred during the construction of the MonkeyKind object.
    //  Can't report them where they occurred because errln() is a method coming from intlTest,
    //  and is not visible outside of RBBITest :-(
    if (U_FAILURE(mk.deferredStatus)) {
        errln("status of \"%s\" in creation of RBBIMonkeyKind.", u_errorName(mk.deferredStatus));
        return;
    }

    // Verify that the character classes all have at least one member.
    for (i=0; i<numCharClasses; i++) {
        UnicodeSet *s = (UnicodeSet *)chClasses->elementAt(i);
        if (s == NULL || s->size() == 0) {
            errln("Character Class #%d is null or of zero size.", i);
            return;
        }
    }

    // For minimizing width of class name output.
    int classNameSize = mk.maxClassNameSize();

    while (loopCount < numIterations || numIterations == -1) {
        if (numIterations == -1 && loopCount % 10 == 0) {
            // If test is running in an infinite loop, display a periodic tic so
            //   we can tell that it is making progress.
            fprintf(stderr, ".");
        }
        // Save current random number seed, so that we can recreate the random numbers
        //   for this loop iteration in event of an error.
        seed = m_seed;

        // Populate a test string with data.
        testText.truncate(0);
        for (i=0; i<TESTSTRINGLEN; i++) {
            int32_t  aClassNum = m_rand() % numCharClasses;
            UnicodeSet *classSet = (UnicodeSet *)chClasses->elementAt(aClassNum);
            int32_t   charIdx = m_rand() % classSet->size();
            UChar32   c = classSet->charAt(charIdx);
            if (c < 0) {   // TODO:  deal with sets containing strings.
                errln("%s:%d c < 0", __FILE__, __LINE__);
                break;
            }
            // Do not assemble a supplementary character from randomly generated separate surrogates.
            //   (It could be a dictionary character)
            if (U16_IS_TRAIL(c) && testText.length() > 0 && U16_IS_LEAD(testText.charAt(testText.length()-1))) {
                continue;
            }

            testText.append(c);
        }

        // Calculate the expected results for this test string and reset applied rules.
        mk.setText(testText);

        memset(expectedBreaks, 0, sizeof(expectedBreaks));
        expectedBreaks[0] = 1;
        int32_t breakPos = 0;
        expectedCount = 0;
        for (;;) {
            breakPos = mk.next(breakPos);
            if (breakPos == -1) {
                break;
            }
            if (breakPos > testText.length()) {
                errln("breakPos > testText.length()");
            }
            expectedBreaks[breakPos] = 1;
            expectedCount++;
            U_ASSERT(expectedCount<testText.length());
        }

        // Find the break positions using forward iteration
        memset(forwardBreaks, 0, sizeof(forwardBreaks));
        if (useUText) {
            UErrorCode status = U_ZERO_ERROR;
            UText *testUText = utext_openReplaceable(NULL, &testText, &status);
            // testUText = utext_openUnicodeString(testUText, &testText, &status);
            bi->setText(testUText, status);
            TEST_ASSERT_SUCCESS(status);
            utext_close(testUText);   // The break iterator does a shallow clone of the UText
                                      //  This UText can be closed immediately, so long as the
                                      //  testText string continues to exist.
        } else {
            bi->setText(testText);
        }

        for (i=bi->first(); i != BreakIterator::DONE; i=bi->next()) {
            if (i < 0 || i > testText.length()) {
                errln("%s break monkey test: Out of range value returned by breakIterator::next()", name);
                break;
            }
            forwardBreaks[i] = 1;
        }

        // Find the break positions using reverse iteration
        memset(reverseBreaks, 0, sizeof(reverseBreaks));
        for (i=bi->last(); i != BreakIterator::DONE; i=bi->previous()) {
            if (i < 0 || i > testText.length()) {
                errln("%s break monkey test: Out of range value returned by breakIterator::next()", name);
                break;
            }
            reverseBreaks[i] = 1;
        }

        // Find the break positions using isBoundary() tests.
        memset(isBoundaryBreaks, 0, sizeof(isBoundaryBreaks));
        U_ASSERT((int32_t)sizeof(isBoundaryBreaks) > testText.length());
        for (i=0; i<=testText.length(); i++) {
            isBoundaryBreaks[i] = bi->isBoundary(i);
        }


        // Find the break positions using the following() function.
        // printf(".");
        memset(followingBreaks, 0, sizeof(followingBreaks));
        int32_t   lastBreakPos = 0;
        followingBreaks[0] = 1;
        for (i=0; i<testText.length(); i++) {
            breakPos = bi->following(i);
            if (breakPos <= i ||
                breakPos < lastBreakPos ||
                breakPos > testText.length() ||
                (breakPos > lastBreakPos && lastBreakPos > i)) {
                errln("%s break monkey test: "
                    "Out of range value returned by BreakIterator::following().\n"
                        "Random seed=%d  index=%d; following returned %d;  lastbreak=%d",
                         name, seed, i, breakPos, lastBreakPos);
                break;
            }
            followingBreaks[breakPos] = 1;
            lastBreakPos = breakPos;
        }

        // Find the break positions using the preceding() function.
        memset(precedingBreaks, 0, sizeof(precedingBreaks));
        lastBreakPos = testText.length();
        precedingBreaks[testText.length()] = 1;
        for (i=testText.length(); i>0; i--) {
            breakPos = bi->preceding(i);
            if (breakPos >= i ||
                breakPos > lastBreakPos ||
                (breakPos < 0 && testText.getChar32Start(i)>0) ||
                (breakPos < lastBreakPos && lastBreakPos < testText.getChar32Start(i)) ) {
                errln("%s break monkey test: "
                    "Out of range value returned by BreakIterator::preceding().\n"
                    "index=%d;  prev returned %d; lastBreak=%d" ,
                    name,  i, breakPos, lastBreakPos);
                if (breakPos >= 0 && breakPos < (int32_t)sizeof(precedingBreaks)) {
                    precedingBreaks[i] = 2;   // Forces an error.
                }
            } else {
                if (breakPos >= 0) {
                    precedingBreaks[breakPos] = 1;
                }
                lastBreakPos = breakPos;
            }
        }

        // Compare the expected and actual results.
        for (i=0; i<=testText.length(); i++) {
            const char *errorType = NULL;
            const char* currentBreakData = NULL;
            if  (forwardBreaks[i] != expectedBreaks[i]) {
                errorType = "next()";
                currentBreakData = forwardBreaks;
            } else if (reverseBreaks[i] != forwardBreaks[i]) {
                errorType = "previous()";
                currentBreakData = reverseBreaks;
           } else if (isBoundaryBreaks[i] != expectedBreaks[i]) {
                errorType = "isBoundary()";
                currentBreakData = isBoundaryBreaks;
            } else if (followingBreaks[i] != expectedBreaks[i]) {
                errorType = "following()";
                currentBreakData = followingBreaks;
            } else if (precedingBreaks[i] != expectedBreaks[i]) {
                errorType = "preceding()";
                currentBreakData = precedingBreaks;
            }

            if (errorType != NULL) {
                // Format a range of the test text that includes the failure as
                //  a data item that can be included in the rbbi test data file.

                // Start of the range is the last point where expected and actual results
                //  both agreed that there was a break position.

                int startContext = i;
                int32_t count = 0;
                for (;;) {
                    if (startContext==0) { break; }
                    startContext --;
                    if (expectedBreaks[startContext] != 0) {
                        if (count == 2) break;
                        count ++;
                    }
                }

                // End of range is two expected breaks past the start position.
                int endContext = i + 1;
                int ci;
                for (ci=0; ci<2; ci++) {  // Number of items to include in error text.
                    for (;;) {
                        if (endContext >= testText.length()) {break;}
                        if (expectedBreaks[endContext-1] != 0) {
                            if (count == 0) break;
                            count --;
                        }
                        endContext ++;
                    }
                }

                // Formatting of each line includes:
                //   character code
                //   reference break: '|' -> a break, '.' -> no break
                //   actual break:    '|' -> a break, '.' -> no break
                //   (name of character clase)
                //   Unicode name of character
                //   '-->' indicates location of the difference.

                MONKEY_ERROR(
                    (expectedBreaks[i] ? "Break expected but not found" :
                       "Break found but not expected"),
                    name, i, seed);

                for (ci=startContext; (ci = testText.moveIndex32(ci, 1));) {
                    UChar32  c;
                    c = testText.char32At(ci);

                    std::string currentLineFlag = "   ";
                    if (ci == i) {
                        currentLineFlag = "-->";  // Error position
                    }

                    // BMP or SMP character in hex
                    char hexCodePoint[12];
                    std::string format = "    \\u%04x";
                    if (c >= 0x10000) {
                        format = "\\U%08x";
                    }
                    sprintf(hexCodePoint, format.c_str(), c);

                    // Get the class name and character name for the character.
                    char cName[200];
                    UErrorCode status = U_ZERO_ERROR;
                    u_charName(c, U_EXTENDED_CHAR_NAME, cName, sizeof(cName), &status);

                    char buffer[200];
                    auto ret = snprintf(buffer, UPRV_LENGTHOF(buffer),
                             "%4s %3i :  %1s  %1s  %10s  %-*s  %-40s  %-40s",
                             currentLineFlag.c_str(),
                             ci,
                             expectedBreaks[ci] == 0 ? "." : "|",  // Reference break
                             currentBreakData[ci] == 0 ? "." : "|",  // Actual break
                             hexCodePoint,
                             classNameSize,
                             mk.classNameFromCodepoint(c).c_str(),
                             mk.getAppliedRule(ci).c_str(), cName);
                    (void)ret;
                    U_ASSERT(0 <= ret && ret < UPRV_LENGTHOF(buffer));

                    // Output the error
                    if (ci == i) {
                        errln(buffer);
                    } else {
                        infoln(buffer);
                    }

                    if (ci >= endContext) { break; }
                }
                break;
            }
        }

        loopCount++;
    }
#endif
}


//  Bug 5532.  UTF-8 based UText fails in dictionary code.
//             This test checks the initial patch,
//             which is to just keep it from crashing.  Correct word boundaries
//             await a proper fix to the dictionary code.
//
void RBBITest::TestBug5532(void)  {
   // Text includes a mixture of Thai and Latin.
   const unsigned char utf8Data[] = {
           0xE0u, 0xB8u, 0x82u, 0xE0u, 0xB8u, 0xB2u, 0xE0u, 0xB8u, 0xA2u, 0xE0u,
           0xB9u, 0x80u, 0xE0u, 0xB8u, 0x84u, 0xE0u, 0xB8u, 0xA3u, 0xE0u, 0xB8u,
           0xB7u, 0xE0u, 0xB9u, 0x88u, 0xE0u, 0xB8u, 0xADu, 0xE0u, 0xB8u, 0x87u,
           0xE0u, 0xB9u, 0x80u, 0xE0u, 0xB8u, 0xA5u, 0xE0u, 0xB9u, 0x88u, 0xE0u,
           0xB8u, 0x99u, 0xE0u, 0xB8u, 0x8Bu, 0xE0u, 0xB8u, 0xB5u, 0xE0u, 0xB8u,
           0x94u, 0xE0u, 0xB8u, 0xB5u, 0x20u, 0x73u, 0x69u, 0x6Du, 0x20u, 0x61u,
           0x75u, 0x64u, 0x69u, 0x6Fu, 0x2Fu, 0x20u, 0x4Du, 0x4Fu, 0x4Fu, 0x4Eu,
           0x20u, 0x65u, 0x63u, 0x6Cu, 0x69u, 0x70u, 0x73u, 0x65u, 0x20u, 0xE0u,
           0xB8u, 0xA3u, 0xE0u, 0xB8u, 0xB2u, 0xE0u, 0xB8u, 0x84u, 0xE0u, 0xB8u,
           0xB2u, 0x20u, 0x34u, 0x37u, 0x30u, 0x30u, 0x20u, 0xE0u, 0xB8u, 0xA2u,
           0xE0u, 0xB8u, 0xB9u, 0xE0u, 0xB9u, 0x82u, 0xE0u, 0xB8u, 0xA3u, 0x00};

    UErrorCode status = U_ZERO_ERROR;
    UText utext=UTEXT_INITIALIZER;
    utext_openUTF8(&utext, (const char *)utf8Data, -1, &status);
    TEST_ASSERT_SUCCESS(status);

    BreakIterator *bi = BreakIterator::createWordInstance(Locale("th"), status);
    TEST_ASSERT_SUCCESS(status);
    if (U_SUCCESS(status)) {
        bi->setText(&utext, status);
        TEST_ASSERT_SUCCESS(status);

        int32_t breakCount = 0;
        int32_t previousBreak = -1;
        for (bi->first(); bi->next() != BreakIterator::DONE; breakCount++) {
            // For now, just make sure that the break iterator doesn't hang.
            TEST_ASSERT(previousBreak < bi->current());
            previousBreak = bi->current();
        }
        TEST_ASSERT(breakCount > 0);
    }
    delete bi;
    utext_close(&utext);
}


void RBBITest::TestBug9983(void)  {
    UnicodeString text = UnicodeString("\\u002A"  // * Other
                                       "\\uFF65"  //   Other
                                       "\\u309C"  //   Katakana
                                       "\\uFF9F"  //   Extend
                                       "\\uFF65"  //   Other
                                       "\\u0020"  //   Other
                                       "\\u0000").unescape();

    UErrorCode status = U_ZERO_ERROR;
    LocalPointer<RuleBasedBreakIterator> brkiter(static_cast<RuleBasedBreakIterator *>(
        BreakIterator::createWordInstance(Locale::getRoot(), status)));
    TEST_ASSERT_SUCCESS(status);
    LocalPointer<RuleBasedBreakIterator> brkiterPOSIX(static_cast<RuleBasedBreakIterator *>(
        BreakIterator::createWordInstance(Locale::createFromName("en_US_POSIX"), status)));
    TEST_ASSERT_SUCCESS(status);
    if (U_FAILURE(status)) {
        return;
    }
    int32_t offset, rstatus, iterationCount;

    brkiter->setText(text);
    brkiter->last();
    iterationCount = 0;
    while ( (offset = brkiter->previous()) != UBRK_DONE ) {
        iterationCount++;
        rstatus = brkiter->getRuleStatus();
        (void)rstatus;     // Suppress set but not used warning.
        if (iterationCount >= 10) {
           break;
        }
    }
    TEST_ASSERT(iterationCount == 6);

    brkiterPOSIX->setText(text);
    brkiterPOSIX->last();
    iterationCount = 0;
    while ( (offset = brkiterPOSIX->previous()) != UBRK_DONE ) {
        iterationCount++;
        rstatus = brkiterPOSIX->getRuleStatus();
        (void)rstatus;     // Suppress set but not used warning.
        if (iterationCount >= 10) {
           break;
        }
    }
    TEST_ASSERT(iterationCount == 6);
}

// Bug 7547 - verify that building a break itereator from empty rules produces an error.
//
void RBBITest::TestBug7547() {
    UnicodeString rules;
    UErrorCode status = U_ZERO_ERROR;
    UParseError parseError;
    RuleBasedBreakIterator breakIterator(rules, parseError, status);
    if (status != U_BRK_RULE_SYNTAX) {
        errln("%s:%d Expected U_BRK_RULE_SYNTAX, got %s", __FILE__, __LINE__, u_errorName(status));
    }
    if (parseError.line != 1 || parseError.offset != 0) {
        errln("parseError (line, offset) expected (1, 0), got (%d, %d)", parseError.line, parseError.offset);
    }
}


void RBBITest::TestBug12797() {
    UnicodeString rules = "!!chain; !!forward; $v=b c; a b; $v; !!reverse; .*;";
    UErrorCode status = U_ZERO_ERROR;
    UParseError parseError;
    RuleBasedBreakIterator bi(rules, parseError, status);
    if (U_FAILURE(status)) {
        errln("%s:%s status = %s", __FILE__, __LINE__, u_errorName(status));
        return;
    }
    UnicodeString text = "abc";
    bi.setText(text);
    bi.first();
    int32_t boundary = bi.next();
    if (boundary != 3) {
        errln("%s:%d expected boundary==3, got %d", __FILE__, __LINE__, boundary);
    }
}

void RBBITest::TestBug12918() {
    // This test triggers an assertion failure in dictbe.cpp
    const UChar *crasherString = u"\u3325\u4a16";
    UErrorCode status = U_ZERO_ERROR;
    UBreakIterator* iter = ubrk_open(UBRK_WORD, NULL, crasherString, -1, &status);
    if (U_FAILURE(status)) {
        dataerrln("%s:%d status = %s", __FILE__, __LINE__, u_errorName(status));
        return;
    }
    ubrk_first(iter);
    int32_t pos = 0;
    int32_t lastPos = -1;
    while((pos = ubrk_next(iter)) != UBRK_DONE) {
        if (pos <= lastPos) {
            errln("%s:%d (pos, lastPos) = (%d, %d)", __FILE__, __LINE__, pos, lastPos);
            break;
        }
    }
    ubrk_close(iter);
}

void RBBITest::TestBug12932() {
    // Node Stack overflow in the RBBI rule parser caused a seg fault.
    UnicodeString ruleStr(
            "((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((("
            "((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((("
            "(((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((()))"
            ")))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))"
            ")))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))"
            ")))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))");

    UErrorCode status = U_ZERO_ERROR;
    UParseError parseError;
    RuleBasedBreakIterator rbbi(ruleStr, parseError, status);
    if (status != U_BRK_RULE_SYNTAX) {
        errln("%s:%d expected U_BRK_RULE_SYNTAX, got %s",
                __FILE__, __LINE__, u_errorName(status));
    }
}


// Emoji Test. Verify that the sequences defined in the Unicode data file emoji-test.txt
//             remain undevided by ICU char, word and line break.
void RBBITest::TestEmoji() {
#if !UCONFIG_NO_REGULAR_EXPRESSIONS
    UErrorCode  status = U_ZERO_ERROR;

    CharString testFileName;
    testFileName.append(IntlTest::getSourceTestData(status), status);
    testFileName.appendPathPart("emoji-test.txt", status);
    if (U_FAILURE(status)) {
        errln("%s:%s %s while opening emoji-test.txt", __FILE__, __LINE__, u_errorName(status));
        return;
    }
    logln("Opening data file %s\n", testFileName.data());

    int    len;
    UChar *testFile = ReadAndConvertFile(testFileName.data(), len, "UTF-8", status);
    if (U_FAILURE(status) || testFile == NULL) {
        errln("%s:%s %s while opening emoji-test.txt", __FILE__, __LINE__, u_errorName(status));
        return;
    }
    UnicodeString testFileAsString(testFile, len);
    delete [] testFile;

    RegexMatcher lineMatcher(u"^.*?$", testFileAsString, UREGEX_MULTILINE, status);
    RegexMatcher hexMatcher(u"\\s*([a-f0-9]*)", UREGEX_CASE_INSENSITIVE, status);
    //           hexMatcher group(1) is a hex number, or empty string if no hex number present.
    int32_t lineNumber = 0;

    LocalPointer<BreakIterator> charBreaks(BreakIterator::createCharacterInstance(Locale::getEnglish(), status), status);
    LocalPointer<BreakIterator> wordBreaks(BreakIterator::createWordInstance(Locale::getEnglish(), status), status);
    LocalPointer<BreakIterator> lineBreaks(BreakIterator::createLineInstance(Locale::getEnglish(), status), status);
    if (U_FAILURE(status)) {
        dataerrln("%s:%d %s while opening break iterators", __FILE__, __LINE__, u_errorName(status));
        return;
    }

    while (lineMatcher.find()) {
        ++lineNumber;
        UnicodeString line = lineMatcher.group(status);
        hexMatcher.reset(line);
        UnicodeString testString;   // accumulates the emoji sequence.
        while (hexMatcher.find() && hexMatcher.group(1, status).length() > 0) {
            UnicodeString hex = hexMatcher.group(1, status);
            if (hex.length() > 8) {
                errln("%s:%d emoji-test.txt:%d invalid code point %s", __FILE__, __LINE__, lineNumber, CStr(hex)());
                break;
            }
            CharString hex8;
            hex8.appendInvariantChars(hex, status);
            UChar32 c = (UChar32)strtol(hex8.data(), NULL, 16);
            if (c<=0x10ffff) {
                testString.append(c);
            } else {
                errln("%s:%d emoji-test.txt:%d Error: Unicode Character %s value out of range.",
                        __FILE__, __LINE__, lineNumber, hex8.data());
                break;
            }
        }

        if (testString.length() > 1) {
            charBreaks->setText(testString);
            charBreaks->first();
            int32_t firstBreak = charBreaks->next();
            if (testString.length() != firstBreak) {
                errln("%s:%d  emoji-test.txt:%d Error, uexpected break at offset %d",
                        __FILE__, __LINE__, lineNumber, firstBreak);
            }
            wordBreaks->setText(testString);
            wordBreaks->first();
            firstBreak = wordBreaks->next();
            if (testString.length() != firstBreak) {
                errln("%s:%d  emoji-test.txt:%d Error, uexpected break at offset %d",
                        __FILE__, __LINE__, lineNumber, firstBreak);
            }
            lineBreaks->setText(testString);
            lineBreaks->first();
            firstBreak = lineBreaks->next();
            if (testString.length() != firstBreak) {
                errln("%s:%d  emoji-test.txt:%d Error, uexpected break at offset %d",
                        __FILE__, __LINE__, lineNumber, firstBreak);
            }
        }
    }
#endif
}


// TestBug12519  -  Correct handling of Locales by assignment / copy / clone

void RBBITest::TestBug12519() {
    UErrorCode status = U_ZERO_ERROR;
    LocalPointer<RuleBasedBreakIterator> biEn((RuleBasedBreakIterator *)BreakIterator::createWordInstance(Locale::getEnglish(), status));
    LocalPointer<RuleBasedBreakIterator> biFr((RuleBasedBreakIterator *)BreakIterator::createWordInstance(Locale::getFrance(), status));
    if (!assertSuccess(WHERE, status)) {
        dataerrln("%s %d status = %s", __FILE__, __LINE__, u_errorName(status));
        return;
    }
    assertTrue(WHERE, Locale::getEnglish() == biEn->getLocale(ULOC_VALID_LOCALE, status));

    assertTrue(WHERE, Locale::getFrench() == biFr->getLocale(ULOC_VALID_LOCALE, status));
    assertTrue(WHERE "Locales do not participate in BreakIterator equality.", *biEn == *biFr);

    LocalPointer<RuleBasedBreakIterator>cloneEn(biEn->clone());
    assertTrue(WHERE, *biEn == *cloneEn);
    assertTrue(WHERE, Locale::getEnglish() == cloneEn->getLocale(ULOC_VALID_LOCALE, status));

    LocalPointer<RuleBasedBreakIterator>cloneFr(biFr->clone());
    assertTrue(WHERE, *biFr == *cloneFr);
    assertTrue(WHERE, Locale::getFrench() == cloneFr->getLocale(ULOC_VALID_LOCALE, status));

    LocalPointer<RuleBasedBreakIterator>biDe((RuleBasedBreakIterator *)BreakIterator::createLineInstance(Locale::getGerman(), status));
    UnicodeString text("Hallo Welt");
    biDe->setText(text);
    assertTrue(WHERE "before assignment of \"biDe = biFr\", they should be different, but are equal.", *biFr != *biDe);
    *biDe = *biFr;
    assertTrue(WHERE "after assignment of \"biDe = biFr\", they should be equal, but are not.", *biFr == *biDe);
}

void RBBITest::TestBug12677() {
    // Check that stripping of comments from rules for getRules() is not confused by
    // the presence of '#' characters in the rules that do not introduce comments.
    UnicodeString rules(u"!!forward; \n"
                         "$x = [ab#];  # a set with a # literal. \n"
                         " # .;        # a comment that looks sort of like a rule.   \n"
                         " '#' '?';    # a rule with a quoted #   \n"
                       );

    UErrorCode status = U_ZERO_ERROR;
    UParseError pe;
    RuleBasedBreakIterator bi(rules, pe, status);
    assertSuccess(WHERE, status);
    UnicodeString rtRules = bi.getRules();
    assertEquals(WHERE, UnicodeString(u"!!forward;$x=[ab#];'#''?';"),  rtRules);
}


void RBBITest::TestTableRedundancies() {
    UErrorCode status = U_ZERO_ERROR;

    LocalPointer<RuleBasedBreakIterator> bi (
        (RuleBasedBreakIterator *)BreakIterator::createLineInstance(Locale::getEnglish(), status));
    assertSuccess(WHERE, status);
    if (U_FAILURE(status)) return;

    RBBIDataWrapper *dw = bi->fData;
    const RBBIStateTable *fwtbl = dw->fForwardTable;
    UBool in8Bits = fwtbl->fFlags & RBBI_8BITS_ROWS;
    int32_t numCharClasses = dw->fHeader->fCatCount;
    // printf("Char Classes: %d     states: %d\n", numCharClasses, fwtbl->fNumStates);

    // Check for duplicate columns (character categories)

    std::vector<UnicodeString> columns;
    for (int32_t column = 0; column < numCharClasses; column++) {
        UnicodeString s;
        for (int32_t r = 1; r < (int32_t)fwtbl->fNumStates; r++) {
            RBBIStateTableRow  *row = (RBBIStateTableRow *) (fwtbl->fTableData + (fwtbl->fRowLen * r));
            s.append(in8Bits ? row->r8.fNextState[column] : row->r16.fNextState[column]);
        }
        columns.push_back(s);
    }
    // Ignore column (char class) 0 while checking; it's special, and may have duplicates.
    for (int c1=1; c1<numCharClasses; c1++) {
        int limit = c1 < (int)fwtbl->fDictCategoriesStart ? fwtbl->fDictCategoriesStart : numCharClasses;
        for (int c2 = c1+1; c2 < limit; c2++) {
            if (columns.at(c1) == columns.at(c2)) {
                errln("%s:%d Duplicate columns (%d, %d)\n", __FILE__, __LINE__, c1, c2);
                goto out;
            }
        }
    }
  out:

    // Check for duplicate states
    std::vector<UnicodeString> rows;
    for (int32_t r=0; r < (int32_t)fwtbl->fNumStates; r++) {
        UnicodeString s;
        RBBIStateTableRow  *row = (RBBIStateTableRow *) (fwtbl->fTableData + (fwtbl->fRowLen * r));
        if (in8Bits) {
            s.append(row->r8.fAccepting);
            s.append(row->r8.fLookAhead);
            s.append(row->r8.fTagsIdx);
            for (int32_t column = 0; column < numCharClasses; column++) {
                s.append(row->r8.fNextState[column]);
            }
        } else {
            s.append(row->r16.fAccepting);
            s.append(row->r16.fLookAhead);
            s.append(row->r16.fTagsIdx);
            for (int32_t column = 0; column < numCharClasses; column++) {
                s.append(row->r16.fNextState[column]);
            }
        }
        rows.push_back(s);
    }
    for (int r1=0; r1 < (int32_t)fwtbl->fNumStates; r1++) {
        for (int r2 = r1+1; r2 < (int32_t)fwtbl->fNumStates; r2++) {
            if (rows.at(r1) == rows.at(r2)) {
                errln("%s:%d Duplicate rows (%d, %d)\n", __FILE__, __LINE__, r1, r2);
                return;
            }
        }
    }
}

// Bug 13447: verify that getRuleStatus() returns the value corresponding to current(),
//            even after next() has returned DONE.

void RBBITest::TestBug13447() {
    UErrorCode status = U_ZERO_ERROR;
    LocalPointer<RuleBasedBreakIterator> bi(
        (RuleBasedBreakIterator *)BreakIterator::createWordInstance(Locale::getEnglish(), status));
    assertSuccess(WHERE, status);
    if (U_FAILURE(status)) return;
    UnicodeString data(u"1234");
    bi->setText(data);
    assertEquals(WHERE, UBRK_WORD_NONE, bi->getRuleStatus());
    assertEquals(WHERE, 4, bi->next());
    assertEquals(WHERE, UBRK_WORD_NUMBER, bi->getRuleStatus());
    assertEquals(WHERE, UBRK_DONE, bi->next());
    assertEquals(WHERE, 4, bi->current());
    assertEquals(WHERE, UBRK_WORD_NUMBER, bi->getRuleStatus());
}

//  TestReverse exercises both the synthesized safe reverse rules and the logic
//  for filling the break iterator cache when starting from random positions
//  in the text.
//
//  It's a monkey test, working on random data, with the expected data obtained
//  from forward iteration (no safe rules involved), comparing with results
//  when indexing into the interior of the string (safe rules needed).

void RBBITest::TestReverse() {
    UErrorCode status = U_ZERO_ERROR;

    TestReverse(std::unique_ptr<RuleBasedBreakIterator>((RuleBasedBreakIterator *)
            BreakIterator::createCharacterInstance(Locale::getEnglish(), status)));
    assertSuccess(WHERE, status, true);
    status = U_ZERO_ERROR;
    TestReverse(std::unique_ptr<RuleBasedBreakIterator>((RuleBasedBreakIterator *)
            BreakIterator::createWordInstance(Locale::getEnglish(), status)));
    assertSuccess(WHERE, status, true);
    status = U_ZERO_ERROR;
    TestReverse(std::unique_ptr<RuleBasedBreakIterator>((RuleBasedBreakIterator *)
            BreakIterator::createLineInstance(Locale::getEnglish(), status)));
    assertSuccess(WHERE, status, true);
    status = U_ZERO_ERROR;
    TestReverse(std::unique_ptr<RuleBasedBreakIterator>((RuleBasedBreakIterator *)
            BreakIterator::createSentenceInstance(Locale::getEnglish(), status)));
    assertSuccess(WHERE, status, true);
}

void RBBITest::TestReverse(std::unique_ptr<RuleBasedBreakIterator>bi) {
    if (!bi) {
        return;
    }

    // From the mapping trie in the break iterator's internal data, create a
    // vector of UnicodeStrings, one for each character category, containing
    // all of the code points that map to that category. Unicode planes 0 and 1 only,
    // to avoid an execess of unassigned code points.

    RBBIDataWrapper *data = bi->fData;
    int32_t categoryCount = data->fHeader->fCatCount;
    UCPTrie *trie = data->fTrie;
    bool use8BitsTrie = ucptrie_getValueWidth(trie) == UCPTRIE_VALUE_BITS_8;
    uint32_t dictBit = use8BitsTrie ? 0x0080 : 0x4000;

    std::vector<UnicodeString> strings(categoryCount, UnicodeString());
    for (int cp=0; cp<0x1fff0; ++cp) {
        int cat = ucptrie_get(trie, cp);
        cat &= ~dictBit;    // And off the dictionary bit from the category.
        assertTrue(WHERE, cat < categoryCount && cat >= 0);
        if (cat < 0 || cat >= categoryCount) return;
        strings[cat].append(cp);
    }

    icu_rand randomGen;
    const int testStringLength = 10000;
    UnicodeString testString;

    for (int i=0; i<testStringLength; ++i) {
        int charClass = randomGen() % categoryCount;
        if (strings[charClass].length() > 0) {
            int cp = strings[charClass].char32At(randomGen() % strings[charClass].length());
            testString.append(cp);
        }
    }

    typedef std::pair<UBool, int32_t> Result;
    std::vector<Result> expectedResults;
    bi->setText(testString);
    for (int i=0; i<testString.length(); ++i) {
        bool isboundary = bi->isBoundary(i);
        int  ruleStatus = bi->getRuleStatus();
        expectedResults.push_back(std::make_pair(isboundary, ruleStatus));
    }

    for (int i=testString.length()-1; i>=0; --i) {
        bi->setText(testString);   // clears the internal break cache
        Result expected = expectedResults[i];
        assertEquals(WHERE, expected.first, bi->isBoundary(i));
        assertEquals(WHERE, expected.second, bi->getRuleStatus());
    }
}


// Ticket 13692 - finding word boundaries in very large numbers or words could
//                be very time consuming. When the problem was present, this void test
//                would run more than fifteen minutes, which is to say, the failure was noticeale.

void RBBITest::TestBug13692() {
    UErrorCode status = U_ZERO_ERROR;
    LocalPointer<RuleBasedBreakIterator> bi ((RuleBasedBreakIterator *)
            BreakIterator::createWordInstance(Locale::getEnglish(), status), status);
    if (!assertSuccess(WHERE, status, true)) {
        return;
    }
    constexpr int32_t LENGTH = 1000000;
    UnicodeString longNumber(LENGTH, (UChar32)u'3', LENGTH);
    for (int i=0; i<20; i+=2) {
        longNumber.setCharAt(i, u' ');
    }
    bi->setText(longNumber);
    assertFalse(WHERE, bi->isBoundary(LENGTH-5));
    assertSuccess(WHERE, status);
}


void RBBITest::TestProperties() {
    UErrorCode errorCode = U_ZERO_ERROR;
    UnicodeSet prependSet(UNICODE_STRING_SIMPLE("[:GCB=Prepend:]"), errorCode);
    if (!prependSet.isEmpty()) {
        errln(
            "[:GCB=Prepend:] is not empty any more. "
            "Uncomment relevant lines in source/data/brkitr/char.txt and "
            "change this test to the opposite condition.");
    }
}


//
//  TestDebug    -  A place-holder test for debugging purposes.
//                  For putting in fragments of other tests that can be invoked
//                  for tracing  without a lot of unwanted extra stuff happening.
//
void RBBITest::TestDebug(void) {
    UErrorCode status = U_ZERO_ERROR;
    LocalPointer<RuleBasedBreakIterator> bi ((RuleBasedBreakIterator *)
            BreakIterator::createCharacterInstance(Locale::getEnglish(), status), status);
    if (!assertSuccess(WHERE, status, true)) {
        return;
    }
    const UnicodeString &rules = bi->getRules();
    UParseError pe;
    LocalPointer<RuleBasedBreakIterator> newbi(new RuleBasedBreakIterator(rules, pe, status));
    assertSuccess(WHERE, status);
}


//
//  TestDebugRules   A stub test for use in debugging rule compilation problems.
//                   Can be freely altered as needed or convenient.
//                   Leave disabled - #ifdef'ed out - when not activley debugging. The rule source
//                   data files may not be available in all environments.
//                   Any permanent test cases should be moved to rbbitst.txt
//                   (see Bug 20303 in that file, for example), or to another test function in this file.
//
void RBBITest::TestDebugRules() {
#if 0
    const char16_t *rules = u""
        "!!quoted_literals_only; \n"
        "!!chain; \n"
        "!!lookAheadHardBreak; \n"
        " \n"
        // "[a] / ; \n"
        "[a] [b] / [c] [d]; \n"
        "[a] [b] / [c] [d] {100}; \n"
        "[x] [a] [b] / [c] [d] {100}; \n"
        "[a] [b] [c] / [d] {100}; \n"
        //" [c] [d] / [e] [f]; \n"
        //"[a] [b] / [c]; \n"
        ;

    UErrorCode status = U_ZERO_ERROR;
    CharString path(pathToDataDirectory(), status);
    path.appendPathPart("brkitr", status);
    path.appendPathPart("rules", status);
    path.appendPathPart("line.txt", status);
    int    len;
    std::unique_ptr<UChar []> testFile(ReadAndConvertFile(path.data(), len, "UTF-8", status));
    if (!assertSuccess(WHERE, status)) {
        return;
    }

    UParseError pe;
    // rules = testFile.get();
    RuleBasedBreakIterator *bi = new RuleBasedBreakIterator(rules, pe, status);

    if (!assertSuccess(WHERE, status)) {
        delete bi;
        return;
    }
    // bi->dumpTables();

    delete bi;
#endif
}

void RBBITest::testTrieStateTable(int32_t numChar, bool expectedTrieWidthIn8Bits, bool expectedStateRowIn8Bits) {
    UCPTrieValueWidth expectedTrieWidth = expectedTrieWidthIn8Bits ? UCPTRIE_VALUE_BITS_8 : UCPTRIE_VALUE_BITS_16;
    int32_t expectedStateRowBits = expectedStateRowIn8Bits ? RBBI_8BITS_ROWS : 0;
    // Text are duplicate characters from U+4E00 to U+4FFF
    UnicodeString text;
    for (UChar c = 0x4e00; c < 0x5000; c++) {
        text.append(c).append(c);
    }
    // Generate rule which will caused length+4 character classes and
    // length+3 states
    UnicodeString rules(u"!!quoted_literals_only;");
    for (UChar c = 0x4e00; c < 0x4e00 + numChar; c++) {
        rules.append(u'\'').append(c).append(c).append(u"';");
    }
    rules.append(u".;");
    UErrorCode status = U_ZERO_ERROR;
    UParseError parseError;
    RuleBasedBreakIterator bi(rules, parseError, status);

    assertEquals(WHERE, numChar + 4, bi.fData->fHeader->fCatCount);
    assertEquals(WHERE, numChar + 3, bi.fData->fForwardTable->fNumStates);
    assertEquals(WHERE, expectedTrieWidth, ucptrie_getValueWidth(bi.fData->fTrie));
    assertEquals(WHERE, expectedStateRowBits, bi.fData->fForwardTable->fFlags & RBBI_8BITS_ROWS);
    assertEquals(WHERE, expectedStateRowBits, bi.fData->fReverseTable->fFlags & RBBI_8BITS_ROWS);

    bi.setText(text);

    int32_t pos;
    int32_t i = 0;
    while ((pos = bi.next()) > 0) {
        // The first numChar should not break between the pair
        if (i++ < numChar) {
            assertEquals(WHERE, i * 2, pos);
        } else {
            // After the first numChar next(), break on each character.
            assertEquals(WHERE, i + numChar, pos);
        }
    }
    while ((pos = bi.previous()) > 0) {
        // The first numChar should not break between the pair
        if (--i < numChar) {
            assertEquals(WHERE, i * 2, pos);
        } else {
            // After the first numChar next(), break on each character.
            assertEquals(WHERE, i + numChar, pos);
        }
    }
}

void RBBITest::Test8BitsTrieWith8BitStateTable() {
    testTrieStateTable(251, true /* expectedTrieWidthIn8Bits */, true /* expectedStateRowIn8Bits */);
}

void RBBITest::Test16BitsTrieWith8BitStateTable() {
    testTrieStateTable(252, false /* expectedTrieWidthIn8Bits */, true /* expectedStateRowIn8Bits */);
}

void RBBITest::Test16BitsTrieWith16BitStateTable() {
    testTrieStateTable(253, false /* expectedTrieWidthIn8Bits */, false /* expectedStateRowIn8Bits */);
}

void RBBITest::Test8BitsTrieWith16BitStateTable() {
    // Test UCPTRIE_VALUE_BITS_8 with 16 bits rows. Use a different approach to
    // create state table in 16 bits.

    // Generate 510 'a' as text
    UnicodeString text;
    for (int32_t i = 0; i < 510; i++) {
        text.append(u'a');
    }

    UnicodeString rules(u"!!quoted_literals_only;'");
    // 254 'a' in the rule will cause 256 states
    for (int32_t i = 0; i < 254; i++) {
        rules.append(u'a');
    }
    rules.append(u"';.;");

    UErrorCode status = U_ZERO_ERROR;
    UParseError parseError;
    LocalPointer<RuleBasedBreakIterator> bi(new RuleBasedBreakIterator(rules, parseError, status));

    assertEquals(WHERE, 256, bi->fData->fForwardTable->fNumStates);
    assertEquals(WHERE, UCPTRIE_VALUE_BITS_8, ucptrie_getValueWidth(bi->fData->fTrie));
    assertEquals(WHERE,
                 false, RBBI_8BITS_ROWS == (bi->fData->fForwardTable->fFlags & RBBI_8BITS_ROWS));
    bi->setText(text);

    // break positions:
    // 254, 508, 509, ... 510
    assertEquals("next()", 254, bi->next());
    int32_t i = 0;
    int32_t pos;
    while ((pos = bi->next()) > 0) {
        assertEquals(WHERE, 508 + i , pos);
        i++;
    }
    i = 0;
    while ((pos = bi->previous()) > 0) {
        i++;
        if (pos >= 508) {
            assertEquals(WHERE, 510 - i , pos);
        } else {
            assertEquals(WHERE, 254 , pos);
        }
    }
}

// Test that both compact (8 bit) and full sized (16 bit) rbbi tables work, and
// that there are no problems with rules at the size that transitions between the two.
//
// A rule that matches a literal string, like 'abcdefghij', will require one state and
// one character class per character in the string. So we can make a rule to tickle the
// boundaries by using literal strings of various lengths.
//
// For both the number of states and the number of character classes, the eight bit format
// only has 7 bits available, allowing for 128 values. For both, a few values are reserved,
// leaving 120 something available. This test runs the string over the range of 120 - 130,
// which allows some margin for changes to the number of values reserved by the rule builder
// without breaking the test.

void RBBITest::TestTable_8_16_Bits() {

    // testStr serves as both the source of the rule string (truncated to the desired length)
    // and as test data to check matching behavior. A break rule consisting of the first 120
    // characters of testStr will match the first 120 chars of the full-length testStr.
    UnicodeString testStr;
    for (UChar c=0x3000; c<0x3200; ++c) {
        testStr.append(c);
    }

    const int32_t startLength = 120;   // The shortest rule string to test.
    const int32_t endLength = 260;     // The longest rule string to test
    const int32_t increment = this->quick ? endLength - startLength : 1;

    for (int32_t ruleLen=startLength; ruleLen <= endLength; ruleLen += increment) {
        UParseError parseError;
        UErrorCode status = U_ZERO_ERROR;

        UnicodeString ruleString{u"!!quoted_literals_only; '#';"};
        ruleString.findAndReplace(UnicodeString(u"#"), UnicodeString(testStr, 0, ruleLen));
        RuleBasedBreakIterator bi(ruleString, parseError, status);
        if (!assertSuccess(WHERE, status)) {
            errln(ruleString);
            break;
        }
        // bi.dumpTables();

        // Verify that the break iterator is functioning - that the first boundary found
        // in testStr is at the length of the rule string.
        bi.setText(testStr);
        assertEquals(WHERE, ruleLen, bi.next());

        // Reverse iteration. Do a setText() first, to flush the break iterator's internal cache
        // of previously detected boundaries, thus forcing the engine to run the safe reverse rules.
        bi.setText(testStr);
        int32_t result = bi.preceding(ruleLen);
        assertEquals(WHERE, 0, result);

        // Verify that the range of rule lengths being tested cover the translations
        // from 8 to 16 bit data.
        bool has8BitRowData = bi.fData->fForwardTable->fFlags & RBBI_8BITS_ROWS;
        bool has8BitsTrie = ucptrie_getValueWidth(bi.fData->fTrie) == UCPTRIE_VALUE_BITS_8;

        if (ruleLen == startLength) {
            assertEquals(WHERE, true, has8BitRowData);
            assertEquals(WHERE, true, has8BitsTrie);
        }
        if (ruleLen == endLength) {
            assertEquals(WHERE, false, has8BitRowData);
            assertEquals(WHERE, false, has8BitsTrie);
        }
    }
}

/* Test handling of a large number of look-ahead rules.
 * The number of rules in the test exceeds the implementation limits prior to the
 * improvements introduced with #13590.
 *
 * The test look-ahead rules have the form "AB / CE"; "CD / EG"; ...
 * The text being matched is sequential, "ABCDEFGHI..."
 *
 * The upshot is that the look-ahead rules all match on their preceding context,
 * and consequently must save a potential result, but then fail to match on their
 * trailing context, so that they don't actually cause a boundary.
 *
 * Additionally, add a ".*" rule, so there are no boundaries unless a
 * look-ahead hard-break rule forces one.
 */
void RBBITest::TestBug13590() {
    UnicodeString rules {u"!!quoted_literals_only; !!chain; .*;\n"};

    const int NUM_LOOKAHEAD_RULES = 50;
    const char16_t STARTING_CHAR = u'\u5000';
    char16_t firstChar;
    for (int ruleNum = 0; ruleNum < NUM_LOOKAHEAD_RULES; ++ruleNum) {
        firstChar = STARTING_CHAR + ruleNum*2;
        rules.append(u'\'') .append(firstChar) .append(firstChar+1) .append(u'\'')
             .append(u' ') .append(u'/') .append(u' ')
             .append(u'\'') .append(firstChar+2) .append(firstChar+4) .append(u'\'')
             .append(u';') .append(u'\n');
    }

    // Change the last rule added from the form "UV / WY" to "UV / WX".
    // Changes the rule so that it will match - all 4 chars are in ascending sequence.
    rules.findAndReplace(UnicodeString(firstChar+4), UnicodeString(firstChar+3));

    UErrorCode status = U_ZERO_ERROR;
    UParseError parseError;
    RuleBasedBreakIterator bi(rules, parseError, status);
    if (!assertSuccess(WHERE, status)) {
        errln(rules);
        return;
    }
    // bi.dumpTables();

    UnicodeString testString;
    for (char16_t c = STARTING_CHAR-200; c < STARTING_CHAR + NUM_LOOKAHEAD_RULES*4; ++c) {
        testString.append(c);
    }
    bi.setText(testString);

    int breaksFound = 0;
    while (bi.next() != UBRK_DONE) {
        ++breaksFound;
    }

    // Two matches are expected, one from the last rule that was explicitly modified,
    // and one at the end of the text.
    assertEquals(WHERE, 2, breaksFound);
}


#if U_ENABLE_TRACING
static std::vector<std::string> gData;
static std::vector<int32_t> gEntryFn;
static std::vector<int32_t> gExitFn;
static std::vector<int32_t> gDataFn;

static void U_CALLCONV traceData(
        const void*,
        int32_t fnNumber,
        int32_t,
        const char *,
        va_list args) {
    if (UTRACE_UBRK_START <= fnNumber && fnNumber <= UTRACE_UBRK_LIMIT) {
        const char* data = va_arg(args, const char*);
        gDataFn.push_back(fnNumber);
        gData.push_back(data);
    }
}

static void traceEntry(const void *, int32_t fnNumber) {
    if (UTRACE_UBRK_START <= fnNumber && fnNumber <= UTRACE_UBRK_LIMIT) {
        gEntryFn.push_back(fnNumber);
    }
}

static void traceExit(const void *, int32_t fnNumber, const char *, va_list) {
    if (UTRACE_UBRK_START <= fnNumber && fnNumber <= UTRACE_UBRK_LIMIT) {
        gExitFn.push_back(fnNumber);
    }
}


void RBBITest::assertTestTraceResult(int32_t fnNumber, const char* expectedData) {
    assertEquals("utrace_entry should be called ", 1, gEntryFn.size());
    assertEquals("utrace_entry should be called with ", fnNumber, gEntryFn[0]);
    assertEquals("utrace_exit should be called ", 1, gExitFn.size());
    assertEquals("utrace_exit should be called with ", fnNumber, gExitFn[0]);

    if (expectedData == nullptr) {
      assertEquals("utrace_data should not be called ", 0, gDataFn.size());
      assertEquals("utrace_data should not be called ", 0, gData.size());
    } else {
      assertEquals("utrace_data should be called ", 1, gDataFn.size());
      assertEquals("utrace_data should be called with ", fnNumber, gDataFn[0]);
      assertEquals("utrace_data should be called ", 1, gData.size());
      assertEquals("utrace_data should pass in ", expectedData, gData[0].c_str());
    }
}

void SetupTestTrace() {
    gEntryFn.clear();
    gExitFn.clear();
    gDataFn.clear();
    gData.clear();

    const void* context = nullptr;
    utrace_setFunctions(context, traceEntry, traceExit, traceData);
    utrace_setLevel(UTRACE_INFO);
}

void RBBITest::TestTraceCreateCharacter(void) {
    SetupTestTrace();
    IcuTestErrorCode status(*this, "TestTraceCreateCharacter");
    LocalPointer<BreakIterator> brkitr(
        BreakIterator::createCharacterInstance("zh-CN", status));
    status.errIfFailureAndReset();
    assertTestTraceResult(UTRACE_UBRK_CREATE_CHARACTER, nullptr);
}

void RBBITest::TestTraceCreateTitle(void) {
    SetupTestTrace();
    IcuTestErrorCode status(*this, "TestTraceCreateTitle");
    LocalPointer<BreakIterator> brkitr(
        BreakIterator::createTitleInstance("zh-CN", status));
    status.errIfFailureAndReset();
    assertTestTraceResult(UTRACE_UBRK_CREATE_TITLE, nullptr);
}

void RBBITest::TestTraceCreateSentence(void) {
    SetupTestTrace();
    IcuTestErrorCode status(*this, "TestTraceCreateSentence");
    LocalPointer<BreakIterator> brkitr(
        BreakIterator::createSentenceInstance("zh-CN", status));
    status.errIfFailureAndReset();
    assertTestTraceResult(UTRACE_UBRK_CREATE_SENTENCE, nullptr);
}

void RBBITest::TestTraceCreateWord(void) {
    SetupTestTrace();
    IcuTestErrorCode status(*this, "TestTraceCreateWord");
    LocalPointer<BreakIterator> brkitr(
        BreakIterator::createWordInstance("zh-CN", status));
    status.errIfFailureAndReset();
    assertTestTraceResult(UTRACE_UBRK_CREATE_WORD, nullptr);
}

void RBBITest::TestTraceCreateLine(void) {
    SetupTestTrace();
    IcuTestErrorCode status(*this, "TestTraceCreateLine");
    LocalPointer<BreakIterator> brkitr(
        BreakIterator::createLineInstance("zh-CN", status));
    status.errIfFailureAndReset();
    assertTestTraceResult(UTRACE_UBRK_CREATE_LINE, "");
}

void RBBITest::TestTraceCreateLineStrict(void) {
    SetupTestTrace();
    IcuTestErrorCode status(*this, "TestTraceCreateLineStrict");
    LocalPointer<BreakIterator> brkitr(
        BreakIterator::createLineInstance("zh-CN-u-lb-strict", status));
    status.errIfFailureAndReset();
    assertTestTraceResult(UTRACE_UBRK_CREATE_LINE, "strict");
}

void RBBITest::TestTraceCreateLineNormal(void) {
    SetupTestTrace();
    IcuTestErrorCode status(*this, "TestTraceCreateLineNormal");
    LocalPointer<BreakIterator> brkitr(
        BreakIterator::createLineInstance("zh-CN-u-lb-normal", status));
    status.errIfFailureAndReset();
    assertTestTraceResult(UTRACE_UBRK_CREATE_LINE, "normal");
}

void RBBITest::TestTraceCreateLineLoose(void) {
    SetupTestTrace();
    IcuTestErrorCode status(*this, "TestTraceCreateLineLoose");
    LocalPointer<BreakIterator> brkitr(
        BreakIterator::createLineInstance("zh-CN-u-lb-loose", status));
    status.errIfFailureAndReset();
    assertTestTraceResult(UTRACE_UBRK_CREATE_LINE, "loose");
}

void RBBITest::TestTraceCreateBreakEngine(void) {
    rbbi_cleanup();
    SetupTestTrace();
    IcuTestErrorCode status(*this, "TestTraceCreateBreakEngine");
    LocalPointer<BreakIterator> brkitr(
        BreakIterator::createWordInstance("zh-CN", status));
    status.errIfFailureAndReset();
    assertTestTraceResult(UTRACE_UBRK_CREATE_WORD, nullptr);

    // To word break the following text, BreakIterator will create 5 dictionary
    // break engine internally.
    brkitr->setText(
        u"test "
        u"測試 " // Hani
        u"សាកល្បង " // Khmr
        u"ທົດສອບ " // Laoo
        u"စမ်းသပ်မှု " // Mymr
        u"ทดสอบ " // Thai
        u"test "
    );

    // Loop through all the text.
    while (brkitr->next() > 0) ;

    assertEquals("utrace_entry should be called ", 6, gEntryFn.size());
    assertEquals("utrace_exit should be called ", 6, gExitFn.size());
    assertEquals("utrace_data should be called ", 5, gDataFn.size());

    for (std::vector<int>::size_type i = 0; i < gDataFn.size(); i++) {
        assertEquals("utrace_entry should be called ",
                     UTRACE_UBRK_CREATE_BREAK_ENGINE, gEntryFn[i+1]);
        assertEquals("utrace_exit should be called ",
                     UTRACE_UBRK_CREATE_BREAK_ENGINE, gExitFn[i+1]);
        assertEquals("utrace_data should be called ",
                     UTRACE_UBRK_CREATE_BREAK_ENGINE, gDataFn[i]);
    }

    assertEquals("utrace_data should pass ", "Hani", gData[0].c_str());
    assertEquals("utrace_data should pass ", "Khmr", gData[1].c_str());
    assertEquals("utrace_data should pass ", "Laoo", gData[2].c_str());
    assertEquals("utrace_data should pass ", "Mymr", gData[3].c_str());
    assertEquals("utrace_data should pass ", "Thai", gData[4].c_str());

}
#endif

void RBBITest::TestUnpairedSurrogate() {
    UnicodeString rules(u"ab;");

    UErrorCode status = U_ZERO_ERROR;
    UParseError pe;
    RuleBasedBreakIterator bi1(rules, pe, status);
    assertSuccess(WHERE, status);
    UnicodeString rtRules = bi1.getRules();
    // make sure the simple one work first.
    assertEquals(WHERE, rules,  rtRules);


    rules = UnicodeString(u"a\\ud800b;").unescape();
    pe.line = 0;
    pe.offset = 0;
    RuleBasedBreakIterator bi2(rules, pe, status);
    assertEquals(WHERE "unpaired lead surrogate", U_ILLEGAL_CHAR_FOUND , status);
    if (pe.line != 1 || pe.offset != 1) {
        errln("pe (line, offset) expected (1, 1), got (%d, %d)", pe.line, pe.offset);
    }

    status = U_ZERO_ERROR;
    rules = UnicodeString(u"a\\ude00b;").unescape();
    pe.line = 0;
    pe.offset = 0;
    RuleBasedBreakIterator bi3(rules, pe, status);
    assertEquals(WHERE "unpaired tail surrogate", U_ILLEGAL_CHAR_FOUND , status);
    if (pe.line != 1 || pe.offset != 1) {
        errln("pe (line, offset) expected (1, 1), got (%d, %d)", pe.line, pe.offset);
    }

    // make sure the surrogate one work too.
    status = U_ZERO_ERROR;
    rules = UnicodeString(u"a😀b;");
    RuleBasedBreakIterator bi4(rules, pe, status);
    rtRules = bi4.getRules();
    assertEquals(WHERE, rules, rtRules);
}

// Read file generated by
// https://github.com/unicode-org/lstm_word_segmentation/blob/master/segment_text.py
// as test cases and compare the Output.
// Format of the file
//   Model:\t[Model Name (such as 'Thai_graphclust_model4_heavy')]
//   Embedding:\t[Embedding type (such as 'grapheme_clusters_tf')]
//   Input:\t[source text]
//   Output:\t[expected output separated by | ]
//   Input: ...
//   Output: ...

void RBBITest::runLSTMTestFromFile(const char* filename, UScriptCode script) {
    // The expectation in this test depends on LSTM, skip the test if the
    // configuration is not build with LSTM data.
    if (skipLSTMTest()) {
        return;
    }
    UErrorCode   status = U_ZERO_ERROR;
    LocalPointer<BreakIterator> iterator(BreakIterator::createWordInstance(Locale(), status));
    if (U_FAILURE(status)) {
        errln("%s:%d Error %s Cannot create Word BreakIterator", __FILE__, __LINE__, u_errorName(status));
        return;
    }
    //  Open and read the test data file.
    const char *testDataDirectory = IntlTest::getSourceTestData(status);
    CharString testFileName(testDataDirectory, -1, status);
    testFileName.append(filename, -1, status);

    int len;
    UChar *testFile = ReadAndConvertFile(testFileName.data(), len, "UTF-8", status);
    if (U_FAILURE(status)) {
        errln("%s:%d Error %s opening test file %s", __FILE__, __LINE__, u_errorName(status), filename);
        return;
    }

    //  Put the test data into a UnicodeString
    UnicodeString testString(FALSE, testFile, len);

    int32_t start = 0;

    UnicodeString line;
    int32_t end;
    std::string actual_sep_str;
    int32_t caseNum = 0;
    // Iterate through all the lines in the test file.
    do {
        int32_t cr = testString.indexOf(u'\r', start);
        int32_t lf = testString.indexOf(u'\n', start);
        end = cr >= 0 ? (lf >= 0 ? std::min(cr, lf) : cr) : lf;
        line = testString.tempSubString(start, end < 0 ? INT32_MAX : end - start);
        if (line.length() > 0) {
            // Separate each line to key and value by TAB.
            int32_t tab = line.indexOf(u'\t');
            UnicodeString key = line.tempSubString(0, tab);
            const UnicodeString value = line.tempSubString(tab+1);

            if (key == "Model:") {
                // Verify the expectation in the test file match the LSTM model
                // we are using now.
                const LSTMData* data = CreateLSTMDataForScript(script, status);
                if (U_FAILURE(status)) {
                    dataerrln("%s:%d Error %s Cannot create LSTM data for script %s",
                              __FILE__, __LINE__, u_errorName(status), uscript_getName(script));
                    return;
                }
                UnicodeString name(LSTMDataName(data));
                DeleteLSTMData(data);
                if (value != name) {
                    std::string utf8Name, utf8Value;
                    dataerrln("%s:%d Error %s The LSTM data for script %s is %s instead of %s",
                              __FILE__, __LINE__, u_errorName(status), uscript_getName(script),
                              name.toUTF8String<std::string>(utf8Name).c_str(),
                              value.toUTF8String<std::string>(utf8Value).c_str());
                    return;
                }
            } else if (key == "Input:") {
                UnicodeString input("prefix ");
                input += value + " suffix";
                std::stringstream ss;

                // Construct the UText which is expected by the the engine as
                // input from the UnicodeString.
                UText ut = UTEXT_INITIALIZER;
                utext_openConstUnicodeString(&ut, &input, &status);
                if (U_FAILURE(status)) {
                    dataerrln("Could not utext_openConstUnicodeString for " + value + UnicodeString(u_errorName(status)));
                    return;
                }

                iterator->setText(&ut, status);
                if (U_FAILURE(status)) {
                    errln("%s:%d Error %s Could not setText to BreakIterator", __FILE__, __LINE__, u_errorName(status));
                    return;
                }

                int32_t bp;
                for (bp = iterator->first(); bp != BreakIterator::DONE; bp = iterator->next()) {
                    ss << bp;
                    if (bp != input.length()) {
                        ss << ", ";
                    }
                }

                utext_close(&ut);
                // Turn the break points into a string for easy comparison
                // output.
                actual_sep_str = "{" + ss.str() + "}";
            } else if (key == "Output:" && !actual_sep_str.empty()) {
                UnicodeString input("prefix| |");
                input += value + "| |suffix";
                std::string d;
                int32_t sep;
                int32_t start = 0;
                int32_t curr = 0;
                std::stringstream ss;
                // Include 0 as the break point.
                ss << "0, ";
                while ((sep = input.indexOf(u'|', start)) >= 0) {
                    int32_t len = sep - start;
                    if (len > 0) {
                        if (curr > 0) {
                            ss << ", ";
                        }
                        curr += len;
                        ss << curr;
                    }
                    start = sep + 1;
                }
                // Include end of the string as break point.
                ss << ", " << curr + input.length() - start;
                // Turn the break points into a string for easy comparison
                // output.
                std::string expected = "{" + ss.str() + "}";
                std::string utf8;

                assertEquals((input + " Test Case#" + caseNum).toUTF8String<std::string>(utf8).c_str(),
                             expected.c_str(), actual_sep_str.c_str());
                actual_sep_str.clear();
            }
        }
        start = std::max(cr, lf) + 1;
    } while (end >= 0);

    delete [] testFile;
}

void RBBITest::TestLSTMThai() {
    runLSTMTestFromFile("Thai_graphclust_model4_heavy_Test.txt", USCRIPT_THAI);
}

void RBBITest::TestLSTMBurmese() {
    runLSTMTestFromFile("Burmese_graphclust_model5_heavy_Test.txt", USCRIPT_MYANMAR);
}

#endif // #if !UCONFIG_NO_BREAK_ITERATION