• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1 // © 2016 and later: Unicode, Inc. and others.
2 // License & terms of use: http://www.unicode.org/copyright.html
3 /********************************************************************
4  * COPYRIGHT:
5  * Copyright (c) 2012-2016, International Business Machines Corporation
6  * and others. All Rights Reserved.
7  ********************************************************************/
8 //
9 //   file:  alphaindextst.cpp
10 //          Alphabetic Index Tests.
11 //
12 //   Note: please... no character literals cast to UChars.. use (UChar)0xZZZZ
13 
14 #include <stdio.h>  // for sprintf
15 
16 #include "intltest.h"
17 #include "alphaindextst.h"
18 #include "cmemory.h"
19 
20 #include "unicode/alphaindex.h"
21 #include "unicode/coll.h"
22 #include "unicode/localpointer.h"
23 #include "unicode/tblcoll.h"
24 #include "unicode/uniset.h"
25 #include "unicode/uscript.h"
26 
27 #if !UCONFIG_NO_COLLATION && !UCONFIG_NO_NORMALIZATION
28 
29 // #include <string>
30 // #include <iostream>
31 
32 namespace {
33 
joinLabelsAndAppend(AlphabeticIndex::ImmutableIndex & index,UnicodeString & dest)34 UnicodeString joinLabelsAndAppend(AlphabeticIndex::ImmutableIndex &index, UnicodeString &dest) {
35     int32_t oldLength = dest.length();
36     const AlphabeticIndex::Bucket *bucket;
37     for (int32_t i = 0; (bucket = index.getBucket(i)) != NULL; ++i) {
38         if (dest.length() > oldLength) {
39             dest.append((UChar)0x3A);  // ':'
40         }
41         dest.append(bucket->getLabel());
42     }
43     return dest;
44 }
45 
46 }  // namespace
47 
AlphabeticIndexTest()48 AlphabeticIndexTest::AlphabeticIndexTest() {
49 }
50 
~AlphabeticIndexTest()51 AlphabeticIndexTest::~AlphabeticIndexTest() {
52 }
53 
runIndexedTest(int32_t index,UBool exec,const char * & name,char *)54 void AlphabeticIndexTest::runIndexedTest( int32_t index, UBool exec, const char* &name, char* /*par*/ )
55 {
56     if (exec) logln("TestSuite AlphabeticIndex: ");
57     TESTCASE_AUTO_BEGIN;
58     TESTCASE_AUTO(APITest);
59     TESTCASE_AUTO(ManyLocalesTest);
60     TESTCASE_AUTO(HackPinyinTest);
61     TESTCASE_AUTO(TestBug9009);
62     TESTCASE_AUTO(TestIndexCharactersList);
63     TESTCASE_AUTO(TestHaniFirst);
64     TESTCASE_AUTO(TestPinyinFirst);
65     TESTCASE_AUTO(TestSchSt);
66     TESTCASE_AUTO(TestNoLabels);
67     TESTCASE_AUTO(TestChineseZhuyin);
68     TESTCASE_AUTO(TestJapaneseKanji);
69     TESTCASE_AUTO(TestChineseUnihan);
70     TESTCASE_AUTO(testHasBuckets);
71     TESTCASE_AUTO_END;
72 }
73 
74 #define TEST_CHECK_STATUS UPRV_BLOCK_MACRO_BEGIN { \
75     if (U_FAILURE(status)) { \
76         dataerrln("%s:%d: Test failure.  status=%s", \
77                   __FILE__, __LINE__, u_errorName(status)); \
78         return; \
79     } \
80 } UPRV_BLOCK_MACRO_END
81 
82 #define TEST_ASSERT(expr) UPRV_BLOCK_MACRO_BEGIN { \
83     if ((expr)==FALSE) { \
84         errln("%s:%d: Test failure \n", __FILE__, __LINE__); \
85     } \
86 } UPRV_BLOCK_MACRO_END
87 
88 //
89 //  APITest.   Invoke every function at least once, and check that it does something.
90 //             Does not attempt to check complete functionality.
91 //
APITest()92 void AlphabeticIndexTest::APITest() {
93     //
94     //  Simple constructor and destructor,  getBucketCount()
95     //
96     UErrorCode status = U_ZERO_ERROR;
97     int32_t lc = 0;
98     int32_t i  = 0;
99     LocalPointer<AlphabeticIndex> index(new AlphabeticIndex(Locale::getEnglish(), status));
100     TEST_CHECK_STATUS;
101     lc = index->getBucketCount(status);
102     TEST_CHECK_STATUS;
103     TEST_ASSERT(28 == lc);    // 26 letters plus two under/overflow labels.
104     //printf("getBucketCount() == %d\n", lc);
105     index.adoptInstead(nullptr);
106 
107     // Constructor from a Collator
108     //
109     status = U_ZERO_ERROR;
110     LocalPointer<RuleBasedCollator> coll(dynamic_cast<RuleBasedCollator *>(
111         Collator::createInstance(Locale::getGerman(), status)), status);
112     TEST_CHECK_STATUS;
113     TEST_ASSERT(coll.isValid());
114     RuleBasedCollator *originalColl = coll.getAlias();
115     index.adoptInstead(new AlphabeticIndex(coll.orphan(), status));
116     TEST_CHECK_STATUS;
117     TEST_ASSERT(originalColl == &index->getCollator());
118     assertEquals("only the underflow label in an index built from a collator",
119                  1, index->getBucketCount(status));
120     TEST_CHECK_STATUS;
121     index.adoptInstead(nullptr);
122 
123 
124     // addLabels()
125 
126     status = U_ZERO_ERROR;
127     index.adoptInstead(new AlphabeticIndex(Locale::getEnglish(), status));
128     TEST_CHECK_STATUS;
129     UnicodeSet additions;
130     additions.add((UChar32)0x410).add((UChar32)0x415);   // A couple of Cyrillic letters
131     index->addLabels(additions, status);
132     TEST_CHECK_STATUS;
133     lc = index->getBucketCount(status);
134     TEST_CHECK_STATUS;
135     assertEquals("underflow, A-Z, inflow, 2 Cyrillic, overflow",
136                  31, index->getBucketCount(status));
137     // std::cout << lc << std::endl;
138     index.adoptInstead(nullptr);
139 
140 
141     // addLabels(Locale)
142 
143     status = U_ZERO_ERROR;
144     index.adoptInstead(new AlphabeticIndex(Locale::getEnglish(), status));
145     TEST_CHECK_STATUS;
146     AlphabeticIndex &aip = index->addLabels(Locale::getJapanese(), status);
147     TEST_ASSERT(&aip == index.getAlias());
148     TEST_CHECK_STATUS;
149     lc = index->getBucketCount(status);
150     TEST_CHECK_STATUS;
151     TEST_ASSERT(35 < lc);  // Japanese should add a bunch.  Don't rely on the exact value.
152     index.adoptInstead(nullptr);
153 
154     // GetCollator(),  Get under/in/over flow labels
155 
156     status = U_ZERO_ERROR;
157     index.adoptInstead(new AlphabeticIndex(Locale::getGerman(), status));
158     TEST_CHECK_STATUS;
159     LocalPointer<Collator> germanCol(Collator::createInstance(Locale::getGerman(), status));
160     TEST_CHECK_STATUS;
161     const RuleBasedCollator &indexCol = index->getCollator();
162     TEST_ASSERT(*germanCol == indexCol);
163 
164     UnicodeString ELLIPSIS;  ELLIPSIS.append((UChar32)0x2026);
165     UnicodeString s = index->getUnderflowLabel();
166     TEST_ASSERT(ELLIPSIS == s);
167     s = index->getOverflowLabel();
168     TEST_ASSERT(ELLIPSIS == s);
169     s = index->getInflowLabel();
170     TEST_ASSERT(ELLIPSIS == s);
171     index->setOverflowLabel(UNICODE_STRING_SIMPLE("O"), status);
172     index->setUnderflowLabel(UNICODE_STRING_SIMPLE("U"), status).setInflowLabel(UNICODE_STRING_SIMPLE("I"), status);
173     s = index->getUnderflowLabel();
174     TEST_ASSERT(UNICODE_STRING_SIMPLE("U") == s);
175     s = index->getOverflowLabel();
176     TEST_ASSERT(UNICODE_STRING_SIMPLE("O") == s);
177     s = index->getInflowLabel();
178     TEST_ASSERT(UNICODE_STRING_SIMPLE("I") == s);
179 
180 
181 
182 
183     index.adoptInstead(nullptr);
184 
185 
186 
187     const UnicodeString adam = UNICODE_STRING_SIMPLE("Adam");
188     const UnicodeString baker = UNICODE_STRING_SIMPLE("Baker");
189     const UnicodeString charlie = UNICODE_STRING_SIMPLE("Charlie");
190     const UnicodeString chad = UNICODE_STRING_SIMPLE("Chad");
191     const UnicodeString zed  = UNICODE_STRING_SIMPLE("Zed");
192     const UnicodeString Cyrillic = UNICODE_STRING_SIMPLE("\\u0410\\u0443\\u0435").unescape();
193 
194     // addRecord(), verify that it comes back out.
195     //
196     status = U_ZERO_ERROR;
197     index.adoptInstead(new AlphabeticIndex(Locale::getEnglish(), status));
198     TEST_CHECK_STATUS;
199     index->addRecord(UnicodeString("Adam"), this, status);
200     UBool   b;
201     TEST_CHECK_STATUS;
202     index->resetBucketIterator(status);
203     TEST_CHECK_STATUS;
204     index->nextBucket(status);  // Move to underflow label
205     index->nextBucket(status);  // Move to "A"
206     TEST_CHECK_STATUS;
207     const UnicodeString &label2 = index->getBucketLabel();
208     UnicodeString A_STR = UNICODE_STRING_SIMPLE("A");
209     TEST_ASSERT(A_STR == label2);
210 
211     b = index->nextRecord(status);
212     TEST_CHECK_STATUS;
213     TEST_ASSERT(b);
214     const UnicodeString &itemName = index->getRecordName();
215     TEST_ASSERT(adam == itemName);
216 
217     const void *itemContext = index->getRecordData();
218     TEST_ASSERT(itemContext == this);
219 
220     index.adoptInstead(nullptr);
221 
222     // clearRecords, addRecord(), Iteration
223 
224     status = U_ZERO_ERROR;
225     index.adoptInstead(new AlphabeticIndex(Locale::getEnglish(), status));
226     TEST_CHECK_STATUS;
227     while (index->nextBucket(status)) {
228         TEST_CHECK_STATUS;
229         while (index->nextRecord(status)) {
230             TEST_CHECK_STATUS;
231             TEST_ASSERT(FALSE);   // No items have been added.
232         }
233         TEST_CHECK_STATUS;
234     }
235 
236     index->addRecord(adam, NULL, status);
237     index->addRecord(baker, NULL, status);
238     index->addRecord(charlie, NULL, status);
239     index->addRecord(chad, NULL, status);
240     TEST_CHECK_STATUS;
241     int itemCount = 0;
242     index->resetBucketIterator(status);
243     while (index->nextBucket(status)) {
244         TEST_CHECK_STATUS;
245         while (index->nextRecord(status)) {
246             TEST_CHECK_STATUS;
247             ++itemCount;
248         }
249     }
250     TEST_CHECK_STATUS;
251     TEST_ASSERT(itemCount == 4);
252 
253     TEST_ASSERT(index->nextBucket(status) == FALSE);
254     index->resetBucketIterator(status);
255     TEST_CHECK_STATUS;
256     TEST_ASSERT(index->nextBucket(status) == TRUE);
257 
258     index->clearRecords(status);
259     TEST_CHECK_STATUS;
260     index->resetBucketIterator(status);
261     while (index->nextBucket(status)) {
262         TEST_CHECK_STATUS;
263         while (index->nextRecord(status)) {
264             TEST_ASSERT(FALSE);   // No items have been added.
265         }
266     }
267     TEST_CHECK_STATUS;
268     index.adoptInstead(nullptr);
269 
270     // getBucketLabel(), getBucketType()
271 
272     status = U_ZERO_ERROR;
273     index.adoptInstead(new AlphabeticIndex(Locale::getEnglish(), status));
274     TEST_CHECK_STATUS;
275     index->setUnderflowLabel(adam, status).setOverflowLabel(charlie, status);
276     TEST_CHECK_STATUS;
277     for (i=0; index->nextBucket(status); i++) {
278         TEST_CHECK_STATUS;
279         UnicodeString label = index->getBucketLabel();
280         UAlphabeticIndexLabelType type = index->getBucketLabelType();
281         if (i == 0) {
282             TEST_ASSERT(type == U_ALPHAINDEX_UNDERFLOW);
283             TEST_ASSERT(label == adam);
284         } else if (i <= 26) {
285             // Labels A - Z for English locale
286             TEST_ASSERT(type == U_ALPHAINDEX_NORMAL);
287             UnicodeString expectedLabel((UChar)(0x40 + i));
288             TEST_ASSERT(expectedLabel == label);
289         } else if (i == 27) {
290             TEST_ASSERT(type == U_ALPHAINDEX_OVERFLOW);
291             TEST_ASSERT(label == charlie);
292         } else {
293             TEST_ASSERT(FALSE);
294         }
295     }
296     TEST_ASSERT(i==28);
297     index.adoptInstead(nullptr);
298 
299     // getBucketIndex()
300 
301     status = U_ZERO_ERROR;
302     index.adoptInstead(new AlphabeticIndex(Locale::getEnglish(), status));
303     TEST_CHECK_STATUS;
304     int32_t n = index->getBucketIndex(adam, status);
305     TEST_CHECK_STATUS;
306     TEST_ASSERT(n == 1);    /*  Label #0 is underflow, 1 is A, etc. */
307     n = index->getBucketIndex(baker, status);
308     TEST_ASSERT(n == 2);
309     n = index->getBucketIndex(Cyrillic, status);
310     TEST_ASSERT(n == 27);   // Overflow label
311     n = index->getBucketIndex(zed, status);
312     TEST_ASSERT(n == 26);
313 
314     for (i=0; index->nextBucket(status); i++) {
315         n = index->getBucketIndex();
316         TEST_ASSERT(n == i);
317         UnicodeString label = index->getBucketLabel();
318         TEST_ASSERT(n == i);
319     }
320     TEST_ASSERT(i == 28);
321 
322     index.adoptInstead(nullptr);
323     index.adoptInstead(new AlphabeticIndex(Locale::createFromName("ru"), status));
324     TEST_CHECK_STATUS;
325     assertEquals("Russian index.getBucketCount()", 32, index->getBucketCount(status));
326     // Latin-script names should go into the underflow label (0)
327     // if the Russian collation does not use script reordering,
328     // but into the overflow label (getBucketCount()-1)
329     // if Russian sorts Cyrillic first.
330     int32_t reorderCodes[20];
331     int32_t expectedLatinIndex = 0;
332     if (index->getCollator().getReorderCodes(reorderCodes, UPRV_LENGTHOF(reorderCodes), status) > 0) {
333         expectedLatinIndex = index->getBucketCount(status) - 1;
334     }
335     n = index->getBucketIndex(adam, status);
336     TEST_CHECK_STATUS;
337     assertEquals("Russian index.getBucketIndex(adam)", expectedLatinIndex, n);
338     n = index->getBucketIndex(baker, status);
339     assertEquals("Russian index.getBucketIndex(baker)", expectedLatinIndex, n);
340     n = index->getBucketIndex(Cyrillic, status);
341     assertEquals("Russian index.getBucketIndex(Cyrillic)", 1, n);
342     n = index->getBucketIndex(zed, status);
343     assertEquals("Russian index.getBucketIndex(zed)", expectedLatinIndex, n);
344 
345 }
346 
347 
348 static const char * KEY_LOCALES[] = {
349             "en", "es", "de", "fr", "ja", "it", "tr", "pt", "zh", "nl",
350             "pl", "ar", "ru", "zh_Hant", "ko", "th", "sv", "fi", "da",
351             "he", "nb", "el", "hr", "bg", "sk", "lt", "vi", "lv", "sr",
352             "pt_PT", "ro", "hu", "cs", "id", "sl", "fil", "fa", "uk",
353             "ca", "hi", "et", "eu", "is", "sw", "ms", "bn", "am", "ta",
354             "te", "mr", "ur", "ml", "kn", "gu", "or", ""};
355 
356 
ManyLocalesTest()357 void AlphabeticIndexTest::ManyLocalesTest() {
358     UErrorCode status = U_ZERO_ERROR;
359     int32_t  lc = 0;
360 
361     for (int i=0; ; ++i) {
362         status = U_ZERO_ERROR;
363         const char *localeName = KEY_LOCALES[i];
364         if (localeName[0] == 0) {
365             break;
366         }
367         // std::cout <<  localeName << "  ";
368         Locale loc = Locale::createFromName(localeName);
369         AlphabeticIndex index(loc, status);
370         TEST_CHECK_STATUS;
371         lc = index.getBucketCount(status);
372         TEST_CHECK_STATUS;
373         // std::cout << "getBucketCount() == " << lc << std::endl;
374 
375         LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status));
376         TEST_CHECK_STATUS;
377         TEST_ASSERT(lc == immIndex->getBucketCount());
378 
379         assertEquals("initial bucket index", -1, index.getBucketIndex());
380         int32_t bucketIndex = 0;
381         while (index.nextBucket(status)) {
382             TEST_CHECK_STATUS;
383             assertEquals("bucket index", bucketIndex, index.getBucketIndex());
384             const UnicodeString &label = index.getBucketLabel();
385             TEST_ASSERT(label.length()>0);
386             // std::string ss;
387             // std::cout << ":" << label.toUTF8String(ss);
388             const AlphabeticIndex::Bucket *bucket = immIndex->getBucket(bucketIndex);
389             TEST_ASSERT(bucket != NULL);
390             assertEquals("bucket label vs. immutable: locale=" + UnicodeString(localeName) +
391                          " index=" + bucketIndex,
392                          label, bucket->getLabel());
393             TEST_ASSERT(&label != &bucket->getLabel());  // not the same pointers
394             UAlphabeticIndexLabelType labelType = index.getBucketLabelType();
395             TEST_ASSERT(labelType == bucket->getLabelType());
396             ++bucketIndex;
397         }
398         // std::cout << ":" << std::endl;
399 
400         TEST_ASSERT(immIndex->getBucketCount() == bucketIndex);
401         TEST_ASSERT(immIndex->getBucket(-1) == NULL);
402         TEST_ASSERT(immIndex->getBucket(bucketIndex) == NULL);
403     }
404 }
405 
406 
407 // Test data for Pinyin based indexes.
408 //  The Chinese characters should be distributed under latin labels in
409 //  an index.
410 
411 static const char *pinyinTestData[] = {
412         "\\u0101", "\\u5416", "\\u58ba", //
413         "b", "\\u516b", "\\u62d4", "\\u8500", //
414         "c", "\\u5693", "\\u7938", "\\u9e7e", //
415         "d", "\\u5491", "\\u8fcf", "\\u964a", //
416         "\\u0113","\\u59b8", "\\u92e8", "\\u834b", //
417         "f", "\\u53d1", "\\u9197", "\\u99a5", //
418         "g", "\\u7324", "\\u91d3", "\\u8142", //
419         "h", "\\u598e", "\\u927f", "\\u593b", //
420         "j", "\\u4e0c", "\\u6785", "\\u9d58", //
421         "k", "\\u5494", "\\u958b", "\\u7a52", //
422         "l", "\\u5783", "\\u62c9", "\\u9ba5", //
423         "m", "\\u5638", "\\u9ebb", "\\u65c0", //
424         "n", "\\u62ff", "\\u80ad", "\\u685b", //
425         "\\u014D", "\\u5662", "\\u6bee", "\\u8bb4", //
426         "p", "\\u5991", "\\u8019", "\\u8c31", //
427         "q", "\\u4e03", "\\u6053", "\\u7f56", //
428         "r", "\\u5465", "\\u72aa", "\\u6e03", //
429         "s", "\\u4ee8", "\\u9491", "\\u93c1", //
430         "t", "\\u4ed6", "\\u9248", "\\u67dd", //
431         "w", "\\u5c72", "\\u5558", "\\u5a7a", //
432         "x", "\\u5915", "\\u5438", "\\u6bbe", //
433         "y", "\\u4e2b", "\\u82bd", "\\u8574", //
434         "z", "\\u5e00", "\\u707d", "\\u5c0a",
435         NULL
436     };
437 
HackPinyinTest()438 void AlphabeticIndexTest::HackPinyinTest() {
439     UErrorCode status = U_ZERO_ERROR;
440     AlphabeticIndex aindex(Locale::createFromName("zh"), status);
441     TEST_CHECK_STATUS;
442 
443     UnicodeString names[UPRV_LENGTHOF(pinyinTestData)];
444     int32_t  nameCount;
445     for (nameCount=0; pinyinTestData[nameCount] != NULL; nameCount++) {
446         names[nameCount] = UnicodeString(pinyinTestData[nameCount], -1, UnicodeString::kInvariant).unescape();
447         aindex.addRecord(names[nameCount], &names[nameCount], status);
448         TEST_CHECK_STATUS;
449         if (U_FAILURE(status)) {
450             return;
451         }
452     }
453     TEST_ASSERT(nameCount == aindex.getRecordCount(status));
454 
455     // Weak checking:  make sure that none of the Chinese names landed in the overflow bucket
456     //   of the index, and that the names are distributed among several buckets.
457     //   (Exact expected data would be subject to change with evolution of the collation rules.)
458 
459     int32_t bucketCount = 0;
460     int32_t filledBucketCount = 0;
461     while (aindex.nextBucket(status)) {
462         bucketCount++;
463         UnicodeString label = aindex.getBucketLabel();
464         // std::string s;
465         // std::cout << label.toUTF8String(s) << ":  ";
466 
467         UBool  bucketHasContents = FALSE;
468         while (aindex.nextRecord(status)) {
469             bucketHasContents = TRUE;
470             UnicodeString name = aindex.getRecordName();
471             if (aindex.getBucketLabelType() != U_ALPHAINDEX_NORMAL) {
472                 errln("File %s, Line %d, Name \"\\u%x\" is in an under or overflow bucket.",
473                     __FILE__, __LINE__, name.char32At(0));
474             }
475             // s.clear();
476             // std::cout << aindex.getRecordName().toUTF8String(s) << " ";
477         }
478         if (bucketHasContents) {
479             filledBucketCount++;
480         }
481         // std::cout << std::endl;
482     }
483     TEST_ASSERT(bucketCount > 25);
484     TEST_ASSERT(filledBucketCount > 15);
485 }
486 
487 
TestBug9009()488 void AlphabeticIndexTest::TestBug9009() {
489     UErrorCode status = U_ZERO_ERROR;
490     Locale loc("root");
491     AlphabeticIndex aindex(loc, status);
492     TEST_CHECK_STATUS;
493     aindex.nextBucket(status);  // Crash here before bug was fixed.
494     TEST_CHECK_STATUS;
495 }
496 
497 static const char *localeAndIndexCharactersLists[][2] = {
498     /* Arabic*/ {"ar", "\\u0627:\\u0628:\\u062A:\\u062B:\\u062C:\\u062D:\\u062E:\\u062F:\\u0630:\\u0631:\\u0632:\\u0633:\\u0634:\\u0635:\\u0636:\\u0637:\\u0638:\\u0639:\\u063A:\\u0641:\\u0642:\\u0643:\\u0644:\\u0645:\\u0646:\\u0647:\\u0648:\\u064A"},
499     /* Bulgarian*/  {"bg", "\\u0410:\\u0411:\\u0412:\\u0413:\\u0414:\\u0415:\\u0416:\\u0417:\\u0418:\\u0419:\\u041A:\\u041B:\\u041C:\\u041D:\\u041E:\\u041F:\\u0420:\\u0421:\\u0422:\\u0423:\\u0424:\\u0425:\\u0426:\\u0427:\\u0428:\\u0429:\\u042E:\\u042F"},
500     /* Catalan*/    {"ca", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"},
501     /* Czech*/  {"cs", "A:B:C:\\u010C:D:E:F:G:H:CH:I:J:K:L:M:N:O:P:Q:R:\\u0158:S:\\u0160:T:U:V:W:X:Y:Z:\\u017D"},
502     /* Danish*/ {"da", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z:\\u00C6:\\u00D8:\\u00C5"},
503     /* German*/ {"de", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"},
504     /* Greek*/  {"el", "\\u0391:\\u0392:\\u0393:\\u0394:\\u0395:\\u0396:\\u0397:\\u0398:\\u0399:\\u039A:\\u039B:\\u039C:\\u039D:\\u039E:\\u039F:\\u03A0:\\u03A1:\\u03A3:\\u03A4:\\u03A5:\\u03A6:\\u03A7:\\u03A8:\\u03A9"},
505     /* English*/    {"en", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"},
506     /* Spanish*/    {"es", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:\\u00D1:O:P:Q:R:S:T:U:V:W:X:Y:Z"},
507     /* Estonian*/   {"et", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:\\u0160:Z:\\u017D:T:U:V:W:\\u00D5:\\u00C4:\\u00D6:\\u00DC:X:Y"},
508     /* Basque*/ {"eu", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"},
509     /* Finnish*/    {"fi", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z:\\u00C5:\\u00C4:\\u00D6"},
510     /* Filipino*/   {"fil", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:\\u00D1:Ng:O:P:Q:R:S:T:U:V:W:X:Y:Z"},
511     /* French*/ {"fr", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"},
512     /* Hebrew*/ {"he", "\\u05D0:\\u05D1:\\u05D2:\\u05D3:\\u05D4:\\u05D5:\\u05D6:\\u05D7:\\u05D8:\\u05D9:\\u05DB:\\u05DC:\\u05DE:\\u05E0:\\u05E1:\\u05E2:\\u05E4:\\u05E6:\\u05E7:\\u05E8:\\u05E9:\\u05EA"},
513     /* Icelandic*/  {"is", "A:\\u00C1:B:C:D:\\u00D0:E:\\u00C9:F:G:H:I:\\u00CD:J:K:L:M:N:O:\\u00D3:P:Q:R:S:T:U:\\u00DA:V:W:X:Y:\\u00DD:Z:\\u00DE:\\u00C6:\\u00D6"},
514     /* Italian*/    {"it", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"},
515     /* Japanese*/   {"ja", "\\u3042:\\u304B:\\u3055:\\u305F:\\u306A:\\u306F:\\u307E:\\u3084:\\u3089:\\u308F"},
516     /* Korean*/ {"ko", "\\u3131:\\u3134:\\u3137:\\u3139:\\u3141:\\u3142:\\u3145:\\u3147:\\u3148:\\u314A:\\u314B:\\u314C:\\u314D:\\u314E"},
517     /* Lithuanian*/ {"lt", "A:B:C:\\u010C:D:E:F:G:H:I:J:K:L:M:N:O:P:R:S:\\u0160:T:U:V:Z:\\u017D"},
518     /* Latvian*/    {"lv", "A:B:C:\\u010C:D:E:F:G:\\u0122:H:I:J:K:\\u0136:L:\\u013B:M:N:\\u0145:O:P:Q:R:S:\\u0160:T:U:V:W:X:Z:\\u017D"},
519     /* Norwegian Bokm\\u00E5l*/  {"nb", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z:\\u00C6:\\u00D8:\\u00C5"},
520     /* Dutch*/  {"nl", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"},
521     /* Polish*/ {"pl", "A:B:C:\\u0106:D:E:F:G:H:I:J:K:L:\\u0141:M:N:O:\\u00D3:P:Q:R:S:\\u015A:T:U:V:W:X:Y:Z:\\u0179:\\u017B"},
522     /* Portuguese*/ {"pt", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"},
523     /* Romanian*/   {"ro", "A:\\u0102:\\u00C2:B:C:D:E:F:G:H:I:\\u00CE:J:K:L:M:N:O:P:Q:R:S:\\u0218:T:\\u021A:U:V:W:X:Y:Z"},
524     /* Russian*/    {"ru", "\\u0410:\\u0411:\\u0412:\\u0413:\\u0414:\\u0415:\\u0416:\\u0417:\\u0418:\\u0419:\\u041A:\\u041B:\\u041C:\\u041D:\\u041E:\\u041F:\\u0420:\\u0421:\\u0422:\\u0423:\\u0424:\\u0425:\\u0426:\\u0427:\\u0428:\\u0429:\\u042B:\\u042D:\\u042E:\\u042F"},
525     /* Slovak*/ {"sk", "A:\\u00C4:B:C:\\u010C:D:E:F:G:H:CH:I:J:K:L:M:N:O:\\u00D4:P:Q:R:S:\\u0160:T:U:V:W:X:Y:Z:\\u017D"},
526     /* Slovenian*/  {"sl", "A:B:C:\\u010C:\\u0106:D:\\u0110:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:\\u0160:T:U:V:W:X:Y:Z:\\u017D"},
527     /* Serbian*/    {"sr", "\\u0410:\\u0411:\\u0412:\\u0413:\\u0414:\\u0402:\\u0415:\\u0416:\\u0417:\\u0418:\\u0408:\\u041A:\\u041B:\\u0409:\\u041C:\\u041D:\\u040A:\\u041E:\\u041F:\\u0420:\\u0421:\\u0422:\\u040B:\\u0423:\\u0424:\\u0425:\\u0426:\\u0427:\\u040F:\\u0428"},
528     /* Swedish*/    {"sv", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z:\\u00C5:\\u00C4:\\u00D6"},
529     /* Turkish*/    {"tr", "A:B:C:\\u00C7:D:E:F:G:H:I:\\u0130:J:K:L:M:N:O:\\u00D6:P:Q:R:S:\\u015E:T:U:\\u00DC:V:W:X:Y:Z"},
530     /* Ukrainian*/  {"uk", "\\u0410:\\u0411:\\u0412:\\u0413:\\u0490:\\u0414:\\u0415:\\u0404:\\u0416:\\u0417:\\u0418:\\u0406:\\u0407:\\u0419:\\u041A:\\u041B:\\u041C:\\u041D:\\u041E:\\u041F:\\u0420:\\u0421:\\u0422:\\u0423:\\u0424:\\u0425:\\u0426:\\u0427:\\u0428:\\u0429:\\u042E:\\u042F"},
531     /* Vietnamese*/ {"vi", "A:\\u0102:\\u00C2:B:C:D:\\u0110:E:\\u00CA:F:G:H:I:J:K:L:M:N:O:\\u00D4:\\u01A0:P:Q:R:S:T:U:\\u01AF:V:W:X:Y:Z"},
532     /* Chinese*/    {"zh", "A:B:C:D:E:F:G:H:I:J:K:L:M:N:O:P:Q:R:S:T:U:V:W:X:Y:Z"},
533     /* Chinese (Traditional Han)*/  {"zh_Hant", "1\\u5283:2\\u5283:3\\u5283:4\\u5283:5\\u5283:6\\u5283:7\\u5283:8\\u5283:9\\u5283:10\\u5283:11\\u5283:12\\u5283:13\\u5283:14\\u5283:15\\u5283:16\\u5283:17\\u5283:18\\u5283:19\\u5283:20\\u5283:21\\u5283:22\\u5283:23\\u5283:24\\u5283:25\\u5283:26\\u5283:27\\u5283:28\\u5283:29\\u5283:30\\u5283:31\\u5283:32\\u5283:33\\u5283:35\\u5283:36\\u5283:39\\u5283:48\\u5283"},
534 };
535 
TestIndexCharactersList()536 void AlphabeticIndexTest::TestIndexCharactersList() {
537     UErrorCode status = U_ZERO_ERROR;
538     for (int32_t i = 0; i < UPRV_LENGTHOF(localeAndIndexCharactersLists); ++i) {
539         const char *(&localeAndIndexCharacters)[2] = localeAndIndexCharactersLists[i];
540         const char *locale = localeAndIndexCharacters[0];
541         UnicodeString expectedIndexCharacters
542             = (UnicodeString("\\u2026:") + localeAndIndexCharacters[1] + ":\\u2026").unescape();
543         AlphabeticIndex index(locale, status);
544         TEST_CHECK_STATUS;
545         LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status));
546         TEST_CHECK_STATUS;
547 
548         // Join the elements of the list to a string with delimiter ":"
549         UnicodeString actualIndexCharacters;
550         assertEquals(locale,
551                      expectedIndexCharacters,
552                      joinLabelsAndAppend(*immIndex, actualIndexCharacters));
553         logln(locale + UnicodeString(": ") + actualIndexCharacters);
554     }
555 }
556 
TestHaniFirst()557 void AlphabeticIndexTest::TestHaniFirst() {
558     UErrorCode status = U_ZERO_ERROR;
559     LocalPointer<RuleBasedCollator> coll(
560         static_cast<RuleBasedCollator *>(Collator::createInstance(Locale::getRoot(), status)));
561 
562     if (U_FAILURE(status)) {
563         dataerrln("Failed Collator::createInstance call - %s", u_errorName(status));
564         return;
565     }
566     int32_t reorderCodes[] = { USCRIPT_HAN };
567     coll->setReorderCodes(reorderCodes, UPRV_LENGTHOF(reorderCodes), status);
568     TEST_CHECK_STATUS;
569     AlphabeticIndex index(coll.orphan(), status);
570     TEST_CHECK_STATUS;
571     assertEquals("getBucketCount()", 1, index.getBucketCount(status));   // ... (underflow only)
572     index.addLabels(Locale::getEnglish(), status);
573     assertEquals("getBucketCount()", 28, index.getBucketCount(status));  // ... A-Z ...
574     int32_t bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x897f), status);
575     assertEquals("getBucketIndex(U+897F)", 0, bucketIndex);  // underflow bucket
576     bucketIndex = index.getBucketIndex("i", status);
577     assertEquals("getBucketIndex(i)", 9, bucketIndex);
578     bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x03B1), status);
579     assertEquals("getBucketIndex(Greek alpha)", 27, bucketIndex);
580     // U+50005 is an unassigned code point which sorts at the end, independent of the Hani group.
581     bucketIndex = index.getBucketIndex(UnicodeString((UChar32)0x50005), status);
582     assertEquals("getBucketIndex(U+50005)", 27, bucketIndex);
583     bucketIndex = index.getBucketIndex(UnicodeString((UChar)0xFFFF), status);
584     assertEquals("getBucketIndex(U+FFFF)", 27, bucketIndex);
585 }
586 
TestPinyinFirst()587 void AlphabeticIndexTest::TestPinyinFirst() {
588     UErrorCode status = U_ZERO_ERROR;
589     LocalPointer<RuleBasedCollator> coll(
590         static_cast<RuleBasedCollator *>(Collator::createInstance(Locale::getChinese(), status)));
591     if (U_FAILURE(status)) {
592         dataerrln("Failed Collator::createInstance call - %s", u_errorName(status));
593         return;
594     }
595     int32_t reorderCodes[] = { USCRIPT_HAN };
596     coll->setReorderCodes(reorderCodes, UPRV_LENGTHOF(reorderCodes), status);
597     TEST_CHECK_STATUS;
598     AlphabeticIndex index(coll.orphan(), status);
599     TEST_CHECK_STATUS;
600     assertEquals("getBucketCount()", 28, index.getBucketCount(status));   // ... A-Z ...
601     index.addLabels(Locale::getChinese(), status);
602     assertEquals("getBucketCount()", 28, index.getBucketCount(status));  // ... A-Z ...
603     int32_t bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x897f), status);
604     assertEquals("getBucketIndex(U+897F)", (int32_t)((UChar)0x0058/*X*/ - (UChar)0x0041/*A*/ + 1), bucketIndex);
605     bucketIndex = index.getBucketIndex("i", status);
606     assertEquals("getBucketIndex(i)", 9, bucketIndex);
607     bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x03B1), status);
608     assertEquals("getBucketIndex(Greek alpha)", (int32_t)27, bucketIndex);
609     // U+50005 is an unassigned code point which sorts at the end, independent of the Hani group.
610     bucketIndex = index.getBucketIndex(UnicodeString((UChar32)0x50005), status);
611     assertEquals("getBucketIndex(U+50005)", 27, bucketIndex);
612     bucketIndex = index.getBucketIndex(UnicodeString((UChar)0xFFFF), status);
613     assertEquals("getBucketIndex(U+FFFF)", 27, bucketIndex);
614 }
615 
TestSchSt()616 void AlphabeticIndexTest::TestSchSt() {
617     UErrorCode status = U_ZERO_ERROR;
618     AlphabeticIndex index(Locale::getGerman(), status);
619     index.addLabels(UnicodeSet("[\\u00C6{Sch*}{St*}]", status), status);
620     TEST_CHECK_STATUS;
621     // ... A AE-ligature B-R S Sch St T-Z ...
622     LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status));
623     TEST_CHECK_STATUS;
624     assertEquals("getBucketCount()", 31, index.getBucketCount(status));
625     assertEquals("immutable getBucketCount()", 31, immIndex->getBucketCount());
626     static const struct TestCase {
627         const char *name;
628         int32_t bucketIndex;
629         const char *bucketLabel;
630     } testCases[] = {
631         // name, bucket index, bucket label
632         { "Adelbert", 1, "A" },
633         { "Afrika", 1, "A" },
634         { "\\u00C6sculap", 2, "\\u00C6" },
635         { "Aesthet", 2, "\\u00C6" },
636         { "Berlin", 3, "B" },
637         { "Rilke", 19, "R" },
638         { "Sacher", 20, "S" },
639         { "Seiler", 20, "S" },
640         { "Sultan", 20, "S" },
641         { "Schiller", 21, "Sch" },
642         { "Steiff", 22, "St" },
643         { "Thomas", 23, "T" }
644     };
645     for (int32_t i = 0; i < UPRV_LENGTHOF(testCases); ++i) {
646         const TestCase &testCase = testCases[i];
647         UnicodeString name = UnicodeString(testCase.name).unescape();
648         UnicodeString label = UnicodeString(testCase.bucketLabel).unescape();
649         char msg[100];
650         sprintf(msg, "getBucketIndex(%s)", testCase.name);
651         assertEquals(msg, testCase.bucketIndex, index.getBucketIndex(name, status));
652         sprintf(msg, "immutable getBucketIndex(%s)", testCase.name);
653         assertEquals(msg, testCase.bucketIndex, immIndex->getBucketIndex(name, status));
654         sprintf(msg, "immutable bucket label (%s)", testCase.name);
655         assertEquals(msg, label, immIndex->getBucket(testCase.bucketIndex)->getLabel());
656     }
657 }
658 
TestNoLabels()659 void AlphabeticIndexTest::TestNoLabels() {
660     UErrorCode status = U_ZERO_ERROR;
661     LocalPointer<RuleBasedCollator> coll(
662         static_cast<RuleBasedCollator *>(Collator::createInstance(Locale::getRoot(), status)));
663     TEST_CHECK_STATUS;
664     AlphabeticIndex index(coll.orphan(), status);
665     TEST_CHECK_STATUS;
666     index.addRecord(UnicodeString((UChar)0x897f), NULL, status);
667     index.addRecord("i", NULL, status);
668     index.addRecord(UnicodeString((UChar)0x03B1), NULL, status);
669     assertEquals("getBucketCount()", 1, index.getBucketCount(status));  // ...
670     TEST_ASSERT(index.nextBucket(status));
671     assertEquals("underflow label type", (int32_t)U_ALPHAINDEX_UNDERFLOW, index.getBucketLabelType());
672     assertEquals("all records in the underflow bucket", (int32_t)3, index.getBucketRecordCount());
673 }
674 
TestChineseZhuyin()675 void AlphabeticIndexTest::TestChineseZhuyin() {
676     UErrorCode status = U_ZERO_ERROR;
677     char loc[100];
678     uloc_forLanguageTag("zh-u-co-zhuyin", loc, UPRV_LENGTHOF(loc), NULL, &status);
679     AlphabeticIndex index(loc, status);
680     LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status));
681     TEST_CHECK_STATUS;
682     assertEquals("getBucketCount()", 38, immIndex->getBucketCount());
683     assertEquals("label 1", UnicodeString((UChar)0x3105), immIndex->getBucket(1)->getLabel());
684     assertEquals("label 2", UnicodeString((UChar)0x3106), immIndex->getBucket(2)->getLabel());
685     assertEquals("label 3", UnicodeString((UChar)0x3107), immIndex->getBucket(3)->getLabel());
686     assertEquals("label 4", UnicodeString((UChar)0x3108), immIndex->getBucket(4)->getLabel());
687     assertEquals("label 5", UnicodeString((UChar)0x3109), immIndex->getBucket(5)->getLabel());
688 }
689 
TestJapaneseKanji()690 void AlphabeticIndexTest::TestJapaneseKanji() {
691     UErrorCode status = U_ZERO_ERROR;
692     AlphabeticIndex index(Locale::getJapanese(), status);
693     LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status));
694     TEST_CHECK_STATUS;
695     // There are no index characters for Kanji in the Japanese standard collator.
696     // They should all go into the overflow bucket.
697     static const UChar32 kanji[] = { 0x4E9C, 0x95C7, 0x4E00, 0x58F1 };
698     int32_t overflowIndex = immIndex->getBucketCount() - 1;
699     for(int32_t i = 0; i < UPRV_LENGTHOF(kanji); ++i) {
700         char msg[40];
701         sprintf(msg, "kanji[%d]=U+%04lX in overflow bucket", (int)i, (long)kanji[i]);
702         assertEquals(msg, overflowIndex, immIndex->getBucketIndex(UnicodeString(kanji[i]), status));
703         TEST_CHECK_STATUS;
704     }
705 }
706 
TestChineseUnihan()707 void AlphabeticIndexTest::TestChineseUnihan() {
708     UErrorCode status = U_ZERO_ERROR;
709     AlphabeticIndex index("zh-u-co-unihan", status);
710     if(U_FAILURE(status)) {
711         dataerrln("unable create an AlphabeticIndex for Chinese/unihan: %s", u_errorName(status));
712         return;
713     }
714     index.setMaxLabelCount(500, status);  // ICU 54 default is 99.
715     LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status));
716     TEST_CHECK_STATUS;
717     int32_t bucketCount = immIndex->getBucketCount();
718     if(bucketCount < 216) {
719         // There should be at least an underflow and overflow label,
720         // and one for each of 214 radicals,
721         // and maybe additional labels for simplified radicals.
722         dataerrln("too few buckets/labels for Chinese/unihan: %d (is zh/unihan data available?)",
723                   bucketCount);
724         return;
725     } else {
726         logln("Chinese/unihan has %d buckets/labels", bucketCount);
727     }
728     // bucketIndex = radical number, adjusted for simplified radicals in lower buckets.
729     int32_t bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x4e5d), status);
730     assertEquals("getBucketIndex(U+4E5D)", 5, bucketIndex);
731     // radical 100, and there is a 90' since Unicode 8
732     bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x7527), status);
733     assertEquals("getBucketIndex(U+7527)", 101, bucketIndex);
734 }
735 
testHasBuckets()736 void AlphabeticIndexTest::testHasBuckets() {
737     checkHasBuckets(Locale("am"), USCRIPT_ETHIOPIC);
738     checkHasBuckets(Locale("haw"), USCRIPT_LATIN);
739     checkHasBuckets(Locale("hy"), USCRIPT_ARMENIAN);
740     checkHasBuckets(Locale("vai"), USCRIPT_VAI);
741 }
742 
checkHasBuckets(const Locale & locale,UScriptCode script)743 void AlphabeticIndexTest::checkHasBuckets(const Locale &locale, UScriptCode script) {
744     IcuTestErrorCode errorCode(*this, "checkHasBuckets");
745     AlphabeticIndex aindex(locale, errorCode);
746     LocalPointer<AlphabeticIndex::ImmutableIndex> index(aindex.buildImmutableIndex(errorCode), errorCode);
747     if (U_FAILURE(errorCode)) {
748       dataerrln("%s %d  Error in index creation",  __FILE__, __LINE__);
749       return;
750     }
751     UnicodeString loc = locale.getName();
752     assertTrue(loc + u" at least 3 buckets", index->getBucketCount() >= 3);
753     const AlphabeticIndex::Bucket *bucket = index->getBucket(1);
754     assertEquals(loc + u" real bucket", U_ALPHAINDEX_NORMAL, bucket->getLabelType());
755     assertEquals(loc + u" expected script", script,
756             uscript_getScript(bucket->getLabel().char32At(0), errorCode));
757 }
758 
759 #endif
760