void AlphabeticIndexTest::TestChineseUnihan() {
UErrorCode status = U_ZERO_ERROR;
AlphabeticIndex index("zh-u-co-unihan", status);
if(U_FAILURE(status)) {
dataerrln("unable create an AlphabeticIndex for Chinese/unihan: %s", u_errorName(status));
return;
}
index.setMaxLabelCount(500, status); // ICU 54 default is 99.
LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status));
TEST_CHECK_STATUS;
int32_t bucketCount = immIndex->getBucketCount();
if(bucketCount < 216) {
// There should be at least an underflow and overflow label,
// and one for each of 214 radicals,
// and maybe additional labels for simplified radicals.
dataerrln("too few buckets/labels for Chinese/unihan: %d (is zh/unihan data available?)",
bucketCount);
return;
} else {
logln("Chinese/unihan has %d buckets/labels", bucketCount);
}
// bucketIndex = radical number, adjusted for simplified radicals in lower buckets.
int32_t bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x4e5d), status);
assertEquals("getBucketIndex(U+4E5D)", 5, bucketIndex);
// radical 100, and there is a 90' since Unicode 8
bucketIndex = index.getBucketIndex(UnicodeString((UChar)0x7527), status);
assertEquals("getBucketIndex(U+7527)", 101, bucketIndex);
}
void AlphabeticIndexTest::TestChineseZhuyin() {
UErrorCode status = U_ZERO_ERROR;
char loc[100];
uloc_forLanguageTag("zh-u-co-zhuyin", loc, LENGTHOF(loc), NULL, &status);
AlphabeticIndex index(loc, status);
LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status));
TEST_CHECK_STATUS;
assertEquals("getBucketCount()", 38, immIndex->getBucketCount());
assertEquals("label 1", UnicodeString((UChar)0x3105), immIndex->getBucket(1)->getLabel());
assertEquals("label 2", UnicodeString((UChar)0x3106), immIndex->getBucket(2)->getLabel());
assertEquals("label 3", UnicodeString((UChar)0x3107), immIndex->getBucket(3)->getLabel());
assertEquals("label 4", UnicodeString((UChar)0x3108), immIndex->getBucket(4)->getLabel());
assertEquals("label 5", UnicodeString((UChar)0x3109), immIndex->getBucket(5)->getLabel());
}
void AlphabeticIndexTest::ManyLocalesTest() {
UErrorCode status = U_ZERO_ERROR;
int32_t lc = 0;
for (int i=0; ; ++i) {
status = U_ZERO_ERROR;
const char *localeName = KEY_LOCALES[i];
if (localeName[0] == 0) {
break;
}
// std::cout << localeName << " ";
Locale loc = Locale::createFromName(localeName);
AlphabeticIndex index(loc, status);
TEST_CHECK_STATUS;
lc = index.getBucketCount(status);
TEST_CHECK_STATUS;
// std::cout << "getBucketCount() == " << lc << std::endl;
LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status));
TEST_CHECK_STATUS;
TEST_ASSERT(lc == immIndex->getBucketCount());
assertEquals("initial bucket index", -1, index.getBucketIndex());
int32_t bucketIndex = 0;
while (index.nextBucket(status)) {
TEST_CHECK_STATUS;
assertEquals("bucket index", bucketIndex, index.getBucketIndex());
const UnicodeString &label = index.getBucketLabel();
TEST_ASSERT(label.length()>0);
// std::string ss;
// std::cout << ":" << label.toUTF8String(ss);
const AlphabeticIndex::Bucket *bucket = immIndex->getBucket(bucketIndex);
TEST_ASSERT(bucket != NULL);
assertEquals("bucket label vs. immutable: locale=" + UnicodeString(localeName) +
" index=" + bucketIndex,
label, bucket->getLabel());
TEST_ASSERT(&label != &bucket->getLabel()); // not the same pointers
UAlphabeticIndexLabelType labelType = index.getBucketLabelType();
TEST_ASSERT(labelType == bucket->getLabelType());
++bucketIndex;
}
// std::cout << ":" << std::endl;
TEST_ASSERT(immIndex->getBucketCount() == bucketIndex);
TEST_ASSERT(immIndex->getBucket(-1) == NULL);
TEST_ASSERT(immIndex->getBucket(bucketIndex) == NULL);
}
}
void AlphabeticIndexTest::TestJapaneseKanji() {
UErrorCode status = U_ZERO_ERROR;
AlphabeticIndex index(Locale::getJapanese(), status);
LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status));
TEST_CHECK_STATUS;
// There are no index characters for Kanji in the Japanese standard collator.
// They should all go into the overflow bucket.
static const UChar32 kanji[] = { 0x4E9C, 0x95C7, 0x4E00, 0x58F1 };
int32_t overflowIndex = immIndex->getBucketCount() - 1;
for(int32_t i = 0; i < LENGTHOF(kanji); ++i) {
char msg[40];
sprintf(msg, "kanji[%d]=U+%04lX in overflow bucket", (int)i, (long)kanji[i]);
assertEquals(msg, overflowIndex, immIndex->getBucketIndex(UnicodeString(kanji[i]), status));
TEST_CHECK_STATUS;
}
}
void AlphabeticIndexTest::TestSchSt() {
UErrorCode status = U_ZERO_ERROR;
AlphabeticIndex index(Locale::getGerman(), status);
index.addLabels(UnicodeSet("[\\u00C6{Sch*}{St*}]", status), status);
TEST_CHECK_STATUS;
// ... A AE-ligature B-R S Sch St T-Z ...
LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status));
TEST_CHECK_STATUS;
assertEquals("getBucketCount()", 31, index.getBucketCount(status));
assertEquals("immutable getBucketCount()", 31, immIndex->getBucketCount());
static const struct TestCase {
const char *name;
int32_t bucketIndex;
const char *bucketLabel;
} testCases[] = {
// name, bucket index, bucket label
{ "Adelbert", 1, "A" },
{ "Afrika", 1, "A" },
{ "\\u00C6sculap", 2, "\\u00C6" },
{ "Aesthet", 2, "\\u00C6" },
{ "Berlin", 3, "B" },
{ "Rilke", 19, "R" },
{ "Sacher", 20, "S" },
{ "Seiler", 20, "S" },
{ "Sultan", 20, "S" },
{ "Schiller", 21, "Sch" },
{ "Steiff", 22, "St" },
{ "Thomas", 23, "T" }
};
for (int32_t i = 0; i < LENGTHOF(testCases); ++i) {
const TestCase &testCase = testCases[i];
UnicodeString name = UnicodeString(testCase.name).unescape();
UnicodeString label = UnicodeString(testCase.bucketLabel).unescape();
char msg[100];
sprintf(msg, "getBucketIndex(%s)", testCase.name);
assertEquals(msg, testCase.bucketIndex, index.getBucketIndex(name, status));
sprintf(msg, "immutable getBucketIndex(%s)", testCase.name);
assertEquals(msg, testCase.bucketIndex, immIndex->getBucketIndex(name, status));
sprintf(msg, "immutable bucket label (%s)", testCase.name);
assertEquals(msg, label, immIndex->getBucket(testCase.bucketIndex)->getLabel());
}
}
void AlphabeticIndexTest::TestIndexCharactersList() {
UErrorCode status = U_ZERO_ERROR;
for (int32_t i = 0; i < LENGTHOF(localeAndIndexCharactersLists); ++i) {
const char *(&localeAndIndexCharacters)[2] = localeAndIndexCharactersLists[i];
const char *locale = localeAndIndexCharacters[0];
UnicodeString expectedIndexCharacters
= (UnicodeString("\\u2026:") + localeAndIndexCharacters[1] + ":\\u2026").unescape();
AlphabeticIndex index(locale, status);
TEST_CHECK_STATUS;
LocalPointer<AlphabeticIndex::ImmutableIndex> immIndex(index.buildImmutableIndex(status));
TEST_CHECK_STATUS;
// Join the elements of the list to a string with delimiter ":"
UnicodeString actualIndexCharacters;
assertEquals(locale,
expectedIndexCharacters,
joinLabelsAndAppend(*immIndex, actualIndexCharacters));
logln(locale + UnicodeString(": ") + actualIndexCharacters);
}
}
