void UnicodeTest::TestScriptMetadata() {
IcuTestErrorCode errorCode(*this, "TestScriptMetadata()");
UnicodeSet rtl("[[:bc=R:][:bc=AL:]-[:Cn:]-[:sc=Common:]]", errorCode);
// So far, sample characters are uppercase.
// Georgian is special.
UnicodeSet cased("[[:Lu:]-[:sc=Common:]-[:sc=Geor:]]", errorCode);
for(int32_t sci = 0; sci < USCRIPT_CODE_LIMIT; ++sci) {
UScriptCode sc = (UScriptCode)sci;
// Run the test with -v to see which script has failures:
// .../intltest$ make && ./intltest utility/UnicodeTest/TestScriptMetadata -v | grep -C 3 FAIL
logln(uscript_getShortName(sc));
UScriptUsage usage = uscript_getUsage(sc);
UnicodeString sample = uscript_getSampleUnicodeString(sc);
UnicodeSet scriptSet;
scriptSet.applyIntPropertyValue(UCHAR_SCRIPT, sc, errorCode);
if(usage == USCRIPT_USAGE_NOT_ENCODED) {
assertTrue("not encoded, no sample", sample.isEmpty());
assertFalse("not encoded, not RTL", uscript_isRightToLeft(sc));
assertFalse("not encoded, not LB letters", uscript_breaksBetweenLetters(sc));
assertFalse("not encoded, not cased", uscript_isCased(sc));
assertTrue("not encoded, no characters", scriptSet.isEmpty());
} else {
assertFalse("encoded, has a sample character", sample.isEmpty());
UChar32 firstChar = sample.char32At(0);
UScriptCode charScript = getCharScript(sc);
assertEquals("script(sample(script))",
charScript, uscript_getScript(firstChar, errorCode));
assertEquals("RTL vs. set", rtl.contains(firstChar), uscript_isRightToLeft(sc));
assertEquals("cased vs. set", cased.contains(firstChar), uscript_isCased(sc));
assertEquals("encoded, has characters", sc == charScript, !scriptSet.isEmpty());
if(uscript_isRightToLeft(sc)) {
rtl.removeAll(scriptSet);
}
if(uscript_isCased(sc)) {
cased.removeAll(scriptSet);
}
}
}
UnicodeString pattern;
assertEquals("no remaining RTL characters",
UnicodeString("[]"), rtl.toPattern(pattern));
assertEquals("no remaining cased characters",
UnicodeString("[]"), cased.toPattern(pattern));
assertTrue("Hani breaks between letters", uscript_breaksBetweenLetters(USCRIPT_HAN));
assertTrue("Thai breaks between letters", uscript_breaksBetweenLetters(USCRIPT_THAI));
assertFalse("Latn does not break between letters", uscript_breaksBetweenLetters(USCRIPT_LATIN));
}
// Implemented here because this calls uloc_addLikelySubtags().
U_CAPI UBool U_EXPORT2
uloc_isRightToLeft(const char *locale) {
UErrorCode errorCode = U_ZERO_ERROR;
char script[8];
int32_t scriptLength = uloc_getScript(locale, script, UPRV_LENGTHOF(script), &errorCode);
if (U_FAILURE(errorCode) || errorCode == U_STRING_NOT_TERMINATED_WARNING ||
scriptLength == 0) {
// Fastpath: We know the likely scripts and their writing direction
// for some common languages.
errorCode = U_ZERO_ERROR;
char lang[8];
int32_t langLength = uloc_getLanguage(locale, lang, UPRV_LENGTHOF(lang), &errorCode);
if (U_FAILURE(errorCode) || errorCode == U_STRING_NOT_TERMINATED_WARNING ||
langLength == 0) {
return FALSE;
}
const char* langPtr = uprv_strstr(LANG_DIR_STRING, lang);
if (langPtr != NULL) {
switch (langPtr[langLength]) {
case '-': return FALSE;
case '+': return TRUE;
default: break; // partial match of a longer code
}
}
// Otherwise, find the likely script.
errorCode = U_ZERO_ERROR;
char likely[ULOC_FULLNAME_CAPACITY];
(void)uloc_addLikelySubtags(locale, likely, UPRV_LENGTHOF(likely), &errorCode);
if (U_FAILURE(errorCode) || errorCode == U_STRING_NOT_TERMINATED_WARNING) {
return FALSE;
}
scriptLength = uloc_getScript(likely, script, UPRV_LENGTHOF(script), &errorCode);
if (U_FAILURE(errorCode) || errorCode == U_STRING_NOT_TERMINATED_WARNING ||
scriptLength == 0) {
return FALSE;
}
}
UScriptCode scriptCode = (UScriptCode)u_getPropertyValueEnum(UCHAR_SCRIPT, script);
return uscript_isRightToLeft(scriptCode);
}
void TestScriptMetadataAPI() {
/* API & code coverage. More testing in intltest/ucdtest.cpp. */
UErrorCode errorCode=U_ZERO_ERROR;
UChar sample[8];
if(uscript_getSampleString(USCRIPT_LATIN, sample, UPRV_LENGTHOF(sample), &errorCode)!=1 ||
U_FAILURE(errorCode) ||
uscript_getScript(sample[0], &errorCode)!=USCRIPT_LATIN ||
sample[1]!=0) {
log_err("uscript_getSampleString(Latn) failed - %s\n", u_errorName(errorCode));
}
sample[0]=0xfffe;
if(uscript_getSampleString(USCRIPT_LATIN, sample, 0, &errorCode)!=1 ||
errorCode!=U_BUFFER_OVERFLOW_ERROR ||
sample[0]!=0xfffe) {
log_err("uscript_getSampleString(Latn, capacity=0) failed - %s\n", u_errorName(errorCode));
}
errorCode=U_ZERO_ERROR;
if(uscript_getSampleString(USCRIPT_INVALID_CODE, sample, UPRV_LENGTHOF(sample), &errorCode)!=0 ||
U_FAILURE(errorCode) ||
sample[0]!=0) {
log_err("uscript_getSampleString(invalid) failed - %s\n", u_errorName(errorCode));
}
sample[0]=0xfffe;
if(uscript_getSampleString(USCRIPT_CODE_LIMIT, sample, 0, &errorCode)!=0 ||
errorCode!=U_STRING_NOT_TERMINATED_WARNING ||
sample[0]!=0xfffe) {
log_err("uscript_getSampleString(limit, capacity=0) failed - %s\n", u_errorName(errorCode));
}
if(uscript_getUsage(USCRIPT_LATIN)!=USCRIPT_USAGE_RECOMMENDED ||
uscript_getUsage(USCRIPT_YI)!=USCRIPT_USAGE_ASPIRATIONAL ||
uscript_getUsage(USCRIPT_CHEROKEE)!=USCRIPT_USAGE_LIMITED_USE ||
uscript_getUsage(USCRIPT_COPTIC)!=USCRIPT_USAGE_EXCLUDED ||
uscript_getUsage(USCRIPT_CIRTH)!=USCRIPT_USAGE_NOT_ENCODED ||
uscript_getUsage(USCRIPT_INVALID_CODE)!=USCRIPT_USAGE_NOT_ENCODED ||
uscript_getUsage(USCRIPT_CODE_LIMIT)!=USCRIPT_USAGE_NOT_ENCODED) {
log_err("uscript_getUsage() failed\n");
}
if(uscript_isRightToLeft(USCRIPT_LATIN) ||
uscript_isRightToLeft(USCRIPT_CIRTH) ||
!uscript_isRightToLeft(USCRIPT_ARABIC) ||
!uscript_isRightToLeft(USCRIPT_HEBREW)) {
log_err("uscript_isRightToLeft() failed\n");
}
if(uscript_breaksBetweenLetters(USCRIPT_LATIN) ||
uscript_breaksBetweenLetters(USCRIPT_CIRTH) ||
!uscript_breaksBetweenLetters(USCRIPT_HAN) ||
!uscript_breaksBetweenLetters(USCRIPT_THAI)) {
log_err("uscript_breaksBetweenLetters() failed\n");
}
if(uscript_isCased(USCRIPT_CIRTH) ||
uscript_isCased(USCRIPT_HAN) ||
!uscript_isCased(USCRIPT_LATIN) ||
!uscript_isCased(USCRIPT_GREEK)) {
log_err("uscript_isCased() failed\n");
}
}
