void
print_stderr_OChar_color(BashColors fg, BashColors bg, BashAttr attr, const OChar *text)
{
U_STRING_INIT(colorstr, "%c[%d;%d;%dm%s", 14);
u_fprintf_u(u_stderr, colorstr, 0x1B, attr, fg + 30, bg + 40, text);
reset();
}
U_CFUNC UResourceBundle *u_wmsg_setPath(const char *path, UErrorCode *err)
{
if(U_FAILURE(*err))
{
return 0;
}
if(gBundle != NULL)
{
*err = U_ILLEGAL_ARGUMENT_ERROR;
return 0;
}
else
{
UResourceBundle *b = NULL;
b = ures_open(path, NULL, err);
if(U_FAILURE(*err))
{
return 0;
}
gBundle = b;
U_STRING_INIT(gNoFormatting, " (UCONFIG_NO_FORMATTING see uconfig.h)", 38);
}
return gBundle;
}
static void
demoUnicodeStringInit() {
// *** Make sure to read about invariant characters in utypes.h! ***
// Initialization of Unicode strings from C literals works _only_ for
// invariant characters!
printf("\n* demoUnicodeStringInit() ---------- ***\n\n");
// the string literal is 32 chars long - this must be counted for the macro
UnicodeString invariantOnly=UNICODE_STRING("such characters are safe 123 %-.", 32);
/*
* In C, we need two macros: one to declare the UChar[] array, and
* one to populate it; the second one is a noop on platforms where
* wchar_t is compatible with UChar and ASCII-based.
* The length of the string literal must be counted for both macros.
*/
/* declare the invString array for the string */
U_STRING_DECL(invString, "such characters are safe 123 %-.", 32);
/* populate it with the characters */
U_STRING_INIT(invString, "such characters are safe 123 %-.", 32);
// compare the C and C++ strings
printf("C and C++ Unicode strings are equal: %d\n", invariantOnly==UnicodeString(TRUE, invString, 32));
/*
* convert between char * and UChar * strings that
* contain only invariant characters
*/
static const char *cs1="such characters are safe 123 %-.";
static UChar us1[40];
static char cs2[40];
u_charsToUChars(cs1, us1, 33); /* include the terminating NUL */
u_UCharsToChars(us1, cs2, 33);
printf("char * -> UChar * -> char * with only "
"invariant characters: \"%s\"\n",
cs2);
// initialize a UnicodeString from a string literal that contains
// escape sequences written with invariant characters
// do not forget to duplicate the backslashes for ICU to see them
// then, count each double backslash only once!
UnicodeString german=UNICODE_STRING(
"Sch\\u00f6nes Auto: \\u20ac 11240.\\fPrivates Zeichen: \\U00102345\\n", 64).
unescape();
printUnicodeString("german UnicodeString from unescaping:\n ", german);
/*
* C: convert and unescape a char * string with only invariant
* characters to fill a UChar * string
*/
UChar buffer[200];
int32_t length;
length=u_unescape(
"Sch\\u00f6nes Auto: \\u20ac 11240.\\fPrivates Zeichen: \\U00102345\\n",
buffer, UPRV_LENGTHOF(buffer));
printf("german C Unicode string from char * unescaping: (length %d)\n ", length);
printUnicodeString("", UnicodeString(buffer));
}
static void TestMessageWithUnusedArgNumber() {
UErrorCode errorCode = U_ZERO_ERROR;
U_STRING_DECL(pattern, "abc {1} def", 11);
UChar x[2] = { 0x78, 0 }; // "x"
UChar y[2] = { 0x79, 0 }; // "y"
U_STRING_DECL(expected, "abc y def", 9);
UChar result[20];
int32_t length;
U_STRING_INIT(pattern, "abc {1} def", 11);
U_STRING_INIT(expected, "abc y def", 9);
length = u_formatMessage("en", pattern, -1, result, LENGTHOF(result), &errorCode, x, y);
if (U_FAILURE(errorCode) || length != u_strlen(expected) || u_strcmp(result, expected) != 0) {
log_err("u_formatMessage(pattern with only {1}, 2 args) failed: result length %d, UErrorCode %s \n",
(int)length, u_errorName(errorCode));
}
}
static void TestBadPattern(void) {
UErrorCode status = U_ZERO_ERROR;
USet *pat;
U_STRING_DECL(pattern, "[", 1);
U_STRING_INIT(pattern, "[", 1);
pat = uset_openPatternOptions(pattern, u_strlen(pattern), 0, &status);
if (pat != NULL || U_SUCCESS(status)) {
log_err("uset_openPatternOptions did not fail as expected %s\n", u_errorName(status));
}
}
void
StringTest::Test_U_STRING() {
U_STRING_INIT(ustringVar, "aZ0 -", 5);
if( sizeof(ustringVar)/sizeof(*ustringVar)!=6 ||
ustringVar[0]!=0x61 ||
ustringVar[1]!=0x5a ||
ustringVar[2]!=0x30 ||
ustringVar[3]!=0x20 ||
ustringVar[4]!=0x2d ||
ustringVar[5]!=0
) {
errln("Test_U_STRING: U_STRING_DECL with U_STRING_INIT does not work right! "
"See putil.h and utypes.h with platform.h.");
}
}
static void
TestSerialized() {
uint16_t buffer[1000];
USerializedSet sset;
USet *set;
UErrorCode errorCode;
UChar32 c;
int32_t length;
/* use a pattern that generates both BMP and supplementary code points */
U_STRING_DECL(pattern, "[:Cf:]", 6);
U_STRING_INIT(pattern, "[:Cf:]", 6);
errorCode=U_ZERO_ERROR;
set=uset_openPattern(pattern, -1, &errorCode);
if(U_FAILURE(errorCode)) {
log_data_err("uset_openPattern([:Cf:]) failed - %s (Are you missing data?)\n", u_errorName(errorCode));
return;
}
length=uset_serialize(set, buffer, UPRV_LENGTHOF(buffer), &errorCode);
if(U_FAILURE(errorCode)) {
log_err("unable to uset_serialize([:Cf:]) - %s\n", u_errorName(errorCode));
uset_close(set);
return;
}
uset_getSerializedSet(&sset, buffer, length);
for(c=0; c<=0x10ffff; ++c) {
if(uset_contains(set, c)!=uset_serializedContains(&sset, c)) {
log_err("uset_contains(U+%04x)!=uset_serializedContains(U+%04x)\n", c);
break;
}
}
uset_close(set);
}
void SSearchTest::monkeyTest(char *params)
{
// ook!
UErrorCode status = U_ZERO_ERROR;
//UCollator *coll = ucol_open(NULL, &status);
UCollator *coll = ucol_openFromShortString("S1", FALSE, NULL, &status);
if (U_FAILURE(status)) {
errcheckln(status, "Failed to create collator in MonkeyTest! - %s", u_errorName(status));
return;
}
CollData *monkeyData = new CollData(coll, status);
USet *expansions = uset_openEmpty();
USet *contractions = uset_openEmpty();
ucol_getContractionsAndExpansions(coll, contractions, expansions, FALSE, &status);
U_STRING_DECL(letter_pattern, "[[:letter:]-[:ideographic:]-[:hangul:]]", 39);
U_STRING_INIT(letter_pattern, "[[:letter:]-[:ideographic:]-[:hangul:]]", 39);
USet *letters = uset_openPattern(letter_pattern, 39, &status);
SetMonkey letterMonkey(letters);
StringSetMonkey contractionMonkey(contractions, coll, monkeyData);
StringSetMonkey expansionMonkey(expansions, coll, monkeyData);
UnicodeString testCase;
UnicodeString alternate;
UnicodeString pattern, altPattern;
UnicodeString prefix, altPrefix;
UnicodeString suffix, altSuffix;
Monkey *monkeys[] = {
&letterMonkey,
&contractionMonkey,
&expansionMonkey,
&contractionMonkey,
&expansionMonkey,
&contractionMonkey,
&expansionMonkey,
&contractionMonkey,
&expansionMonkey};
int32_t monkeyCount = sizeof(monkeys) / sizeof(monkeys[0]);
// int32_t nonMatchCount = 0;
UCollationStrength strengths[] = {UCOL_PRIMARY, UCOL_SECONDARY, UCOL_TERTIARY};
const char *strengthNames[] = {"primary", "secondary", "tertiary"};
int32_t strengthCount = sizeof(strengths) / sizeof(strengths[0]);
int32_t loopCount = quick? 1000 : 10000;
int32_t firstStrength = 0;
int32_t lastStrength = strengthCount - 1; //*/ 0;
if (params != NULL) {
#if !UCONFIG_NO_REGULAR_EXPRESSIONS
UnicodeString p(params);
loopCount = getIntParam("loop", p, loopCount);
m_seed = getIntParam("seed", p, m_seed);
RegexMatcher m(" *strength *= *(primary|secondary|tertiary) *", p, 0, status);
if (m.find()) {
UnicodeString breakType = m.group(1, status);
for (int32_t s = 0; s < strengthCount; s += 1) {
if (breakType == strengthNames[s]) {
firstStrength = lastStrength = s;
break;
}
}
m.reset();
p = m.replaceFirst("", status);
}
if (RegexMatcher("\\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;
}
#else
infoln("SSearchTest built with UCONFIG_NO_REGULAR_EXPRESSIONS: ignoring parameters.");
#endif
}
for(int32_t s = firstStrength; s <= lastStrength; s += 1) {
int32_t notFoundCount = 0;
logln("Setting strength to %s.", strengthNames[s]);
ucol_setStrength(coll, strengths[s]);
// TODO: try alternate prefix and suffix too?
// TODO: alterntaes are only equal at primary strength. Is this OK?
for(int32_t t = 0; t < loopCount; t += 1) {
uint32_t seed = m_seed;
// int32_t nmc = 0;
generateTestCase(coll, monkeys, monkeyCount, pattern, altPattern);
generateTestCase(coll, monkeys, monkeyCount, prefix, altPrefix);
generateTestCase(coll, monkeys, monkeyCount, suffix, altSuffix);
// pattern
notFoundCount += monkeyTestCase(coll, pattern, pattern, altPattern, "pattern", strengthNames[s], seed);
testCase.remove();
testCase.append(prefix);
testCase.append(/*alt*/pattern);
// prefix + pattern
notFoundCount += monkeyTestCase(coll, testCase, pattern, altPattern, "prefix + pattern", strengthNames[s], seed);
testCase.append(suffix);
// prefix + pattern + suffix
notFoundCount += monkeyTestCase(coll, testCase, pattern, altPattern, "prefix + pattern + suffix", strengthNames[s], seed);
testCase.remove();
testCase.append(pattern);
testCase.append(suffix);
// pattern + suffix
notFoundCount += monkeyTestCase(coll, testCase, pattern, altPattern, "pattern + suffix", strengthNames[s], seed);
}
logln("For strength %s the not found count is %d.", strengthNames[s], notFoundCount);
}
uset_close(contractions);
uset_close(expansions);
uset_close(letters);
delete monkeyData;
ucol_close(coll);
}
CollData::CollData(UCollator *collator, UErrorCode &status)
: coll(NULL), ceToCharsStartingWith(NULL)
{
// [:c:] == [[:cn:][:cc:][:co:][:cf:][:cs:]]
// i.e. other, control, private use, format, surrogate
U_STRING_DECL(test_pattern, "[[:assigned:]-[:c:]]", 20);
U_STRING_INIT(test_pattern, "[[:assigned:]-[:c:]]", 20);
USet *charsToTest = uset_openPattern(test_pattern, 20, &status);
// Han ext. A, Han, Jamo, Hangul, Han Ext. B
// i.e. all the characers we handle implicitly
U_STRING_DECL(remove_pattern, "[[\\u3400-\\u9FFF][\\u1100-\\u11F9][\\uAC00-\\uD7AF][\\U00020000-\\U0002A6DF]]", 70);
U_STRING_INIT(remove_pattern, "[[\\u3400-\\u9FFF][\\u1100-\\u11F9][\\uAC00-\\uD7AF][\\U00020000-\\U0002A6DF]]", 70);
USet *charsToRemove = uset_openPattern(remove_pattern, 70, &status);
if (U_FAILURE(status)) {
return;
}
USet *expansions = uset_openEmpty();
USet *contractions = uset_openEmpty();
int32_t itemCount;
ceToCharsStartingWith = new CEToStringsMap(status);
if (U_FAILURE(status)) {
goto bail;
}
#ifdef CLONE_COLLATOR
coll = ucol_safeClone(collator, NULL, NULL, &status);
if (U_FAILURE(status)) {
goto bail;
}
#else
coll = collator;
#endif
ucol_getContractionsAndExpansions(coll, contractions, expansions, FALSE, &status);
uset_addAll(charsToTest, contractions);
uset_addAll(charsToTest, expansions);
uset_removeAll(charsToTest, charsToRemove);
itemCount = uset_getItemCount(charsToTest);
for(int32_t item = 0; item < itemCount; item += 1) {
UChar32 start = 0, end = 0;
UChar buffer[16];
int32_t len = uset_getItem(charsToTest, item, &start, &end,
buffer, 16, &status);
if (len == 0) {
for (UChar32 ch = start; ch <= end; ch += 1) {
UnicodeString *st = new UnicodeString(ch);
if (st == NULL) {
status = U_MEMORY_ALLOCATION_ERROR;
break;
}
CEList *ceList = new CEList(coll, *st, status);
ceToCharsStartingWith->put(ceList->get(0), st, status);
delete ceList;
delete st;
}
} else if (len > 0) {
UnicodeString *st = new UnicodeString(buffer, len);
if (st == NULL) {
status = U_MEMORY_ALLOCATION_ERROR;
break;
}
CEList *ceList = new CEList(coll, *st, status);
ceToCharsStartingWith->put(ceList->get(0), st, status);
delete ceList;
delete st;
} else {
// shouldn't happen...
}
if (U_FAILURE(status)) {
break;
}
}
bail:
uset_close(contractions);
uset_close(expansions);
uset_close(charsToRemove);
uset_close(charsToTest);
if (U_FAILURE(status)) {
return;
}
UnicodeSet hanRanges(UNICODE_STRING_SIMPLE("[:Unified_Ideograph:]"), status);
if (U_FAILURE(status)) {
return;
}
UnicodeSetIterator hanIter(hanRanges);
UnicodeString hanString;
while(hanIter.nextRange()) {
hanString.append(hanIter.getCodepoint());
hanString.append(hanIter.getCodepointEnd());
}
// TODO: Why U+11FF? The old code had an outdated UCOL_LAST_T_JAMO=0x11F9,
// but as of Unicode 6.3 the 11xx block is filled,
// and there are also more Jamo T at U+D7CB..U+D7FB.
// Maybe use [:HST=T:] and look for the end of the last range?
// Maybe use script boundary mappings instead of this code??
UChar jamoRanges[] = {Hangul::JAMO_L_BASE, Hangul::JAMO_V_BASE, Hangul::JAMO_T_BASE + 1, 0x11FF};
UnicodeString jamoString(FALSE, jamoRanges, UPRV_LENGTHOF(jamoRanges));
CEList hanList(coll, hanString, status);
CEList jamoList(coll, jamoString, status);
int32_t j = 0;
if (U_FAILURE(status)) {
return;
}
for (int32_t c = 0; c < jamoList.size(); c += 1) {
uint32_t jce = jamoList[c];
if (! isContinuation(jce)) {
jamoLimits[j++] = jce;
}
}
jamoLimits[3] += (1 << UCOL_PRIMARYORDERSHIFT);
minHan = 0xFFFFFFFF;
maxHan = 0;
for(int32_t h = 0; h < hanList.size(); h += 2) {
uint32_t han = (uint32_t) hanList[h];
if (han < minHan) {
minHan = han;
}
if (han > maxHan) {
maxHan = han;
}
}
maxHan += (1 << UCOL_PRIMARYORDERSHIFT);
}
void CharIterTest::TestConstructionAndEqualityUChariter() {
U_STRING_DECL(testText, "Now is the time for all good men to come to the aid of their country.", 69);
U_STRING_DECL(testText2, "Don't bother using this string.", 31);
U_STRING_INIT(testText, "Now is the time for all good men to come to the aid of their country.", 69);
U_STRING_INIT(testText2, "Don't bother using this string.", 31);
UnicodeString result, result4, result5;
UCharCharacterIterator* test1 = new UCharCharacterIterator(testText, u_strlen(testText));
UCharCharacterIterator* test2 = new UCharCharacterIterator(testText, u_strlen(testText), 5);
UCharCharacterIterator* test3 = new UCharCharacterIterator(testText, u_strlen(testText), 2, 20, 5);
UCharCharacterIterator* test4 = new UCharCharacterIterator(testText2, u_strlen(testText2));
UCharCharacterIterator* test5 = (UCharCharacterIterator*)test1->clone();
UCharCharacterIterator* test6 = new UCharCharacterIterator(*test1);
// j785: length=-1 will use u_strlen()
UCharCharacterIterator* test7a = new UCharCharacterIterator(testText, -1);
UCharCharacterIterator* test7b = new UCharCharacterIterator(testText, -1);
UCharCharacterIterator* test7c = new UCharCharacterIterator(testText, -1, 2, 20, 5);
// Bad parameters.
UCharCharacterIterator* test8a = new UCharCharacterIterator(testText, -1, -1, 20, 5);
UCharCharacterIterator* test8b = new UCharCharacterIterator(testText, -1, 2, 100, 5);
UCharCharacterIterator* test8c = new UCharCharacterIterator(testText, -1, 2, 20, 100);
if (test8a->startIndex() < 0)
errln("Construction failed: startIndex is negative");
if (test8b->endIndex() != u_strlen(testText))
errln("Construction failed: endIndex is different from the text length");
if (test8c->getIndex() < test8c->startIndex() || test8c->endIndex() < test8c->getIndex())
errln("Construction failed: index is invalid");
if (*test1 == *test2 || *test1 == *test3 || *test1 == *test4 )
errln("Construction or operator== failed: Unequal objects compared equal");
if (*test1 != *test5 )
errln("clone() or equals() failed: Two clones tested unequal");
if (*test6 != *test1 )
errln("copy construction or equals() failed: Two copies tested unequal");
if (test1->hashCode() == test2->hashCode() || test1->hashCode() == test3->hashCode()
|| test1->hashCode() == test4->hashCode())
errln("hashCode() failed: different objects have same hash code");
if (test1->hashCode() != test5->hashCode())
errln("hashCode() failed: identical objects have different hash codes");
test7a->getText(result);
test7b->getText(result4);
test7c->getText(result5);
if(result != UnicodeString(testText) || result4 != result || result5 != result)
errln("error in construction");
test1->getText(result);
test4->getText(result4);
test5->getText(result5);
if(result != result5 || result == result4)
errln("getText() failed");
test5->setText(testText2, u_strlen(testText2));
test5->getText(result5);
if(result == result5 || result4 != result5)
errln("setText() or getText() failed");
test5->setText(testText, u_strlen(testText));
test5->getText(result5);
if(result != result5 || result == result4)
errln("setText() or getText() round-trip failed");
test1->setIndex(5);
if (*test1 != *test2 || *test1 == *test5)
errln("setIndex() failed");
test8b->setIndex32(5);
if (test8b->getIndex()!=5)
errln("setIndex32() failed");
*test1 = *test3;
if (*test1 != *test3 || *test1 == *test5)
errln("operator= failed");
delete test1;
delete test2;
delete test3;
delete test4;
delete test5;
delete test6;
delete test7a;
delete test7b;
delete test7c;
delete test8a;
delete test8b;
delete test8c;
}
static USet *openIDSet() {
UErrorCode errorCode = U_ZERO_ERROR;
U_STRING_DECL(pattern, "[:ID_Continue:]", 15);
U_STRING_INIT(pattern, "[:ID_Continue:]", 15);
return uset_openPattern(pattern, 15, &errorCode);
}
static void TestRegisterUnregister() {
UErrorCode status=U_ZERO_ERROR;
UTransliterator* t1=NULL;
UTransliterator* rules=NULL, *rules2;
UTransliterator* inverse1=NULL;
UChar rule[]= { 0x0061, 0x003c, 0x003e, 0x0063}; /*a<>b*/
U_STRING_DECL(ID, "TestA-TestB", 11);
U_STRING_INIT(ID, "TestA-TestB", 11);
/* Make sure it doesn't exist */
t1=utrans_open("TestA-TestB", UTRANS_FORWARD,NULL,0,NULL, &status);
if(t1 != NULL || U_SUCCESS(status)) {
log_err("FAIL: TestA-TestB already registered\n");
return;
}
status=U_ZERO_ERROR;
/* Check inverse too */
inverse1=utrans_open("TestA-TestB", UTRANS_REVERSE, NULL,0,NULL,&status);
if(inverse1 != NULL || U_SUCCESS(status)) {
log_err("FAIL: TestA-TestB already registered\n");
return;
}
status=U_ZERO_ERROR;
/* Create it */
rules=utrans_open("TestA-TestB",UTRANS_FORWARD, rule, 4, NULL, &status);
if(U_FAILURE(status)) {
log_err("FAIL: utrans_openRules(a<>B) failed with error=%s\n", myErrorName(status));
return;
}
/* clone it so we can register it a second time */
rules2=utrans_clone(rules, &status);
if(U_FAILURE(status)) {
log_err("FAIL: utrans_clone(a<>B) failed with error=%s\n", myErrorName(status));
return;
}
status=U_ZERO_ERROR;
/* Register it */
utrans_register(rules, &status);
if(U_FAILURE(status)) {
log_err("FAIL: utrans_register failed with error=%s\n", myErrorName(status));
return;
}
status=U_ZERO_ERROR;
/* Now check again -- should exist now*/
t1= utrans_open("TestA-TestB", UTRANS_FORWARD, NULL,0,NULL,&status);
if(U_FAILURE(status) || t1 == NULL) {
log_err("FAIL: TestA-TestB not registered\n");
return;
}
utrans_close(t1);
/*unregister the instance*/
status=U_ZERO_ERROR;
utrans_unregister("TestA-TestB");
/* now Make sure it doesn't exist */
t1=utrans_open("TestA-TestB", UTRANS_FORWARD,NULL,0,NULL, &status);
if(U_SUCCESS(status) || t1 != NULL) {
log_err("FAIL: TestA-TestB isn't unregistered\n");
return;
}
utrans_close(t1);
/* now with utrans_unregisterID(const UChar *) */
status=U_ZERO_ERROR;
utrans_register(rules2, &status);
if(U_FAILURE(status)) {
log_err("FAIL: 2nd utrans_register failed with error=%s\n", myErrorName(status));
return;
}
status=U_ZERO_ERROR;
/* Now check again -- should exist now*/
t1= utrans_open("TestA-TestB", UTRANS_FORWARD, NULL,0,NULL,&status);
if(U_FAILURE(status) || t1 == NULL) {
log_err("FAIL: 2nd TestA-TestB not registered\n");
return;
}
utrans_close(t1);
/*unregister the instance*/
status=U_ZERO_ERROR;
utrans_unregisterID(ID, -1);
/* now Make sure it doesn't exist */
t1=utrans_openU(ID, -1, UTRANS_FORWARD,NULL,0,NULL, &status);
if(U_SUCCESS(status) || t1 != NULL) {
log_err("FAIL: 2nd TestA-TestB isn't unregistered\n");
return;
}
utrans_close(t1);
utrans_close(inverse1);
}