void
JamoTest::TestRealText() {
// Test text taken from the Unicode web site
static const char* const WHAT_IS_UNICODE[] = {
"\\uc720\\ub2c8\\ucf54\\ub4dc\\uc5d0", "\\ub300\\ud574", "?",
"\\uc5b4\\ub5a4", "\\ud50c\\ub7ab\\ud3fc,", "\\uc5b4\\ub5a4",
"\\ud504\\ub85c\\uadf8\\ub7a8,", "\\uc5b4\\ub5a4", "\\uc5b8\\uc5b4\\uc5d0\\ub3c4",
"\\uc0c1\\uad00\\uc5c6\\uc774", "\\uc720\\ub2c8\\ucf54\\ub4dc\\ub294", "\\ubaa8\\ub4e0",
"\\ubb38\\uc790\\uc5d0", "\\ub300\\ud574", "\\uace0\\uc720", "\\ubc88\\ud638\\ub97c",
"\\uc81c\\uacf5\\ud569\\ub2c8\\ub2e4.",
"\\uae30\\ubcf8\\uc801\\uc73c\\ub85c", "\\ucef4\\ud4e8\\ud130\\ub294",
"\\uc22b\\uc790\\ub9cc", "\\ucc98\\ub9ac\\ud569\\ub2c8\\ub2e4.", "\\uae00\\uc790\\ub098",
"\\ub2e4\\ub978", "\\ubb38\\uc790\\uc5d0\\ub3c4", "\\uc22b\\uc790\\ub97c",
"\\uc9c0\\uc815\\ud558\\uc5ec",
"\\uc800\\uc7a5\\ud569\\ub2c8\\ub2e4.", "\\uc720\\ub2c8\\ucf54\\ub4dc\\uac00",
"\\uac1c\\ubc1c\\ub418\\uae30", "\\uc804\\uc5d0\\ub294", "\\uc774\\ub7ec\\ud55c",
"\\uc22b\\uc790\\ub97c", "\\uc9c0\\uc815\\ud558\\uae30", "\\uc704\\ud574", "\\uc218\\ubc31",
"\\uac00\\uc9c0\\uc758", "\\ub2e4\\ub978", "\\uae30\\ud638\\ud654",
"\\uc2dc\\uc2a4\\ud15c\\uc744",
"\\uc0ac\\uc6a9\\ud588\\uc2b5\\ub2c8\\ub2e4.", "\\ub2e8\\uc77c", "\\uae30\\ud638\\ud654",
"\\ubc29\\ubc95\\uc73c\\ub85c\\ub294", "\\ubaa8\\ub4e0", "\\ubb38\\uc790\\ub97c",
"\\ud3ec\\ud568\\ud560", "\\uc218", "\\uc5c6\\uc5c8\\uc2b5\\ub2c8\\ub2e4.", "\\uc608\\ub97c",
"\\ub4e4\\uc5b4", "\\uc720\\ub7fd", "\\uc5f0\\ud569\\uc5d0\\uc11c\\ub9cc",
"\\ubcf4\\ub354\\ub77c\\ub3c4", "\\ubaa8\\ub4e0", "\\uac01", "\\ub098\\ub77c\\ubcc4",
"\\uc5b8\\uc5b4\\ub97c", "\\ucc98\\ub9ac\\ud558\\ub824\\uba74", "\\uc5ec\\ub7ec",
"\\uac1c\\uc758", "\\ub2e4\\ub978", "\\uae30\\ud638\\ud654", "\\ubc29\\ubc95\\uc774",
"\\ud544\\uc694\\ud569\\ub2c8\\ub2e4.", "\\uc601\\uc5b4\\uc640", "\\uac19\\uc740",
"\\ub2e8\\uc77c", "\\uc5b8\\uc5b4\\uc758", "\\uacbd\\uc6b0\\ub3c4",
"\\uacf5\\ud1b5\\uc801\\uc73c\\ub85c", "\\uc0ac\\uc6a9\\ub418\\ub294", "\\ubaa8\\ub4e0",
"\\uae00\\uc790,", "\\ubb38\\uc7a5", "\\ubd80\\ud638", "\\ubc0f",
"\\ud14c\\ud06c\\ub2c8\\uceec", "\\uae30\\ud638\\uc5d0", "\\ub9de\\ub294", "\\ub2e8\\uc77c",
"\\uae30\\ud638\\ud654", "\\ubc29\\ubc95\\uc744", "\\uac16\\uace0", "\\uc788\\uc9c0",
"\\ubabb\\ud558\\uc600\\uc2b5\\ub2c8\\ub2e4.",
"\\uc774\\ub7ec\\ud55c", "\\uae30\\ud638\\ud654", "\\uc2dc\\uc2a4\\ud15c\\uc740",
"\\ub610\\ud55c", "\\ub2e4\\ub978", "\\uae30\\ud638\\ud654", "\\uc2dc\\uc2a4\\ud15c\\uacfc",
"\\ucda9\\ub3cc\\ud569\\ub2c8\\ub2e4.", "\\uc989", "\\ub450", "\\uac00\\uc9c0",
"\\uae30\\ud638\\ud654", "\\ubc29\\ubc95\\uc774", "\\ub450", "\\uac1c\\uc758", "\\ub2e4\\ub978",
"\\ubb38\\uc790\\uc5d0", "\\ub300\\ud574", "\\uac19\\uc740", "\\ubc88\\ud638\\ub97c",
"\\uc0ac\\uc6a9\\ud558\\uac70\\ub098", "\\uac19\\uc740", "\\ubb38\\uc790\\uc5d0",
"\\ub300\\ud574", "\\ub2e4\\ub978", "\\ubc88\\ud638\\ub97c", "\\uc0ac\\uc6a9\\ud560", "\\uc218",
"\\uc788\\uc2b5\\ub2c8\\ub2e4.", "\\uc8fc\\uc5b4\\uc9c4", "\\ubaa8\\ub4e0",
"\\ucef4\\ud4e8\\ud130(\\ud2b9\\ud788", "\\uc11c\\ubc84)\\ub294", "\\uc11c\\ub85c",
"\\ub2e4\\ub978", "\\uc5ec\\ub7ec", "\\uac00\\uc9c0", "\\uae30\\ud638\\ud654",
"\\ubc29\\ubc95\\uc744", "\\uc9c0\\uc6d0\\ud574\\uc57c",
"\\ud569\\ub2c8\\ub2e4.", "\\uadf8\\ub7ec\\ub098,", "\\ub370\\uc774\\ud130\\ub97c",
"\\uc11c\\ub85c", "\\ub2e4\\ub978", "\\uae30\\ud638\\ud654", "\\ubc29\\ubc95\\uc774\\ub098",
"\\ud50c\\ub7ab\\ud3fc", "\\uac04\\uc5d0", "\\uc804\\ub2ec\\ud560", "\\ub54c\\ub9c8\\ub2e4",
"\\uadf8", "\\ub370\\uc774\\ud130\\ub294", "\\ud56d\\uc0c1", "\\uc190\\uc0c1\\uc758",
"\\uc704\\ud5d8\\uc744", "\\uacaa\\uac8c", "\\ub429\\ub2c8\\ub2e4.",
"\\uc720\\ub2c8\\ucf54\\ub4dc\\ub85c", "\\ubaa8\\ub4e0", "\\uac83\\uc744",
"\\ud574\\uacb0\\ud560", "\\uc218", "\\uc788\\uc2b5\\ub2c8\\ub2e4!",
"\\uc720\\ub2c8\\ucf54\\ub4dc\\ub294", "\\uc0ac\\uc6a9", "\\uc911\\uc778",
"\\ud50c\\ub7ab\\ud3fc,", "\\ud504\\ub85c\\uadf8\\ub7a8,", "\\uc5b8\\uc5b4\\uc5d0",
"\\uad00\\uacc4\\uc5c6\\uc774", "\\ubb38\\uc790\\ub9c8\\ub2e4", "\\uace0\\uc720\\ud55c",
"\\uc22b\\uc790\\ub97c",
"\\uc81c\\uacf5\\ud569\\ub2c8\\ub2e4.", "\\uc720\\ub2c8\\ucf54\\ub4dc",
"\\ud45c\\uc900\\uc740", // "Apple, HP, IBM, JustSystem, Microsoft, Oracle, SAP, "
// "Sun, Sybase, Unisys "
"\\ubc0f", "\\uae30\\ud0c0", "\\uc5ec\\ub7ec",
"\\ud68c\\uc0ac\\uc640", "\\uac19\\uc740", "\\uc5c5\\uacc4",
"\\uc120\\ub450\\uc8fc\\uc790\\uc5d0", "\\uc758\\ud574",
"\\ucc44\\ud0dd\\ub418\\uc5c8\\uc2b5\\ub2c8\\ub2e4.", "\\uc720\\ub2c8\\ucf54\\ub4dc\\ub294",
// "XML, Java, ECMAScript(JavaScript), LDAP, CORBA 3.0, WML "
"\\ub4f1\\uacfc",
"\\uac19\\uc774", "\\ud604\\uc7ac", "\\ub110\\ub9ac", "\\uc0ac\\uc6a9\\ub418\\ub294",
"\\ud45c\\uc900\\uc5d0\\uc11c", "\\ud544\\uc694\\ud558\\uba70", "\\uc774\\ub294", // "ISO/IEC",
"10646\\uc744", "\\uad6c\\ud604\\ud558\\ub294", "\\uacf5\\uc2dd\\uc801\\uc778",
"\\ubc29\\ubc95\\uc785\\ub2c8\\ub2e4.", "\\uc774\\ub294", "\\ub9ce\\uc740", "\\uc6b4\\uc601",
"\\uccb4\\uc81c,", "\\uc694\\uc998", "\\uc0ac\\uc6a9\\ub418\\ub294", "\\ubaa8\\ub4e0",
"\\ube0c\\ub77c\\uc6b0\\uc800", "\\ubc0f", "\\uae30\\ud0c0", "\\ub9ce\\uc740",
"\\uc81c\\ud488\\uc5d0\\uc11c",
"\\uc9c0\\uc6d0\\ub429\\ub2c8\\ub2e4.", "\\uc720\\ub2c8\\ucf54\\ub4dc",
"\\ud45c\\uc900\\uc758", "\\ubd80\\uc0c1\\uacfc", "\\uc774\\ub97c",
"\\uc9c0\\uc6d0\\ud558\\ub294", "\\ub3c4\\uad6c\\uc758", "\\uac00\\uc6a9\\uc131\\uc740",
"\\ucd5c\\uadfc", "\\uc804", "\\uc138\\uacc4\\uc5d0", "\\ubd88\\uace0", "\\uc788\\ub294",
"\\uae30\\uc220", "\\uacbd\\ud5a5\\uc5d0\\uc11c", "\\uac00\\uc7a5", "\\uc911\\uc694\\ud55c",
"\\ubd80\\ubd84\\uc744", "\\ucc28\\uc9c0\\ud558\\uace0", "\\uc788\\uc2b5\\ub2c8\\ub2e4.",
"\\uc720\\ub2c8\\ucf54\\ub4dc\\ub97c",
// Replaced a hyphen with a space to make the test case work with CLDR1.5
//"\\ud074\\ub77c\\uc774\\uc5b8\\ud2b8-\\uc11c\\ubc84", "\\ub610\\ub294",
"\\ud074\\ub77c\\uc774\\uc5b8\\ud2b8 \\uc11c\\ubc84", "\\ub610\\ub294",
// Replaced a hyphen with a space.
//"\\ub2e4\\uc911-\\uc5f0\\uacb0", "\\uc751\\uc6a9", "\\ud504\\ub85c\\uadf8\\ub7a8\\uacfc",
"\\ub2e4\\uc911 \\uc5f0\\uacb0", "\\uc751\\uc6a9", "\\ud504\\ub85c\\uadf8\\ub7a8\\uacfc",
"\\uc6f9", "\\uc0ac\\uc774\\ud2b8\\uc5d0", "\\ud1b5\\ud569\\ud558\\uba74",
"\\ub808\\uac70\\uc2dc", "\\ubb38\\uc790", "\\uc138\\ud2b8", "\\uc0ac\\uc6a9\\uc5d0",
"\\uc788\\uc5b4\\uc11c", "\\uc0c1\\ub2f9\\ud55c", "\\ube44\\uc6a9", "\\uc808\\uac10",
"\\ud6a8\\uacfc\\uac00",
"\\ub098\\ud0c0\\ub0a9\\ub2c8\\ub2e4.", "\\uc720\\ub2c8\\ucf54\\ub4dc\\ub97c",
"\\ud1b5\\ud574", "\\ub9ac\\uc5d4\\uc9c0\\ub2c8\\uc5b4\\ub9c1", "\\uc5c6\\uc774",
"\\ub2e4\\uc911", "\\ud50c\\ub7ab\\ud3fc,", "\\uc5b8\\uc5b4", "\\ubc0f", "\\uad6d\\uac00",
"\\uac04\\uc5d0", "\\ub2e8\\uc77c", "\\uc18c\\ud504\\ud2b8\\uc6e8\\uc5b4",
"\\ud50c\\ub7ab\\ud3fc", "\\ub610\\ub294", "\\ub2e8\\uc77c", "\\uc6f9",
"\\uc0ac\\uc774\\ud2b8\\ub97c", "\\ubaa9\\ud45c\\ub85c", "\\uc0bc\\uc744", "\\uc218",
"\\uc788\\uc2b5\\ub2c8\\ub2e4.", "\\uc774\\ub97c", "\\uc0ac\\uc6a9\\ud558\\uba74",
"\\ub370\\uc774\\ud130\\ub97c", "\\uc190\\uc0c1", "\\uc5c6\\uc774", "\\uc5ec\\ub7ec",
"\\uc2dc\\uc2a4\\ud15c\\uc744", "\\ud1b5\\ud574", "\\uc804\\uc1a1\\ud560", "\\uc218",
"\\uc788\\uc2b5\\ub2c8\\ub2e4.",
"\\uc720\\ub2c8\\ucf54\\ub4dc", "\\ucf58\\uc18c\\uc2dc\\uc5c4\\uc5d0", "\\ub300\\ud574",
"\\uc720\\ub2c8\\ucf54\\ub4dc", "\\ucf58\\uc18c\\uc2dc\\uc5c4\\uc740",
"\\ube44\\uc601\\ub9ac", "\\uc870\\uc9c1\\uc73c\\ub85c\\uc11c", "\\ud604\\ub300",
"\\uc18c\\ud504\\ud2b8\\uc6e8\\uc5b4", "\\uc81c\\ud488\\uacfc",
"\\ud45c\\uc900\\uc5d0\\uc11c", "\\ud14d\\uc2a4\\ud2b8\\uc758", "\\ud45c\\ud604\\uc744",
"\\uc9c0\\uc815\\ud558\\ub294", "\\uc720\\ub2c8\\ucf54\\ub4dc", "\\ud45c\\uc900\\uc758",
"\\uc0ac\\uc6a9\\uc744", "\\uac1c\\ubc1c\\ud558\\uace0", "\\ud655\\uc7a5\\ud558\\uba70",
"\\uc7a5\\ub824\\ud558\\uae30", "\\uc704\\ud574",
"\\uc138\\uc6cc\\uc84c\\uc2b5\\ub2c8\\ub2e4.", "\\ucf58\\uc18c\\uc2dc\\uc5c4",
"\\uba64\\ubc84\\uc27d\\uc740", "\\ucef4\\ud4e8\\ud130\\uc640", "\\uc815\\ubcf4",
"\\ucc98\\ub9ac", "\\uc0b0\\uc5c5\\uc5d0", "\\uc885\\uc0ac\\ud558\\uace0", "\\uc788\\ub294",
"\\uad11\\ubc94\\uc704\\ud55c", "\\ud68c\\uc0ac", "\\ubc0f", "\\uc870\\uc9c1\\uc758",
"\\ubc94\\uc704\\ub97c",
"\\ub098\\ud0c0\\ub0c5\\ub2c8\\ub2e4.", "\\ucf58\\uc18c\\uc2dc\\uc5c4\\uc758",
"\\uc7ac\\uc815\\uc740", "\\uc804\\uc801\\uc73c\\ub85c", "\\ud68c\\ube44\\uc5d0",
"\\uc758\\ud574", "\\ucda9\\ub2f9\\ub429\\ub2c8\\ub2e4.", "\\uc720\\ub2c8\\ucf54\\ub4dc",
"\\ucee8\\uc18c\\uc2dc\\uc5c4\\uc5d0\\uc11c\\uc758", "\\uba64\\ubc84\\uc27d\\uc740",
"\\uc804", "\\uc138\\uacc4", "\\uc5b4\\ub290", "\\uacf3\\uc5d0\\uc11c\\ub098",
"\\uc720\\ub2c8\\ucf54\\ub4dc", "\\ud45c\\uc900\\uc744", "\\uc9c0\\uc6d0\\ud558\\uace0",
"\\uadf8", "\\ud655\\uc7a5\\uacfc", "\\uad6c\\ud604\\uc744",
"\\uc9c0\\uc6d0\\ud558\\uace0\\uc790\\ud558\\ub294", "\\uc870\\uc9c1\\uacfc",
"\\uac1c\\uc778\\uc5d0\\uac8c", "\\uac1c\\ubc29\\ub418\\uc5b4",
"\\uc788\\uc2b5\\ub2c8\\ub2e4.",
"\\ub354", "\\uc790\\uc138\\ud55c", "\\ub0b4\\uc6a9\\uc740", "\\uc6a9\\uc5b4\\uc9d1,",
"\\uc608\\uc81c", "\\uc720\\ub2c8\\ucf54\\ub4dc", "\\uc0ac\\uc6a9", "\\uac00\\ub2a5",
"\\uc81c\\ud488,", "\\uae30\\uc220", "\\uc815\\ubcf4", "\\ubc0f", "\\uae30\\ud0c0",
"\\uc720\\uc6a9\\ud55c", "\\uc815\\ubcf4\\ub97c",
"\\ucc38\\uc870\\ud558\\uc2ed\\uc2dc\\uc624."
};
enum { WHAT_IS_UNICODE_length = UPRV_LENGTHOF(WHAT_IS_UNICODE) };
UParseError parseError;
UErrorCode status = U_ZERO_ERROR;
Transliterator* latinJamo = Transliterator::createInstance("Latin-Jamo", UTRANS_FORWARD, parseError, status);
Transliterator* jamoHangul = Transliterator::createInstance("NFC(NFD)", UTRANS_FORWARD, parseError, status);
if (latinJamo == 0 || jamoHangul == 0 || U_FAILURE(status)) {
delete latinJamo;
delete jamoHangul;
dataerrln("FAIL: createInstance returned NULL - %s", u_errorName(status));
return;
}
Transliterator* jamoLatin = latinJamo->createInverse(status);
Transliterator* hangulJamo = jamoHangul->createInverse(status);
if (jamoLatin == 0 || hangulJamo == 0) {
errln("FAIL: createInverse returned NULL");
delete latinJamo;
delete jamoLatin;
delete jamoHangul;
delete hangulJamo;
return;
}
Transliterator* tarray[4] =
{ hangulJamo, jamoLatin, latinJamo, jamoHangul };
CompoundTransliterator rt(tarray, 4);
UnicodeString buf;
int32_t total = 0;
int32_t errors = 0;
int32_t i;
for (i=0; i < WHAT_IS_UNICODE_length; ++i) {
++total;
UnicodeString hangul = UnicodeString(WHAT_IS_UNICODE[i], -1, US_INV);
hangul = hangul.unescape(); // Parse backslash-u escapes
UnicodeString hangulX = hangul;
rt.transliterate(hangulX);
if (hangul != hangulX) {
++errors;
UnicodeString jamo = hangul; hangulJamo->transliterate(jamo);
UnicodeString latin = jamo; jamoLatin->transliterate(latin);
UnicodeString jamo2 = latin; latinJamo->transliterate(jamo2);
UnicodeString hangul2 = jamo2; jamoHangul->transliterate(hangul2);
buf.remove(0);
buf.append("FAIL: ");
if (hangul2 != hangulX) {
buf.append((UnicodeString)"(Weird: " + hangulX + " != " + hangul2 + ")");
}
// The Hangul-Jamo conversion is not usually the
// bug here, so we hide it from display.
// Uncomment lines to see the Hangul.
buf.append(//hangul + " => " +
jamoToName(jamo) + " => " +
latin + " => " + jamoToName(jamo2)
//+ " => " + hangul2
);
errln(prettify(buf));
}
}
if (errors != 0) {
errln((UnicodeString)"Test word failures: " + errors + " out of " + total);
} else {
logln((UnicodeString)"All " + total + " test words passed");
}
delete latinJamo;
delete jamoLatin;
delete jamoHangul;
delete hangulJamo;
}
void
PluralFormat::applyPattern(const UnicodeString & newPattern, UErrorCode & status)
{
if (U_FAILURE(status))
{
return;
}
this->pattern = newPattern;
UnicodeString token;
int32_t braceCount = 0;
fmtToken type;
UBool spaceIncluded = FALSE;
if (fParsedValuesHash == NULL)
{
fParsedValuesHash = new Hashtable(TRUE, status);
if (U_FAILURE(status))
{
return;
}
fParsedValuesHash->setValueDeleter(deleteHashStrings);
}
UBool getKeyword = TRUE;
UnicodeString hashKeyword;
UnicodeString * hashPattern;
for (int32_t i = 0; i < pattern.length(); ++i)
{
UChar ch = pattern.charAt(i);
if (!inRange(ch, type))
{
if (getKeyword)
{
status = U_ILLEGAL_CHARACTER;
return;
}
else
{
token += ch;
continue;
}
}
switch (type)
{
case tSpace:
if (token.length() == 0)
{
continue;
}
if (getKeyword)
{
// space after keyword
spaceIncluded = TRUE;
}
else
{
token += ch;
}
break;
case tLeftBrace:
if (getKeyword)
{
if (fParsedValuesHash->get(token) != NULL)
{
status = U_DUPLICATE_KEYWORD;
return;
}
if (token.length() == 0)
{
status = U_PATTERN_SYNTAX_ERROR;
return;
}
if (!pluralRules->isKeyword(token))
{
status = U_UNDEFINED_KEYWORD;
return;
}
hashKeyword = token;
getKeyword = FALSE;
token.remove();
}
else
{
if (braceCount == 0)
{
status = U_UNEXPECTED_TOKEN;
return;
}
else
{
token += ch;
}
}
braceCount++;
spaceIncluded = FALSE;
break;
case tRightBrace:
if (getKeyword)
{
status = U_UNEXPECTED_TOKEN;
return;
}
else
{
hashPattern = new UnicodeString(token);
fParsedValuesHash->put(hashKeyword, hashPattern, status);
if (U_FAILURE(status))
{
return;
}
braceCount--;
if (braceCount == 0)
{
getKeyword = TRUE;
hashKeyword.remove();
hashPattern = NULL;
token.remove();
}
else
{
token += ch;
}
}
spaceIncluded = FALSE;
break;
case tLetter:
case tNumberSign:
if (spaceIncluded)
{
status = U_PATTERN_SYNTAX_ERROR;
return;
}
default:
token += ch;
break;
}
}
if (checkSufficientDefinition())
{
return;
}
else
{
status = U_DEFAULT_KEYWORD_MISSING;
return;
}
}
void
CalendarLimitTest::doLimitsTest(Calendar& cal,
const int32_t* fieldsToTest,
UDate startDate,
int32_t testDuration) {
static const int32_t FIELDS[] = {
UCAL_ERA,
UCAL_YEAR,
UCAL_MONTH,
UCAL_WEEK_OF_YEAR,
UCAL_WEEK_OF_MONTH,
UCAL_DAY_OF_MONTH,
UCAL_DAY_OF_YEAR,
UCAL_DAY_OF_WEEK_IN_MONTH,
UCAL_YEAR_WOY,
UCAL_EXTENDED_YEAR,
-1,
};
static const char* FIELD_NAME[] = {
"ERA", "YEAR", "MONTH", "WEEK_OF_YEAR", "WEEK_OF_MONTH",
"DAY_OF_MONTH", "DAY_OF_YEAR", "DAY_OF_WEEK",
"DAY_OF_WEEK_IN_MONTH", "AM_PM", "HOUR", "HOUR_OF_DAY",
"MINUTE", "SECOND", "MILLISECOND", "ZONE_OFFSET",
"DST_OFFSET", "YEAR_WOY", "DOW_LOCAL", "EXTENDED_YEAR",
"JULIAN_DAY", "MILLISECONDS_IN_DAY",
"IS_LEAP_MONTH"
};
UErrorCode status = U_ZERO_ERROR;
int32_t i, j;
UnicodeString ymd;
GregorianCalendar greg(status);
if (failure(status, "new GregorianCalendar")) {
return;
}
greg.setTime(startDate, status);
if (failure(status, "GregorianCalendar::setTime")) {
return;
}
logln((UnicodeString)"Start: " + startDate);
if (fieldsToTest == NULL) {
fieldsToTest = FIELDS;
}
// Keep a record of minima and maxima that we actually see.
// These are kept in an array of arrays of hashes.
int32_t limits[UCAL_FIELD_COUNT][4];
for (j = 0; j < UCAL_FIELD_COUNT; j++) {
limits[j][0] = INT32_MAX;
limits[j][1] = INT32_MIN;
limits[j][2] = INT32_MAX;
limits[j][3] = INT32_MIN;
}
// This test can run for a long time; show progress.
UDate millis = ucal_getNow();
UDate mark = millis + 5000; // 5 sec
millis -= testDuration * 1000; // stop time if testDuration<0
for (i = 0;
testDuration > 0 ? i < testDuration
: ucal_getNow() < millis;
++i) {
if (ucal_getNow() >= mark) {
logln((UnicodeString)"(" + i + " days)");
mark += 5000; // 5 sec
}
UDate testMillis = greg.getTime(status);
cal.setTime(testMillis, status);
cal.setMinimalDaysInFirstWeek(1);
if (failure(status, "Calendar set/getTime")) {
return;
}
for (j = 0; fieldsToTest[j] >= 0; ++j) {
UCalendarDateFields f = (UCalendarDateFields)fieldsToTest[j];
int32_t v = cal.get(f, status);
int32_t minActual = cal.getActualMinimum(f, status);
int32_t maxActual = cal.getActualMaximum(f, status);
int32_t minLow = cal.getMinimum(f);
int32_t minHigh = cal.getGreatestMinimum(f);
int32_t maxLow = cal.getLeastMaximum(f);
int32_t maxHigh = cal.getMaximum(f);
if (limits[j][0] > minActual) {
// the minimum
limits[j][0] = minActual;
}
if (limits[j][1] < minActual) {
// the greatest minimum
limits[j][1] = minActual;
}
if (limits[j][2] > maxActual) {
// the least maximum
limits[j][2] = maxActual;
}
if (limits[j][3] < maxActual) {
// the maximum
limits[j][3] = maxActual;
}
if (minActual < minLow || minActual > minHigh) {
errln((UnicodeString)"Fail: [" + cal.getType() + "] " +
ymdToString(cal, ymd) +
" Range for min of " + FIELD_NAME[f] + "(" + f +
")=" + minLow + ".." + minHigh +
", actual_min=" + minActual);
}
if (maxActual < maxLow || maxActual > maxHigh) {
errln((UnicodeString)"Fail: [" + cal.getType() + "] " +
ymdToString(cal, ymd) +
" Range for max of " + FIELD_NAME[f] + "(" + f +
")=" + maxLow + ".." + maxHigh +
", actual_max=" + maxActual);
}
if (v < minActual || v > maxActual) {
// timebomb per #9967, fix with #9972
if ( uprv_strcmp(cal.getType(), "dangi") == 0 &&
testMillis >= 1865635198000.0 &&
logKnownIssue("9972", "as per #9967")) { // Feb 2029 gregorian, end of dangi 4361
logln((UnicodeString)"Fail: [" + cal.getType() + "] " +
ymdToString(cal, ymd) +
" " + FIELD_NAME[f] + "(" + f + ")=" + v +
", actual=" + minActual + ".." + maxActual +
", allowed=(" + minLow + ".." + minHigh + ")..(" +
maxLow + ".." + maxHigh + ")");
} else {
errln((UnicodeString)"Fail: [" + cal.getType() + "] " +
ymdToString(cal, ymd) +
" " + FIELD_NAME[f] + "(" + f + ")=" + v +
", actual=" + minActual + ".." + maxActual +
", allowed=(" + minLow + ".." + minHigh + ")..(" +
maxLow + ".." + maxHigh + ")");
}
}
}
greg.add(UCAL_DAY_OF_YEAR, 1, status);
if (failure(status, "Calendar::add")) {
return;
}
}
// Check actual maxima and minima seen against ranges returned
// by API.
UnicodeString buf;
for (j = 0; fieldsToTest[j] >= 0; ++j) {
int32_t rangeLow, rangeHigh;
UBool fullRangeSeen = TRUE;
UCalendarDateFields f = (UCalendarDateFields)fieldsToTest[j];
buf.remove();
buf.append((UnicodeString)"[" + cal.getType() + "] " + FIELD_NAME[f]);
// Minumum
rangeLow = cal.getMinimum(f);
rangeHigh = cal.getGreatestMinimum(f);
if (limits[j][0] != rangeLow || limits[j][1] != rangeHigh) {
fullRangeSeen = FALSE;
}
buf.append((UnicodeString)" minima range=" + rangeLow + ".." + rangeHigh);
buf.append((UnicodeString)" minima actual=" + limits[j][0] + ".." + limits[j][1]);
// Maximum
rangeLow = cal.getLeastMaximum(f);
rangeHigh = cal.getMaximum(f);
if (limits[j][2] != rangeLow || limits[j][3] != rangeHigh) {
fullRangeSeen = FALSE;
}
buf.append((UnicodeString)" maxima range=" + rangeLow + ".." + rangeHigh);
buf.append((UnicodeString)" maxima actual=" + limits[j][2] + ".." + limits[j][3]);
if (fullRangeSeen) {
logln((UnicodeString)"OK: " + buf);
} else {
// This may or may not be an error -- if the range of dates
// we scan over doesn't happen to contain a minimum or
// maximum, it doesn't mean some other range won't.
logln((UnicodeString)"Warning: " + buf);
}
}
logln((UnicodeString)"End: " + greg.getTime(status));
}
inline void markAsSet() {
prefix.remove();
}
void
SelectFormat::applyPattern(const UnicodeString& newPattern, UErrorCode& status) {
if (U_FAILURE(status)) {
return;
}
this->pattern = newPattern;
enum State{ startState, keywordState, pastKeywordState, phraseState};
//Initialization
UnicodeString keyword ;
UnicodeString phrase ;
UnicodeString* ptrPhrase ;
int32_t braceCount = 0;
delete parsedValuesHash;
this->parsedValuesHash = NULL;
parsedValuesHash = new Hashtable(TRUE, status);
if (U_FAILURE(status)) {
return;
}
parsedValuesHash->setValueDeleter(uhash_deleteUnicodeString);
//Process the state machine
State state = startState;
for (int32_t i = 0; i < pattern.length(); ++i) {
//Get the character and check its type
UChar ch = pattern.charAt(i);
CharacterClass type = classifyCharacter(ch);
//Allow any character in phrase but nowhere else
if ( type == tOther ) {
if ( state == phraseState ){
phrase += ch;
continue;
}else {
status = U_PATTERN_SYNTAX_ERROR;
return;
}
}
//Process the state machine
switch (state) {
//At the start of pattern
case startState:
switch (type) {
case tSpace:
break;
case tStartKeyword:
state = keywordState;
keyword += ch;
break;
//If anything else is encountered, it's a syntax error
default:
status = U_PATTERN_SYNTAX_ERROR;
return;
}//end of switch(type)
break;
//Handle the keyword state
case keywordState:
switch (type) {
case tSpace:
state = pastKeywordState;
break;
case tStartKeyword:
case tContinueKeyword:
keyword += ch;
break;
case tLeftBrace:
state = phraseState;
break;
//If anything else is encountered, it's a syntax error
default:
status = U_PATTERN_SYNTAX_ERROR;
return;
}//end of switch(type)
break;
//Handle the pastkeyword state
case pastKeywordState:
switch (type) {
case tSpace:
break;
case tLeftBrace:
state = phraseState;
break;
//If anything else is encountered, it's a syntax error
default:
status = U_PATTERN_SYNTAX_ERROR;
return;
}//end of switch(type)
break;
//Handle the phrase state
case phraseState:
switch (type) {
case tLeftBrace:
braceCount++;
phrase += ch;
break;
case tRightBrace:
//Matching keyword, phrase pair found
if (braceCount == 0){
//Check validity of keyword
if (parsedValuesHash->get(keyword) != NULL) {
status = U_DUPLICATE_KEYWORD;
return;
}
if (keyword.length() == 0) {
status = U_PATTERN_SYNTAX_ERROR;
return;
}
//Store the keyword, phrase pair in hashTable
ptrPhrase = new UnicodeString(phrase);
parsedValuesHash->put( keyword, ptrPhrase, status);
//Reinitialize
keyword.remove();
phrase.remove();
ptrPhrase = NULL;
state = startState;
}
if (braceCount > 0){
braceCount-- ;
phrase += ch;
}
break;
default:
phrase += ch;
}//end of switch(type)
break;
//Handle the default case of switch(state)
default:
status = U_PATTERN_SYNTAX_ERROR;
return;
}//end of switch(state)
}
//Check if the state machine is back to startState
if ( state != startState){
status = U_PATTERN_SYNTAX_ERROR;
return;
}
//Check if "other" keyword is present
if ( !checkSufficientDefinition() ) {
status = U_DEFAULT_KEYWORD_MISSING;
}
return;
}
virtual void run()
{
fTraceInfo = 1;
LocalPointer<NumberFormat> percentFormatter;
UErrorCode status = U_ZERO_ERROR;
#if 0
// debugging code,
for (int i=0; i<4000; i++) {
status = U_ZERO_ERROR;
UDataMemory *data1 = udata_openChoice(0, "res", "en_US", isAcceptable, 0, &status);
UDataMemory *data2 = udata_openChoice(0, "res", "fr", isAcceptable, 0, &status);
udata_close(data1);
udata_close(data2);
if (U_FAILURE(status)) {
error("udata_openChoice failed.\n");
break;
}
}
return;
#endif
#if 0
// debugging code,
int m;
for (m=0; m<4000; m++) {
status = U_ZERO_ERROR;
UResourceBundle *res = NULL;
const char *localeName = NULL;
Locale loc = Locale::getEnglish();
localeName = loc.getName();
// localeName = "en";
// ResourceBundle bund = ResourceBundle(0, loc, status);
//umtx_lock(&gDebugMutex);
res = ures_open(NULL, localeName, &status);
//umtx_unlock(&gDebugMutex);
//umtx_lock(&gDebugMutex);
ures_close(res);
//umtx_unlock(&gDebugMutex);
if (U_FAILURE(status)) {
error("Resource bundle construction failed.\n");
break;
}
}
return;
#endif
// Keep this data here to avoid static initialization.
FormatThreadTestData kNumberFormatTestData[] =
{
FormatThreadTestData((double)5.0, UnicodeString("5", "")),
FormatThreadTestData( 6.0, UnicodeString("6", "")),
FormatThreadTestData( 20.0, UnicodeString("20", "")),
FormatThreadTestData( 8.0, UnicodeString("8", "")),
FormatThreadTestData( 8.3, UnicodeString("8.3", "")),
FormatThreadTestData( 12345, UnicodeString("12,345", "")),
FormatThreadTestData( 81890.23, UnicodeString("81,890.23", "")),
};
int32_t kNumberFormatTestDataLength = UPRV_LENGTHOF(kNumberFormatTestData);
// Keep this data here to avoid static initialization.
FormatThreadTestData kPercentFormatTestData[] =
{
FormatThreadTestData((double)5.0, CharsToUnicodeString("500\\u00a0%")),
FormatThreadTestData( 1.0, CharsToUnicodeString("100\\u00a0%")),
FormatThreadTestData( 0.26, CharsToUnicodeString("26\\u00a0%")),
FormatThreadTestData(
16384.99, CharsToUnicodeString("1\\u00a0638\\u00a0499\\u00a0%")), // U+00a0 = NBSP
FormatThreadTestData(
81890.23, CharsToUnicodeString("8\\u00a0189\\u00a0023\\u00a0%")),
};
int32_t kPercentFormatTestDataLength = UPRV_LENGTHOF(kPercentFormatTestData);
int32_t iteration;
status = U_ZERO_ERROR;
LocalPointer<NumberFormat> formatter(NumberFormat::createInstance(Locale::getEnglish(),status));
if(U_FAILURE(status)) {
IntlTest::gTest->dataerrln("%s:%d Error %s on NumberFormat::createInstance().",
__FILE__, __LINE__, u_errorName(status));
goto cleanupAndReturn;
}
percentFormatter.adoptInstead(NumberFormat::createPercentInstance(Locale::getFrench(),status));
if(U_FAILURE(status)) {
IntlTest::gTest->errln("%s:%d Error %s on NumberFormat::createPercentInstance().",
__FILE__, __LINE__, u_errorName(status));
goto cleanupAndReturn;
}
for(iteration = 0;!IntlTest::gTest->getErrors() && iteration<kFormatThreadIterations;iteration++)
{
int32_t whichLine = (iteration + fOffset)%kNumberFormatTestDataLength;
UnicodeString output;
formatter->format(kNumberFormatTestData[whichLine].number, output);
if(0 != output.compare(kNumberFormatTestData[whichLine].string)) {
IntlTest::gTest->errln("format().. expected " + kNumberFormatTestData[whichLine].string
+ " got " + output);
goto cleanupAndReturn;
}
// Now check percent.
output.remove();
whichLine = (iteration + fOffset)%kPercentFormatTestDataLength;
percentFormatter->format(kPercentFormatTestData[whichLine].number, output);
if(0 != output.compare(kPercentFormatTestData[whichLine].string))
{
IntlTest::gTest->errln("percent format().. \n" +
showDifference(kPercentFormatTestData[whichLine].string,output));
goto cleanupAndReturn;
}
// Test message error
const int kNumberOfMessageTests = 3;
UErrorCode statusToCheck;
UnicodeString patternToCheck;
Locale messageLocale;
Locale countryToCheck;
double currencyToCheck;
UnicodeString expected;
// load the cases.
switch((iteration+fOffset) % kNumberOfMessageTests)
{
default:
case 0:
statusToCheck= U_FILE_ACCESS_ERROR;
patternToCheck= "0:Someone from {2} is receiving a #{0}"
" error - {1}. Their telephone call is costing "
"{3,number,currency}."; // number,currency
messageLocale= Locale("en","US");
countryToCheck= Locale("","HR");
currencyToCheck= 8192.77;
expected= "0:Someone from Croatia is receiving a #4 error - "
"U_FILE_ACCESS_ERROR. Their telephone call is costing $8,192.77.";
break;
case 1:
statusToCheck= U_INDEX_OUTOFBOUNDS_ERROR;
patternToCheck= "1:A customer in {2} is receiving a #{0} error - {1}. "
"Their telephone call is costing {3,number,currency}."; // number,currency
messageLocale= Locale("de","DE@currency=DEM");
countryToCheck= Locale("","BF");
currencyToCheck= 2.32;
expected= CharsToUnicodeString(
"1:A customer in Burkina Faso is receiving a #8 error - U_INDEX_OUTOFBOUNDS_ERROR. "
"Their telephone call is costing 2,32\\u00A0DM.");
break;
case 2:
statusToCheck= U_MEMORY_ALLOCATION_ERROR;
patternToCheck= "2:user in {2} is receiving a #{0} error - {1}. "
"They insist they just spent {3,number,currency} "
"on memory."; // number,currency
messageLocale= Locale("de","AT@currency=ATS"); // Austrian German
countryToCheck= Locale("","US"); // hmm
currencyToCheck= 40193.12;
expected= CharsToUnicodeString(
"2:user in Vereinigte Staaten is receiving a #7 error"
" - U_MEMORY_ALLOCATION_ERROR. They insist they just spent"
" \\u00f6S\\u00A040\\u00A0193,12 on memory.");
break;
}
UnicodeString result;
UErrorCode status = U_ZERO_ERROR;
formatErrorMessage(status,patternToCheck,messageLocale,statusToCheck,
countryToCheck,currencyToCheck,result);
if(U_FAILURE(status))
{
UnicodeString tmp(u_errorName(status));
IntlTest::gTest->errln("Failure on message format, pattern=" + patternToCheck +
", error = " + tmp);
goto cleanupAndReturn;
}
if(result != expected)
{
IntlTest::gTest->errln("PatternFormat: \n" + showDifference(expected,result));
goto cleanupAndReturn;
}
// test the Thread Safe Format
UnicodeString appendErr;
if(!fTSF->doStuff(fNum, appendErr, status)) {
IntlTest::gTest->errln(appendErr);
goto cleanupAndReturn;
}
} /* end of for loop */
cleanupAndReturn:
fTraceInfo = 2;
}
UXMLElement *
UXMLParser::parseFile(const char *filename, UErrorCode &errorCode) {
char bytes[4096], charsetBuffer[100];
FileStream *f;
const char *charset, *pb;
UnicodeString src;
UConverter *cnv;
UChar *buffer, *pu;
int32_t fileLength, bytesLength, length, capacity;
UBool flush;
if(U_FAILURE(errorCode)) {
return NULL;
}
f=T_FileStream_open(filename, "rb");
if(f==NULL) {
errorCode=U_FILE_ACCESS_ERROR;
return NULL;
}
bytesLength=T_FileStream_read(f, bytes, (int32_t)sizeof(bytes));
if(bytesLength<(int32_t)sizeof(bytes)) {
// we have already read the entire file
fileLength=bytesLength;
} else {
// get the file length
fileLength=T_FileStream_size(f);
}
/*
* get the charset:
* 1. Unicode signature
* 2. treat as ISO-8859-1 and read XML encoding="charser"
* 3. default to UTF-8
*/
charset=ucnv_detectUnicodeSignature(bytes, bytesLength, NULL, &errorCode);
if(U_SUCCESS(errorCode) && charset!=NULL) {
// open converter according to Unicode signature
cnv=ucnv_open(charset, &errorCode);
} else {
// read as Latin-1 and parse the XML declaration and encoding
cnv=ucnv_open("ISO-8859-1", &errorCode);
if(U_FAILURE(errorCode)) {
// unexpected error opening Latin-1 converter
goto exit;
}
buffer=src.getBuffer(bytesLength);
if(buffer==NULL) {
// unexpected failure to reserve some string capacity
errorCode=U_MEMORY_ALLOCATION_ERROR;
goto exit;
}
pb=bytes;
pu=buffer;
ucnv_toUnicode(
cnv,
&pu, buffer+src.getCapacity(),
&pb, bytes+bytesLength,
NULL, TRUE, &errorCode);
src.releaseBuffer(U_SUCCESS(errorCode) ? (int32_t)(pu-buffer) : 0);
ucnv_close(cnv);
cnv=NULL;
if(U_FAILURE(errorCode)) {
// unexpected error in conversion from Latin-1
src.remove();
goto exit;
}
// parse XML declaration
if(mXMLDecl.reset(src).lookingAt(0, errorCode)) {
int32_t declEnd=mXMLDecl.end(errorCode);
// go beyond <?xml
int32_t pos=src.indexOf((UChar)x_l)+1;
mAttrValue.reset(src);
while(pos<declEnd && mAttrValue.lookingAt(pos, errorCode)) { // loop runs once per attribute on this element.
UnicodeString attName = mAttrValue.group(1, errorCode);
UnicodeString attValue = mAttrValue.group(2, errorCode);
// Trim the quotes from the att value. These are left over from the original regex
// that parsed the attribue, which couldn't conveniently strip them.
attValue.remove(0,1); // one char from the beginning
attValue.truncate(attValue.length()-1); // and one from the end.
if(attName==UNICODE_STRING("encoding", 8)) {
length=attValue.extract(0, 0x7fffffff, charsetBuffer, (int32_t)sizeof(charsetBuffer));
charset=charsetBuffer;
break;
}
pos = mAttrValue.end(2, errorCode);
}
if(charset==NULL) {
// default to UTF-8
charset="UTF-8";
}
cnv=ucnv_open(charset, &errorCode);
}
}
if(U_FAILURE(errorCode)) {
// unable to open the converter
goto exit;
}
// convert the file contents
capacity=fileLength; // estimated capacity
src.getBuffer(capacity);
src.releaseBuffer(0); // zero length
flush=FALSE;
for(;;) {
// convert contents of bytes[bytesLength]
pb=bytes;
for(;;) {
length=src.length();
buffer=src.getBuffer(capacity);
if(buffer==NULL) {
// unexpected failure to reserve some string capacity
errorCode=U_MEMORY_ALLOCATION_ERROR;
goto exit;
}
pu=buffer+length;
ucnv_toUnicode(
cnv, &pu, buffer+src.getCapacity(),
&pb, bytes+bytesLength,
NULL, FALSE, &errorCode);
src.releaseBuffer(U_SUCCESS(errorCode) ? (int32_t)(pu-buffer) : 0);
if(errorCode==U_BUFFER_OVERFLOW_ERROR) {
errorCode=U_ZERO_ERROR;
capacity=(3*src.getCapacity())/2; // increase capacity by 50%
} else {
break;
}
}
if(U_FAILURE(errorCode)) {
break; // conversion error
}
if(flush) {
break; // completely converted the file
}
// read next block
bytesLength=T_FileStream_read(f, bytes, (int32_t)sizeof(bytes));
if(bytesLength==0) {
// reached end of file, convert once more to flush the converter
flush=TRUE;
}
};
exit:
ucnv_close(cnv);
T_FileStream_close(f);
if(U_SUCCESS(errorCode)) {
return parse(src, errorCode);
} else {
return NULL;
}
}
void extractBetween(int32_t /*start*/, int32_t /*limit*/, UnicodeString& result) const {
result.remove();
}
