diff options
Diffstat (limited to 'deps/icu-small/source/i18n')
127 files changed, 6621 insertions, 2710 deletions
diff --git a/deps/icu-small/source/i18n/alphaindex.cpp b/deps/icu-small/source/i18n/alphaindex.cpp index f4a082c5b2..99f70114cb 100644 --- a/deps/icu-small/source/i18n/alphaindex.cpp +++ b/deps/icu-small/source/i18n/alphaindex.cpp @@ -511,8 +511,8 @@ BucketList *AlphabeticIndex::createBucketList(UErrorCode &errorCode) const { ces, errorCode) && current.charAt(current.length() - 1) != 0xFFFF /* !current.endsWith("\uffff") */) { // "AE-ligature" or "Sch" etc. - for (int32_t i = bucketList->size() - 2;; --i) { - Bucket *singleBucket = getBucket(*bucketList, i); + for (int32_t j = bucketList->size() - 2;; --j) { + Bucket *singleBucket = getBucket(*bucketList, j); if (singleBucket->labelType_ != U_ALPHAINDEX_NORMAL) { // There is no single-character bucket since the last // underflow or inflow label. @@ -608,8 +608,8 @@ BucketList *AlphabeticIndex::createBucketList(UErrorCode &errorCode) const { } // Do not call publicBucketList->setDeleter(): // This vector shares its objects with the bucketList. - for (int32_t i = 0; i < bucketList->size(); ++i) { - bucket = getBucket(*bucketList, i); + for (int32_t j = 0; j < bucketList->size(); ++j) { + bucket = getBucket(*bucketList, j); if (bucket->displayBucket_ == NULL) { publicBucketList->addElement(bucket, errorCode); } diff --git a/deps/icu-small/source/i18n/anytrans.cpp b/deps/icu-small/source/i18n/anytrans.cpp index d06469e2ae..6e382b824b 100644 --- a/deps/icu-small/source/i18n/anytrans.cpp +++ b/deps/icu-small/source/i18n/anytrans.cpp @@ -391,12 +391,12 @@ void AnyTransliterator::registerIDs() { UnicodeString id; TransliteratorIDParser::STVtoID(UnicodeString(TRUE, ANY, 3), target, variant, id); ec = U_ZERO_ERROR; - AnyTransliterator* t = new AnyTransliterator(id, target, variant, + AnyTransliterator* tl = new AnyTransliterator(id, target, variant, targetScript, ec); if (U_FAILURE(ec)) { - delete t; + delete tl; } else { - Transliterator::_registerInstance(t); + Transliterator::_registerInstance(tl); Transliterator::_registerSpecialInverse(target, UnicodeString(TRUE, NULL_ID, 4), FALSE); } } diff --git a/deps/icu-small/source/i18n/calendar.cpp b/deps/icu-small/source/i18n/calendar.cpp index 61757cb250..24c2fb964e 100644 --- a/deps/icu-small/source/i18n/calendar.cpp +++ b/deps/icu-small/source/i18n/calendar.cpp @@ -3797,7 +3797,7 @@ Calendar::setWeekData(const Locale& desiredLocale, const char *type, UErrorCode& Locale max = Locale::createFromName(maxLocaleID); useLocale = Locale(max.getLanguage(),max.getCountry()); } else { - useLocale = Locale(desiredLocale); + useLocale = desiredLocale; } /* The code here is somewhat of a hack, since week data and weekend data aren't really tied to diff --git a/deps/icu-small/source/i18n/coll.cpp b/deps/icu-small/source/i18n/coll.cpp index b7348a1d16..2a614fac5d 100644 --- a/deps/icu-small/source/i18n/coll.cpp +++ b/deps/icu-small/source/i18n/coll.cpp @@ -448,6 +448,13 @@ Collator* U_EXPORT2 Collator::createInstance(const Locale& desiredLocale, #endif { coll = makeInstance(desiredLocale, status); + // Either returns NULL with U_FAILURE(status), or non-NULL with U_SUCCESS(status) + } + // The use of *coll in setAttributesFromKeywords can cause the NULL check to be + // optimized out of the delete even though setAttributesFromKeywords returns + // immediately if U_FAILURE(status), so we add a check here. + if (U_FAILURE(status)) { + return NULL; } setAttributesFromKeywords(desiredLocale, *coll, status); if (U_FAILURE(status)) { @@ -986,8 +993,8 @@ Collator::internalCompareUTF8(const char *left, int32_t leftLength, return UCOL_EQUAL; } return compareUTF8( - StringPiece(left, (leftLength < 0) ? uprv_strlen(left) : leftLength), - StringPiece(right, (rightLength < 0) ? uprv_strlen(right) : rightLength), + StringPiece(left, (leftLength < 0) ? static_cast<int32_t>(uprv_strlen(left)) : leftLength), + StringPiece(right, (rightLength < 0) ? static_cast<int32_t>(uprv_strlen(right)) : rightLength), errorCode); } diff --git a/deps/icu-small/source/i18n/collationkeys.cpp b/deps/icu-small/source/i18n/collationkeys.cpp index 4b9e6b5907..b5c322fb44 100644 --- a/deps/icu-small/source/i18n/collationkeys.cpp +++ b/deps/icu-small/source/i18n/collationkeys.cpp @@ -403,13 +403,13 @@ CollationKeys::writeSortKeyUpToQuaternary(CollationIterator &iter, uint8_t *secs = secondaries.data(); int32_t last = secondaries.length() - 1; if(secSegmentStart < last) { - uint8_t *p = secs + secSegmentStart; - uint8_t *q = secs + last; + uint8_t *q = secs + secSegmentStart; + uint8_t *r = secs + last; do { - uint8_t b = *p; - *p++ = *q; - *q-- = b; - } while(p < q); + uint8_t b = *q; + *q++ = *r; + *r-- = b; + } while(q < r); } secondaries.appendByte(p == Collation::NO_CE_PRIMARY ? Collation::LEVEL_SEPARATOR_BYTE : Collation::MERGE_SEPARATOR_BYTE); diff --git a/deps/icu-small/source/i18n/csrmbcs.cpp b/deps/icu-small/source/i18n/csrmbcs.cpp index 0c2df594d5..46d626bb3f 100644 --- a/deps/icu-small/source/i18n/csrmbcs.cpp +++ b/deps/icu-small/source/i18n/csrmbcs.cpp @@ -166,7 +166,7 @@ int32_t CharsetRecog_mbcs::match_mbcs(InputText *det, const uint16_t commonChars doubleByteCharCount++; if (commonChars != 0) { - if (binarySearch(commonChars, commonCharsLen, iter.charValue) >= 0){ + if (binarySearch(commonChars, commonCharsLen, static_cast<uint16_t>(iter.charValue)) >= 0){ commonCharCount += 1; } } diff --git a/deps/icu-small/source/i18n/currpinf.cpp b/deps/icu-small/source/i18n/currpinf.cpp index 6b1efd5f4d..f5d27e28d8 100644 --- a/deps/icu-small/source/i18n/currpinf.cpp +++ b/deps/icu-small/source/i18n/currpinf.cpp @@ -17,7 +17,6 @@ #include <iostream> #endif - #include "unicode/locid.h" #include "unicode/plurrule.h" #include "unicode/strenum.h" @@ -30,7 +29,6 @@ U_NAMESPACE_BEGIN - static const UChar gNumberPatternSeparator = 0x3B; // ; U_CDECL_BEGIN @@ -65,66 +63,86 @@ static const char gDecimalFormatTag[]="decimalFormat"; static const char gCurrUnitPtnTag[]="CurrencyUnitPatterns"; CurrencyPluralInfo::CurrencyPluralInfo(UErrorCode& status) -: fPluralCountToCurrencyUnitPattern(NULL), - fPluralRules(NULL), - fLocale(NULL) { +: fPluralCountToCurrencyUnitPattern(nullptr), + fPluralRules(nullptr), + fLocale(nullptr), + fInternalStatus(U_ZERO_ERROR) { initialize(Locale::getDefault(), status); } CurrencyPluralInfo::CurrencyPluralInfo(const Locale& locale, UErrorCode& status) -: fPluralCountToCurrencyUnitPattern(NULL), - fPluralRules(NULL), - fLocale(NULL) { +: fPluralCountToCurrencyUnitPattern(nullptr), + fPluralRules(nullptr), + fLocale(nullptr), + fInternalStatus(U_ZERO_ERROR) { initialize(locale, status); } CurrencyPluralInfo::CurrencyPluralInfo(const CurrencyPluralInfo& info) : UObject(info), - fPluralCountToCurrencyUnitPattern(NULL), - fPluralRules(NULL), - fLocale(NULL) { + fPluralCountToCurrencyUnitPattern(nullptr), + fPluralRules(nullptr), + fLocale(nullptr), + fInternalStatus(U_ZERO_ERROR) { *this = info; } - CurrencyPluralInfo& CurrencyPluralInfo::operator=(const CurrencyPluralInfo& info) { if (this == &info) { return *this; } + fInternalStatus = info.fInternalStatus; + if (U_FAILURE(fInternalStatus)) { + // bail out early if the object we were copying from was already 'invalid'. + return *this; + } + deleteHash(fPluralCountToCurrencyUnitPattern); - UErrorCode status = U_ZERO_ERROR; - fPluralCountToCurrencyUnitPattern = initHash(status); + fPluralCountToCurrencyUnitPattern = initHash(fInternalStatus); copyHash(info.fPluralCountToCurrencyUnitPattern, - fPluralCountToCurrencyUnitPattern, status); - if ( U_FAILURE(status) ) { + fPluralCountToCurrencyUnitPattern, fInternalStatus); + if ( U_FAILURE(fInternalStatus) ) { return *this; } delete fPluralRules; + fPluralRules = nullptr; delete fLocale; - if (info.fPluralRules) { + fLocale = nullptr; + + if (info.fPluralRules != nullptr) { fPluralRules = info.fPluralRules->clone(); - } else { - fPluralRules = NULL; + if (fPluralRules == nullptr) { + fInternalStatus = U_MEMORY_ALLOCATION_ERROR; + return *this; + } } - if (info.fLocale) { + if (info.fLocale != nullptr) { fLocale = info.fLocale->clone(); - } else { - fLocale = NULL; + if (fLocale == nullptr) { + // Note: If clone had an error parameter, then we could check/set that instead. + fInternalStatus = U_MEMORY_ALLOCATION_ERROR; + return *this; + } + // If the other locale wasn't bogus, but our clone'd locale is bogus, then OOM happened + // during the call to clone(). + if (!info.fLocale->isBogus() && fLocale->isBogus()) { + fInternalStatus = U_MEMORY_ALLOCATION_ERROR; + return *this; + } } return *this; } - CurrencyPluralInfo::~CurrencyPluralInfo() { deleteHash(fPluralCountToCurrencyUnitPattern); - fPluralCountToCurrencyUnitPattern = NULL; + fPluralCountToCurrencyUnitPattern = nullptr; delete fPluralRules; delete fLocale; - fPluralRules = NULL; - fLocale = NULL; + fPluralRules = nullptr; + fLocale = nullptr; } UBool @@ -148,7 +166,14 @@ CurrencyPluralInfo::operator==(const CurrencyPluralInfo& info) const { CurrencyPluralInfo* CurrencyPluralInfo::clone() const { - return new CurrencyPluralInfo(*this); + CurrencyPluralInfo* newObj = new CurrencyPluralInfo(*this); + // Since clone doesn't have a 'status' parameter, the best we can do is return nullptr + // if the new object was not full constructed properly (an error occurred). + if (newObj != nullptr && U_FAILURE(newObj->fInternalStatus)) { + delete newObj; + newObj = nullptr; + } + return newObj; } const PluralRules* @@ -161,15 +186,15 @@ CurrencyPluralInfo::getCurrencyPluralPattern(const UnicodeString& pluralCount, UnicodeString& result) const { const UnicodeString* currencyPluralPattern = (UnicodeString*)fPluralCountToCurrencyUnitPattern->get(pluralCount); - if (currencyPluralPattern == NULL) { + if (currencyPluralPattern == nullptr) { // fall back to "other" if (pluralCount.compare(gPluralCountOther, 5)) { currencyPluralPattern = (UnicodeString*)fPluralCountToCurrencyUnitPattern->get(UnicodeString(TRUE, gPluralCountOther, 5)); } - if (currencyPluralPattern == NULL) { + if (currencyPluralPattern == nullptr) { // no currencyUnitPatterns defined, - // fallback to predefined defult. + // fallback to predefined default. // This should never happen when ICU resource files are // available, since currencyUnitPattern of "other" is always // defined in root. @@ -190,14 +215,11 @@ void CurrencyPluralInfo::setPluralRules(const UnicodeString& ruleDescription, UErrorCode& status) { if (U_SUCCESS(status)) { - if (fPluralRules) { - delete fPluralRules; - } + delete fPluralRules; fPluralRules = PluralRules::createRules(ruleDescription, status); } } - void CurrencyPluralInfo::setCurrencyPluralPattern(const UnicodeString& pluralCount, const UnicodeString& pattern, @@ -206,63 +228,77 @@ CurrencyPluralInfo::setCurrencyPluralPattern(const UnicodeString& pluralCount, UnicodeString* oldValue = static_cast<UnicodeString*>( fPluralCountToCurrencyUnitPattern->get(pluralCount)); delete oldValue; - fPluralCountToCurrencyUnitPattern->put(pluralCount, new UnicodeString(pattern), status); + LocalPointer<UnicodeString> p(new UnicodeString(pattern), status); + if (U_SUCCESS(status)) { + // the p object allocated above will be owned by fPluralCountToCurrencyUnitPattern + // after the call to put(), even if the method returns failure. + fPluralCountToCurrencyUnitPattern->put(pluralCount, p.orphan(), status); + } } } - void CurrencyPluralInfo::setLocale(const Locale& loc, UErrorCode& status) { initialize(loc, status); } - void CurrencyPluralInfo::initialize(const Locale& loc, UErrorCode& status) { if (U_FAILURE(status)) { return; } delete fLocale; + fLocale = nullptr; + delete fPluralRules; + fPluralRules = nullptr; + fLocale = loc.clone(); - if (fPluralRules) { - delete fPluralRules; + if (fLocale == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + return; + } + // If the locale passed in wasn't bogus, but our clone'd locale is bogus, then OOM happened + // during the call to loc.clone(). + if (!loc.isBogus() && fLocale->isBogus()) { + status = U_MEMORY_ALLOCATION_ERROR; + return; } fPluralRules = PluralRules::forLocale(loc, status); setupCurrencyPluralPattern(loc, status); } - void CurrencyPluralInfo::setupCurrencyPluralPattern(const Locale& loc, UErrorCode& status) { if (U_FAILURE(status)) { return; } - if (fPluralCountToCurrencyUnitPattern) { - deleteHash(fPluralCountToCurrencyUnitPattern); - } + deleteHash(fPluralCountToCurrencyUnitPattern); fPluralCountToCurrencyUnitPattern = initHash(status); if (U_FAILURE(status)) { return; } - NumberingSystem *ns = NumberingSystem::createInstance(loc,status); + LocalPointer<NumberingSystem> ns(NumberingSystem::createInstance(loc, status), status); + if (U_FAILURE(status)) { + return; + } UErrorCode ec = U_ZERO_ERROR; - UResourceBundle *rb = ures_open(NULL, loc.getName(), &ec); - UResourceBundle *numElements = ures_getByKeyWithFallback(rb, gNumberElementsTag, NULL, &ec); - rb = ures_getByKeyWithFallback(numElements, ns->getName(), rb, &ec); - rb = ures_getByKeyWithFallback(rb, gPatternsTag, rb, &ec); + LocalUResourceBundlePointer rb(ures_open(nullptr, loc.getName(), &ec)); + LocalUResourceBundlePointer numElements(ures_getByKeyWithFallback(rb.getAlias(), gNumberElementsTag, nullptr, &ec)); + ures_getByKeyWithFallback(numElements.getAlias(), ns->getName(), rb.getAlias(), &ec); + ures_getByKeyWithFallback(rb.getAlias(), gPatternsTag, rb.getAlias(), &ec); int32_t ptnLen; - const UChar* numberStylePattern = ures_getStringByKeyWithFallback(rb, gDecimalFormatTag, &ptnLen, &ec); + const UChar* numberStylePattern = ures_getStringByKeyWithFallback(rb.getAlias(), gDecimalFormatTag, &ptnLen, &ec); // Fall back to "latn" if num sys specific pattern isn't there. - if ( ec == U_MISSING_RESOURCE_ERROR && uprv_strcmp(ns->getName(),gLatnTag)) { + if ( ec == U_MISSING_RESOURCE_ERROR && (uprv_strcmp(ns->getName(), gLatnTag) != 0)) { ec = U_ZERO_ERROR; - rb = ures_getByKeyWithFallback(numElements, gLatnTag, rb, &ec); - rb = ures_getByKeyWithFallback(rb, gPatternsTag, rb, &ec); - numberStylePattern = ures_getStringByKeyWithFallback(rb, gDecimalFormatTag, &ptnLen, &ec); + ures_getByKeyWithFallback(numElements.getAlias(), gLatnTag, rb.getAlias(), &ec); + ures_getByKeyWithFallback(rb.getAlias(), gPatternsTag, rb.getAlias(), &ec); + numberStylePattern = ures_getStringByKeyWithFallback(rb.getAlias(), gDecimalFormatTag, &ptnLen, &ec); } int32_t numberStylePatternLen = ptnLen; - const UChar* negNumberStylePattern = NULL; + const UChar* negNumberStylePattern = nullptr; int32_t negNumberStylePatternLen = 0; // TODO: Java // parse to check whether there is ";" separator in the numberStylePattern @@ -279,127 +315,127 @@ CurrencyPluralInfo::setupCurrencyPluralPattern(const Locale& loc, UErrorCode& st } } - ures_close(numElements); - ures_close(rb); - delete ns; - if (U_FAILURE(ec)) { + // If OOM occurred during the above code, then we want to report that back to the caller. + if (ec == U_MEMORY_ALLOCATION_ERROR) { + status = ec; + } return; } - UResourceBundle *currRb = ures_open(U_ICUDATA_CURR, loc.getName(), &ec); - UResourceBundle *currencyRes = ures_getByKeyWithFallback(currRb, gCurrUnitPtnTag, NULL, &ec); + LocalUResourceBundlePointer currRb(ures_open(U_ICUDATA_CURR, loc.getName(), &ec)); + LocalUResourceBundlePointer currencyRes(ures_getByKeyWithFallback(currRb.getAlias(), gCurrUnitPtnTag, nullptr, &ec)); #ifdef CURRENCY_PLURAL_INFO_DEBUG std::cout << "in set up\n"; #endif - StringEnumeration* keywords = fPluralRules->getKeywords(ec); + LocalPointer<StringEnumeration> keywords(fPluralRules->getKeywords(ec), ec); if (U_SUCCESS(ec)) { const char* pluralCount; - while ((pluralCount = keywords->next(NULL, ec)) != NULL) { - if ( U_SUCCESS(ec) ) { - int32_t ptnLen; - UErrorCode err = U_ZERO_ERROR; - const UChar* patternChars = ures_getStringByKeyWithFallback( - currencyRes, pluralCount, &ptnLen, &err); - if (U_SUCCESS(err) && ptnLen > 0) { - UnicodeString* pattern = new UnicodeString(patternChars, ptnLen); + while (((pluralCount = keywords->next(nullptr, ec)) != nullptr) && U_SUCCESS(ec)) { + int32_t ptnLength; + UErrorCode err = U_ZERO_ERROR; + const UChar* patternChars = ures_getStringByKeyWithFallback(currencyRes.getAlias(), pluralCount, &ptnLength, &err); + if (err == U_MEMORY_ALLOCATION_ERROR || patternChars == nullptr) { + ec = err; + break; + } + if (U_SUCCESS(err) && ptnLength > 0) { + UnicodeString* pattern = new UnicodeString(patternChars, ptnLength); + if (pattern == nullptr) { + ec = U_MEMORY_ALLOCATION_ERROR; + break; + } #ifdef CURRENCY_PLURAL_INFO_DEBUG - char result_1[1000]; - pattern->extract(0, pattern->length(), result_1, "UTF-8"); - std::cout << "pluralCount: " << pluralCount << "; pattern: " << result_1 << "\n"; + char result_1[1000]; + pattern->extract(0, pattern->length(), result_1, "UTF-8"); + std::cout << "pluralCount: " << pluralCount << "; pattern: " << result_1 << "\n"; #endif - pattern->findAndReplace(UnicodeString(TRUE, gPart0, 3), - UnicodeString(numberStylePattern, numberStylePatternLen)); - pattern->findAndReplace(UnicodeString(TRUE, gPart1, 3), UnicodeString(TRUE, gTripleCurrencySign, 3)); - - if (hasSeparator) { - UnicodeString negPattern(patternChars, ptnLen); - negPattern.findAndReplace(UnicodeString(TRUE, gPart0, 3), - UnicodeString(negNumberStylePattern, negNumberStylePatternLen)); - negPattern.findAndReplace(UnicodeString(TRUE, gPart1, 3), UnicodeString(TRUE, gTripleCurrencySign, 3)); - pattern->append(gNumberPatternSeparator); - pattern->append(negPattern); - } + pattern->findAndReplace(UnicodeString(TRUE, gPart0, 3), + UnicodeString(numberStylePattern, numberStylePatternLen)); + pattern->findAndReplace(UnicodeString(TRUE, gPart1, 3), UnicodeString(TRUE, gTripleCurrencySign, 3)); + + if (hasSeparator) { + UnicodeString negPattern(patternChars, ptnLength); + negPattern.findAndReplace(UnicodeString(TRUE, gPart0, 3), + UnicodeString(negNumberStylePattern, negNumberStylePatternLen)); + negPattern.findAndReplace(UnicodeString(TRUE, gPart1, 3), UnicodeString(TRUE, gTripleCurrencySign, 3)); + pattern->append(gNumberPatternSeparator); + pattern->append(negPattern); + } #ifdef CURRENCY_PLURAL_INFO_DEBUG - pattern->extract(0, pattern->length(), result_1, "UTF-8"); - std::cout << "pluralCount: " << pluralCount << "; pattern: " << result_1 << "\n"; + pattern->extract(0, pattern->length(), result_1, "UTF-8"); + std::cout << "pluralCount: " << pluralCount << "; pattern: " << result_1 << "\n"; #endif - - fPluralCountToCurrencyUnitPattern->put(UnicodeString(pluralCount, -1, US_INV), pattern, status); - } + // the 'pattern' object allocated above will be owned by the fPluralCountToCurrencyUnitPattern after the call to + // put(), even if the method returns failure. + fPluralCountToCurrencyUnitPattern->put(UnicodeString(pluralCount, -1, US_INV), pattern, status); } } } - delete keywords; - ures_close(currencyRes); - ures_close(currRb); + // If OOM occurred during the above code, then we want to report that back to the caller. + if (ec == U_MEMORY_ALLOCATION_ERROR) { + status = ec; + } } - - void -CurrencyPluralInfo::deleteHash(Hashtable* hTable) -{ - if ( hTable == NULL ) { +CurrencyPluralInfo::deleteHash(Hashtable* hTable) { + if ( hTable == nullptr ) { return; } int32_t pos = UHASH_FIRST; - const UHashElement* element = NULL; - while ( (element = hTable->nextElement(pos)) != NULL ) { + const UHashElement* element = nullptr; + while ( (element = hTable->nextElement(pos)) != nullptr ) { const UHashTok valueTok = element->value; const UnicodeString* value = (UnicodeString*)valueTok.pointer; delete value; } delete hTable; - hTable = NULL; + hTable = nullptr; } - Hashtable* CurrencyPluralInfo::initHash(UErrorCode& status) { - if ( U_FAILURE(status) ) { - return NULL; - } - Hashtable* hTable; - if ( (hTable = new Hashtable(TRUE, status)) == NULL ) { - status = U_MEMORY_ALLOCATION_ERROR; - return NULL; + if (U_FAILURE(status)) { + return nullptr; } - if ( U_FAILURE(status) ) { - delete hTable; - return NULL; + LocalPointer<Hashtable> hTable(new Hashtable(TRUE, status), status); + if (U_FAILURE(status)) { + return nullptr; } hTable->setValueComparator(ValueComparator); - return hTable; + return hTable.orphan(); } - void CurrencyPluralInfo::copyHash(const Hashtable* source, Hashtable* target, UErrorCode& status) { - if ( U_FAILURE(status) ) { + if (U_FAILURE(status)) { return; } int32_t pos = UHASH_FIRST; - const UHashElement* element = NULL; - if ( source ) { - while ( (element = source->nextElement(pos)) != NULL ) { + const UHashElement* element = nullptr; + if (source) { + while ( (element = source->nextElement(pos)) != nullptr ) { const UHashTok keyTok = element->key; const UnicodeString* key = (UnicodeString*)keyTok.pointer; const UHashTok valueTok = element->value; const UnicodeString* value = (UnicodeString*)valueTok.pointer; - UnicodeString* copy = new UnicodeString(*value); - target->put(UnicodeString(*key), copy, status); - if ( U_FAILURE(status) ) { + LocalPointer<UnicodeString> copy(new UnicodeString(*value), status); + if (U_FAILURE(status)) { + return; + } + // The HashTable owns the 'copy' object after the call to put(). + target->put(UnicodeString(*key), copy.orphan(), status); + if (U_FAILURE(status)) { return; } } } } - U_NAMESPACE_END #endif diff --git a/deps/icu-small/source/i18n/currunit.cpp b/deps/icu-small/source/i18n/currunit.cpp index 6d8d1cd6c6..2ece508751 100644 --- a/deps/icu-small/source/i18n/currunit.cpp +++ b/deps/icu-small/source/i18n/currunit.cpp @@ -27,9 +27,14 @@ CurrencyUnit::CurrencyUnit(ConstChar16Ptr _isoCode, UErrorCode& ec) { // The constructor always leaves the CurrencyUnit in a valid state (with a 3-character currency code). // Note: in ICU4J Currency.getInstance(), we check string length for 3, but in ICU4C we allow a // non-NUL-terminated string to be passed as an argument, so it is not possible to check length. + // However, we allow a NUL-terminated empty string, which should have the same behavior as nullptr. + // Consider NUL-terminated strings of length 1 or 2 as invalid. const char16_t* isoCodeToUse; - if (U_FAILURE(ec) || _isoCode == nullptr) { + if (U_FAILURE(ec) || _isoCode == nullptr || _isoCode[0] == 0) { isoCodeToUse = kDefaultCurrency; + } else if (_isoCode[1] == 0 || _isoCode[2] == 0) { + isoCodeToUse = kDefaultCurrency; + ec = U_ILLEGAL_ARGUMENT_ERROR; } else if (!uprv_isInvariantUString(_isoCode, 3)) { // TODO: Perform a more strict ASCII check like in ICU4J isAlpha3Code? isoCodeToUse = kDefaultCurrency; diff --git a/deps/icu-small/source/i18n/dcfmtsym.cpp b/deps/icu-small/source/i18n/dcfmtsym.cpp index 5a432aec8e..04113785f2 100644 --- a/deps/icu-small/source/i18n/dcfmtsym.cpp +++ b/deps/icu-small/source/i18n/dcfmtsym.cpp @@ -436,7 +436,7 @@ DecimalFormatSymbols::initialize(const Locale& loc, UErrorCode& status, sink.resolveMissingMonetarySeparators(fSymbols); // Resolve codePointZero - UChar32 tempCodePointZero; + UChar32 tempCodePointZero = -1; for (int32_t i=0; i<=9; i++) { const UnicodeString& stringDigit = getConstDigitSymbol(i); if (stringDigit.countChar32() != 1) { diff --git a/deps/icu-small/source/i18n/decimfmt.cpp b/deps/icu-small/source/i18n/decimfmt.cpp index a2638bb742..edd8910d9d 100644 --- a/deps/icu-small/source/i18n/decimfmt.cpp +++ b/deps/icu-small/source/i18n/decimfmt.cpp @@ -1057,12 +1057,19 @@ UBool DecimalFormat::areSignificantDigitsUsed() const { void DecimalFormat::setSignificantDigitsUsed(UBool useSignificantDigits) { // These are the default values from the old implementation. + if (useSignificantDigits) { + if (fields->properties->minimumSignificantDigits != -1 || + fields->properties->maximumSignificantDigits != -1) { + return; + } + } else { + if (fields->properties->minimumSignificantDigits == -1 && + fields->properties->maximumSignificantDigits == -1) { + return; + } + } int32_t minSig = useSignificantDigits ? 1 : -1; int32_t maxSig = useSignificantDigits ? 6 : -1; - if (fields->properties->minimumSignificantDigits == minSig && - fields->properties->maximumSignificantDigits == maxSig) { - return; - } fields->properties->minimumSignificantDigits = minSig; fields->properties->maximumSignificantDigits = maxSig; touchNoError(); @@ -1175,7 +1182,12 @@ void DecimalFormat::setPropertiesFromPattern(const UnicodeString& pattern, int32 } const numparse::impl::NumberParserImpl* DecimalFormat::getParser(UErrorCode& status) const { - if (U_FAILURE(status)) { return nullptr; } + // TODO: Move this into umutex.h? (similar logic also in numrange_fluent.cpp) + // See ICU-20146 + + if (U_FAILURE(status)) { + return nullptr; + } // First try to get the pre-computed parser auto* ptr = fields->atomicParser.load(); @@ -1185,13 +1197,17 @@ const numparse::impl::NumberParserImpl* DecimalFormat::getParser(UErrorCode& sta // Try computing the parser on our own auto* temp = NumberParserImpl::createParserFromProperties(*fields->properties, *fields->symbols, false, status); + if (U_FAILURE(status)) { + return nullptr; + } if (temp == nullptr) { status = U_MEMORY_ALLOCATION_ERROR; - // although we may still dereference, call sites should be guarded + return nullptr; } - // Note: ptr starts as nullptr; during compare_exchange, it is set to what is actually stored in the - // atomic if another thread beat us to computing the parser object. + // Note: ptr starts as nullptr; during compare_exchange, + // it is set to what is actually stored in the atomic + // if another thread beat us to computing the parser object. auto* nonConstThis = const_cast<DecimalFormat*>(this); if (!nonConstThis->fields->atomicParser.compare_exchange_strong(ptr, temp)) { // Another thread beat us to computing the parser diff --git a/deps/icu-small/source/i18n/dtfmtsym.cpp b/deps/icu-small/source/i18n/dtfmtsym.cpp index af421d41c7..c9dfa04583 100644 --- a/deps/icu-small/source/i18n/dtfmtsym.cpp +++ b/deps/icu-small/source/i18n/dtfmtsym.cpp @@ -1311,7 +1311,7 @@ DateFormatSymbols::initZoneStringsArray(void) { UDate now = Calendar::getNow(); UnicodeString tzDispName; - while ((tzid = tzids->snext(status))) { + while ((tzid = tzids->snext(status)) != 0) { if (U_FAILURE(status)) { break; } @@ -2224,8 +2224,8 @@ DateFormatSymbols::initializeData(const Locale& locale, const char *type, UError ++typeMapPtr; } if (typeMapPtr->usageTypeName != NULL && compResult == 0) { - fCapitalization[typeMapPtr->usageTypeEnumValue][0] = intVector[0]; - fCapitalization[typeMapPtr->usageTypeEnumValue][1] = intVector[1]; + fCapitalization[typeMapPtr->usageTypeEnumValue][0] = static_cast<UBool>(intVector[0]); + fCapitalization[typeMapPtr->usageTypeEnumValue][1] = static_cast<UBool>(intVector[1]); } } } diff --git a/deps/icu-small/source/i18n/dtitvfmt.cpp b/deps/icu-small/source/i18n/dtitvfmt.cpp index 743b534fc8..d952cbf509 100644 --- a/deps/icu-small/source/i18n/dtitvfmt.cpp +++ b/deps/icu-small/source/i18n/dtitvfmt.cpp @@ -877,8 +877,7 @@ DateIntervalFormat::getDateTimeSkeleton(const UnicodeString& skeleton, if ( MCount < 3 ) { normalizedDateSkeleton.append(CAP_M); } else { - int32_t i; - for ( i = 0; i < MCount && i < MAX_M_COUNT; ++i ) { + for ( int32_t j = 0; j < MCount && j < MAX_M_COUNT; ++j) { normalizedDateSkeleton.append(CAP_M); } } @@ -887,8 +886,7 @@ DateIntervalFormat::getDateTimeSkeleton(const UnicodeString& skeleton, if ( ECount <= 3 ) { normalizedDateSkeleton.append(CAP_E); } else { - int32_t i; - for ( i = 0; i < ECount && i < MAX_E_COUNT; ++i ) { + for ( int32_t j = 0; j < ECount && j < MAX_E_COUNT; ++j ) { normalizedDateSkeleton.append(CAP_E); } } diff --git a/deps/icu-small/source/i18n/dtitvinf.cpp b/deps/icu-small/source/i18n/dtitvinf.cpp index c863a683a5..a289fc79c8 100644 --- a/deps/icu-small/source/i18n/dtitvinf.cpp +++ b/deps/icu-small/source/i18n/dtitvinf.cpp @@ -594,7 +594,7 @@ DateIntervalInfo::getBestSkeleton(const UnicodeString& skeleton, const UHashElement* elem = NULL; while ( (elem = fIntervalPatterns->nextElement(pos)) != NULL ) { const UHashTok keyTok = elem->key; - UnicodeString* skeleton = (UnicodeString*)keyTok.pointer; + UnicodeString* newSkeleton = (UnicodeString*)keyTok.pointer; #ifdef DTITVINF_DEBUG skeleton->extract(0, skeleton->length(), result, "UTF-8"); sprintf(mesg, "available skeletons: skeleton: %s; \n", result); @@ -606,7 +606,7 @@ DateIntervalInfo::getBestSkeleton(const UnicodeString& skeleton, for ( i = 0; i < fieldLength; ++i ) { skeletonFieldWidth[i] = 0; } - parseSkeleton(*skeleton, skeletonFieldWidth); + parseSkeleton(*newSkeleton, skeletonFieldWidth); // calculate distance int32_t distance = 0; int8_t fieldDifference = 1; @@ -632,7 +632,7 @@ DateIntervalInfo::getBestSkeleton(const UnicodeString& skeleton, } } if ( distance < bestDistance ) { - bestSkeleton = skeleton; + bestSkeleton = newSkeleton; bestDistance = distance; bestMatchDistanceInfo = fieldDifference; } diff --git a/deps/icu-small/source/i18n/dtptngen.cpp b/deps/icu-small/source/i18n/dtptngen.cpp index aefd70464e..b44daf37bf 100644 --- a/deps/icu-small/source/i18n/dtptngen.cpp +++ b/deps/icu-small/source/i18n/dtptngen.cpp @@ -18,6 +18,7 @@ #include "unicode/decimfmt.h" #include "unicode/dtfmtsym.h" #include "unicode/dtptngen.h" +#include "unicode/localpointer.h" #include "unicode/simpleformatter.h" #include "unicode/smpdtfmt.h" #include "unicode/udat.h" @@ -88,17 +89,17 @@ static void ures_a_open(UResourceBundleAIterator *aiter, UResourceBundle *bund, aiter->num = ures_getSize(aiter->bund); aiter->cursor = 0; #if !defined(U_SORT_ASCII_BUNDLE_ITERATOR) - aiter->entries = NULL; + aiter->entries = nullptr; #else aiter->entries = (UResAEntry*)uprv_malloc(sizeof(UResAEntry)*aiter->num); for(int i=0;i<aiter->num;i++) { - aiter->entries[i].item = ures_getByIndex(aiter->bund, i, NULL, status); + aiter->entries[i].item = ures_getByIndex(aiter->bund, i, nullptr, status); const char *akey = ures_getKey(aiter->entries[i].item); int32_t len = uprv_strlen(akey)+1; aiter->entries[i].key = (UChar*)uprv_malloc(len*sizeof(UChar)); u_charsToUChars(akey, aiter->entries[i].key, len); } - uprv_sortArray(aiter->entries, aiter->num, sizeof(UResAEntry), ures_a_codepointSort, NULL, TRUE, status); + uprv_sortArray(aiter->entries, aiter->num, sizeof(UResAEntry), ures_a_codepointSort, nullptr, TRUE, status); #endif } @@ -115,7 +116,7 @@ static const UChar *ures_a_getNextString(UResourceBundleAIterator *aiter, int32_ #if !defined(U_SORT_ASCII_BUNDLE_ITERATOR) return ures_getNextString(aiter->bund, len, key, err); #else - if(U_FAILURE(*err)) return NULL; + if(U_FAILURE(*err)) return nullptr; UResourceBundle *item = aiter->entries[aiter->cursor].item; const UChar* ret = ures_getString(item, len, err); *key = ures_getKey(item); @@ -302,49 +303,48 @@ DateTimePatternGenerator::createInstance(UErrorCode& status) { DateTimePatternGenerator* U_EXPORT2 DateTimePatternGenerator::createInstance(const Locale& locale, UErrorCode& status) { if (U_FAILURE(status)) { - return NULL; + return nullptr; } LocalPointer<DateTimePatternGenerator> result( new DateTimePatternGenerator(locale, status), status); - return U_SUCCESS(status) ? result.orphan() : NULL; + return U_SUCCESS(status) ? result.orphan() : nullptr; } DateTimePatternGenerator* U_EXPORT2 DateTimePatternGenerator::createEmptyInstance(UErrorCode& status) { - DateTimePatternGenerator *result = new DateTimePatternGenerator(status); - if (result == NULL) { - status = U_MEMORY_ALLOCATION_ERROR; - } if (U_FAILURE(status)) { - delete result; - result = NULL; + return nullptr; } - return result; + LocalPointer<DateTimePatternGenerator> result( + new DateTimePatternGenerator(status), status); + return U_SUCCESS(status) ? result.orphan() : nullptr; } DateTimePatternGenerator::DateTimePatternGenerator(UErrorCode &status) : - skipMatcher(NULL), - fAvailableFormatKeyHash(NULL) + skipMatcher(nullptr), + fAvailableFormatKeyHash(nullptr), + internalErrorCode(U_ZERO_ERROR) { fp = new FormatParser(); dtMatcher = new DateTimeMatcher(); distanceInfo = new DistanceInfo(); patternMap = new PatternMap(); - if (fp == NULL || dtMatcher == NULL || distanceInfo == NULL || patternMap == NULL) { - status = U_MEMORY_ALLOCATION_ERROR; + if (fp == nullptr || dtMatcher == nullptr || distanceInfo == nullptr || patternMap == nullptr) { + internalErrorCode = status = U_MEMORY_ALLOCATION_ERROR; } } DateTimePatternGenerator::DateTimePatternGenerator(const Locale& locale, UErrorCode &status) : - skipMatcher(NULL), - fAvailableFormatKeyHash(NULL) + skipMatcher(nullptr), + fAvailableFormatKeyHash(nullptr), + internalErrorCode(U_ZERO_ERROR) { fp = new FormatParser(); dtMatcher = new DateTimeMatcher(); distanceInfo = new DistanceInfo(); patternMap = new PatternMap(); - if (fp == NULL || dtMatcher == NULL || distanceInfo == NULL || patternMap == NULL) { - status = U_MEMORY_ALLOCATION_ERROR; + if (fp == nullptr || dtMatcher == nullptr || distanceInfo == nullptr || patternMap == nullptr) { + internalErrorCode = status = U_MEMORY_ALLOCATION_ERROR; } else { initData(locale, status); @@ -353,13 +353,17 @@ DateTimePatternGenerator::DateTimePatternGenerator(const Locale& locale, UErrorC DateTimePatternGenerator::DateTimePatternGenerator(const DateTimePatternGenerator& other) : UObject(), - skipMatcher(NULL), - fAvailableFormatKeyHash(NULL) + skipMatcher(nullptr), + fAvailableFormatKeyHash(nullptr), + internalErrorCode(U_ZERO_ERROR) { fp = new FormatParser(); dtMatcher = new DateTimeMatcher(); distanceInfo = new DistanceInfo(); patternMap = new PatternMap(); + if (fp == nullptr || dtMatcher == nullptr || distanceInfo == nullptr || patternMap == nullptr) { + internalErrorCode = U_MEMORY_ALLOCATION_ERROR; + } *this=other; } @@ -369,6 +373,7 @@ DateTimePatternGenerator::operator=(const DateTimePatternGenerator& other) { if (&other == this) { return *this; } + internalErrorCode = other.internalErrorCode; pLocale = other.pLocale; fDefaultHourFormatChar = other.fDefaultHourFormatChar; *fp = *(other.fp); @@ -380,11 +385,16 @@ DateTimePatternGenerator::operator=(const DateTimePatternGenerator& other) { dateTimeFormat.getTerminatedBuffer(); decimal.getTerminatedBuffer(); delete skipMatcher; - if ( other.skipMatcher == NULL ) { - skipMatcher = NULL; + if ( other.skipMatcher == nullptr ) { + skipMatcher = nullptr; } else { skipMatcher = new DateTimeMatcher(*other.skipMatcher); + if (skipMatcher == nullptr) + { + internalErrorCode = U_MEMORY_ALLOCATION_ERROR; + return *this; + } } for (int32_t i=0; i< UDATPG_FIELD_COUNT; ++i ) { appendItemFormats[i] = other.appendItemFormats[i]; @@ -394,9 +404,8 @@ DateTimePatternGenerator::operator=(const DateTimePatternGenerator& other) { fieldDisplayNames[i][j].getTerminatedBuffer(); // NUL-terminate for the C API. } } - UErrorCode status = U_ZERO_ERROR; - patternMap->copyFrom(*other.patternMap, status); - copyHashtable(other.fAvailableFormatKeyHash, status); + patternMap->copyFrom(*other.patternMap, internalErrorCode); + copyHashtable(other.fAvailableFormatKeyHash, internalErrorCode); return *this; } @@ -431,21 +440,21 @@ DateTimePatternGenerator::operator!=(const DateTimePatternGenerator& other) cons } DateTimePatternGenerator::~DateTimePatternGenerator() { - if (fAvailableFormatKeyHash!=NULL) { + if (fAvailableFormatKeyHash!=nullptr) { delete fAvailableFormatKeyHash; } - if (fp != NULL) delete fp; - if (dtMatcher != NULL) delete dtMatcher; - if (distanceInfo != NULL) delete distanceInfo; - if (patternMap != NULL) delete patternMap; - if (skipMatcher != NULL) delete skipMatcher; + if (fp != nullptr) delete fp; + if (dtMatcher != nullptr) delete dtMatcher; + if (distanceInfo != nullptr) delete distanceInfo; + if (patternMap != nullptr) delete patternMap; + if (skipMatcher != nullptr) delete skipMatcher; } namespace { UInitOnce initOnce = U_INITONCE_INITIALIZER; -UHashtable *localeToAllowedHourFormatsMap = NULL; +UHashtable *localeToAllowedHourFormatsMap = nullptr; // Value deleter for hashmap. U_CFUNC void U_CALLCONV deleteAllowedHourFormats(void *ptr) { @@ -474,8 +483,8 @@ void DateTimePatternGenerator::initData(const Locale& locale, UErrorCode &status) { //const char *baseLangName = locale.getBaseName(); // unused - skipMatcher = NULL; - fAvailableFormatKeyHash=NULL; + skipMatcher = nullptr; + fAvailableFormatKeyHash=nullptr; addCanonicalItems(status); addICUPatterns(locale, status); addCLDRData(locale, status); @@ -483,6 +492,8 @@ DateTimePatternGenerator::initData(const Locale& locale, UErrorCode &status) { setDecimalSymbols(locale, status); umtx_initOnce(initOnce, loadAllowedHourFormatsData, status); getAllowedHourFormats(locale, status); + // If any of the above methods failed then the object is in an invalid state. + internalErrorCode = status; } // DateTimePatternGenerator::initData namespace { @@ -505,7 +516,7 @@ struct AllowedHourFormatsSink : public ResourceSink { LocalMemory<int32_t> list; int32_t length; if (value.getType() == URES_STRING) { - if (list.allocateInsteadAndReset(2) == NULL) { + if (list.allocateInsteadAndReset(2) == nullptr) { errorCode = U_MEMORY_ALLOCATION_ERROR; return; } @@ -515,7 +526,7 @@ struct AllowedHourFormatsSink : public ResourceSink { else { ResourceArray allowedFormats = value.getArray(errorCode); length = allowedFormats.getSize(); - if (list.allocateInsteadAndReset(length + 1) == NULL) { + if (list.allocateInsteadAndReset(length + 1) == nullptr) { errorCode = U_MEMORY_ALLOCATION_ERROR; return; } @@ -555,9 +566,14 @@ AllowedHourFormatsSink::~AllowedHourFormatsSink() {} U_CFUNC void U_CALLCONV DateTimePatternGenerator::loadAllowedHourFormatsData(UErrorCode &status) { if (U_FAILURE(status)) { return; } localeToAllowedHourFormatsMap = uhash_open( - uhash_hashChars, uhash_compareChars, NULL, &status); + uhash_hashChars, uhash_compareChars, nullptr, &status); + if (U_FAILURE(status)) { return; } + uhash_setValueDeleter(localeToAllowedHourFormatsMap, deleteAllowedHourFormats); - LocalUResourceBundlePointer rb(ures_openDirect(NULL, "supplementalData", &status)); + ucln_i18n_registerCleanup(UCLN_I18N_ALLOWED_HOUR_FORMATS, allowedHourFormatsCleanup); + + LocalUResourceBundlePointer rb(ures_openDirect(nullptr, "supplementalData", &status)); + if (U_FAILURE(status)) { return; } AllowedHourFormatsSink sink; // TODO: Currently in the enumeration each table allocates a new array. @@ -567,8 +583,6 @@ U_CFUNC void U_CALLCONV DateTimePatternGenerator::loadAllowedHourFormatsData(UEr // vector (at index enum*2) for easy data sharing, copy sub-arrays into runtime // object. Remember to clean up the vector, too. ures_getAllItemsWithFallback(rb.getAlias(), "timeData", sink, status); - - ucln_i18n_registerCleanup(UCLN_I18N_ALLOWED_HOUR_FORMATS, allowedHourFormatsCleanup); } void DateTimePatternGenerator::getAllowedHourFormats(const Locale &locale, UErrorCode &status) { @@ -589,17 +603,17 @@ void DateTimePatternGenerator::getAllowedHourFormats(const Locale &locale, UErro const char *language = maxLocale.getLanguage(); CharString langCountry; - langCountry.append(language, uprv_strlen(language), status); + langCountry.append(language, static_cast<int32_t>(uprv_strlen(language)), status); langCountry.append('_', status); - langCountry.append(country, uprv_strlen(country), status); + langCountry.append(country, static_cast<int32_t>(uprv_strlen(country)), status); int32_t *allowedFormats; allowedFormats = (int32_t *)uhash_get(localeToAllowedHourFormatsMap, langCountry.data()); - if (allowedFormats == NULL) { + if (allowedFormats == nullptr) { allowedFormats = (int32_t *)uhash_get(localeToAllowedHourFormatsMap, const_cast<char *>(country)); } - if (allowedFormats != NULL) { // Lookup is successful + if (allowedFormats != nullptr) { // Lookup is successful for (int32_t i = 0; i < UPRV_LENGTHOF(fAllowedHourFormats); ++i) { fAllowedHourFormats[i] = allowedFormats[i]; if (allowedFormats[i] == ALLOWED_HOUR_FORMAT_UNKNOWN) { @@ -615,10 +629,10 @@ void DateTimePatternGenerator::getAllowedHourFormats(const Locale &locale, UErro UnicodeString DateTimePatternGenerator::getSkeleton(const UnicodeString& pattern, UErrorCode& /*status*/) { - FormatParser fp; + FormatParser fp2; DateTimeMatcher matcher; PtnSkeleton localSkeleton; - matcher.set(pattern, &fp, localSkeleton); + matcher.set(pattern, &fp2, localSkeleton); return localSkeleton.getSkeleton(); } @@ -634,10 +648,10 @@ DateTimePatternGenerator::staticGetSkeleton( UnicodeString DateTimePatternGenerator::getBaseSkeleton(const UnicodeString& pattern, UErrorCode& /*status*/) { - FormatParser fp; + FormatParser fp2; DateTimeMatcher matcher; PtnSkeleton localSkeleton; - matcher.set(pattern, &fp, localSkeleton); + matcher.set(pattern, &fp2, localSkeleton); return localSkeleton.getBaseSkeleton(); } @@ -663,7 +677,7 @@ DateTimePatternGenerator::addICUPatterns(const Locale& locale, UErrorCode& statu DateFormat::EStyle style = (DateFormat::EStyle)i; df = DateFormat::createDateInstance(style, locale); SimpleDateFormat* sdf; - if (df != NULL && (sdf = dynamic_cast<SimpleDateFormat*>(df)) != NULL) { + if (df != nullptr && (sdf = dynamic_cast<SimpleDateFormat*>(df)) != nullptr) { sdf->toPattern(dfPattern); addPattern(dfPattern, FALSE, conflictingString, status); } @@ -672,7 +686,7 @@ DateTimePatternGenerator::addICUPatterns(const Locale& locale, UErrorCode& statu if (U_FAILURE(status)) { return; } df = DateFormat::createTimeInstance(style, locale); - if (df != NULL && (sdf = dynamic_cast<SimpleDateFormat*>(df)) != NULL) { + if (df != nullptr && (sdf = dynamic_cast<SimpleDateFormat*>(df)) != nullptr) { sdf->toPattern(dfPattern); addPattern(dfPattern, FALSE, conflictingString, status); @@ -747,13 +761,14 @@ DateTimePatternGenerator::getCalendarTypeToUse(const Locale& locale, CharString& ures_getFunctionalEquivalent( localeWithCalendarKey, ULOC_LOCALE_IDENTIFIER_CAPACITY, - NULL, + nullptr, "calendar", "calendar", locale.getName(), - NULL, + nullptr, FALSE, &err); + if (U_FAILURE(err)) { return; } localeWithCalendarKey[ULOC_LOCALE_IDENTIFIER_CAPACITY-1] = 0; // ensure null termination // now get the calendar key value from that locale char calendarType[ULOC_KEYWORDS_CAPACITY]; @@ -763,7 +778,8 @@ DateTimePatternGenerator::getCalendarTypeToUse(const Locale& locale, CharString& calendarType, ULOC_KEYWORDS_CAPACITY, &err); - if (U_SUCCESS(err) && calendarTypeLen < ULOC_KEYWORDS_CAPACITY) { + if (U_FAILURE(err)) { return; } + if (calendarTypeLen < ULOC_KEYWORDS_CAPACITY) { destination.clear().append(calendarType, -1, err); if (U_FAILURE(err)) { return; } } @@ -774,7 +790,7 @@ DateTimePatternGenerator::getCalendarTypeToUse(const Locale& locale, CharString& void DateTimePatternGenerator::consumeShortTimePattern(const UnicodeString& shortTimePattern, UErrorCode& status) { - + if (U_FAILURE(status)) { return; } // set fDefaultHourFormatChar to the hour format character from this pattern int32_t tfIdx, tfLen = shortTimePattern.length(); UBool ignoreChars = FALSE; @@ -782,7 +798,7 @@ DateTimePatternGenerator::consumeShortTimePattern(const UnicodeString& shortTime UChar tfChar = shortTimePattern.charAt(tfIdx); if ( tfChar == SINGLE_QUOTE ) { ignoreChars = !ignoreChars; // toggle (handle quoted literals & '' for single quote) - } else if ( !ignoreChars && u_strchr(hourFormatChars, tfChar) != NULL ) { + } else if ( !ignoreChars && u_strchr(hourFormatChars, tfChar) != nullptr ) { fDefaultHourFormatChar = tfChar; break; } @@ -872,9 +888,9 @@ struct DateTimePatternGenerator::AppendItemNamesSink : public ResourceSink { valueStr.getTerminatedBuffer(); } for (int32_t j = 1; j < UDATPG_WIDTH_COUNT; j++) { - UnicodeString& valueStr = dtpg.getMutableFieldDisplayName((UDateTimePatternField)i, (UDateTimePGDisplayWidth)j); - if (valueStr.isEmpty()) { - valueStr = dtpg.getFieldDisplayName((UDateTimePatternField)i, (UDateTimePGDisplayWidth)(j-1)); + UnicodeString& valueStr2 = dtpg.getMutableFieldDisplayName((UDateTimePatternField)i, (UDateTimePGDisplayWidth)j); + if (valueStr2.isEmpty()) { + valueStr2 = dtpg.getFieldDisplayName((UDateTimePatternField)i, (UDateTimePGDisplayWidth)(j-1)); } } } @@ -921,7 +937,7 @@ DateTimePatternGenerator::addCLDRData(const Locale& locale, UErrorCode& errorCod UnicodeString rbPattern, value, field; CharString path; - LocalUResourceBundlePointer rb(ures_open(NULL, locale.getName(), &errorCode)); + LocalUResourceBundlePointer rb(ures_open(nullptr, locale.getName(), &errorCode)); if (U_FAILURE(errorCode)) { return; } CharString calendarTypeToUse; // to be filled in with the type to use, if all goes well @@ -966,12 +982,13 @@ DateTimePatternGenerator::addCLDRData(const Locale& locale, UErrorCode& errorCod void DateTimePatternGenerator::initHashtable(UErrorCode& err) { - if (fAvailableFormatKeyHash!=NULL) { + if (U_FAILURE(err)) { return; } + if (fAvailableFormatKeyHash!=nullptr) { return; } - if ((fAvailableFormatKeyHash = new Hashtable(FALSE, err))==NULL) { - err=U_MEMORY_ALLOCATION_ERROR; - return; + LocalPointer<Hashtable> hash(new Hashtable(FALSE, err), err); + if (U_SUCCESS(err)) { + fAvailableFormatKeyHash = hash.orphan(); } } @@ -1028,7 +1045,14 @@ DateTimePatternGenerator::getBestPattern(const UnicodeString& patternForm, UErro UnicodeString DateTimePatternGenerator::getBestPattern(const UnicodeString& patternForm, UDateTimePatternMatchOptions options, UErrorCode& status) { - const UnicodeString *bestPattern=NULL; + if (U_FAILURE(status)) { + return UnicodeString(); + } + if (U_FAILURE(internalErrorCode)) { + status = internalErrorCode; + return UnicodeString(); + } + const UnicodeString *bestPattern = nullptr; UnicodeString dtFormat; UnicodeString resultPattern; int32_t flags = kDTPGNoFlags; @@ -1044,16 +1068,23 @@ DateTimePatternGenerator::getBestPattern(const UnicodeString& patternForm, UDate resultPattern.remove(); dtMatcher->set(patternFormMapped, fp); - const PtnSkeleton* specifiedSkeleton=NULL; - bestPattern=getBestRaw(*dtMatcher, -1, distanceInfo, &specifiedSkeleton); + const PtnSkeleton* specifiedSkeleton = nullptr; + bestPattern=getBestRaw(*dtMatcher, -1, distanceInfo, status, &specifiedSkeleton); + if (U_FAILURE(status)) { + return UnicodeString(); + } + if ( distanceInfo->missingFieldMask==0 && distanceInfo->extraFieldMask==0 ) { resultPattern = adjustFieldTypes(*bestPattern, specifiedSkeleton, flags, options); return resultPattern; } int32_t neededFields = dtMatcher->getFieldMask(); - UnicodeString datePattern=getBestAppending(neededFields & dateMask, flags, options); - UnicodeString timePattern=getBestAppending(neededFields & timeMask, flags, options); + UnicodeString datePattern=getBestAppending(neededFields & dateMask, flags, status, options); + UnicodeString timePattern=getBestAppending(neededFields & timeMask, flags, status, options); + if (U_FAILURE(status)) { + return UnicodeString(); + } if (datePattern.length()==0) { if (timePattern.length()==0) { resultPattern.remove(); @@ -1074,7 +1105,7 @@ DateTimePatternGenerator::getBestPattern(const UnicodeString& patternForm, UDate /* * Map a skeleton that may have metacharacters jJC to one without, by replacing - * the metacharacters with locale-appropriate fields of of h/H/k/K and of a/b/B + * the metacharacters with locale-appropriate fields of h/H/k/K and of a/b/B * (depends on fDefaultHourFormatChar and fAllowedHourFormats being set, which in * turn depends on initData having been run). This method also updates the flags * as necessary. Returns the updated skeleton. @@ -1159,9 +1190,16 @@ UnicodeString DateTimePatternGenerator::replaceFieldTypes(const UnicodeString& pattern, const UnicodeString& skeleton, UDateTimePatternMatchOptions options, - UErrorCode& /*status*/) { + UErrorCode& status) { + if (U_FAILURE(status)) { + return UnicodeString(); + } + if (U_FAILURE(internalErrorCode)) { + status = internalErrorCode; + return UnicodeString(); + } dtMatcher->set(skeleton, fp); - UnicodeString result = adjustFieldTypes(pattern, NULL, kDTPGNoFlags, options); + UnicodeString result = adjustFieldTypes(pattern, nullptr, kDTPGNoFlags, options); return result; } @@ -1204,20 +1242,24 @@ DateTimePatternGenerator::getDateTimeFormat() const { void DateTimePatternGenerator::setDateTimeFromCalendar(const Locale& locale, UErrorCode& status) { + if (U_FAILURE(status)) { return; } + const UChar *resStr; int32_t resStrLen = 0; - Calendar* fCalendar = Calendar::createInstance(locale, status); + LocalPointer<Calendar> fCalendar(Calendar::createInstance(locale, status), status); if (U_FAILURE(status)) { return; } - LocalUResourceBundlePointer calData(ures_open(NULL, locale.getBaseName(), &status)); + LocalUResourceBundlePointer calData(ures_open(nullptr, locale.getBaseName(), &status)); + if (U_FAILURE(status)) { return; } ures_getByKey(calData.getAlias(), DT_DateTimeCalendarTag, calData.getAlias(), &status); + if (U_FAILURE(status)) { return; } LocalUResourceBundlePointer dateTimePatterns; - if (fCalendar != NULL && fCalendar->getType() != NULL && *fCalendar->getType() != '\0' + if (fCalendar->getType() != nullptr && *fCalendar->getType() != '\0' && uprv_strcmp(fCalendar->getType(), DT_DateTimeGregorianTag) != 0) { dateTimePatterns.adoptInstead(ures_getByKeyWithFallback(calData.getAlias(), fCalendar->getType(), - NULL, &status)); + nullptr, &status)); ures_getByKeyWithFallback(dateTimePatterns.getAlias(), DT_DateTimePatternsTag, dateTimePatterns.getAlias(), &status); } @@ -1238,8 +1280,6 @@ DateTimePatternGenerator::setDateTimeFromCalendar(const Locale& locale, UErrorCo } resStr = ures_getStringByIndex(dateTimePatterns.getAlias(), (int32_t)DateFormat::kDateTime, &resStrLen, &status); setDateTimeFormat(UnicodeString(TRUE, resStr, resStrLen)); - - delete fCalendar; } void @@ -1259,7 +1299,12 @@ DateTimePatternGenerator::addPattern( UnicodeString &conflictingPattern, UErrorCode& status) { - return addPatternWithSkeleton(pattern, NULL, override, conflictingPattern, status); + if (U_FAILURE(internalErrorCode)) { + status = internalErrorCode; + return UDATPG_NO_CONFLICT; + } + + return addPatternWithSkeleton(pattern, nullptr, override, conflictingPattern, status); } // For DateTimePatternGenerator::addPatternWithSkeleton - @@ -1280,13 +1325,17 @@ DateTimePatternGenerator::addPatternWithSkeleton( UnicodeString& conflictingPattern, UErrorCode& status) { + if (U_FAILURE(internalErrorCode)) { + status = internalErrorCode; + return UDATPG_NO_CONFLICT; + } UnicodeString basePattern; PtnSkeleton skeleton; UDateTimePatternConflict conflictingStatus = UDATPG_NO_CONFLICT; DateTimeMatcher matcher; - if ( skeletonToUse == NULL ) { + if ( skeletonToUse == nullptr ) { matcher.set(pattern, fp, skeleton); matcher.getBasePattern(basePattern); } else { @@ -1302,7 +1351,7 @@ DateTimePatternGenerator::addPatternWithSkeleton( // availableFormats items from root, which should not override any previous entry with the same base. UBool entryHadSpecifiedSkeleton; const UnicodeString *duplicatePattern = patternMap->getPatternFromBasePattern(basePattern, entryHadSpecifiedSkeleton); - if (duplicatePattern != NULL && (!entryHadSpecifiedSkeleton || (skeletonToUse != NULL && !override))) { + if (duplicatePattern != nullptr && (!entryHadSpecifiedSkeleton || (skeletonToUse != nullptr && !override))) { conflictingStatus = UDATPG_BASE_CONFLICT; conflictingPattern = *duplicatePattern; if (!override) { @@ -1313,16 +1362,16 @@ DateTimePatternGenerator::addPatternWithSkeleton( // items from CLDR data. In that case, we don't want an item from a parent locale to replace an item with // same skeleton from the specified locale, so skip the current item if skeletonWasSpecified is true for // the previously-specified conflicting item. - const PtnSkeleton* entrySpecifiedSkeleton = NULL; + const PtnSkeleton* entrySpecifiedSkeleton = nullptr; duplicatePattern = patternMap->getPatternFromSkeleton(skeleton, &entrySpecifiedSkeleton); - if (duplicatePattern != NULL ) { + if (duplicatePattern != nullptr ) { conflictingStatus = UDATPG_CONFLICT; conflictingPattern = *duplicatePattern; - if (!override || (skeletonToUse != NULL && entrySpecifiedSkeleton != NULL)) { + if (!override || (skeletonToUse != nullptr && entrySpecifiedSkeleton != nullptr)) { return conflictingStatus; } } - patternMap->add(basePattern, skeleton, pattern, skeletonToUse != NULL, status); + patternMap->add(basePattern, skeleton, pattern, skeletonToUse != nullptr, status); if(U_FAILURE(status)) { return conflictingStatus; } @@ -1369,13 +1418,16 @@ const UnicodeString* DateTimePatternGenerator::getBestRaw(DateTimeMatcher& source, int32_t includeMask, DistanceInfo* missingFields, + UErrorCode &status, const PtnSkeleton** specifiedSkeletonPtr) { int32_t bestDistance = 0x7fffffff; DistanceInfo tempInfo; - const UnicodeString *bestPattern=NULL; - const PtnSkeleton* specifiedSkeleton=NULL; + const UnicodeString *bestPattern=nullptr; + const PtnSkeleton* specifiedSkeleton=nullptr; + + PatternMapIterator it(status); + if (U_FAILURE(status)) { return nullptr; } - PatternMapIterator it; for (it.set(*patternMap); it.hasNext(); ) { DateTimeMatcher trial = it.next(); if (trial.equals(skipMatcher)) { @@ -1485,8 +1537,8 @@ DateTimePatternGenerator::adjustFieldTypes(const UnicodeString& pattern, c = fDefaultHourFormatChar; } field.remove(); - for (int32_t i=adjFieldLen; i>0; --i) { - field+=c; + for (int32_t j=adjFieldLen; j>0; --j) { + field += c; } } newPattern+=field; @@ -1496,14 +1548,21 @@ DateTimePatternGenerator::adjustFieldTypes(const UnicodeString& pattern, } UnicodeString -DateTimePatternGenerator::getBestAppending(int32_t missingFields, int32_t flags, UDateTimePatternMatchOptions options) { +DateTimePatternGenerator::getBestAppending(int32_t missingFields, int32_t flags, UErrorCode &status, UDateTimePatternMatchOptions options) { + if (U_FAILURE(status)) { + return UnicodeString(); + } UnicodeString resultPattern, tempPattern; - UErrorCode err=U_ZERO_ERROR; + const UnicodeString* tempPatternPtr; int32_t lastMissingFieldMask=0; if (missingFields!=0) { resultPattern=UnicodeString(); - const PtnSkeleton* specifiedSkeleton=NULL; - tempPattern = *getBestRaw(*dtMatcher, missingFields, distanceInfo, &specifiedSkeleton); + const PtnSkeleton* specifiedSkeleton=nullptr; + tempPatternPtr = getBestRaw(*dtMatcher, missingFields, distanceInfo, status, &specifiedSkeleton); + if (U_FAILURE(status)) { + return UnicodeString(); + } + tempPattern = *tempPatternPtr; resultPattern = adjustFieldTypes(tempPattern, specifiedSkeleton, flags, options); if ( distanceInfo->missingFieldMask==0 ) { return resultPattern; @@ -1519,19 +1578,26 @@ DateTimePatternGenerator::getBestAppending(int32_t missingFields, int32_t flags, continue; } int32_t startingMask = distanceInfo->missingFieldMask; - tempPattern = *getBestRaw(*dtMatcher, distanceInfo->missingFieldMask, distanceInfo, &specifiedSkeleton); + tempPatternPtr = getBestRaw(*dtMatcher, distanceInfo->missingFieldMask, distanceInfo, status, &specifiedSkeleton); + if (U_FAILURE(status)) { + return UnicodeString(); + } + tempPattern = *tempPatternPtr; tempPattern = adjustFieldTypes(tempPattern, specifiedSkeleton, flags, options); int32_t foundMask=startingMask& ~distanceInfo->missingFieldMask; int32_t topField=getTopBitNumber(foundMask); - UnicodeString appendName; - getAppendName((UDateTimePatternField)topField, appendName); - const UnicodeString *values[3] = { - &resultPattern, - &tempPattern, - &appendName - }; - SimpleFormatter(appendItemFormats[topField], 2, 3, err). - formatAndReplace(values, 3, resultPattern, NULL, 0, err); + + if (appendItemFormats[topField].length() != 0) { + UnicodeString appendName; + getAppendName((UDateTimePatternField)topField, appendName); + const UnicodeString *values[3] = { + &resultPattern, + &tempPattern, + &appendName + }; + SimpleFormatter(appendItemFormats[topField], 2, 3, status). + formatAndReplace(values, 3, resultPattern, nullptr, 0, status); + } lastMissingFieldMask = distanceInfo->missingFieldMask; } } @@ -1539,7 +1605,7 @@ DateTimePatternGenerator::getBestAppending(int32_t missingFields, int32_t flags, } int32_t -DateTimePatternGenerator::getTopBitNumber(int32_t foundMask) { +DateTimePatternGenerator::getTopBitNumber(int32_t foundMask) const { if ( foundMask==0 ) { return 0; } @@ -1568,22 +1634,21 @@ DateTimePatternGenerator::isAvailableFormatSet(const UnicodeString &key) const { void DateTimePatternGenerator::copyHashtable(Hashtable *other, UErrorCode &status) { - - if (other == NULL) { + if (other == nullptr || U_FAILURE(status)) { return; } - if (fAvailableFormatKeyHash != NULL) { + if (fAvailableFormatKeyHash != nullptr) { delete fAvailableFormatKeyHash; - fAvailableFormatKeyHash = NULL; + fAvailableFormatKeyHash = nullptr; } initHashtable(status); if(U_FAILURE(status)){ return; } int32_t pos = UHASH_FIRST; - const UHashElement* elem = NULL; + const UHashElement* elem = nullptr; // walk through the hash table and create a deep clone - while((elem = other->nextElement(pos))!= NULL){ + while((elem = other->nextElement(pos))!= nullptr){ const UHashTok otherKeyTok = elem->key; UnicodeString* otherKey = (UnicodeString*)otherKeyTok.pointer; fAvailableFormatKeyHash->puti(*otherKey, 1, status); @@ -1595,8 +1660,17 @@ DateTimePatternGenerator::copyHashtable(Hashtable *other, UErrorCode &status) { StringEnumeration* DateTimePatternGenerator::getSkeletons(UErrorCode& status) const { - StringEnumeration* skeletonEnumerator = new DTSkeletonEnumeration(*patternMap, DT_SKELETON, status); - return skeletonEnumerator; + if (U_FAILURE(status)) { + return nullptr; + } + if (U_FAILURE(internalErrorCode)) { + status = internalErrorCode; + return nullptr; + } + LocalPointer<StringEnumeration> skeletonEnumerator( + new DTSkeletonEnumeration(*patternMap, DT_SKELETON, status), status); + + return U_SUCCESS(status) ? skeletonEnumerator.orphan() : nullptr; } const UnicodeString& @@ -1607,47 +1681,70 @@ DateTimePatternGenerator::getPatternForSkeleton(const UnicodeString& skeleton) c return emptyString; } curElem = patternMap->getHeader(skeleton.charAt(0)); - while ( curElem != NULL ) { + while ( curElem != nullptr ) { if ( curElem->skeleton->getSkeleton()==skeleton ) { return curElem->pattern; } - curElem=curElem->next; + curElem = curElem->next.getAlias(); } return emptyString; } StringEnumeration* DateTimePatternGenerator::getBaseSkeletons(UErrorCode& status) const { - StringEnumeration* baseSkeletonEnumerator = new DTSkeletonEnumeration(*patternMap, DT_BASESKELETON, status); - return baseSkeletonEnumerator; + if (U_FAILURE(status)) { + return nullptr; + } + if (U_FAILURE(internalErrorCode)) { + status = internalErrorCode; + return nullptr; + } + LocalPointer<StringEnumeration> baseSkeletonEnumerator( + new DTSkeletonEnumeration(*patternMap, DT_BASESKELETON, status), status); + + return U_SUCCESS(status) ? baseSkeletonEnumerator.orphan() : nullptr; } StringEnumeration* DateTimePatternGenerator::getRedundants(UErrorCode& status) { - StringEnumeration* output = new DTRedundantEnumeration(); + if (U_FAILURE(status)) { return nullptr; } + if (U_FAILURE(internalErrorCode)) { + status = internalErrorCode; + return nullptr; + } + LocalPointer<StringEnumeration> output(new DTRedundantEnumeration(), status); + if (U_FAILURE(status)) { return nullptr; } const UnicodeString *pattern; - PatternMapIterator it; + PatternMapIterator it(status); + if (U_FAILURE(status)) { return nullptr; } + for (it.set(*patternMap); it.hasNext(); ) { DateTimeMatcher current = it.next(); pattern = patternMap->getPatternFromSkeleton(*(it.getSkeleton())); if ( isCanonicalItem(*pattern) ) { continue; } - if ( skipMatcher == NULL ) { + if ( skipMatcher == nullptr ) { skipMatcher = new DateTimeMatcher(current); + if (skipMatcher == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + return nullptr; + } } else { *skipMatcher = current; } UnicodeString trial = getBestPattern(current.getPattern(), status); + if (U_FAILURE(status)) { return nullptr; } if (trial == *pattern) { - ((DTRedundantEnumeration *)output)->add(*pattern, status); + ((DTRedundantEnumeration *)output.getAlias())->add(*pattern, status); + if (U_FAILURE(status)) { return nullptr; } } if (current.equals(skipMatcher)) { continue; } } - return output; + return output.orphan(); } UBool @@ -1671,45 +1768,54 @@ DateTimePatternGenerator::clone() const { PatternMap::PatternMap() { for (int32_t i=0; i < MAX_PATTERN_ENTRIES; ++i ) { - boot[i]=NULL; + boot[i] = nullptr; } isDupAllowed = TRUE; } void PatternMap::copyFrom(const PatternMap& other, UErrorCode& status) { + if (U_FAILURE(status)) { + return; + } this->isDupAllowed = other.isDupAllowed; - for (int32_t bootIndex=0; bootIndex<MAX_PATTERN_ENTRIES; ++bootIndex ) { - PtnElem *curElem, *otherElem, *prevElem=NULL; + for (int32_t bootIndex = 0; bootIndex < MAX_PATTERN_ENTRIES; ++bootIndex) { + PtnElem *curElem, *otherElem, *prevElem=nullptr; otherElem = other.boot[bootIndex]; - while (otherElem!=NULL) { - if ((curElem = new PtnElem(otherElem->basePattern, otherElem->pattern))==NULL) { - // out of memory - status = U_MEMORY_ALLOCATION_ERROR; - return; + while (otherElem != nullptr) { + LocalPointer<PtnElem> newElem(new PtnElem(otherElem->basePattern, otherElem->pattern), status); + if (U_FAILURE(status)) { + return; // out of memory } - if ( this->boot[bootIndex]== NULL ) { - this->boot[bootIndex] = curElem; + newElem->skeleton.adoptInsteadAndCheckErrorCode(new PtnSkeleton(*(otherElem->skeleton)), status); + if (U_FAILURE(status)) { + return; // out of memory } - if ((curElem->skeleton=new PtnSkeleton(*(otherElem->skeleton))) == NULL ) { - // out of memory - status = U_MEMORY_ALLOCATION_ERROR; - return; - } - curElem->skeletonWasSpecified = otherElem->skeletonWasSpecified; - if (prevElem!=NULL) { - prevElem->next=curElem; + newElem->skeletonWasSpecified = otherElem->skeletonWasSpecified; + + // Release ownership from the LocalPointer of the PtnElem object. + // The PtnElem will now be owned by either the boot (for the first entry in the linked-list) + // or owned by the previous PtnElem object in the linked-list. + curElem = newElem.orphan(); + + if (this->boot[bootIndex] == nullptr) { + this->boot[bootIndex] = curElem; + } else { + if (prevElem != nullptr) { + prevElem->next.adoptInstead(curElem); + } else { + U_ASSERT(false); + } } - curElem->next=NULL; prevElem = curElem; - otherElem = otherElem->next; + otherElem = otherElem->next.getAlias(); } } } PtnElem* -PatternMap::getHeader(UChar baseChar) { +PatternMap::getHeader(UChar baseChar) const { PtnElem* curElem; if ( (baseChar >= CAP_A) && (baseChar <= CAP_Z) ) { @@ -1720,7 +1826,7 @@ PatternMap::getHeader(UChar baseChar) { curElem = boot[26+baseChar-LOW_A]; } else { - return NULL; + return nullptr; } } return curElem; @@ -1728,9 +1834,9 @@ PatternMap::getHeader(UChar baseChar) { PatternMap::~PatternMap() { for (int32_t i=0; i < MAX_PATTERN_ENTRIES; ++i ) { - if (boot[i]!=NULL ) { + if (boot[i] != nullptr ) { delete boot[i]; - boot[i]=NULL; + boot[i] = nullptr; } } } // PatternMap destructor @@ -1759,39 +1865,45 @@ PatternMap::add(const UnicodeString& basePattern, } } - if (baseElem == NULL) { - if ((curElem = new PtnElem(basePattern, value)) == NULL ) { - // out of memory - status = U_MEMORY_ALLOCATION_ERROR; - return; + if (baseElem == nullptr) { + LocalPointer<PtnElem> newElem(new PtnElem(basePattern, value), status); + if (U_FAILURE(status)) { + return; // out of memory } + newElem->skeleton.adoptInsteadAndCheckErrorCode(new PtnSkeleton(skeleton), status); + if (U_FAILURE(status)) { + return; // out of memory + } + newElem->skeletonWasSpecified = skeletonWasSpecified; if (baseChar >= LOW_A) { - boot[26 + (baseChar-LOW_A)] = curElem; + boot[26 + (baseChar - LOW_A)] = newElem.orphan(); // the boot array now owns the PtnElem. } else { - boot[baseChar-CAP_A] = curElem; + boot[baseChar - CAP_A] = newElem.orphan(); // the boot array now owns the PtnElem. } - curElem->skeleton = new PtnSkeleton(skeleton); - curElem->skeletonWasSpecified = skeletonWasSpecified; } - if ( baseElem != NULL ) { + if ( baseElem != nullptr ) { curElem = getDuplicateElem(basePattern, skeleton, baseElem); - if (curElem == NULL) { + if (curElem == nullptr) { // add new element to the list. curElem = baseElem; - while( curElem -> next != NULL ) + while( curElem -> next != nullptr ) { - curElem = curElem->next; + curElem = curElem->next.getAlias(); } - if ((curElem->next = new PtnElem(basePattern, value)) == NULL ) { - // out of memory - status = U_MEMORY_ALLOCATION_ERROR; - return; + + LocalPointer<PtnElem> newElem(new PtnElem(basePattern, value), status); + if (U_FAILURE(status)) { + return; // out of memory } - curElem=curElem->next; - curElem->skeleton = new PtnSkeleton(skeleton); - curElem->skeletonWasSpecified = skeletonWasSpecified; + newElem->skeleton.adoptInsteadAndCheckErrorCode(new PtnSkeleton(skeleton), status); + if (U_FAILURE(status)) { + return; // out of memory + } + newElem->skeletonWasSpecified = skeletonWasSpecified; + curElem->next.adoptInstead(newElem.orphan()); + curElem = curElem->next.getAlias(); } else { // Pattern exists in the list already. @@ -1809,11 +1921,11 @@ PatternMap::add(const UnicodeString& basePattern, // Find the pattern from the given basePattern string. const UnicodeString * -PatternMap::getPatternFromBasePattern(UnicodeString& basePattern, UBool& skeletonWasSpecified) { // key to search for +PatternMap::getPatternFromBasePattern(const UnicodeString& basePattern, UBool& skeletonWasSpecified) const { // key to search for PtnElem *curElem; - if ((curElem=getHeader(basePattern.charAt(0)))==NULL) { - return NULL; // no match + if ((curElem=getHeader(basePattern.charAt(0)))==nullptr) { + return nullptr; // no match } do { @@ -1821,10 +1933,10 @@ PatternMap::getPatternFromBasePattern(UnicodeString& basePattern, UBool& skeleto skeletonWasSpecified = curElem->skeletonWasSpecified; return &(curElem->pattern); } - curElem=curElem->next; - }while (curElem != NULL); + curElem = curElem->next.getAlias(); + } while (curElem != nullptr); - return NULL; + return nullptr; } // PatternMap::getFromBasePattern @@ -1835,69 +1947,69 @@ PatternMap::getPatternFromBasePattern(UnicodeString& basePattern, UBool& skeleto // optimum distance value in getBestRaw. When this is called from public getRedundants (specifiedSkeletonPtr is NULL), // for now it will continue to compare based on baseOriginal so as not to change the behavior unnecessarily. const UnicodeString * -PatternMap::getPatternFromSkeleton(PtnSkeleton& skeleton, const PtnSkeleton** specifiedSkeletonPtr) { // key to search for +PatternMap::getPatternFromSkeleton(const PtnSkeleton& skeleton, const PtnSkeleton** specifiedSkeletonPtr) const { // key to search for PtnElem *curElem; if (specifiedSkeletonPtr) { - *specifiedSkeletonPtr = NULL; + *specifiedSkeletonPtr = nullptr; } // find boot entry UChar baseChar = skeleton.getFirstChar(); - if ((curElem=getHeader(baseChar))==NULL) { - return NULL; // no match + if ((curElem=getHeader(baseChar))==nullptr) { + return nullptr; // no match } do { UBool equal; - if (specifiedSkeletonPtr != NULL) { // called from DateTimePatternGenerator::getBestRaw or addPattern, use original + if (specifiedSkeletonPtr != nullptr) { // called from DateTimePatternGenerator::getBestRaw or addPattern, use original equal = curElem->skeleton->original == skeleton.original; } else { // called from DateTimePatternGenerator::getRedundants, use baseOriginal equal = curElem->skeleton->baseOriginal == skeleton.baseOriginal; } if (equal) { if (specifiedSkeletonPtr && curElem->skeletonWasSpecified) { - *specifiedSkeletonPtr = curElem->skeleton; + *specifiedSkeletonPtr = curElem->skeleton.getAlias(); } return &(curElem->pattern); } - curElem=curElem->next; - }while (curElem != NULL); + curElem = curElem->next.getAlias(); + } while (curElem != nullptr); - return NULL; + return nullptr; } UBool -PatternMap::equals(const PatternMap& other) { +PatternMap::equals(const PatternMap& other) const { if ( this==&other ) { return TRUE; } - for (int32_t bootIndex=0; bootIndex<MAX_PATTERN_ENTRIES; ++bootIndex ) { - if ( boot[bootIndex]==other.boot[bootIndex] ) { + for (int32_t bootIndex = 0; bootIndex < MAX_PATTERN_ENTRIES; ++bootIndex) { + if (boot[bootIndex] == other.boot[bootIndex]) { continue; } - if ( (boot[bootIndex]==NULL)||(other.boot[bootIndex]==NULL) ) { + if ((boot[bootIndex] == nullptr) || (other.boot[bootIndex] == nullptr)) { return FALSE; } PtnElem *otherElem = other.boot[bootIndex]; PtnElem *myElem = boot[bootIndex]; - while ((otherElem!=NULL) || (myElem!=NULL)) { + while ((otherElem != nullptr) || (myElem != nullptr)) { if ( myElem == otherElem ) { break; } - if ((otherElem==NULL) || (myElem==NULL)) { + if ((otherElem == nullptr) || (myElem == nullptr)) { return FALSE; } if ( (myElem->basePattern != otherElem->basePattern) || (myElem->pattern != otherElem->pattern) ) { return FALSE; } - if ((myElem->skeleton!=otherElem->skeleton)&& + if ((myElem->skeleton.getAlias() != otherElem->skeleton.getAlias()) && !myElem->skeleton->equals(*(otherElem->skeleton))) { return FALSE; } - myElem = myElem->next; - otherElem=otherElem->next; + myElem = myElem->next.getAlias(); + otherElem = otherElem->next.getAlias(); } } return TRUE; @@ -1909,21 +2021,21 @@ PtnElem* PatternMap::getDuplicateElem( const UnicodeString &basePattern, const PtnSkeleton &skeleton, - PtnElem *baseElem) { + PtnElem *baseElem) { PtnElem *curElem; - if ( baseElem == (PtnElem *)NULL ) { - return (PtnElem*)NULL; + if ( baseElem == nullptr ) { + return nullptr; } else { curElem = baseElem; } do { if ( basePattern.compare(curElem->basePattern)==0 ) { - UBool isEqual=TRUE; - for (int32_t i=0; i<UDATPG_FIELD_COUNT; ++i) { + UBool isEqual = TRUE; + for (int32_t i = 0; i < UDATPG_FIELD_COUNT; ++i) { if (curElem->skeleton->type[i] != skeleton.type[i] ) { - isEqual=FALSE; + isEqual = FALSE; break; } } @@ -1931,11 +2043,11 @@ PatternMap::getDuplicateElem( return curElem; } } - curElem = curElem->next; - } while( curElem != (PtnElem *)NULL ); + curElem = curElem->next.getAlias(); + } while( curElem != nullptr ); // end of the list - return (PtnElem*)NULL; + return nullptr; } // PatternMap::getDuplicateElem @@ -1976,7 +2088,7 @@ DateTimeMatcher::set(const UnicodeString& pattern, FormatParser* fp, PtnSkeleton continue; } int32_t canonicalIndex = fp->getCanonicalIndex(value); - if (canonicalIndex < 0 ) { + if (canonicalIndex < 0) { continue; } const dtTypeElem *row = &dtTypes[canonicalIndex]; @@ -1986,8 +2098,9 @@ DateTimeMatcher::set(const UnicodeString& pattern, FormatParser* fp, PtnSkeleton int32_t repeatCount = row->minLen; skeletonResult.baseOriginal.populate(field, repeatChar, repeatCount); int16_t subField = row->type; - if ( row->type > 0) { - subField += value.length(); + if (row->type > 0) { + U_ASSERT(value.length() < INT16_MAX); + subField += static_cast<int16_t>(value.length()); } skeletonResult.type[field] = subField; } @@ -2031,8 +2144,8 @@ DateTimeMatcher::getPattern() { } int32_t -DateTimeMatcher::getDistance(const DateTimeMatcher& other, int32_t includeMask, DistanceInfo& distanceInfo) { - int32_t result=0; +DateTimeMatcher::getDistance(const DateTimeMatcher& other, int32_t includeMask, DistanceInfo& distanceInfo) const { + int32_t result = 0; distanceInfo.clear(); for (int32_t i=0; i<UDATPG_FIELD_COUNT; ++i ) { int32_t myType = (includeMask&(1<<i))==0 ? 0 : skeleton.type[i]; @@ -2071,13 +2184,13 @@ DateTimeMatcher::copyFrom() { UBool DateTimeMatcher::equals(const DateTimeMatcher* other) const { - if (other==NULL) { return FALSE; } + if (other==nullptr) { return FALSE; } return skeleton.original == other->skeleton.original; } int32_t -DateTimeMatcher::getFieldMask() { - int32_t result=0; +DateTimeMatcher::getFieldMask() const { + int32_t result = 0; for (int32_t i=0; i<UDATPG_FIELD_COUNT; ++i) { if (skeleton.type[i]!=0) { @@ -2094,7 +2207,7 @@ DateTimeMatcher::getSkeletonPtr() { FormatParser::FormatParser () { status = START; - itemNumber=0; + itemNumber = 0; } @@ -2106,7 +2219,7 @@ FormatParser::~FormatParser () { // Note: the startPos may FormatParser::TokenStatus FormatParser::setTokens(const UnicodeString& pattern, int32_t startPos, int32_t *len) { - int32_t curLoc = startPos; + int32_t curLoc = startPos; if ( curLoc >= pattern.length()) { return DONE; } @@ -2132,10 +2245,10 @@ FormatParser::setTokens(const UnicodeString& pattern, int32_t startPos, int32_t void FormatParser::set(const UnicodeString& pattern) { - int32_t startPos=0; - TokenStatus result=START; - int32_t len=0; - itemNumber =0; + int32_t startPos = 0; + TokenStatus result = START; + int32_t len = 0; + itemNumber = 0; do { result = setTokens( pattern, startPos, &len ); @@ -2186,14 +2299,14 @@ FormatParser::getCanonicalIndex(const UnicodeString& s, UBool strict) { UBool FormatParser::isQuoteLiteral(const UnicodeString& s) { - return (UBool)(s.charAt(0)==SINGLE_QUOTE); + return (UBool)(s.charAt(0) == SINGLE_QUOTE); } -// This function aussumes the current itemIndex points to the quote literal. +// This function assumes the current itemIndex points to the quote literal. // Please call isQuoteLiteral prior to this function. void FormatParser::getQuoteLiteral(UnicodeString& quote, int32_t *itemIndex) { - int32_t i=*itemIndex; + int32_t i = *itemIndex; quote.remove(); if (items[i].charAt(0)==SINGLE_QUOTE) { @@ -2222,7 +2335,7 @@ FormatParser::getQuoteLiteral(UnicodeString& quote, int32_t *itemIndex) { } UBool -FormatParser::isPatternSeparator(UnicodeString& field) { +FormatParser::isPatternSeparator(const UnicodeString& field) const { for (int32_t i=0; i<field.length(); ++i ) { UChar c= field.charAt(i); if ( (c==SINGLE_QUOTE) || (c==BACKSLASH) || (c==SPACE) || (c==COLON) || @@ -2239,21 +2352,19 @@ FormatParser::isPatternSeparator(UnicodeString& field) { DistanceInfo::~DistanceInfo() {} void -DistanceInfo::setTo(DistanceInfo &other) { +DistanceInfo::setTo(const DistanceInfo& other) { missingFieldMask = other.missingFieldMask; extraFieldMask= other.extraFieldMask; } -PatternMapIterator::PatternMapIterator() { - bootIndex = 0; - nodePtr = NULL; - patternMap=NULL; - matcher= new DateTimeMatcher(); +PatternMapIterator::PatternMapIterator(UErrorCode& status) : + bootIndex(0), nodePtr(nullptr), matcher(nullptr), patternMap(nullptr) +{ + if (U_FAILURE(status)) { return; } + matcher.adoptInsteadAndCheckErrorCode(new DateTimeMatcher(), status); } - PatternMapIterator::~PatternMapIterator() { - delete matcher; } void @@ -2262,36 +2373,36 @@ PatternMapIterator::set(PatternMap& newPatternMap) { } PtnSkeleton* -PatternMapIterator::getSkeleton() { - if ( nodePtr == NULL ) { - return NULL; +PatternMapIterator::getSkeleton() const { + if ( nodePtr == nullptr ) { + return nullptr; } else { - return nodePtr->skeleton; + return nodePtr->skeleton.getAlias(); } } UBool -PatternMapIterator::hasNext() { - int32_t headIndex=bootIndex; - PtnElem *curPtr=nodePtr; +PatternMapIterator::hasNext() const { + int32_t headIndex = bootIndex; + PtnElem *curPtr = nodePtr; - if (patternMap==NULL) { + if (patternMap==nullptr) { return FALSE; } while ( headIndex < MAX_PATTERN_ENTRIES ) { - if ( curPtr != NULL ) { - if ( curPtr->next != NULL ) { + if ( curPtr != nullptr ) { + if ( curPtr->next != nullptr ) { return TRUE; } else { headIndex++; - curPtr=NULL; + curPtr=nullptr; continue; } } else { - if ( patternMap->boot[headIndex] != NULL ) { + if ( patternMap->boot[headIndex] != nullptr ) { return TRUE; } else { @@ -2299,7 +2410,6 @@ PatternMapIterator::hasNext() { continue; } } - } return FALSE; } @@ -2307,19 +2417,19 @@ PatternMapIterator::hasNext() { DateTimeMatcher& PatternMapIterator::next() { while ( bootIndex < MAX_PATTERN_ENTRIES ) { - if ( nodePtr != NULL ) { - if ( nodePtr->next != NULL ) { - nodePtr = nodePtr->next; + if ( nodePtr != nullptr ) { + if ( nodePtr->next != nullptr ) { + nodePtr = nodePtr->next.getAlias(); break; } else { bootIndex++; - nodePtr=NULL; + nodePtr=nullptr; continue; } } else { - if ( patternMap->boot[bootIndex] != NULL ) { + if ( patternMap->boot[bootIndex] != nullptr ) { nodePtr = patternMap->boot[bootIndex]; break; } @@ -2329,7 +2439,7 @@ PatternMapIterator::next() { } } } - if (nodePtr!=NULL) { + if (nodePtr!=nullptr) { matcher->copyFrom(*nodePtr->skeleton); } else { @@ -2468,36 +2578,28 @@ PtnSkeleton::~PtnSkeleton() { } PtnElem::PtnElem(const UnicodeString &basePat, const UnicodeString &pat) : -basePattern(basePat), -skeleton(NULL), -pattern(pat), -next(NULL) + basePattern(basePat), skeleton(nullptr), pattern(pat), next(nullptr) { } PtnElem::~PtnElem() { - - if (next!=NULL) { - delete next; - } - delete skeleton; } -DTSkeletonEnumeration::DTSkeletonEnumeration(PatternMap &patternMap, dtStrEnum type, UErrorCode& status) { +DTSkeletonEnumeration::DTSkeletonEnumeration(PatternMap& patternMap, dtStrEnum type, UErrorCode& status) : fSkeletons(nullptr) { PtnElem *curElem; PtnSkeleton *curSkeleton; UnicodeString s; int32_t bootIndex; pos=0; - fSkeletons = new UVector(status); + fSkeletons.adoptInsteadAndCheckErrorCode(new UVector(status), status); if (U_FAILURE(status)) { - delete fSkeletons; return; } + for (bootIndex=0; bootIndex<MAX_PATTERN_ENTRIES; ++bootIndex ) { curElem = patternMap.boot[bootIndex]; - while (curElem!=NULL) { + while (curElem!=nullptr) { switch(type) { case DT_BASESKELETON: s=curElem->basePattern; @@ -2506,32 +2608,36 @@ DTSkeletonEnumeration::DTSkeletonEnumeration(PatternMap &patternMap, dtStrEnum t s=curElem->pattern; break; case DT_SKELETON: - curSkeleton=curElem->skeleton; + curSkeleton=curElem->skeleton.getAlias(); s=curSkeleton->getSkeleton(); break; } if ( !isCanonicalItem(s) ) { - fSkeletons->addElement(new UnicodeString(s), status); + LocalPointer<UnicodeString> newElem(new UnicodeString(s), status); if (U_FAILURE(status)) { - delete fSkeletons; - fSkeletons = NULL; return; } + fSkeletons->addElement(newElem.getAlias(), status); + if (U_FAILURE(status)) { + fSkeletons.adoptInstead(nullptr); + return; + } + newElem.orphan(); // fSkeletons vector now owns the UnicodeString. } - curElem = curElem->next; + curElem = curElem->next.getAlias(); } } - if ((bootIndex==MAX_PATTERN_ENTRIES) && (curElem!=NULL) ) { + if ((bootIndex==MAX_PATTERN_ENTRIES) && (curElem!=nullptr) ) { status = U_BUFFER_OVERFLOW_ERROR; } } const UnicodeString* DTSkeletonEnumeration::snext(UErrorCode& status) { - if (U_SUCCESS(status) && pos < fSkeletons->size()) { + if (U_SUCCESS(status) && fSkeletons.isValid() && pos < fSkeletons->size()) { return (const UnicodeString*)fSkeletons->elementAt(pos++); } - return NULL; + return nullptr; } void @@ -2541,7 +2647,7 @@ DTSkeletonEnumeration::reset(UErrorCode& /*status*/) { int32_t DTSkeletonEnumeration::count(UErrorCode& /*status*/) const { - return (fSkeletons==NULL) ? 0 : fSkeletons->size(); + return (fSkeletons.isNull()) ? 0 : fSkeletons->size(); } UBool @@ -2559,44 +2665,45 @@ DTSkeletonEnumeration::isCanonicalItem(const UnicodeString& item) { DTSkeletonEnumeration::~DTSkeletonEnumeration() { UnicodeString *s; - for (int32_t i=0; i<fSkeletons->size(); ++i) { - if ((s=(UnicodeString *)fSkeletons->elementAt(i))!=NULL) { - delete s; + if (fSkeletons.isValid()) { + for (int32_t i = 0; i < fSkeletons->size(); ++i) { + if ((s = (UnicodeString *)fSkeletons->elementAt(i)) != nullptr) { + delete s; + } } } - delete fSkeletons; } -DTRedundantEnumeration::DTRedundantEnumeration() { - pos=0; - fPatterns = NULL; +DTRedundantEnumeration::DTRedundantEnumeration() : pos(0), fPatterns(nullptr) { } void DTRedundantEnumeration::add(const UnicodeString& pattern, UErrorCode& status) { - if (U_FAILURE(status)) return; - if (fPatterns == NULL) { - fPatterns = new UVector(status); + if (U_FAILURE(status)) { return; } + if (fPatterns.isNull()) { + fPatterns.adoptInsteadAndCheckErrorCode(new UVector(status), status); if (U_FAILURE(status)) { - delete fPatterns; - fPatterns = NULL; return; } } - fPatterns->addElement(new UnicodeString(pattern), status); + LocalPointer<UnicodeString> newElem(new UnicodeString(pattern), status); + if (U_FAILURE(status)) { + return; + } + fPatterns->addElement(newElem.getAlias(), status); if (U_FAILURE(status)) { - delete fPatterns; - fPatterns = NULL; + fPatterns.adoptInstead(nullptr); return; } + newElem.orphan(); // fPatterns now owns the string. } const UnicodeString* DTRedundantEnumeration::snext(UErrorCode& status) { - if (U_SUCCESS(status) && pos < fPatterns->size()) { + if (U_SUCCESS(status) && fPatterns.isValid() && pos < fPatterns->size()) { return (const UnicodeString*)fPatterns->elementAt(pos++); } - return NULL; + return nullptr; } void @@ -2606,11 +2713,11 @@ DTRedundantEnumeration::reset(UErrorCode& /*status*/) { int32_t DTRedundantEnumeration::count(UErrorCode& /*status*/) const { - return (fPatterns==NULL) ? 0 : fPatterns->size(); + return (fPatterns.isNull()) ? 0 : fPatterns->size(); } UBool -DTRedundantEnumeration::isCanonicalItem(const UnicodeString& item) { +DTRedundantEnumeration::isCanonicalItem(const UnicodeString& item) const { if ( item.length() != 1 ) { return FALSE; } @@ -2624,12 +2731,13 @@ DTRedundantEnumeration::isCanonicalItem(const UnicodeString& item) { DTRedundantEnumeration::~DTRedundantEnumeration() { UnicodeString *s; - for (int32_t i=0; i<fPatterns->size(); ++i) { - if ((s=(UnicodeString *)fPatterns->elementAt(i))!=NULL) { - delete s; + if (fPatterns.isValid()) { + for (int32_t i = 0; i < fPatterns->size(); ++i) { + if ((s = (UnicodeString *)fPatterns->elementAt(i)) != nullptr) { + delete s; + } } } - delete fPatterns; } U_NAMESPACE_END diff --git a/deps/icu-small/source/i18n/dtptngen_impl.h b/deps/icu-small/source/i18n/dtptngen_impl.h index 2ea31a75c4..95219f0ba2 100644 --- a/deps/icu-small/source/i18n/dtptngen_impl.h +++ b/deps/icu-small/source/i18n/dtptngen_impl.h @@ -116,7 +116,7 @@ typedef struct dtTypeElem { int16_t type; int16_t minLen; int16_t weight; -}dtTypeElem; +} dtTypeElem; // A compact storage mechanism for skeleton field strings. Several dozen of these will be created // for a typical DateTimePatternGenerator instance. @@ -172,30 +172,28 @@ public: virtual ~PtnSkeleton(); }; - class PtnElem : public UMemory { public: UnicodeString basePattern; - PtnSkeleton *skeleton; + LocalPointer<PtnSkeleton> skeleton; UnicodeString pattern; UBool skeletonWasSpecified; // if specified in availableFormats, not derived - PtnElem *next; + LocalPointer<PtnElem> next; PtnElem(const UnicodeString &basePattern, const UnicodeString &pattern); virtual ~PtnElem(); - }; class FormatParser : public UMemory { public: UnicodeString items[MAX_DT_TOKEN]; - int32_t itemNumber; + int32_t itemNumber; FormatParser(); virtual ~FormatParser(); void set(const UnicodeString& patternString); void getQuoteLiteral(UnicodeString& quote, int32_t *itemIndex); - UBool isPatternSeparator(UnicodeString& field); + UBool isPatternSeparator(const UnicodeString& field) const; static UBool isQuoteLiteral(const UnicodeString& s); static int32_t getCanonicalIndex(const UnicodeString& s) { return getCanonicalIndex(s, TRUE); } static int32_t getCanonicalIndex(const UnicodeString& s, UBool strict); @@ -206,7 +204,7 @@ private: ADD_TOKEN, SYNTAX_ERROR, DONE - } ToeknStatus; + } TokenStatus; TokenStatus status; virtual TokenStatus setTokens(const UnicodeString& pattern, int32_t startPos, int32_t *len); @@ -220,7 +218,7 @@ public: DistanceInfo() {} virtual ~DistanceInfo(); void clear() { missingFieldMask = extraFieldMask = 0; } - void setTo(DistanceInfo& other); + void setTo(const DistanceInfo& other); void addMissing(int32_t field) { missingFieldMask |= (1<<field); } void addExtra(int32_t field) { extraFieldMask |= (1<<field); } }; @@ -237,11 +235,11 @@ public: void copyFrom(); PtnSkeleton* getSkeletonPtr(); UBool equals(const DateTimeMatcher* other) const; - int32_t getDistance(const DateTimeMatcher& other, int32_t includeMask, DistanceInfo& distanceInfo); + int32_t getDistance(const DateTimeMatcher& other, int32_t includeMask, DistanceInfo& distanceInfo) const; DateTimeMatcher(); DateTimeMatcher(const DateTimeMatcher& other); virtual ~DateTimeMatcher(); - int32_t getFieldMask(); + int32_t getFieldMask() const; }; class PatternMap : public UMemory { @@ -250,34 +248,34 @@ public: PatternMap(); virtual ~PatternMap(); void add(const UnicodeString& basePattern, const PtnSkeleton& skeleton, const UnicodeString& value, UBool skeletonWasSpecified, UErrorCode& status); - const UnicodeString* getPatternFromBasePattern(UnicodeString& basePattern, UBool& skeletonWasSpecified); - const UnicodeString* getPatternFromSkeleton(PtnSkeleton& skeleton, const PtnSkeleton** specifiedSkeletonPtr = 0); + const UnicodeString* getPatternFromBasePattern(const UnicodeString& basePattern, UBool& skeletonWasSpecified) const; + const UnicodeString* getPatternFromSkeleton(const PtnSkeleton& skeleton, const PtnSkeleton** specifiedSkeletonPtr = 0) const; void copyFrom(const PatternMap& other, UErrorCode& status); - PtnElem* getHeader(UChar baseChar); - UBool equals(const PatternMap& other); + PtnElem* getHeader(UChar baseChar) const; + UBool equals(const PatternMap& other) const; private: UBool isDupAllowed; - PtnElem* getDuplicateElem(const UnicodeString &basePattern, const PtnSkeleton& skeleton, PtnElem *baseElem); + PtnElem* getDuplicateElem(const UnicodeString& basePattern, const PtnSkeleton& skeleton, PtnElem *baseElem); }; // end PatternMap class PatternMapIterator : public UMemory { public: - PatternMapIterator(); + PatternMapIterator(UErrorCode &status); virtual ~PatternMapIterator(); void set(PatternMap& patternMap); - PtnSkeleton* getSkeleton(); - UBool hasNext(); + PtnSkeleton* getSkeleton() const; + UBool hasNext() const; DateTimeMatcher& next(); private: int32_t bootIndex; PtnElem *nodePtr; - DateTimeMatcher *matcher; + LocalPointer<DateTimeMatcher> matcher; PatternMap *patternMap; }; class DTSkeletonEnumeration : public StringEnumeration { public: - DTSkeletonEnumeration(PatternMap &patternMap, dtStrEnum type, UErrorCode& status); + DTSkeletonEnumeration(PatternMap& patternMap, dtStrEnum type, UErrorCode& status); virtual ~DTSkeletonEnumeration(); static UClassID U_EXPORT2 getStaticClassID(void); virtual UClassID getDynamicClassID(void) const; @@ -287,7 +285,7 @@ public: private: int32_t pos; UBool isCanonicalItem(const UnicodeString& item); - UVector *fSkeletons; + LocalPointer<UVector> fSkeletons; }; class DTRedundantEnumeration : public StringEnumeration { @@ -302,8 +300,8 @@ public: void add(const UnicodeString &pattern, UErrorCode& status); private: int32_t pos; - UBool isCanonicalItem(const UnicodeString& item); - UVector *fPatterns; + UBool isCanonicalItem(const UnicodeString& item) const; + LocalPointer<UVector> fPatterns; }; U_NAMESPACE_END diff --git a/deps/icu-small/source/i18n/erarules.cpp b/deps/icu-small/source/i18n/erarules.cpp new file mode 100644 index 0000000000..669f844230 --- /dev/null +++ b/deps/icu-small/source/i18n/erarules.cpp @@ -0,0 +1,307 @@ +// © 2018 and later: Unicode, Inc. and others. +// License & terms of use: http://www.unicode.org/copyright.html + +#include "unicode/utypes.h" + +#if !UCONFIG_NO_FORMATTING + +#include <stdlib.h> +#include "unicode/ucal.h" +#include "unicode/ures.h" +#include "unicode/ustring.h" +#include "cmemory.h" +#include "cstring.h" +#include "erarules.h" +#include "gregoimp.h" +#include "uassert.h" + +U_NAMESPACE_BEGIN + +static const int32_t MAX_ENCODED_START_YEAR = 32767; +static const int32_t MIN_ENCODED_START_YEAR = -32768; +static const int32_t MIN_ENCODED_START = -2147483391; // encodeDate(MIN_ENCODED_START_YEAR, 1, 1, ...); + +static const int32_t YEAR_MASK = 0xFFFF0000; +static const int32_t MONTH_MASK = 0x0000FF00; +static const int32_t DAY_MASK = 0x000000FF; + +static const int32_t MAX_INT32 = 0x7FFFFFFF; +static const int32_t MIN_INT32 = 0xFFFFFFFF; + +static const UChar VAL_FALSE[] = {0x66, 0x61, 0x6c, 0x73, 0x65}; // "false" +static const UChar VAL_FALSE_LEN = 5; + +static UBool isSet(int startDate) { + return startDate != 0; +} + +static UBool isValidRuleStartDate(int32_t year, int32_t month, int32_t day) { + return year >= MIN_ENCODED_START_YEAR && year <= MAX_ENCODED_START_YEAR + && month >= 1 && month <= 12 && day >=1 && day <= 31; +} + +/** + * Encode year/month/date to a single integer. + * year is high 16 bits (-32768 to 32767), month is + * next 8 bits and day of month is last 8 bits. + * + * @param year year + * @param month month (1-base) + * @param day day of month + * @return an encoded date. + */ +static int32_t encodeDate(int32_t year, int32_t month, int32_t day) { + return year << 16 | month << 8 | day; +} + +static void decodeDate(int32_t encodedDate, int32_t (&fields)[3]) { + if (encodedDate == MIN_ENCODED_START) { + fields[0] = MIN_INT32; + fields[1] = 1; + fields[2] = 1; + } else { + fields[0] = (encodedDate & YEAR_MASK) >> 16; + fields[1] = (encodedDate & MONTH_MASK) >> 8; + fields[2] = encodedDate & DAY_MASK; + } +} + +/** + * Compare an encoded date with another date specified by year/month/day. + * @param encoded An encoded date + * @param year Year of another date + * @param month Month of another date + * @param day Day of another date + * @return -1 when encoded date is earlier, 0 when two dates are same, + * and 1 when encoded date is later. + */ +static int32_t compareEncodedDateWithYMD(int encoded, int year, int month, int day) { + if (year < MIN_ENCODED_START_YEAR) { + if (encoded == MIN_ENCODED_START) { + if (year > MIN_INT32 || month > 1 || day > 1) { + return -1; + } + return 0; + } else { + return 1; + } + } else if (year > MAX_ENCODED_START_YEAR) { + return -1; + } else { + int tmp = encodeDate(year, month, day); + if (encoded < tmp) { + return -1; + } else if (encoded == tmp) { + return 0; + } else { + return 1; + } + } +} + +EraRules::EraRules(LocalMemory<int32_t>& eraStartDates, int32_t numEras) + : numEras(numEras) { + startDates.moveFrom(eraStartDates); + initCurrentEra(); +} + +EraRules::~EraRules() { +} + +EraRules* EraRules::createInstance(const char *calType, UBool includeTentativeEra, UErrorCode& status) { + if(U_FAILURE(status)) { + return nullptr; + } + LocalUResourceBundlePointer rb(ures_openDirect(nullptr, "supplementalData", &status)); + ures_getByKey(rb.getAlias(), "calendarData", rb.getAlias(), &status); + ures_getByKey(rb.getAlias(), calType, rb.getAlias(), &status); + ures_getByKey(rb.getAlias(), "eras", rb.getAlias(), &status); + + if (U_FAILURE(status)) { + return nullptr; + } + + int32_t numEras = ures_getSize(rb.getAlias()); + int32_t firstTentativeIdx = MAX_INT32; + + LocalMemory<int32_t> startDates(static_cast<int32_t *>(uprv_malloc(numEras * sizeof(int32_t)))); + if (startDates.isNull()) { + status = U_MEMORY_ALLOCATION_ERROR; + return nullptr; + } + uprv_memset(startDates.getAlias(), 0 , numEras * sizeof(int32_t)); + + while (ures_hasNext(rb.getAlias())) { + LocalUResourceBundlePointer eraRuleRes(ures_getNextResource(rb.getAlias(), nullptr, &status)); + if (U_FAILURE(status)) { + return nullptr; + } + const char *eraIdxStr = ures_getKey(eraRuleRes.getAlias()); + char *endp; + int32_t eraIdx = (int32_t)strtol(eraIdxStr, &endp, 10); + if ((size_t)(endp - eraIdxStr) != uprv_strlen(eraIdxStr)) { + status = U_INVALID_FORMAT_ERROR; + return nullptr; + } + if (eraIdx < 0 || eraIdx >= numEras) { + status = U_INVALID_FORMAT_ERROR; + return nullptr; + } + if (isSet(startDates[eraIdx])) { + // start date of the index was already set + status = U_INVALID_FORMAT_ERROR; + return nullptr; + } + + UBool hasName = TRUE; + UBool hasEnd = TRUE; + int32_t len; + while (ures_hasNext(eraRuleRes.getAlias())) { + LocalUResourceBundlePointer res(ures_getNextResource(eraRuleRes.getAlias(), nullptr, &status)); + if (U_FAILURE(status)) { + return nullptr; + } + const char *key = ures_getKey(res.getAlias()); + if (uprv_strcmp(key, "start") == 0) { + const int32_t *fields = ures_getIntVector(res.getAlias(), &len, &status); + if (U_FAILURE(status)) { + return nullptr; + } + if (len != 3 || !isValidRuleStartDate(fields[0], fields[1], fields[2])) { + status = U_INVALID_FORMAT_ERROR; + return nullptr; + } + startDates[eraIdx] = encodeDate(fields[0], fields[1], fields[2]); + } else if (uprv_strcmp(key, "named") == 0) { + const UChar *val = ures_getString(res.getAlias(), &len, &status); + if (u_strncmp(val, VAL_FALSE, VAL_FALSE_LEN) == 0) { + hasName = FALSE; + } + } else if (uprv_strcmp(key, "end") == 0) { + hasEnd = TRUE; + } + } + + if (isSet(startDates[eraIdx])) { + if (hasEnd) { + // This implementation assumes either start or end is available, not both. + // For now, just ignore the end rule. + } + } else { + if (hasEnd) { + if (eraIdx != 0) { + // This implementation does not support end only rule for eras other than + // the first one. + status = U_INVALID_FORMAT_ERROR; + return nullptr; + } + U_ASSERT(eraIdx == 0); + startDates[eraIdx] = MIN_ENCODED_START; + } else { + status = U_INVALID_FORMAT_ERROR; + return nullptr; + } + } + + if (hasName) { + if (eraIdx >= firstTentativeIdx) { + status = U_INVALID_FORMAT_ERROR; + return nullptr; + } + } else { + if (eraIdx < firstTentativeIdx) { + firstTentativeIdx = eraIdx; + } + } + } + + EraRules *result; + if (firstTentativeIdx < MAX_INT32 && !includeTentativeEra) { + result = new EraRules(startDates, firstTentativeIdx); + } else { + result = new EraRules(startDates, numEras); + } + + if (result == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + } + return result; +} + +void EraRules::getStartDate(int32_t eraIdx, int32_t (&fields)[3], UErrorCode& status) const { + if(U_FAILURE(status)) { + return; + } + if (eraIdx < 0 || eraIdx >= numEras) { + status = U_ILLEGAL_ARGUMENT_ERROR; + return; + } + decodeDate(startDates[eraIdx], fields); +} + +int32_t EraRules::getStartYear(int32_t eraIdx, UErrorCode& status) const { + int year = MAX_INT32; // bogus value + if(U_FAILURE(status)) { + return year; + } + if (eraIdx < 0 || eraIdx >= numEras) { + status = U_ILLEGAL_ARGUMENT_ERROR; + return year; + } + int fields[3]; + decodeDate(startDates[eraIdx], fields); + year = fields[0]; + + return year; +} + +int32_t EraRules::getEraIndex(int32_t year, int32_t month, int32_t day, UErrorCode& status) const { + if(U_FAILURE(status)) { + return -1; + } + + if (month < 1 || month > 12 || day < 1 || day > 31) { + status = U_ILLEGAL_ARGUMENT_ERROR; + return -1; + } + int32_t high = numEras; // last index + 1 + int32_t low; + + // Short circuit for recent years. Most modern computations will + // occur in the last few eras. + if (compareEncodedDateWithYMD(startDates[getCurrentEraIndex()], year, month, day) <= 0) { + low = getCurrentEraIndex(); + } else { + low = 0; + } + + // Do binary search + while (low < high - 1) { + int i = (low + high) / 2; + if (compareEncodedDateWithYMD(startDates[i], year, month, day) <= 0) { + low = i; + } else { + high = i; + } + } + return low; +} + +void EraRules::initCurrentEra() { + UDate now = ucal_getNow(); + int year, month0, dom, dow, doy, mid; + Grego::timeToFields(now, year, month0, dom, dow, doy, mid); + int currentEncodedDate = encodeDate(year, month0 + 1 /* changes to 1-base */, dom); + int eraIdx = numEras - 1; + while (eraIdx > 0) { + if (currentEncodedDate >= startDates[eraIdx]) { + break; + } + eraIdx--; + } + // Note: current era could be before the first era. + // In this case, this implementation returns the first era index (0). + currentEra = eraIdx;} + +U_NAMESPACE_END +#endif /* #if !UCONFIG_NO_FORMATTING */ diff --git a/deps/icu-small/source/i18n/erarules.h b/deps/icu-small/source/i18n/erarules.h new file mode 100644 index 0000000000..4ed8640832 --- /dev/null +++ b/deps/icu-small/source/i18n/erarules.h @@ -0,0 +1,92 @@ +// © 2018 and later: Unicode, Inc. and others. +// License & terms of use: http://www.unicode.org/copyright.html + +#ifndef ERARULES_H_ +#define ERARULES_H_ + +#include "unicode/utypes.h" + +#if !UCONFIG_NO_FORMATTING + +#include "unicode/localpointer.h" +#include "unicode/uobject.h" +#include "cmemory.h" + +U_NAMESPACE_BEGIN + +// Export an explicit template instantiation of LocalMemory used as a data member of EraRules. +// When building DLLs for Windows this is required even though no direct access leaks out of the i18n library. +// See digitlst.h, pluralaffix.h, datefmt.h, and others for similar examples. +#if U_PF_WINDOWS <= U_PLATFORM && U_PLATFORM <= U_PF_CYGWIN +// Ignore warning 4661 as LocalPointerBase does not use operator== or operator!= +#pragma warning(suppress: 4661) +template class U_I18N_API LocalPointerBase<int32_t>; +template class U_I18N_API LocalMemory<int32_t>; +#endif + +class U_I18N_API EraRules : public UMemory { +public: + ~EraRules(); + + static EraRules* createInstance(const char *calType, UBool includeTentativeEra, UErrorCode& status); + + /** + * Gets number of effective eras + * @return number of effective eras + */ + inline int32_t getNumberOfEras() const { + return numEras; + } + + /** + * Gets start date of an era + * @param eraIdx Era index + * @param fields Receives date fields. The result includes values of year, month, + * day of month in this order. When an era has no start date, the result + * will be January 1st in year whose value is minimum integer. + * @param status Receives status. + */ + void getStartDate(int32_t eraIdx, int32_t (&fields)[3], UErrorCode& status) const; + + /** + * Gets start year of an era + * @param eraIdx Era index + * @param status Receives status. + * @return The first year of an era. When a era has no start date, minimum int32 + * value is returned. + */ + int32_t getStartYear(int32_t eraIdx, UErrorCode& status) const; + + /** + * Returns era index for the specified year/month/day. + * @param year Year + * @param month Month (1-base) + * @param day Day of month + * @param status Receives status + * @return era index (or 0, when the specified date is before the first era) + */ + int32_t getEraIndex(int32_t year, int32_t month, int32_t day, UErrorCode& status) const; + + /** + * Gets the current era index. This is calculated only once for an instance of + * EraRules. + * + * @return era index of current era (or 0, when current date is before the first era) + */ + inline int32_t getCurrentEraIndex() const { + return currentEra; + } + +private: + EraRules(LocalMemory<int32_t>& eraStartDates, int32_t numEra); + + void initCurrentEra(); + + LocalMemory<int32_t> startDates; + int32_t numEras; + int32_t currentEra; +}; + +U_NAMESPACE_END +#endif /* #if !UCONFIG_NO_FORMATTING */ +#endif /* ERARULES_H_ */ diff --git a/deps/icu-small/source/i18n/fphdlimp.h b/deps/icu-small/source/i18n/fphdlimp.h index 2e9d5622b1..a6827e01e9 100644 --- a/deps/icu-small/source/i18n/fphdlimp.h +++ b/deps/icu-small/source/i18n/fphdlimp.h @@ -10,9 +10,10 @@ #ifndef FPHDLIMP_H #define FPHDLIMP_H +#include "unicode/utypes.h" + #if !UCONFIG_NO_FORMATTING -#include "unicode/utypes.h" #include "unicode/fieldpos.h" #include "unicode/fpositer.h" diff --git a/deps/icu-small/source/i18n/gregocal.cpp b/deps/icu-small/source/i18n/gregocal.cpp index 49c4226049..4db66758df 100644 --- a/deps/icu-small/source/i18n/gregocal.cpp +++ b/deps/icu-small/source/i18n/gregocal.cpp @@ -541,8 +541,8 @@ int32_t GregorianCalendar::handleComputeMonthStart(int32_t eyear, int32_t month, } UBool isLeap = eyear%4 == 0; - int32_t y = eyear-1; - int32_t julianDay = 365*y + ClockMath::floorDivide(y, 4) + (kJan1_1JulianDay - 3); + int64_t y = (int64_t)eyear-1; + int64_t julianDay = 365*y + ClockMath::floorDivide(y, (int64_t)4) + (kJan1_1JulianDay - 3); nonConstThis->fIsGregorian = (eyear >= fGregorianCutoverYear); #if defined (U_DEBUG_CAL) @@ -572,7 +572,7 @@ int32_t GregorianCalendar::handleComputeMonthStart(int32_t eyear, int32_t month, julianDay += isLeap?kLeapNumDays[month]:kNumDays[month]; } - return julianDay; + return static_cast<int32_t>(julianDay); } int32_t GregorianCalendar::handleGetMonthLength(int32_t extendedYear, int32_t month) const diff --git a/deps/icu-small/source/i18n/gregoimp.h b/deps/icu-small/source/i18n/gregoimp.h index 55922a6b40..eb3844f852 100644 --- a/deps/icu-small/source/i18n/gregoimp.h +++ b/deps/icu-small/source/i18n/gregoimp.h @@ -299,8 +299,8 @@ inline int32_t Grego::millisToJulianDay(double millis) { } inline int32_t Grego::gregorianShift(int32_t eyear) { - int32_t y = eyear-1; - int32_t gregShift = ClockMath::floorDivide(y, 400) - ClockMath::floorDivide(y, 100) + 2; + int64_t y = (int64_t)eyear-1; + int32_t gregShift = static_cast<int32_t>(ClockMath::floorDivide(y, (int64_t)400) - ClockMath::floorDivide(y, (int64_t)100) + 2); return gregShift; } diff --git a/deps/icu-small/source/i18n/indiancal.cpp b/deps/icu-small/source/i18n/indiancal.cpp index a2a7f2dcdd..667b6f2d7a 100644 --- a/deps/icu-small/source/i18n/indiancal.cpp +++ b/deps/icu-small/source/i18n/indiancal.cpp @@ -347,12 +347,15 @@ IndianCalendar::inDaylightTime(UErrorCode& status) const return (UBool)(U_SUCCESS(status) ? (internalGet(UCAL_DST_OFFSET) != 0) : FALSE); } -// default century -const UDate IndianCalendar::fgSystemDefaultCentury = DBL_MIN; -const int32_t IndianCalendar::fgSystemDefaultCenturyYear = -1; -UDate IndianCalendar::fgSystemDefaultCenturyStart = DBL_MIN; -int32_t IndianCalendar::fgSystemDefaultCenturyStartYear = -1; +/** + * The system maintains a static default century start date and Year. They are + * initialized the first time they are used. Once the system default century date + * and year are set, they do not change. + */ +static UDate gSystemDefaultCenturyStart = DBL_MIN; +static int32_t gSystemDefaultCenturyStartYear = -1; +static icu::UInitOnce gSystemDefaultCenturyInit = U_INITONCE_INITIALIZER; UBool IndianCalendar::haveDefaultCentury() const @@ -360,87 +363,45 @@ UBool IndianCalendar::haveDefaultCentury() const return TRUE; } -UDate IndianCalendar::defaultCenturyStart() const +static void U_CALLCONV +initializeSystemDefaultCentury() { - return internalGetDefaultCenturyStart(); -} + // initialize systemDefaultCentury and systemDefaultCenturyYear based + // on the current time. They'll be set to 80 years before + // the current time. + UErrorCode status = U_ZERO_ERROR; -int32_t IndianCalendar::defaultCenturyStartYear() const -{ - return internalGetDefaultCenturyStartYear(); -} + IndianCalendar calendar ( Locale ( "@calendar=Indian" ), status); + if ( U_SUCCESS ( status ) ) { + calendar.setTime ( Calendar::getNow(), status ); + calendar.add ( UCAL_YEAR, -80, status ); -UDate -IndianCalendar::internalGetDefaultCenturyStart() const -{ - // lazy-evaluate systemDefaultCenturyStart - UBool needsUpdate; - { - Mutex m; - needsUpdate = (fgSystemDefaultCenturyStart == fgSystemDefaultCentury); - } + UDate newStart = calendar.getTime ( status ); + int32_t newYear = calendar.get ( UCAL_YEAR, status ); - if (needsUpdate) { - initializeSystemDefaultCentury(); + gSystemDefaultCenturyStart = newStart; + gSystemDefaultCenturyStartYear = newYear; } + // We have no recourse upon failure. +} - // use defaultCenturyStart unless it's the flag value; - // then use systemDefaultCenturyStart - return fgSystemDefaultCenturyStart; +UDate +IndianCalendar::defaultCenturyStart() const +{ + // lazy-evaluate systemDefaultCenturyStart + umtx_initOnce(gSystemDefaultCenturyInit, &initializeSystemDefaultCentury); + return gSystemDefaultCenturyStart; } int32_t -IndianCalendar::internalGetDefaultCenturyStartYear() const +IndianCalendar::defaultCenturyStartYear() const { // lazy-evaluate systemDefaultCenturyStartYear - UBool needsUpdate; - { - Mutex m; - - needsUpdate = (fgSystemDefaultCenturyStart == fgSystemDefaultCentury); - } - - if (needsUpdate) { - initializeSystemDefaultCentury(); - } - - // use defaultCenturyStart unless it's the flag value; - // then use systemDefaultCenturyStartYear - - return fgSystemDefaultCenturyStartYear; + umtx_initOnce(gSystemDefaultCenturyInit, &initializeSystemDefaultCentury); + return gSystemDefaultCenturyStartYear; } -void -IndianCalendar::initializeSystemDefaultCentury() -{ - // initialize systemDefaultCentury and systemDefaultCenturyYear based - // on the current time. They'll be set to 80 years before - // the current time. - // No point in locking as it should be idempotent. - if (fgSystemDefaultCenturyStart == fgSystemDefaultCentury) { - UErrorCode status = U_ZERO_ERROR; - - IndianCalendar calendar(Locale("@calendar=Indian"),status); - if (U_SUCCESS(status)) { - calendar.setTime(Calendar::getNow(), status); - calendar.add(UCAL_YEAR, -80, status); - - UDate newStart = calendar.getTime(status); - int32_t newYear = calendar.get(UCAL_YEAR, status); - - { - Mutex m; - - fgSystemDefaultCenturyStart = newStart; - fgSystemDefaultCenturyStartYear = newYear; - } - } - - // We have no recourse upon failure unless we want to propagate the failure - // out. - } -} UOBJECT_DEFINE_RTTI_IMPLEMENTATION(IndianCalendar) diff --git a/deps/icu-small/source/i18n/indiancal.h b/deps/icu-small/source/i18n/indiancal.h index d40af5ad45..ffd4ae8b8a 100644 --- a/deps/icu-small/source/i18n/indiancal.h +++ b/deps/icu-small/source/i18n/indiancal.h @@ -68,7 +68,7 @@ U_NAMESPACE_BEGIN */ -class IndianCalendar : public Calendar { +class U_I18N_API IndianCalendar : public Calendar { public: /** * Useful constants for IndianCalendar. @@ -274,10 +274,10 @@ public: * @return The class ID for all objects of this class. * @internal */ - U_I18N_API static UClassID U_EXPORT2 getStaticClassID(void); + static UClassID U_EXPORT2 getStaticClassID(void); /** - * return the calendar type, "buddhist". + * return the calendar type, "indian". * * @return calendar type * @internal @@ -320,49 +320,6 @@ protected: * @internal */ virtual int32_t defaultCenturyStartYear() const; - - private: // default century stuff. - /** - * The system maintains a static default century start date. This is initialized - * the first time it is used. Before then, it is set to SYSTEM_DEFAULT_CENTURY to - * indicate an uninitialized state. Once the system default century date and year - * are set, they do not change. - */ - static UDate fgSystemDefaultCenturyStart; - - /** - * See documentation for systemDefaultCenturyStart. - */ - static int32_t fgSystemDefaultCenturyStartYear; - - /** - * Default value that indicates the defaultCenturyStartYear is unitialized - */ - static const int32_t fgSystemDefaultCenturyYear; - - /** - * start of default century, as a date - */ - static const UDate fgSystemDefaultCentury; - - /** - * Returns the beginning date of the 100-year window that dates - * with 2-digit years are considered to fall within. - */ - UDate internalGetDefaultCenturyStart(void) const; - - /** - * Returns the first year of the 100-year window that dates with - * 2-digit years are considered to fall within. - */ - int32_t internalGetDefaultCenturyStartYear(void) const; - - /** - * Initializes the 100-year window that dates with 2-digit years - * are considered to fall within so that its start date is 80 years - * before the current time. - */ - static void initializeSystemDefaultCentury(void); }; U_NAMESPACE_END diff --git a/deps/icu-small/source/i18n/japancal.cpp b/deps/icu-small/source/i18n/japancal.cpp index a2738e86a9..056781617d 100644 --- a/deps/icu-small/source/i18n/japancal.cpp +++ b/deps/icu-small/source/i18n/japancal.cpp @@ -16,286 +16,88 @@ #include "unicode/utypes.h" #if !UCONFIG_NO_FORMATTING - +#if U_PLATFORM_HAS_WINUWP_API == 0 +#include <stdlib.h> // getenv() is not available in UWP env +#endif #include "cmemory.h" +#include "erarules.h" #include "japancal.h" #include "unicode/gregocal.h" #include "umutex.h" #include "uassert.h" - -//#define U_DEBUG_JCAL - -#ifdef U_DEBUG_JCAL -#include <stdio.h> -#endif +#include "ucln_in.h" +#include "cstring.h" + +static icu::EraRules * gJapaneseEraRules = nullptr; +static icu::UInitOnce gJapaneseEraRulesInitOnce = U_INITONCE_INITIALIZER; +static int32_t gCurrentEra = 0; + +U_CDECL_BEGIN +static UBool japanese_calendar_cleanup(void) { + if (gJapaneseEraRules) { + delete gJapaneseEraRules; + gJapaneseEraRules = nullptr; + } + gCurrentEra = 0; + gJapaneseEraRulesInitOnce.reset(); + return TRUE; +} +U_CDECL_END U_NAMESPACE_BEGIN UOBJECT_DEFINE_RTTI_IMPLEMENTATION(JapaneseCalendar) -// Gregorian date of each emperor's ascension -// Years are AD, months are 1-based. -static const struct { - int16_t year; - int8_t month; - int8_t day; -} kEraInfo[] = { - // Year Month Day - { 645, 6, 19 }, // Taika 0 - { 650, 2, 15 }, // Hakuchi 1 - { 672, 1, 1 }, // Hakuho 2 - { 686, 7, 20 }, // Shucho 3 - { 701, 3, 21 }, // Taiho 4 - { 704, 5, 10 }, // Keiun 5 - { 708, 1, 11 }, // Wado 6 - { 715, 9, 2 }, // Reiki 7 - { 717, 11, 17 }, // Yoro 8 - { 724, 2, 4 }, // Jinki 9 - { 729, 8, 5 }, // Tempyo 10 - { 749, 4, 14 }, // Tempyo-kampo 11 - { 749, 7, 2 }, // Tempyo-shoho 12 - { 757, 8, 18 }, // Tempyo-hoji 13 - { 765, 1, 7 }, // Tempho-jingo 14 - { 767, 8, 16 }, // Jingo-keiun 15 - { 770, 10, 1 }, // Hoki 16 - { 781, 1, 1 }, // Ten-o 17 - { 782, 8, 19 }, // Enryaku 18 - { 806, 5, 18 }, // Daido 19 - { 810, 9, 19 }, // Konin 20 - { 824, 1, 5 }, // Tencho - { 834, 1, 3 }, // Showa - { 848, 6, 13 }, // Kajo - { 851, 4, 28 }, // Ninju - { 854, 11, 30 }, // Saiko - { 857, 2, 21 }, // Tennan - { 859, 4, 15 }, // Jogan - { 877, 4, 16 }, // Genkei - { 885, 2, 21 }, // Ninna - { 889, 4, 27 }, // Kampyo 30 - { 898, 4, 26 }, // Shotai - { 901, 7, 15 }, // Engi - { 923, 4, 11 }, // Encho - { 931, 4, 26 }, // Shohei - { 938, 5, 22 }, // Tengyo - { 947, 4, 22 }, // Tenryaku - { 957, 10, 27 }, // Tentoku - { 961, 2, 16 }, // Owa - { 964, 7, 10 }, // Koho - { 968, 8, 13 }, // Anna 40 - { 970, 3, 25 }, // Tenroku - { 973, 12, 20 }, // Ten-en - { 976, 7, 13 }, // Jogen - { 978, 11, 29 }, // Tengen - { 983, 4, 15 }, // Eikan - { 985, 4, 27 }, // Kanna - { 987, 4, 5 }, // Ei-en - { 989, 8, 8 }, // Eiso - { 990, 11, 7 }, // Shoryaku - { 995, 2, 22 }, // Chotoku 50 - { 999, 1, 13 }, // Choho - { 1004, 7, 20 }, // Kanko - { 1012, 12, 25 }, // Chowa - { 1017, 4, 23 }, // Kannin - { 1021, 2, 2 }, // Jian - { 1024, 7, 13 }, // Manju - { 1028, 7, 25 }, // Chogen - { 1037, 4, 21 }, // Choryaku - { 1040, 11, 10 }, // Chokyu - { 1044, 11, 24 }, // Kantoku 60 - { 1046, 4, 14 }, // Eisho - { 1053, 1, 11 }, // Tengi - { 1058, 8, 29 }, // Kohei - { 1065, 8, 2 }, // Jiryaku - { 1069, 4, 13 }, // Enkyu - { 1074, 8, 23 }, // Shoho - { 1077, 11, 17 }, // Shoryaku - { 1081, 2, 10 }, // Eiho - { 1084, 2, 7 }, // Otoku - { 1087, 4, 7 }, // Kanji 70 - { 1094, 12, 15 }, // Kaho - { 1096, 12, 17 }, // Eicho - { 1097, 11, 21 }, // Shotoku - { 1099, 8, 28 }, // Kowa - { 1104, 2, 10 }, // Choji - { 1106, 4, 9 }, // Kasho - { 1108, 8, 3 }, // Tennin - { 1110, 7, 13 }, // Ten-ei - { 1113, 7, 13 }, // Eikyu - { 1118, 4, 3 }, // Gen-ei 80 - { 1120, 4, 10 }, // Hoan - { 1124, 4, 3 }, // Tenji - { 1126, 1, 22 }, // Daiji - { 1131, 1, 29 }, // Tensho - { 1132, 8, 11 }, // Chosho - { 1135, 4, 27 }, // Hoen - { 1141, 7, 10 }, // Eiji - { 1142, 4, 28 }, // Koji - { 1144, 2, 23 }, // Tenyo - { 1145, 7, 22 }, // Kyuan 90 - { 1151, 1, 26 }, // Ninpei - { 1154, 10, 28 }, // Kyuju - { 1156, 4, 27 }, // Hogen - { 1159, 4, 20 }, // Heiji - { 1160, 1, 10 }, // Eiryaku - { 1161, 9, 4 }, // Oho - { 1163, 3, 29 }, // Chokan - { 1165, 6, 5 }, // Eiman - { 1166, 8, 27 }, // Nin-an - { 1169, 4, 8 }, // Kao 100 - { 1171, 4, 21 }, // Shoan - { 1175, 7, 28 }, // Angen - { 1177, 8, 4 }, // Jisho - { 1181, 7, 14 }, // Yowa - { 1182, 5, 27 }, // Juei - { 1184, 4, 16 }, // Genryuku - { 1185, 8, 14 }, // Bunji - { 1190, 4, 11 }, // Kenkyu - { 1199, 4, 27 }, // Shoji - { 1201, 2, 13 }, // Kennin 110 - { 1204, 2, 20 }, // Genkyu - { 1206, 4, 27 }, // Ken-ei - { 1207, 10, 25 }, // Shogen - { 1211, 3, 9 }, // Kenryaku - { 1213, 12, 6 }, // Kenpo - { 1219, 4, 12 }, // Shokyu - { 1222, 4, 13 }, // Joo - { 1224, 11, 20 }, // Gennin - { 1225, 4, 20 }, // Karoku - { 1227, 12, 10 }, // Antei 120 - { 1229, 3, 5 }, // Kanki - { 1232, 4, 2 }, // Joei - { 1233, 4, 15 }, // Tempuku - { 1234, 11, 5 }, // Bunryaku - { 1235, 9, 19 }, // Katei - { 1238, 11, 23 }, // Ryakunin - { 1239, 2, 7 }, // En-o - { 1240, 7, 16 }, // Ninji - { 1243, 2, 26 }, // Kangen - { 1247, 2, 28 }, // Hoji 130 - { 1249, 3, 18 }, // Kencho - { 1256, 10, 5 }, // Kogen - { 1257, 3, 14 }, // Shoka - { 1259, 3, 26 }, // Shogen - { 1260, 4, 13 }, // Bun-o - { 1261, 2, 20 }, // Kocho - { 1264, 2, 28 }, // Bun-ei - { 1275, 4, 25 }, // Kenji - { 1278, 2, 29 }, // Koan - { 1288, 4, 28 }, // Shoo 140 - { 1293, 8, 55 }, // Einin - { 1299, 4, 25 }, // Shoan - { 1302, 11, 21 }, // Kengen - { 1303, 8, 5 }, // Kagen - { 1306, 12, 14 }, // Tokuji - { 1308, 10, 9 }, // Enkei - { 1311, 4, 28 }, // Ocho - { 1312, 3, 20 }, // Showa - { 1317, 2, 3 }, // Bunpo - { 1319, 4, 28 }, // Geno 150 - { 1321, 2, 23 }, // Genkyo - { 1324, 12, 9 }, // Shochu - { 1326, 4, 26 }, // Kareki - { 1329, 8, 29 }, // Gentoku - { 1331, 8, 9 }, // Genko - { 1334, 1, 29 }, // Kemmu - { 1336, 2, 29 }, // Engen - { 1340, 4, 28 }, // Kokoku - { 1346, 12, 8 }, // Shohei - { 1370, 7, 24 }, // Kentoku 160 - { 1372, 4, 1 }, // Bunch\u0169 - { 1375, 5, 27 }, // Tenju - { 1379, 3, 22 }, // Koryaku - { 1381, 2, 10 }, // Kowa - { 1384, 4, 28 }, // Gench\u0169 - { 1384, 2, 27 }, // Meitoku - { 1387, 8, 23 }, // Kakei - { 1389, 2, 9 }, // Koo - { 1390, 3, 26 }, // Meitoku - { 1394, 7, 5 }, // Oei 170 - { 1428, 4, 27 }, // Shocho - { 1429, 9, 5 }, // Eikyo - { 1441, 2, 17 }, // Kakitsu - { 1444, 2, 5 }, // Bun-an - { 1449, 7, 28 }, // Hotoku - { 1452, 7, 25 }, // Kyotoku - { 1455, 7, 25 }, // Kosho - { 1457, 9, 28 }, // Choroku - { 1460, 12, 21 }, // Kansho - { 1466, 2, 28 }, // Bunsho 180 - { 1467, 3, 3 }, // Onin - { 1469, 4, 28 }, // Bunmei - { 1487, 7, 29 }, // Chokyo - { 1489, 8, 21 }, // Entoku - { 1492, 7, 19 }, // Meio - { 1501, 2, 29 }, // Bunki - { 1504, 2, 30 }, // Eisho - { 1521, 8, 23 }, // Taiei - { 1528, 8, 20 }, // Kyoroku - { 1532, 7, 29 }, // Tenmon 190 - { 1555, 10, 23 }, // Koji - { 1558, 2, 28 }, // Eiroku - { 1570, 4, 23 }, // Genki - { 1573, 7, 28 }, // Tensho - { 1592, 12, 8 }, // Bunroku - { 1596, 10, 27 }, // Keicho - { 1615, 7, 13 }, // Genwa - { 1624, 2, 30 }, // Kan-ei - { 1644, 12, 16 }, // Shoho - { 1648, 2, 15 }, // Keian 200 - { 1652, 9, 18 }, // Shoo - { 1655, 4, 13 }, // Meiryaku - { 1658, 7, 23 }, // Manji - { 1661, 4, 25 }, // Kanbun - { 1673, 9, 21 }, // Enpo - { 1681, 9, 29 }, // Tenwa - { 1684, 2, 21 }, // Jokyo - { 1688, 9, 30 }, // Genroku - { 1704, 3, 13 }, // Hoei - { 1711, 4, 25 }, // Shotoku 210 - { 1716, 6, 22 }, // Kyoho - { 1736, 4, 28 }, // Genbun - { 1741, 2, 27 }, // Kanpo - { 1744, 2, 21 }, // Enkyo - { 1748, 7, 12 }, // Kan-en - { 1751, 10, 27 }, // Horyaku - { 1764, 6, 2 }, // Meiwa - { 1772, 11, 16 }, // An-ei - { 1781, 4, 2 }, // Tenmei - { 1789, 1, 25 }, // Kansei 220 - { 1801, 2, 5 }, // Kyowa - { 1804, 2, 11 }, // Bunka - { 1818, 4, 22 }, // Bunsei - { 1830, 12, 10 }, // Tenpo - { 1844, 12, 2 }, // Koka - { 1848, 2, 28 }, // Kaei - { 1854, 11, 27 }, // Ansei - { 1860, 3, 18 }, // Man-en - { 1861, 2, 19 }, // Bunkyu - { 1864, 2, 20 }, // Genji 230 - { 1865, 4, 7 }, // Keio 231 - { 1868, 9, 8 }, // Meiji 232 - { 1912, 7, 30 }, // Taisho 233 - { 1926, 12, 25 }, // Showa 234 - { 1989, 1, 8 } // Heisei 235 -}; - -#define kEraCount UPRV_LENGTHOF(kEraInfo) - -/** - * The current era, for reference. - */ -static const int32_t kCurrentEra = (kEraCount-1); // int32_t to match the calendar field type - static const int32_t kGregorianEpoch = 1970; // used as the default value of EXTENDED_YEAR +static const char* TENTATIVE_ERA_VAR_NAME = "ICU_ENABLE_TENTATIVE_ERA"; + +// Initialize global Japanese era data +static void U_CALLCONV initializeEras(UErrorCode &status) { + // Although start date of next Japanese era is planned ahead, a name of + // new era might not be available. This implementation allows tester to + // check a new era without era names by settings below (in priority order). + // By default, such tentative era is disabled. + + // 1. Environment variable ICU_ENABLE_TENTATIVE_ERA=true or false + + UBool includeTentativeEra = FALSE; + +#if U_PLATFORM_HAS_WINUWP_API == 1 + // UWP doesn't allow access to getenv(), but we can call GetEnvironmentVariableW to do the same thing. + UChar varName[26] = {}; + u_charsToUChars(TENTATIVE_ERA_VAR_NAME, varName, static_cast<int32_t>(uprv_strlen(TENTATIVE_ERA_VAR_NAME))); + WCHAR varValue[5] = {}; + DWORD ret = GetEnvironmentVariableW(reinterpret_cast<WCHAR*>(varName), varValue, UPRV_LENGTHOF(varValue)); + if ((ret == 4) && (_wcsicmp(varValue, L"true") == 0)) { + includeTentativeEra = TRUE; + } +#else + char *envVarVal = getenv(TENTATIVE_ERA_VAR_NAME); + if (envVarVal != NULL && uprv_stricmp(envVarVal, "true") == 0) { + includeTentativeEra = TRUE; + } +#endif + gJapaneseEraRules = EraRules::createInstance("japanese", includeTentativeEra, status); + if (U_FAILURE(status)) { + return; + } + gCurrentEra = gJapaneseEraRules->getCurrentEraIndex(); +} + +static void init(UErrorCode &status) { + umtx_initOnce(gJapaneseEraRulesInitOnce, &initializeEras, status); + ucln_i18n_registerCleanup(UCLN_I18N_JAPANESE_CALENDAR, japanese_calendar_cleanup); +} /* Some platforms don't like to export constants, like old Palm OS and some z/OS configurations. */ uint32_t JapaneseCalendar::getCurrentEra() { - return kCurrentEra; + return gCurrentEra; } JapaneseCalendar::JapaneseCalendar(const Locale& aLocale, UErrorCode& success) : GregorianCalendar(aLocale, success) { + init(success); setTimeInMillis(getNow(), success); // Call this again now that the vtable is set up properly. } @@ -306,6 +108,9 @@ JapaneseCalendar::~JapaneseCalendar() JapaneseCalendar::JapaneseCalendar(const JapaneseCalendar& source) : GregorianCalendar(source) { + UErrorCode status = U_ZERO_ERROR; + init(status); + U_ASSERT(U_SUCCESS(status)); } JapaneseCalendar& JapaneseCalendar::operator= ( const JapaneseCalendar& right) @@ -332,10 +137,14 @@ int32_t JapaneseCalendar::getDefaultMonthInYear(int32_t eyear) int32_t month = 0; // Find out if we are at the edge of an era - - if(eyear == kEraInfo[era].year) { + int32_t eraStart[3] = { 0,0,0 }; + UErrorCode status = U_ZERO_ERROR; + gJapaneseEraRules->getStartDate(era, eraStart, status); + U_ASSERT(U_SUCCESS(status)); + if(eyear == eraStart[0]) { // Yes, we're in the first year of this era. - return kEraInfo[era].month-1; + return eraStart[1] // month + -1; // return 0-based month } return month; @@ -346,9 +155,13 @@ int32_t JapaneseCalendar::getDefaultDayInMonth(int32_t eyear, int32_t month) int32_t era = internalGetEra(); int32_t day = 1; - if(eyear == kEraInfo[era].year) { - if(month == (kEraInfo[era].month-1)) { - return kEraInfo[era].day; + int32_t eraStart[3] = { 0,0,0 }; + UErrorCode status = U_ZERO_ERROR; + gJapaneseEraRules->getStartDate(era, eraStart, status); + U_ASSERT(U_SUCCESS(status)); + if(eyear == eraStart[0]) { + if(month == eraStart[1] - 1) { + return eraStart[2]; } } @@ -358,7 +171,7 @@ int32_t JapaneseCalendar::getDefaultDayInMonth(int32_t eyear, int32_t month) int32_t JapaneseCalendar::internalGetEra() const { - return internalGet(UCAL_ERA, kCurrentEra); + return internalGet(UCAL_ERA, gCurrentEra); } int32_t JapaneseCalendar::handleGetExtendedYear() @@ -369,12 +182,18 @@ int32_t JapaneseCalendar::handleGetExtendedYear() if (newerField(UCAL_EXTENDED_YEAR, UCAL_YEAR) == UCAL_EXTENDED_YEAR && newerField(UCAL_EXTENDED_YEAR, UCAL_ERA) == UCAL_EXTENDED_YEAR) { - year = internalGet(UCAL_EXTENDED_YEAR, kGregorianEpoch); - } else { - // Subtract one because year starts at 1 - year = internalGet(UCAL_YEAR) + kEraInfo[internalGetEra()].year - 1; - } - return year; + year = internalGet(UCAL_EXTENDED_YEAR, kGregorianEpoch); + } else { + UErrorCode status = U_ZERO_ERROR; + int32_t eraStartYear = gJapaneseEraRules->getStartYear(internalGet(UCAL_ERA, gCurrentEra), status); + U_ASSERT(U_SUCCESS(status)); + + // extended year is a gregorian year, where 1 = 1AD, 0 = 1BC, -1 = 2BC, etc + year = internalGet(UCAL_YEAR, 1) // pin to minimum of year 1 (first year) + + eraStartYear // add gregorian starting year + - 1; // Subtract one because year starts at 1 + } + return year; } @@ -383,79 +202,10 @@ void JapaneseCalendar::handleComputeFields(int32_t julianDay, UErrorCode& status //Calendar::timeToFields(theTime, quick, status); GregorianCalendar::handleComputeFields(julianDay, status); int32_t year = internalGet(UCAL_EXTENDED_YEAR); // Gregorian year + int32_t eraIdx = gJapaneseEraRules->getEraIndex(year, internalGet(UCAL_MONTH) + 1, internalGet(UCAL_DAY_OF_MONTH), status); - int32_t low = 0; - - // Short circuit for recent years. Most modern computations will - // occur in the current era and won't require the binary search. - // Note that if the year is == the current era year, then we use - // the binary search to handle the month/dom comparison. -#ifdef U_DEBUG_JCAL - fprintf(stderr, "== %d \n", year); -#endif - - if (year > kEraInfo[kCurrentEra].year) { - low = kCurrentEra; -#ifdef U_DEBUG_JCAL - fprintf(stderr, " low=%d (special)\n", low); -#endif - } else { - // Binary search - int32_t high = kEraCount; - -#ifdef U_DEBUG_JCAL - fprintf(stderr, " high=%d\n", high); -#endif - while (low < high - 1) { - int32_t i = (low + high) / 2; - int32_t diff = year - kEraInfo[i].year; - -#ifdef U_DEBUG_JCAL - fprintf(stderr, " d=%d low=%d, high=%d. Considering %d:M%d D%d Y%d. { we are ?:M%d D%d Y%d }\n", - diff,low, high, i, kEraInfo[i].month-1, kEraInfo[i].day, kEraInfo[i].year, internalGet(UCAL_MONTH), internalGet(UCAL_DATE),year); -#endif - - // If years are the same, then compare the months, and if those - // are the same, compare days of month. In the ERAS array - // months are 1-based for easier maintenance. - if (diff == 0) { - diff = internalGet(UCAL_MONTH) - (kEraInfo[i].month - 1); -#ifdef U_DEBUG_JCAL - fprintf(stderr, "diff now %d (M) = %d - %d - 1\n", diff, internalGet(UCAL_MONTH), kEraInfo[i].month); -#endif - if (diff == 0) { - diff = internalGet(UCAL_DATE) - kEraInfo[i].day; -#ifdef U_DEBUG_JCAL - fprintf(stderr, "diff now %d (D)\n", diff); -#endif - } - } - if (diff >= 0) { - low = i; - } else { - high = i; - } -#ifdef U_DEBUG_JCAL - fprintf(stderr, ". low=%d, high=%d, i=%d, diff=%d.. %d\n", low, high, i, diff, year); -#endif - - } - } - -#ifdef U_DEBUG_JCAL - fprintf(stderr, " low[era]=%d,.. %d\n", low, year); -#endif - // Now we've found the last era that starts before this date, so - // adjust the year to count from the start of that era. Note that - // all dates before the first era will fall into the first era by - // the algorithm. - - internalSet(UCAL_ERA, low); - internalSet(UCAL_YEAR, year - kEraInfo[low].year + 1); -#ifdef U_DEBUG_JCAL - fprintf(stderr, " Set ERA=%d, year=%d\n", low, year-kEraInfo[low].year+1); -#endif - + internalSet(UCAL_ERA, eraIdx); + internalSet(UCAL_YEAR, year - gJapaneseEraRules->getStartYear(eraIdx, status) + 1); } /* @@ -483,7 +233,7 @@ int32_t JapaneseCalendar::handleGetLimit(UCalendarDateFields field, ELimitType l if (limitType == UCAL_LIMIT_MINIMUM || limitType == UCAL_LIMIT_GREATEST_MINIMUM) { return 0; } - return kCurrentEra; + return gCurrentEra; case UCAL_YEAR: { switch (limitType) { @@ -494,7 +244,12 @@ int32_t JapaneseCalendar::handleGetLimit(UCalendarDateFields field, ELimitType l return 1; case UCAL_LIMIT_COUNT: //added to avoid warning case UCAL_LIMIT_MAXIMUM: - return GregorianCalendar::handleGetLimit(UCAL_YEAR, UCAL_LIMIT_MAXIMUM) - kEraInfo[kCurrentEra].year; + { + UErrorCode status = U_ZERO_ERROR; + int32_t eraStartYear = gJapaneseEraRules->getStartYear(gCurrentEra, status); + U_ASSERT(U_SUCCESS(status)); + return GregorianCalendar::handleGetLimit(UCAL_YEAR, UCAL_LIMIT_MAXIMUM) - eraStartYear; + } default: return 1; // Error condition, invalid limitType } @@ -510,15 +265,18 @@ int32_t JapaneseCalendar::getActualMaximum(UCalendarDateFields field, UErrorCode if (U_FAILURE(status)) { return 0; // error case... any value } - if (era == kCurrentEra) { + if (era == gCurrentEra) { // TODO: Investigate what value should be used here - revisit after 4.0. return handleGetLimit(UCAL_YEAR, UCAL_LIMIT_MAXIMUM); } else { - int32_t nextEraYear = kEraInfo[era + 1].year; - int32_t nextEraMonth = kEraInfo[era + 1].month; - int32_t nextEraDate = kEraInfo[era + 1].day; - - int32_t maxYear = nextEraYear - kEraInfo[era].year + 1; // 1-base + int32_t nextEraStart[3] = { 0,0,0 }; + gJapaneseEraRules->getStartDate(era + 1, nextEraStart, status); + int32_t nextEraYear = nextEraStart[0]; + int32_t nextEraMonth = nextEraStart[1]; // 1-base + int32_t nextEraDate = nextEraStart[2]; + + int32_t eraStartYear = gJapaneseEraRules->getStartYear(era, status); + int32_t maxYear = nextEraYear - eraStartYear + 1; // 1-base if (nextEraMonth == 1 && nextEraDate == 1) { // Subtract 1, because the next era starts at Jan 1 maxYear--; diff --git a/deps/icu-small/source/i18n/japancal.h b/deps/icu-small/source/i18n/japancal.h index 356351d28d..67d2d70315 100644 --- a/deps/icu-small/source/i18n/japancal.h +++ b/deps/icu-small/source/i18n/japancal.h @@ -49,10 +49,18 @@ U_NAMESPACE_BEGIN * July 30, 1912 (Taisho), December 25, 1926 (Showa), and January 7, 1989 (Heisei). Constants * for these eras, suitable for use in the <code>UCAL_ERA</code> field, are provided * in this class. Note that the <em>number</em> used for each era is more or - * less arbitrary. Currently, the era starting in 1053 AD is era #0; however this - * may change in the future as we add more historical data. Use the predefined - * constants rather than using actual, absolute numbers. + * less arbitrary. Currently, the era starting in 645 AD is era #0; however this + * may change in the future. Use the predefined constants rather than using actual, + * absolute numbers. * <p> + * Since ICU4C 63, start date of each era is imported from CLDR. CLDR era data + * may contain tentative era in near future with placeholder names. By default, + * such era data is not enabled. ICU4C users who want to test the behavior of + * the future era can enable this one of following settings (in the priority + * order): + * <ol> + * <li>Environment variable <code>ICU_ENABLE_TENTATIVE_ERA=true</code>.</li> + * </nl> * @internal */ class JapaneseCalendar : public GregorianCalendar { diff --git a/deps/icu-small/source/i18n/listformatter.cpp b/deps/icu-small/source/i18n/listformatter.cpp new file mode 100644 index 0000000000..d919534852 --- /dev/null +++ b/deps/icu-small/source/i18n/listformatter.cpp @@ -0,0 +1,525 @@ +// © 2016 and later: Unicode, Inc. and others. +// License & terms of use: http://www.unicode.org/copyright.html +/* +******************************************************************************* +* +* Copyright (C) 2013-2016, International Business Machines +* Corporation and others. All Rights Reserved. +* +******************************************************************************* +* file name: listformatter.cpp +* encoding: UTF-8 +* tab size: 8 (not used) +* indentation:4 +* +* created on: 2012aug27 +* created by: Umesh P. Nair +*/ + +#include "cmemory.h" +#include "unicode/fpositer.h" // FieldPositionIterator +#include "unicode/listformatter.h" +#include "unicode/simpleformatter.h" +#include "unicode/ulistformatter.h" +#include "fphdlimp.h" +#include "mutex.h" +#include "hash.h" +#include "cstring.h" +#include "uarrsort.h" +#include "ulocimp.h" +#include "charstr.h" +#include "ucln_in.h" +#include "uresimp.h" +#include "resource.h" + +U_NAMESPACE_BEGIN + +struct ListFormatInternal : public UMemory { + SimpleFormatter twoPattern; + SimpleFormatter startPattern; + SimpleFormatter middlePattern; + SimpleFormatter endPattern; + +ListFormatInternal( + const UnicodeString& two, + const UnicodeString& start, + const UnicodeString& middle, + const UnicodeString& end, + UErrorCode &errorCode) : + twoPattern(two, 2, 2, errorCode), + startPattern(start, 2, 2, errorCode), + middlePattern(middle, 2, 2, errorCode), + endPattern(end, 2, 2, errorCode) {} + +ListFormatInternal(const ListFormatData &data, UErrorCode &errorCode) : + twoPattern(data.twoPattern, errorCode), + startPattern(data.startPattern, errorCode), + middlePattern(data.middlePattern, errorCode), + endPattern(data.endPattern, errorCode) { } + +ListFormatInternal(const ListFormatInternal &other) : + twoPattern(other.twoPattern), + startPattern(other.startPattern), + middlePattern(other.middlePattern), + endPattern(other.endPattern) { } +}; + + + +static Hashtable* listPatternHash = nullptr; +static UMutex listFormatterMutex = U_MUTEX_INITIALIZER; +static const char STANDARD_STYLE[] = "standard"; + +U_CDECL_BEGIN +static UBool U_CALLCONV uprv_listformatter_cleanup() { + delete listPatternHash; + listPatternHash = nullptr; + return TRUE; +} + +static void U_CALLCONV +uprv_deleteListFormatInternal(void *obj) { + delete static_cast<ListFormatInternal *>(obj); +} + +U_CDECL_END + +ListFormatter::ListFormatter(const ListFormatter& other) : + owned(other.owned), data(other.data) { + if (other.owned != nullptr) { + owned = new ListFormatInternal(*other.owned); + data = owned; + } +} + +ListFormatter& ListFormatter::operator=(const ListFormatter& other) { + if (this == &other) { + return *this; + } + delete owned; + if (other.owned) { + owned = new ListFormatInternal(*other.owned); + data = owned; + } else { + owned = nullptr; + data = other.data; + } + return *this; +} + +void ListFormatter::initializeHash(UErrorCode& errorCode) { + if (U_FAILURE(errorCode)) { + return; + } + + listPatternHash = new Hashtable(); + if (listPatternHash == nullptr) { + errorCode = U_MEMORY_ALLOCATION_ERROR; + return; + } + + listPatternHash->setValueDeleter(uprv_deleteListFormatInternal); + ucln_i18n_registerCleanup(UCLN_I18N_LIST_FORMATTER, uprv_listformatter_cleanup); + +} + +const ListFormatInternal* ListFormatter::getListFormatInternal( + const Locale& locale, const char *style, UErrorCode& errorCode) { + if (U_FAILURE(errorCode)) { + return nullptr; + } + CharString keyBuffer(locale.getName(), errorCode); + keyBuffer.append(':', errorCode).append(style, errorCode); + UnicodeString key(keyBuffer.data(), -1, US_INV); + ListFormatInternal* result = nullptr; + { + Mutex m(&listFormatterMutex); + if (listPatternHash == nullptr) { + initializeHash(errorCode); + if (U_FAILURE(errorCode)) { + return nullptr; + } + } + result = static_cast<ListFormatInternal*>(listPatternHash->get(key)); + } + if (result != nullptr) { + return result; + } + result = loadListFormatInternal(locale, style, errorCode); + if (U_FAILURE(errorCode)) { + return nullptr; + } + + { + Mutex m(&listFormatterMutex); + ListFormatInternal* temp = static_cast<ListFormatInternal*>(listPatternHash->get(key)); + if (temp != nullptr) { + delete result; + result = temp; + } else { + listPatternHash->put(key, result, errorCode); + if (U_FAILURE(errorCode)) { + return nullptr; + } + } + } + return result; +} + +static const UChar solidus = 0x2F; +static const UChar aliasPrefix[] = { 0x6C,0x69,0x73,0x74,0x50,0x61,0x74,0x74,0x65,0x72,0x6E,0x2F }; // "listPattern/" +enum { + kAliasPrefixLen = UPRV_LENGTHOF(aliasPrefix), + kStyleLenMax = 24 // longest currently is 14 +}; + +struct ListFormatter::ListPatternsSink : public ResourceSink { + UnicodeString two, start, middle, end; +#if ((U_PLATFORM == U_PF_AIX) || (U_PLATFORM == U_PF_OS390)) && (U_CPLUSPLUS_VERSION < 11) + char aliasedStyle[kStyleLenMax+1]; + ListPatternsSink() { + uprv_memset(aliasedStyle, 0, kStyleLenMax+1); + } +#else + char aliasedStyle[kStyleLenMax+1] = {0}; + + ListPatternsSink() {} +#endif + virtual ~ListPatternsSink(); + + void setAliasedStyle(UnicodeString alias) { + int32_t startIndex = alias.indexOf(aliasPrefix, kAliasPrefixLen, 0); + if (startIndex < 0) { + return; + } + startIndex += kAliasPrefixLen; + int32_t endIndex = alias.indexOf(solidus, startIndex); + if (endIndex < 0) { + endIndex = alias.length(); + } + alias.extract(startIndex, endIndex-startIndex, aliasedStyle, kStyleLenMax+1, US_INV); + aliasedStyle[kStyleLenMax] = 0; + } + + void handleValueForPattern(ResourceValue &value, UnicodeString &pattern, UErrorCode &errorCode) { + if (pattern.isEmpty()) { + if (value.getType() == URES_ALIAS) { + if (aliasedStyle[0] == 0) { + setAliasedStyle(value.getAliasUnicodeString(errorCode)); + } + } else { + pattern = value.getUnicodeString(errorCode); + } + } + } + + virtual void put(const char *key, ResourceValue &value, UBool /*noFallback*/, + UErrorCode &errorCode) { + aliasedStyle[0] = 0; + if (value.getType() == URES_ALIAS) { + setAliasedStyle(value.getAliasUnicodeString(errorCode)); + return; + } + ResourceTable listPatterns = value.getTable(errorCode); + for (int i = 0; U_SUCCESS(errorCode) && listPatterns.getKeyAndValue(i, key, value); ++i) { + if (uprv_strcmp(key, "2") == 0) { + handleValueForPattern(value, two, errorCode); + } else if (uprv_strcmp(key, "end") == 0) { + handleValueForPattern(value, end, errorCode); + } else if (uprv_strcmp(key, "middle") == 0) { + handleValueForPattern(value, middle, errorCode); + } else if (uprv_strcmp(key, "start") == 0) { + handleValueForPattern(value, start, errorCode); + } + } + } +}; + +// Virtual destructors must be defined out of line. +ListFormatter::ListPatternsSink::~ListPatternsSink() {} + +ListFormatInternal* ListFormatter::loadListFormatInternal( + const Locale& locale, const char * style, UErrorCode& errorCode) { + UResourceBundle* rb = ures_open(nullptr, locale.getName(), &errorCode); + rb = ures_getByKeyWithFallback(rb, "listPattern", rb, &errorCode); + if (U_FAILURE(errorCode)) { + ures_close(rb); + return nullptr; + } + ListFormatter::ListPatternsSink sink; + char currentStyle[kStyleLenMax+1]; + uprv_strncpy(currentStyle, style, kStyleLenMax); + currentStyle[kStyleLenMax] = 0; + + for (;;) { + ures_getAllItemsWithFallback(rb, currentStyle, sink, errorCode); + if (U_FAILURE(errorCode) || sink.aliasedStyle[0] == 0 || uprv_strcmp(currentStyle, sink.aliasedStyle) == 0) { + break; + } + uprv_strcpy(currentStyle, sink.aliasedStyle); + } + ures_close(rb); + if (U_FAILURE(errorCode)) { + return nullptr; + } + if (sink.two.isEmpty() || sink.start.isEmpty() || sink.middle.isEmpty() || sink.end.isEmpty()) { + errorCode = U_MISSING_RESOURCE_ERROR; + return nullptr; + } + ListFormatInternal* result = new ListFormatInternal(sink.two, sink.start, sink.middle, sink.end, errorCode); + if (result == nullptr) { + errorCode = U_MEMORY_ALLOCATION_ERROR; + return nullptr; + } + if (U_FAILURE(errorCode)) { + delete result; + return nullptr; + } + return result; +} + +ListFormatter* ListFormatter::createInstance(UErrorCode& errorCode) { + Locale locale; // The default locale. + return createInstance(locale, errorCode); +} + +ListFormatter* ListFormatter::createInstance(const Locale& locale, UErrorCode& errorCode) { + return createInstance(locale, STANDARD_STYLE, errorCode); +} + +ListFormatter* ListFormatter::createInstance(const Locale& locale, const char *style, UErrorCode& errorCode) { + const ListFormatInternal* listFormatInternal = getListFormatInternal(locale, style, errorCode); + if (U_FAILURE(errorCode)) { + return nullptr; + } + ListFormatter* p = new ListFormatter(listFormatInternal); + if (p == nullptr) { + errorCode = U_MEMORY_ALLOCATION_ERROR; + return nullptr; + } + return p; +} + +ListFormatter::ListFormatter(const ListFormatData& listFormatData, UErrorCode &errorCode) { + owned = new ListFormatInternal(listFormatData, errorCode); + data = owned; +} + +ListFormatter::ListFormatter(const ListFormatInternal* listFormatterInternal) : owned(nullptr), data(listFormatterInternal) { +} + +ListFormatter::~ListFormatter() { + delete owned; +} + +/** + * Joins first and second using the pattern pat. + * On entry offset is an offset into first or -1 if offset unspecified. + * On exit offset is offset of second in result if recordOffset was set + * Otherwise if it was >=0 it is set to point into result where it used + * to point into first. On exit, result is the join of first and second + * according to pat. Any previous value of result gets replaced. + */ +static void joinStringsAndReplace( + const SimpleFormatter& pat, + const UnicodeString& first, + const UnicodeString& second, + UnicodeString &result, + UBool recordOffset, + int32_t &offset, + int32_t *offsetFirst, + int32_t *offsetSecond, + UErrorCode& errorCode) { + if (U_FAILURE(errorCode)) { + return; + } + const UnicodeString *params[2] = {&first, &second}; + int32_t offsets[2]; + pat.formatAndReplace( + params, + UPRV_LENGTHOF(params), + result, + offsets, + UPRV_LENGTHOF(offsets), + errorCode); + if (U_FAILURE(errorCode)) { + return; + } + if (offsets[0] == -1 || offsets[1] == -1) { + errorCode = U_INVALID_FORMAT_ERROR; + return; + } + if (recordOffset) { + offset = offsets[1]; + } else if (offset >= 0) { + offset += offsets[0]; + } + if (offsetFirst != nullptr) *offsetFirst = offsets[0]; + if (offsetSecond != nullptr) *offsetSecond = offsets[1]; +} + +UnicodeString& ListFormatter::format( + const UnicodeString items[], + int32_t nItems, + UnicodeString& appendTo, + UErrorCode& errorCode) const { + int32_t offset; + return format(items, nItems, appendTo, -1, offset, errorCode); +} + +#if !UCONFIG_NO_FORMATTING +UnicodeString& ListFormatter::format( + const UnicodeString items[], + int32_t nItems, + UnicodeString & appendTo, + FieldPositionIterator* posIter, + UErrorCode& errorCode) const { + int32_t offset; + FieldPositionIteratorHandler handler(posIter, errorCode); + return format_(items, nItems, appendTo, -1, offset, &handler, errorCode); +}; +#endif + +UnicodeString& ListFormatter::format( + const UnicodeString items[], + int32_t nItems, + UnicodeString& appendTo, + int32_t index, + int32_t &offset, + UErrorCode& errorCode) const { + return format_(items, nItems, appendTo, index, offset, nullptr, errorCode); +} + +UnicodeString& ListFormatter::format_( + const UnicodeString items[], + int32_t nItems, + UnicodeString& appendTo, + int32_t index, + int32_t &offset, + FieldPositionHandler* handler, + UErrorCode& errorCode) const { +#if !UCONFIG_NO_FORMATTING + offset = -1; + if (U_FAILURE(errorCode)) { + return appendTo; + } + if (data == nullptr) { + errorCode = U_INVALID_STATE_ERROR; + return appendTo; + } + + if (nItems <= 0) { + return appendTo; + } + if (nItems == 1) { + if (index == 0) { + offset = appendTo.length(); + } + if (handler != nullptr) { + handler->addAttribute(ULISTFMT_ELEMENT_FIELD, + appendTo.length(), + appendTo.length() + items[0].length()); + } + appendTo.append(items[0]); + return appendTo; + } + UnicodeString result(items[0]); + if (index == 0) { + offset = 0; + } + int32_t offsetFirst; + int32_t offsetSecond; + int32_t prefixLength = 0; + // for n items, there are 2 * (n + 1) boundary including 0 and the upper + // edge. + MaybeStackArray<int32_t, 10> offsets((handler != nullptr) ? 2 * (nItems + 1): 0); + joinStringsAndReplace( + nItems == 2 ? data->twoPattern : data->startPattern, + result, + items[1], + result, + index == 1, + offset, + &offsetFirst, + &offsetSecond, + errorCode); + if (handler != nullptr) { + offsets[0] = 0; + prefixLength += offsetFirst; + offsets[1] = offsetSecond - prefixLength; + } + if (nItems > 2) { + for (int32_t i = 2; i < nItems - 1; ++i) { + joinStringsAndReplace( + data->middlePattern, + result, + items[i], + result, + index == i, + offset, + &offsetFirst, + &offsetSecond, + errorCode); + if (handler != nullptr) { + prefixLength += offsetFirst; + offsets[i] = offsetSecond - prefixLength; + } + } + joinStringsAndReplace( + data->endPattern, + result, + items[nItems - 1], + result, + index == nItems - 1, + offset, + &offsetFirst, + &offsetSecond, + errorCode); + if (handler != nullptr) { + prefixLength += offsetFirst; + offsets[nItems - 1] = offsetSecond - prefixLength; + } + } + if (handler != nullptr) { + // If there are already some data in appendTo, we need to adjust the index + // by shifting that lenght while insert into handler. + int32_t shift = appendTo.length() + prefixLength; + // Output the ULISTFMT_ELEMENT_FIELD in the order of the input elements + for (int32_t i = 0; i < nItems; ++i) { + offsets[i + nItems] = offsets[i] + items[i].length() + shift; + offsets[i] += shift; + handler->addAttribute( + ULISTFMT_ELEMENT_FIELD, // id + offsets[i], // index + offsets[i + nItems]); // limit + } + // The locale pattern may reorder the items (such as in ur-IN locale), + // so we cannot assume the array is in accendning order. + // To handle the edging case, just insert the two ends into the array + // and sort. Then we output ULISTFMT_LITERAL_FIELD if the indecies + // between the even and odd position are not the same in the sorted array. + offsets[2 * nItems] = shift - prefixLength; + offsets[2 * nItems + 1] = result.length() + shift - prefixLength; + uprv_sortArray(offsets.getAlias(), 2 * (nItems + 1), sizeof(int32_t), + uprv_int32Comparator, nullptr, + false, &errorCode); + for (int32_t i = 0; i <= nItems; ++i) { + if (offsets[i * 2] != offsets[i * 2 + 1]) { + handler->addAttribute( + ULISTFMT_LITERAL_FIELD, // id + offsets[i * 2], // index + offsets[i * 2 + 1]); // limit + } + } + } + if (U_SUCCESS(errorCode)) { + if (offset >= 0) { + offset += appendTo.length(); + } + appendTo += result; + } +#endif + return appendTo; +} + +U_NAMESPACE_END diff --git a/deps/icu-small/source/i18n/measfmt.cpp b/deps/icu-small/source/i18n/measfmt.cpp index 996a20c2e0..03974ff4b7 100644 --- a/deps/icu-small/source/i18n/measfmt.cpp +++ b/deps/icu-small/source/i18n/measfmt.cpp @@ -47,7 +47,7 @@ U_NAMESPACE_BEGIN static constexpr int32_t PER_UNIT_INDEX = StandardPlural::COUNT; static constexpr int32_t PATTERN_COUNT = PER_UNIT_INDEX + 1; -static constexpr int32_t MEAS_UNIT_COUNT = 138; // see assertion in MeasureFormatCacheData constructor +static constexpr int32_t MEAS_UNIT_COUNT = 142; // see assertion in MeasureFormatCacheData constructor static constexpr int32_t WIDTH_INDEX_COUNT = UMEASFMT_WIDTH_NARROW + 1; UOBJECT_DEFINE_RTTI_IMPLEMENTATION(MeasureFormat) @@ -618,7 +618,7 @@ MeasureFormat::MeasureFormat( : cache(NULL), numberFormat(NULL), pluralRules(NULL), - width(w), + fWidth(w), listFormatter(NULL) { initMeasureFormat(locale, w, NULL, status); } @@ -631,7 +631,7 @@ MeasureFormat::MeasureFormat( : cache(NULL), numberFormat(NULL), pluralRules(NULL), - width(w), + fWidth(w), listFormatter(NULL) { initMeasureFormat(locale, w, nfToAdopt, status); } @@ -641,7 +641,7 @@ MeasureFormat::MeasureFormat(const MeasureFormat &other) : cache(other.cache), numberFormat(other.numberFormat), pluralRules(other.pluralRules), - width(other.width), + fWidth(other.fWidth), listFormatter(NULL) { cache->addRef(); numberFormat->addRef(); @@ -659,7 +659,7 @@ MeasureFormat &MeasureFormat::operator=(const MeasureFormat &other) { SharedObject::copyPtr(other.cache, cache); SharedObject::copyPtr(other.numberFormat, numberFormat); SharedObject::copyPtr(other.pluralRules, pluralRules); - width = other.width; + fWidth = other.fWidth; delete listFormatter; if (other.listFormatter != NULL) { listFormatter = new ListFormatter(*other.listFormatter); @@ -673,7 +673,7 @@ MeasureFormat::MeasureFormat() : cache(NULL), numberFormat(NULL), pluralRules(NULL), - width(UMEASFMT_WIDTH_SHORT), + fWidth(UMEASFMT_WIDTH_SHORT), listFormatter(NULL) { } @@ -703,7 +703,7 @@ UBool MeasureFormat::operator==(const Format &other) const { // don't have to check it here. // differing widths aren't equivalent - if (width != rhs.width) { + if (fWidth != rhs.fWidth) { return FALSE; } // Width the same check locales. @@ -805,7 +805,7 @@ UnicodeString &MeasureFormat::formatMeasures( if (measureCount == 1) { return formatMeasure(measures[0], **numberFormat, appendTo, pos, status); } - if (width == UMEASFMT_WIDTH_NUMERIC) { + if (fWidth == UMEASFMT_WIDTH_NUMERIC) { Formattable hms[3]; int32_t bitMap = toHMS(measures, measureCount, hms, status); if (bitMap > 0) { @@ -839,7 +839,7 @@ UnicodeString &MeasureFormat::formatMeasures( } UnicodeString MeasureFormat::getUnitDisplayName(const MeasureUnit& unit, UErrorCode& /*status*/) const { - UMeasureFormatWidth width = getRegularWidth(this->width); + UMeasureFormatWidth width = getRegularWidth(fWidth); const UChar* const* styleToDnam = cache->dnams[unit.getIndex()]; const UChar* dnam = styleToDnam[width]; if (dnam == NULL) { @@ -895,11 +895,11 @@ void MeasureFormat::initMeasureFormat( return; } } - width = w; + fWidth = w; delete listFormatter; listFormatter = ListFormatter::createInstance( locale, - listStyles[getRegularWidth(width)], + listStyles[getRegularWidth(fWidth)], status); } @@ -922,7 +922,7 @@ UBool MeasureFormat::setMeasureFormatLocale(const Locale &locale, UErrorCode &st if (U_FAILURE(status) || locale == getLocale(status)) { return FALSE; } - initMeasureFormat(locale, width, NULL, status); + initMeasureFormat(locale, fWidth, NULL, status); return U_SUCCESS(status); } @@ -956,7 +956,7 @@ UnicodeString &MeasureFormat::formatMeasure( if (isCurrency(amtUnit)) { UChar isoCode[4]; u_charsToUChars(amtUnit.getSubtype(), isoCode, 4); - return cache->getCurrencyFormat(width)->format( + return cache->getCurrencyFormat(fWidth)->format( new CurrencyAmount(amtNumber, isoCode, status), appendTo, pos, @@ -965,7 +965,7 @@ UnicodeString &MeasureFormat::formatMeasure( UnicodeString formattedNumber; StandardPlural::Form pluralForm = QuantityFormatter::selectPlural( amtNumber, nf, **pluralRules, formattedNumber, pos, status); - const SimpleFormatter *formatter = getPluralFormatter(amtUnit, width, pluralForm, status); + const SimpleFormatter *formatter = getPluralFormatter(amtUnit, fWidth, pluralForm, status); return QuantityFormatter::format(*formatter, formattedNumber, appendTo, pos, status); } @@ -1016,7 +1016,6 @@ UnicodeString &MeasureFormat::formatNumeric( return appendTo; break; } - return appendTo; } static void appendRange( @@ -1173,7 +1172,7 @@ int32_t MeasureFormat::withPerUnitAndAppend( if (U_FAILURE(status)) { return offset; } - const SimpleFormatter *perUnitFormatter = getFormatterOrNull(perUnit, width, PER_UNIT_INDEX); + const SimpleFormatter *perUnitFormatter = getFormatterOrNull(perUnit, fWidth, PER_UNIT_INDEX); if (perUnitFormatter != NULL) { const UnicodeString *params[] = {&formatted}; perUnitFormatter->formatAndAppend( @@ -1185,9 +1184,9 @@ int32_t MeasureFormat::withPerUnitAndAppend( status); return offset; } - const SimpleFormatter *perFormatter = getPerFormatter(width, status); + const SimpleFormatter *perFormatter = getPerFormatter(fWidth, status); const SimpleFormatter *pattern = - getPluralFormatter(perUnit, width, StandardPlural::ONE, status); + getPluralFormatter(perUnit, fWidth, StandardPlural::ONE, status); if (U_FAILURE(status)) { return offset; } diff --git a/deps/icu-small/source/i18n/measunit.cpp b/deps/icu-small/source/i18n/measunit.cpp index dc156aa720..f6059f8c6d 100644 --- a/deps/icu-small/source/i18n/measunit.cpp +++ b/deps/icu-small/source/i18n/measunit.cpp @@ -39,23 +39,23 @@ static const int32_t gOffsets[] = { 2, 7, 16, - 20, - 24, - 321, - 331, - 342, - 346, - 352, - 356, - 376, - 377, - 388, - 391, - 397, + 22, + 26, + 325, + 336, + 347, + 351, + 357, + 361, + 381, + 382, + 393, + 396, 402, - 406, - 410, - 435 + 408, + 412, + 416, + 441 }; static const int32_t gIndexes[] = { @@ -63,23 +63,23 @@ static const int32_t gIndexes[] = { 2, 7, 16, - 20, - 24, - 24, - 34, - 45, - 49, - 55, - 59, - 79, - 80, - 91, + 22, + 26, + 26, + 37, + 48, + 52, + 58, + 62, + 82, + 83, 94, - 100, - 105, + 97, + 103, 109, 113, - 138 + 117, + 142 }; // Must be sorted alphabetically. @@ -128,6 +128,8 @@ static const char * const gSubTypes[] = { "milligram-per-deciliter", "millimole-per-liter", "part-per-million", + "percent", + "permille", "liter-per-100kilometers", "liter-per-kilometer", "mile-per-gallon", @@ -388,9 +390,11 @@ static const char * const gSubTypes[] = { "UYN", "UYP", "UYU", + "UYW", "UZS", "VEB", "VEF", + "VES", "VNC", "VND", "VUV", @@ -437,6 +441,7 @@ static const char * const gSubTypes[] = { "kilobyte", "megabit", "megabyte", + "petabyte", "terabit", "terabyte", "century", @@ -505,6 +510,7 @@ static const char * const gSubTypes[] = { "megawatt", "milliwatt", "watt", + "atmosphere", "hectopascal", "inch-hg", "millibar", @@ -547,14 +553,14 @@ static const char * const gSubTypes[] = { // Must be sorted by first value and then second value. static int32_t unitPerUnitToSingleUnit[][4] = { - {363, 333, 17, 0}, - {365, 339, 17, 2}, - {367, 333, 17, 3}, - {367, 424, 4, 2}, - {367, 425, 4, 3}, - {382, 422, 3, 1}, - {385, 11, 16, 4}, - {427, 363, 4, 1} + {368, 338, 17, 0}, + {370, 344, 17, 2}, + {372, 338, 17, 3}, + {372, 430, 4, 2}, + {372, 431, 4, 3}, + {387, 428, 3, 1}, + {390, 11, 16, 5}, + {433, 368, 4, 1} }; // Shortcuts to the base unit in order to make the default constructor fast @@ -641,6 +647,14 @@ MeasureUnit *MeasureUnit::createPartPerMillion(UErrorCode &status) { return MeasureUnit::create(3, 3, status); } +MeasureUnit *MeasureUnit::createPercent(UErrorCode &status) { + return MeasureUnit::create(3, 4, status); +} + +MeasureUnit *MeasureUnit::createPermille(UErrorCode &status) { + return MeasureUnit::create(3, 5, status); +} + MeasureUnit *MeasureUnit::createLiterPer100Kilometers(UErrorCode &status) { return MeasureUnit::create(4, 0, status); } @@ -689,14 +703,18 @@ MeasureUnit *MeasureUnit::createMegabyte(UErrorCode &status) { return MeasureUnit::create(6, 7, status); } -MeasureUnit *MeasureUnit::createTerabit(UErrorCode &status) { +MeasureUnit *MeasureUnit::createPetabyte(UErrorCode &status) { return MeasureUnit::create(6, 8, status); } -MeasureUnit *MeasureUnit::createTerabyte(UErrorCode &status) { +MeasureUnit *MeasureUnit::createTerabit(UErrorCode &status) { return MeasureUnit::create(6, 9, status); } +MeasureUnit *MeasureUnit::createTerabyte(UErrorCode &status) { + return MeasureUnit::create(6, 10, status); +} + MeasureUnit *MeasureUnit::createCentury(UErrorCode &status) { return MeasureUnit::create(7, 0, status); } @@ -949,26 +967,30 @@ MeasureUnit *MeasureUnit::createWatt(UErrorCode &status) { return MeasureUnit::create(15, 5, status); } -MeasureUnit *MeasureUnit::createHectopascal(UErrorCode &status) { +MeasureUnit *MeasureUnit::createAtmosphere(UErrorCode &status) { return MeasureUnit::create(16, 0, status); } -MeasureUnit *MeasureUnit::createInchHg(UErrorCode &status) { +MeasureUnit *MeasureUnit::createHectopascal(UErrorCode &status) { return MeasureUnit::create(16, 1, status); } -MeasureUnit *MeasureUnit::createMillibar(UErrorCode &status) { +MeasureUnit *MeasureUnit::createInchHg(UErrorCode &status) { return MeasureUnit::create(16, 2, status); } -MeasureUnit *MeasureUnit::createMillimeterOfMercury(UErrorCode &status) { +MeasureUnit *MeasureUnit::createMillibar(UErrorCode &status) { return MeasureUnit::create(16, 3, status); } -MeasureUnit *MeasureUnit::createPoundPerSquareInch(UErrorCode &status) { +MeasureUnit *MeasureUnit::createMillimeterOfMercury(UErrorCode &status) { return MeasureUnit::create(16, 4, status); } +MeasureUnit *MeasureUnit::createPoundPerSquareInch(UErrorCode &status) { + return MeasureUnit::create(16, 5, status); +} + MeasureUnit *MeasureUnit::createKilometerPerHour(UErrorCode &status) { return MeasureUnit::create(17, 0, status); } diff --git a/deps/icu-small/source/i18n/msgfmt.cpp b/deps/icu-small/source/i18n/msgfmt.cpp index 8b3807e671..8ff86a2cac 100644 --- a/deps/icu-small/source/i18n/msgfmt.cpp +++ b/deps/icu-small/source/i18n/msgfmt.cpp @@ -1078,7 +1078,7 @@ void MessageFormat::format(int32_t msgStart, const void *plNumber, // that formats the number without subtracting the offset. appendTo.formatAndAppend(pluralNumber.formatter, *arg, success); } - } else if ((formatter = getCachedFormatter(i -2))) { + } else if ((formatter = getCachedFormatter(i -2)) != 0) { // Handles all ArgType.SIMPLE, and formatters from setFormat() and its siblings. if (dynamic_cast<const ChoiceFormat*>(formatter) || dynamic_cast<const PluralFormat*>(formatter) || diff --git a/deps/icu-small/source/i18n/nfrule.cpp b/deps/icu-small/source/i18n/nfrule.cpp index 9f5deb3168..b5e7892d5e 100644 --- a/deps/icu-small/source/i18n/nfrule.cpp +++ b/deps/icu-small/source/i18n/nfrule.cpp @@ -39,14 +39,14 @@ NFRule::NFRule(const RuleBasedNumberFormat* _rbnf, const UnicodeString &_ruleTex , radix(10) , exponent(0) , decimalPoint(0) - , ruleText(_ruleText) + , fRuleText(_ruleText) , sub1(NULL) , sub2(NULL) , formatter(_rbnf) , rulePatternFormat(NULL) { - if (!ruleText.isEmpty()) { - parseRuleDescriptor(ruleText, status); + if (!fRuleText.isEmpty()) { + parseRuleDescriptor(fRuleText, status); } } @@ -122,7 +122,7 @@ NFRule::makeRules(UnicodeString& description, status = U_MEMORY_ALLOCATION_ERROR; return; } - description = rule1->ruleText; + description = rule1->fRuleText; // check the description to see whether there's text enclosed // in brackets @@ -314,7 +314,7 @@ NFRule::parseRuleDescriptor(UnicodeString& description, UErrorCode& status) if (c == gSlash) { val = 0; ++p; - int64_t ll_10 = 10; + ll_10 = 10; while (p < descriptorLength) { c = descriptor.charAt(p); if (c >= gZero && c <= gNine) { @@ -418,7 +418,7 @@ NFRule::extractSubstitutions(const NFRuleSet* ruleSet, if (U_FAILURE(status)) { return; } - this->ruleText = ruleText; + fRuleText = ruleText; sub1 = extractSubstitution(ruleSet, predecessor, status); if (sub1 == NULL) { // Small optimization. There is no need to create a redundant NullSubstitution. @@ -427,15 +427,15 @@ NFRule::extractSubstitutions(const NFRuleSet* ruleSet, else { sub2 = extractSubstitution(ruleSet, predecessor, status); } - int32_t pluralRuleStart = this->ruleText.indexOf(gDollarOpenParenthesis, -1, 0); - int32_t pluralRuleEnd = (pluralRuleStart >= 0 ? this->ruleText.indexOf(gClosedParenthesisDollar, -1, pluralRuleStart) : -1); + int32_t pluralRuleStart = fRuleText.indexOf(gDollarOpenParenthesis, -1, 0); + int32_t pluralRuleEnd = (pluralRuleStart >= 0 ? fRuleText.indexOf(gClosedParenthesisDollar, -1, pluralRuleStart) : -1); if (pluralRuleEnd >= 0) { - int32_t endType = this->ruleText.indexOf(gComma, pluralRuleStart); + int32_t endType = fRuleText.indexOf(gComma, pluralRuleStart); if (endType < 0) { status = U_PARSE_ERROR; return; } - UnicodeString type(this->ruleText.tempSubString(pluralRuleStart + 2, endType - pluralRuleStart - 2)); + UnicodeString type(fRuleText.tempSubString(pluralRuleStart + 2, endType - pluralRuleStart - 2)); UPluralType pluralType; if (type.startsWith(UNICODE_STRING_SIMPLE("cardinal"))) { pluralType = UPLURAL_TYPE_CARDINAL; @@ -448,7 +448,7 @@ NFRule::extractSubstitutions(const NFRuleSet* ruleSet, return; } rulePatternFormat = formatter->createPluralFormat(pluralType, - this->ruleText.tempSubString(endType + 1, pluralRuleEnd - endType - 1), status); + fRuleText.tempSubString(endType + 1, pluralRuleEnd - endType - 1), status); } } @@ -484,16 +484,16 @@ NFRule::extractSubstitution(const NFRuleSet* ruleSet, // special-case the ">>>" token, since searching for the > at the // end will actually find the > in the middle - if (ruleText.indexOf(gGreaterGreaterGreater, 3, 0) == subStart) { + if (fRuleText.indexOf(gGreaterGreaterGreater, 3, 0) == subStart) { subEnd = subStart + 2; // otherwise the substitution token ends with the same character // it began with } else { - UChar c = ruleText.charAt(subStart); - subEnd = ruleText.indexOf(c, subStart + 1); + UChar c = fRuleText.charAt(subStart); + subEnd = fRuleText.indexOf(c, subStart + 1); // special case for '<%foo<<' - if (c == gLessThan && subEnd != -1 && subEnd < ruleText.length() - 1 && ruleText.charAt(subEnd+1) == c) { + if (c == gLessThan && subEnd != -1 && subEnd < fRuleText.length() - 1 && fRuleText.charAt(subEnd+1) == c) { // ordinals use "=#,##0==%abbrev=" as their rule. Notice that the '==' in the middle // occurs because of the juxtaposition of two different rules. The check for '<' is a hack // to get around this. Having the duplicate at the front would cause problems with @@ -513,12 +513,12 @@ NFRule::extractSubstitution(const NFRuleSet* ruleSet, // some text bounded by substitution token characters). Use // makeSubstitution() to create the right kind of substitution UnicodeString subToken; - subToken.setTo(ruleText, subStart, subEnd + 1 - subStart); + subToken.setTo(fRuleText, subStart, subEnd + 1 - subStart); result = NFSubstitution::makeSubstitution(subStart, this, predecessor, ruleSet, this->formatter, subToken, status); // remove the substitution from the rule text - ruleText.removeBetween(subStart, subEnd+1); + fRuleText.removeBetween(subStart, subEnd+1); return result; } @@ -601,7 +601,7 @@ NFRule::indexOfAnyRulePrefix() const { int result = -1; for (int i = 0; RULE_PREFIXES[i]; i++) { - int32_t pos = ruleText.indexOf(*RULE_PREFIXES[i]); + int32_t pos = fRuleText.indexOf(*RULE_PREFIXES[i]); if (pos != -1 && (result == -1 || pos < result)) { result = pos; } @@ -637,7 +637,7 @@ NFRule::operator==(const NFRule& rhs) const return baseValue == rhs.baseValue && radix == rhs.radix && exponent == rhs.exponent - && ruleText == rhs.ruleText + && fRuleText == rhs.fRuleText && util_equalSubstitutions(sub1, rhs.sub1) && util_equalSubstitutions(sub2, rhs.sub2); } @@ -690,14 +690,14 @@ NFRule::_appendRuleText(UnicodeString& result) const // if the rule text begins with a space, write an apostrophe // (whitespace after the rule descriptor is ignored; the // apostrophe is used to make the whitespace significant) - if (ruleText.charAt(0) == gSpace && (sub1 == NULL || sub1->getPos() != 0)) { + if (fRuleText.charAt(0) == gSpace && (sub1 == NULL || sub1->getPos() != 0)) { result.append(gTick); } // now, write the rule's rule text, inserting appropriate // substitution tokens in the appropriate places UnicodeString ruleTextCopy; - ruleTextCopy.setTo(ruleText); + ruleTextCopy.setTo(fRuleText); UnicodeString temp; if (sub2 != NULL) { @@ -743,24 +743,24 @@ NFRule::doFormat(int64_t number, UnicodeString& toInsertInto, int32_t pos, int32 // into the right places in toInsertInto (notice we do the // substitutions in reverse order so that the offsets don't get // messed up) - int32_t pluralRuleStart = ruleText.length(); + int32_t pluralRuleStart = fRuleText.length(); int32_t lengthOffset = 0; if (!rulePatternFormat) { - toInsertInto.insert(pos, ruleText); + toInsertInto.insert(pos, fRuleText); } else { - pluralRuleStart = ruleText.indexOf(gDollarOpenParenthesis, -1, 0); - int pluralRuleEnd = ruleText.indexOf(gClosedParenthesisDollar, -1, pluralRuleStart); + pluralRuleStart = fRuleText.indexOf(gDollarOpenParenthesis, -1, 0); + int pluralRuleEnd = fRuleText.indexOf(gClosedParenthesisDollar, -1, pluralRuleStart); int initialLength = toInsertInto.length(); - if (pluralRuleEnd < ruleText.length() - 1) { - toInsertInto.insert(pos, ruleText.tempSubString(pluralRuleEnd + 2)); + if (pluralRuleEnd < fRuleText.length() - 1) { + toInsertInto.insert(pos, fRuleText.tempSubString(pluralRuleEnd + 2)); } toInsertInto.insert(pos, rulePatternFormat->format((int32_t)(number/util64_pow(radix, exponent)), status)); if (pluralRuleStart > 0) { - toInsertInto.insert(pos, ruleText.tempSubString(0, pluralRuleStart)); + toInsertInto.insert(pos, fRuleText.tempSubString(0, pluralRuleStart)); } - lengthOffset = ruleText.length() - (toInsertInto.length() - initialLength); + lengthOffset = fRuleText.length() - (toInsertInto.length() - initialLength); } if (sub2 != NULL) { @@ -789,17 +789,17 @@ NFRule::doFormat(double number, UnicodeString& toInsertInto, int32_t pos, int32_ // [again, we have two copies of this routine that do the same thing // so that we don't sacrifice precision in a long by casting it // to a double] - int32_t pluralRuleStart = ruleText.length(); + int32_t pluralRuleStart = fRuleText.length(); int32_t lengthOffset = 0; if (!rulePatternFormat) { - toInsertInto.insert(pos, ruleText); + toInsertInto.insert(pos, fRuleText); } else { - pluralRuleStart = ruleText.indexOf(gDollarOpenParenthesis, -1, 0); - int pluralRuleEnd = ruleText.indexOf(gClosedParenthesisDollar, -1, pluralRuleStart); + pluralRuleStart = fRuleText.indexOf(gDollarOpenParenthesis, -1, 0); + int pluralRuleEnd = fRuleText.indexOf(gClosedParenthesisDollar, -1, pluralRuleStart); int initialLength = toInsertInto.length(); - if (pluralRuleEnd < ruleText.length() - 1) { - toInsertInto.insert(pos, ruleText.tempSubString(pluralRuleEnd + 2)); + if (pluralRuleEnd < fRuleText.length() - 1) { + toInsertInto.insert(pos, fRuleText.tempSubString(pluralRuleEnd + 2)); } double pluralVal = number; if (0 <= pluralVal && pluralVal < 1) { @@ -812,9 +812,9 @@ NFRule::doFormat(double number, UnicodeString& toInsertInto, int32_t pos, int32_ } toInsertInto.insert(pos, rulePatternFormat->format((int32_t)(pluralVal), status)); if (pluralRuleStart > 0) { - toInsertInto.insert(pos, ruleText.tempSubString(0, pluralRuleStart)); + toInsertInto.insert(pos, fRuleText.tempSubString(0, pluralRuleStart)); } - lengthOffset = ruleText.length() - (toInsertInto.length() - initialLength); + lengthOffset = fRuleText.length() - (toInsertInto.length() - initialLength); } if (sub2 != NULL) { @@ -908,15 +908,15 @@ NFRule::doParse(const UnicodeString& text, ParsePosition pp; UnicodeString workText(text); - int32_t sub1Pos = sub1 != NULL ? sub1->getPos() : ruleText.length(); - int32_t sub2Pos = sub2 != NULL ? sub2->getPos() : ruleText.length(); + int32_t sub1Pos = sub1 != NULL ? sub1->getPos() : fRuleText.length(); + int32_t sub2Pos = sub2 != NULL ? sub2->getPos() : fRuleText.length(); // check to see whether the text before the first substitution // matches the text at the beginning of the string being // parsed. If it does, strip that off the front of workText; // otherwise, dump out with a mismatch UnicodeString prefix; - prefix.setTo(ruleText, 0, sub1Pos); + prefix.setTo(fRuleText, 0, sub1Pos); #ifdef RBNF_DEBUG fprintf(stderr, "doParse %p ", this); @@ -1000,7 +1000,7 @@ NFRule::doParse(const UnicodeString& text, // the substitution, giving us a new partial parse result pp.setIndex(0); - temp.setTo(ruleText, sub1Pos, sub2Pos - sub1Pos); + temp.setTo(fRuleText, sub1Pos, sub2Pos - sub1Pos); double partialResult = matchToDelimiter(workText, start, tempBaseValue, temp, pp, sub1, nonNumericalExecutedRuleMask, @@ -1021,7 +1021,7 @@ NFRule::doParse(const UnicodeString& text, // partial result with whatever it gets back from its // substitution if there's a successful match, giving us // a real result - temp.setTo(ruleText, sub2Pos, ruleText.length() - sub2Pos); + temp.setTo(fRuleText, sub2Pos, fRuleText.length() - sub2Pos); partialResult = matchToDelimiter(workText2, 0, partialResult, temp, pp2, sub2, nonNumericalExecutedRuleMask, @@ -1039,18 +1039,18 @@ NFRule::doParse(const UnicodeString& text, else { // commented out because ParsePosition doesn't have error index in 1.1.x // restored for ICU4C port - int32_t temp = pp2.getErrorIndex() + sub1Pos + pp.getIndex(); - if (temp> parsePosition.getErrorIndex()) { - parsePosition.setErrorIndex(temp); + int32_t i_temp = pp2.getErrorIndex() + sub1Pos + pp.getIndex(); + if (i_temp> parsePosition.getErrorIndex()) { + parsePosition.setErrorIndex(i_temp); } } } else { // commented out because ParsePosition doesn't have error index in 1.1.x // restored for ICU4C port - int32_t temp = sub1Pos + pp.getErrorIndex(); - if (temp > parsePosition.getErrorIndex()) { - parsePosition.setErrorIndex(temp); + int32_t i_temp = sub1Pos + pp.getErrorIndex(); + if (i_temp > parsePosition.getErrorIndex()) { + parsePosition.setErrorIndex(i_temp); } } // keep trying to match things until the outer matchToDelimiter() @@ -1483,11 +1483,11 @@ NFRule::findText(const UnicodeString& str, rulePatternFormat->parseType(str, this, result, position); int start = position.getBeginIndex(); if (start >= 0) { - int32_t pluralRuleStart = ruleText.indexOf(gDollarOpenParenthesis, -1, 0); - int32_t pluralRuleSuffix = ruleText.indexOf(gClosedParenthesisDollar, -1, pluralRuleStart) + 2; + int32_t pluralRuleStart = fRuleText.indexOf(gDollarOpenParenthesis, -1, 0); + int32_t pluralRuleSuffix = fRuleText.indexOf(gClosedParenthesisDollar, -1, pluralRuleStart) + 2; int32_t matchLen = position.getEndIndex() - start; - UnicodeString prefix(ruleText.tempSubString(0, pluralRuleStart)); - UnicodeString suffix(ruleText.tempSubString(pluralRuleSuffix)); + UnicodeString prefix(fRuleText.tempSubString(0, pluralRuleStart)); + UnicodeString suffix(fRuleText.tempSubString(pluralRuleSuffix)); if (str.compare(start - prefix.length(), prefix.length(), prefix, 0, prefix.length()) == 0 && str.compare(start + matchLen, suffix.length(), suffix, 0, suffix.length()) == 0) { diff --git a/deps/icu-small/source/i18n/nfrule.h b/deps/icu-small/source/i18n/nfrule.h index 843a4a0762..2b030390ea 100644 --- a/deps/icu-small/source/i18n/nfrule.h +++ b/deps/icu-small/source/i18n/nfrule.h @@ -109,7 +109,7 @@ private: int32_t radix; int16_t exponent; UChar decimalPoint; - UnicodeString ruleText; + UnicodeString fRuleText; NFSubstitution* sub1; NFSubstitution* sub2; const RuleBasedNumberFormat* formatter; diff --git a/deps/icu-small/source/i18n/number_compact.cpp b/deps/icu-small/source/i18n/number_compact.cpp index 40278e1a01..10942c35f5 100644 --- a/deps/icu-small/source/i18n/number_compact.cpp +++ b/deps/icu-small/source/i18n/number_compact.cpp @@ -273,13 +273,13 @@ void CompactHandler::processQuantity(DecimalQuantity &quantity, MicroProps &micr if (U_FAILURE(status)) { return; } // Treat zero as if it had magnitude 0 - int magnitude; + int32_t magnitude; if (quantity.isZero()) { magnitude = 0; micros.rounder.apply(quantity, status); } else { // TODO: Revisit chooseMultiplierAndApply - int multiplier = micros.rounder.chooseMultiplierAndApply(quantity, data, status); + int32_t multiplier = micros.rounder.chooseMultiplierAndApply(quantity, data, status); magnitude = quantity.isZero() ? 0 : quantity.getMagnitude(); magnitude -= multiplier; } diff --git a/deps/icu-small/source/i18n/number_decimalquantity.cpp b/deps/icu-small/source/i18n/number_decimalquantity.cpp index 9d80e3349c..2c4182b1c6 100644 --- a/deps/icu-small/source/i18n/number_decimalquantity.cpp +++ b/deps/icu-small/source/i18n/number_decimalquantity.cpp @@ -1154,8 +1154,31 @@ const char16_t* DecimalQuantity::checkHealth() const { } bool DecimalQuantity::operator==(const DecimalQuantity& other) const { - // FIXME: Make a faster implementation. - return toString() == other.toString(); + bool basicEquals = + scale == other.scale + && precision == other.precision + && flags == other.flags + && lOptPos == other.lOptPos + && lReqPos == other.lReqPos + && rReqPos == other.rReqPos + && rOptPos == other.rOptPos + && isApproximate == other.isApproximate; + if (!basicEquals) { + return false; + } + + if (precision == 0) { + return true; + } else if (isApproximate) { + return origDouble == other.origDouble && origDelta == other.origDelta; + } else { + for (int m = getUpperDisplayMagnitude(); m >= getLowerDisplayMagnitude(); m--) { + if (getDigit(m) != other.getDigit(m)) { + return false; + } + } + return true; + } } UnicodeString DecimalQuantity::toString() const { diff --git a/deps/icu-small/source/i18n/number_decimfmtprops.cpp b/deps/icu-small/source/i18n/number_decimfmtprops.cpp index 6754fe19ec..12fe7060e2 100644 --- a/deps/icu-small/source/i18n/number_decimfmtprops.cpp +++ b/deps/icu-small/source/i18n/number_decimfmtprops.cpp @@ -15,6 +15,7 @@ using namespace icu::number::impl; namespace { +alignas(DecimalFormatProperties) char kRawDefaultProperties[sizeof(DecimalFormatProperties)]; icu::UInitOnce gDefaultPropertiesInitOnce = U_INITONCE_INITIALIZER; diff --git a/deps/icu-small/source/i18n/number_fluent.cpp b/deps/icu-small/source/i18n/number_fluent.cpp index 687adb6b5b..a66e3bd0f2 100644 --- a/deps/icu-small/source/i18n/number_fluent.cpp +++ b/deps/icu-small/source/i18n/number_fluent.cpp @@ -363,6 +363,7 @@ UnlocalizedNumberFormatter::UnlocalizedNumberFormatter(const NFS<UNF>& other) // No additional fields to assign } +// Make default copy constructor call the NumberFormatterSettings copy constructor. UnlocalizedNumberFormatter::UnlocalizedNumberFormatter(UNF&& src) U_NOEXCEPT : UNF(static_cast<NFS<UNF>&&>(src)) {} @@ -383,6 +384,7 @@ UnlocalizedNumberFormatter& UnlocalizedNumberFormatter::operator=(UNF&& src) U_N return *this; } +// Make default copy constructor call the NumberFormatterSettings copy constructor. LocalizedNumberFormatter::LocalizedNumberFormatter(const LNF& other) : LNF(static_cast<const NFS<LNF>&>(other)) {} @@ -405,7 +407,8 @@ LocalizedNumberFormatter::LocalizedNumberFormatter(NFS<LNF>&& src) U_NOEXCEPT LocalizedNumberFormatter& LocalizedNumberFormatter::operator=(const LNF& other) { NFS<LNF>::operator=(static_cast<const NFS<LNF>&>(other)); - // No additional fields to assign (let call count and compiled formatter reset to defaults) + // Reset to default values. + clear(); return *this; } @@ -417,20 +420,26 @@ LocalizedNumberFormatter& LocalizedNumberFormatter::operator=(LNF&& src) U_NOEXC // Formatter is compiled lnfMoveHelper(static_cast<LNF&&>(src)); } else { - // Reset to default values. - auto* callCount = reinterpret_cast<u_atomic_int32_t*>(fUnsafeCallCount); - umtx_storeRelease(*callCount, 0); - fCompiled = nullptr; + clear(); } return *this; } +void LocalizedNumberFormatter::clear() { + // Reset to default values. + auto* callCount = reinterpret_cast<u_atomic_int32_t*>(fUnsafeCallCount); + umtx_storeRelease(*callCount, 0); + delete fCompiled; + fCompiled = nullptr; +} + void LocalizedNumberFormatter::lnfMoveHelper(LNF&& src) { // Copy over the compiled formatter and set call count to INT32_MIN as in computeCompiled(). // Don't copy the call count directly because doing so requires a loadAcquire/storeRelease. // The bits themselves appear to be platform-dependent, so copying them might not be safe. auto* callCount = reinterpret_cast<u_atomic_int32_t*>(fUnsafeCallCount); umtx_storeRelease(*callCount, INT32_MIN); + delete fCompiled; fCompiled = src.fCompiled; // Reset the source object to leave it in a safe state. auto* srcCallCount = reinterpret_cast<u_atomic_int32_t*>(src.fUnsafeCallCount); @@ -657,9 +666,9 @@ LocalizedNumberFormatter::formatDecimalQuantity(const DecimalQuantity& dq, UErro void LocalizedNumberFormatter::formatImpl(impl::UFormattedNumberData* results, UErrorCode& status) const { if (computeCompiled(status)) { - fCompiled->apply(results->quantity, results->string, status); + fCompiled->format(results->quantity, results->string, status); } else { - NumberFormatterImpl::applyStatic(fMacros, results->quantity, results->string, status); + NumberFormatterImpl::formatStatic(fMacros, results->quantity, results->string, status); } } @@ -706,7 +715,11 @@ bool LocalizedNumberFormatter::computeCompiled(UErrorCode& status) const { if (currentCount == fMacros.threshold && fMacros.threshold > 0) { // Build the data structure and then use it (slow to fast path). - const NumberFormatterImpl* compiled = NumberFormatterImpl::fromMacros(fMacros, status); + const NumberFormatterImpl* compiled = new NumberFormatterImpl(fMacros, status); + if (compiled == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + return false; + } U_ASSERT(fCompiled == nullptr); const_cast<LocalizedNumberFormatter*>(this)->fCompiled = compiled; umtx_storeRelease(*callCount, INT32_MIN); @@ -776,7 +789,7 @@ Appendable& FormattedNumber::appendTo(Appendable& appendable) { return appendTo(appendable, localStatus); } -Appendable& FormattedNumber::appendTo(Appendable& appendable, UErrorCode& status) { +Appendable& FormattedNumber::appendTo(Appendable& appendable, UErrorCode& status) const { if (U_FAILURE(status)) { return appendable; } diff --git a/deps/icu-small/source/i18n/number_formatimpl.cpp b/deps/icu-small/source/i18n/number_formatimpl.cpp index 3f887128bc..60c18ee284 100644 --- a/deps/icu-small/source/i18n/number_formatimpl.cpp +++ b/deps/icu-small/source/i18n/number_formatimpl.cpp @@ -67,14 +67,18 @@ getCurrencyFormatInfo(const Locale& locale, const char* isoCode, UErrorCode& sta MicroPropsGenerator::~MicroPropsGenerator() = default; -NumberFormatterImpl* NumberFormatterImpl::fromMacros(const MacroProps& macros, UErrorCode& status) { - return new NumberFormatterImpl(macros, true, status); +NumberFormatterImpl::NumberFormatterImpl(const MacroProps& macros, UErrorCode& status) + : NumberFormatterImpl(macros, true, status) { } -void NumberFormatterImpl::applyStatic(const MacroProps& macros, DecimalQuantity& inValue, - NumberStringBuilder& outString, UErrorCode& status) { +int32_t NumberFormatterImpl::formatStatic(const MacroProps& macros, DecimalQuantity& inValue, + NumberStringBuilder& outString, UErrorCode& status) { NumberFormatterImpl impl(macros, false, status); - impl.applyUnsafe(inValue, outString, status); + MicroProps& micros = impl.preProcessUnsafe(inValue, status); + if (U_FAILURE(status)) { return 0; } + int32_t length = writeNumber(micros, inValue, outString, 0, status); + length += writeAffixes(micros, outString, 0, length, status); + return length; } int32_t NumberFormatterImpl::getPrefixSuffixStatic(const MacroProps& macros, int8_t signum, @@ -89,22 +93,40 @@ int32_t NumberFormatterImpl::getPrefixSuffixStatic(const MacroProps& macros, int // The "unsafe" method simply re-uses fMicros, eliminating the extra copy operation. // See MicroProps::processQuantity() for details. -void NumberFormatterImpl::apply(DecimalQuantity& inValue, NumberStringBuilder& outString, +int32_t NumberFormatterImpl::format(DecimalQuantity& inValue, NumberStringBuilder& outString, UErrorCode& status) const { - if (U_FAILURE(status)) { return; } MicroProps micros; - if (!fMicroPropsGenerator) { return; } - fMicroPropsGenerator->processQuantity(inValue, micros, status); - if (U_FAILURE(status)) { return; } - microsToString(micros, inValue, outString, status); + preProcess(inValue, micros, status); + if (U_FAILURE(status)) { return 0; } + int32_t length = writeNumber(micros, inValue, outString, 0, status); + length += writeAffixes(micros, outString, 0, length, status); + return length; } -void NumberFormatterImpl::applyUnsafe(DecimalQuantity& inValue, NumberStringBuilder& outString, - UErrorCode& status) { +void NumberFormatterImpl::preProcess(DecimalQuantity& inValue, MicroProps& microsOut, + UErrorCode& status) const { if (U_FAILURE(status)) { return; } + if (fMicroPropsGenerator == nullptr) { + status = U_INTERNAL_PROGRAM_ERROR; + return; + } + fMicroPropsGenerator->processQuantity(inValue, microsOut, status); + microsOut.rounder.apply(inValue, status); + microsOut.integerWidth.apply(inValue, status); +} + +MicroProps& NumberFormatterImpl::preProcessUnsafe(DecimalQuantity& inValue, UErrorCode& status) { + if (U_FAILURE(status)) { + return fMicros; // must always return a value + } + if (fMicroPropsGenerator == nullptr) { + status = U_INTERNAL_PROGRAM_ERROR; + return fMicros; // must always return a value + } fMicroPropsGenerator->processQuantity(inValue, fMicros, status); - if (U_FAILURE(status)) { return; } - microsToString(fMicros, inValue, outString, status); + fMicros.rounder.apply(inValue, status); + fMicros.integerWidth.apply(inValue, status); + return fMicros; } int32_t NumberFormatterImpl::getPrefixSuffix(int8_t signum, StandardPlural::Form plural, @@ -115,7 +137,7 @@ int32_t NumberFormatterImpl::getPrefixSuffix(int8_t signum, StandardPlural::Form const Modifier* modifier = fImmutablePatternModifier->getModifier(signum, plural); modifier->apply(outString, 0, 0, status); if (U_FAILURE(status)) { return 0; } - return modifier->getPrefixLength(status); + return modifier->getPrefixLength(); } int32_t NumberFormatterImpl::getPrefixSuffixUnsafe(int8_t signum, StandardPlural::Form plural, @@ -126,7 +148,7 @@ int32_t NumberFormatterImpl::getPrefixSuffixUnsafe(int8_t signum, StandardPlural fPatternModifier->setNumberProperties(signum, plural); fPatternModifier->apply(outString, 0, 0, status); if (U_FAILURE(status)) { return 0; } - return fPatternModifier->getPrefixLength(status); + return fPatternModifier->getPrefixLength(); } NumberFormatterImpl::NumberFormatterImpl(const MacroProps& macros, bool safe, UErrorCode& status) { @@ -344,25 +366,23 @@ NumberFormatterImpl::macrosToMicroGenerator(const MacroProps& macros, bool safe, // Outer modifier (CLDR units and currency long names) if (isCldrUnit) { fLongNameHandler.adoptInstead( - new LongNameHandler( - LongNameHandler::forMeasureUnit( - macros.locale, - macros.unit, - macros.perUnit, - unitWidth, - resolvePluralRules(macros.rules, macros.locale, status), - chain, - status))); + LongNameHandler::forMeasureUnit( + macros.locale, + macros.unit, + macros.perUnit, + unitWidth, + resolvePluralRules(macros.rules, macros.locale, status), + chain, + status)); chain = fLongNameHandler.getAlias(); } else if (isCurrency && unitWidth == UNUM_UNIT_WIDTH_FULL_NAME) { fLongNameHandler.adoptInstead( - new LongNameHandler( - LongNameHandler::forCurrencyLongNames( - macros.locale, - currency, - resolvePluralRules(macros.rules, macros.locale, status), - chain, - status))); + LongNameHandler::forCurrencyLongNames( + macros.locale, + currency, + resolvePluralRules(macros.rules, macros.locale, status), + chain, + status)); chain = fLongNameHandler.getAlias(); } else { // No outer modifier required @@ -404,50 +424,46 @@ NumberFormatterImpl::resolvePluralRules(const PluralRules* rulesPtr, const Local return fRules.getAlias(); } -int32_t NumberFormatterImpl::microsToString(const MicroProps& micros, DecimalQuantity& quantity, - NumberStringBuilder& string, UErrorCode& status) { - micros.rounder.apply(quantity, status); - micros.integerWidth.apply(quantity, status); - int32_t length = writeNumber(micros, quantity, string, status); - // NOTE: When range formatting is added, these modifiers can bubble up. - // For now, apply them all here at once. +int32_t NumberFormatterImpl::writeAffixes(const MicroProps& micros, NumberStringBuilder& string, + int32_t start, int32_t end, UErrorCode& status) { // Always apply the inner modifier (which is "strong"). - length += micros.modInner->apply(string, 0, length, status); + int32_t length = micros.modInner->apply(string, start, end, status); if (micros.padding.isValid()) { length += micros.padding - .padAndApply(*micros.modMiddle, *micros.modOuter, string, 0, length, status); + .padAndApply(*micros.modMiddle, *micros.modOuter, string, start, length + end, status); } else { - length += micros.modMiddle->apply(string, 0, length, status); - length += micros.modOuter->apply(string, 0, length, status); + length += micros.modMiddle->apply(string, start, length + end, status); + length += micros.modOuter->apply(string, start, length + end, status); } return length; } int32_t NumberFormatterImpl::writeNumber(const MicroProps& micros, DecimalQuantity& quantity, - NumberStringBuilder& string, UErrorCode& status) { + NumberStringBuilder& string, int32_t index, + UErrorCode& status) { int32_t length = 0; if (quantity.isInfinite()) { length += string.insert( - length, + length + index, micros.symbols->getSymbol(DecimalFormatSymbols::ENumberFormatSymbol::kInfinitySymbol), UNUM_INTEGER_FIELD, status); } else if (quantity.isNaN()) { length += string.insert( - length, + length + index, micros.symbols->getSymbol(DecimalFormatSymbols::ENumberFormatSymbol::kNaNSymbol), UNUM_INTEGER_FIELD, status); } else { // Add the integer digits - length += writeIntegerDigits(micros, quantity, string, status); + length += writeIntegerDigits(micros, quantity, string, length + index, status); // Add the decimal point if (quantity.getLowerDisplayMagnitude() < 0 || micros.decimal == UNUM_DECIMAL_SEPARATOR_ALWAYS) { length += string.insert( - length, + length + index, micros.useCurrency ? micros.symbols->getSymbol( DecimalFormatSymbols::ENumberFormatSymbol::kMonetarySeparatorSymbol) : micros .symbols @@ -458,21 +474,22 @@ int32_t NumberFormatterImpl::writeNumber(const MicroProps& micros, DecimalQuanti } // Add the fraction digits - length += writeFractionDigits(micros, quantity, string, status); + length += writeFractionDigits(micros, quantity, string, length + index, status); } return length; } int32_t NumberFormatterImpl::writeIntegerDigits(const MicroProps& micros, DecimalQuantity& quantity, - NumberStringBuilder& string, UErrorCode& status) { + NumberStringBuilder& string, int32_t index, + UErrorCode& status) { int length = 0; int integerCount = quantity.getUpperDisplayMagnitude() + 1; for (int i = 0; i < integerCount; i++) { // Add grouping separator if (micros.grouping.groupAtPosition(i, quantity)) { length += string.insert( - 0, + index, micros.useCurrency ? micros.symbols->getSymbol( DecimalFormatSymbols::ENumberFormatSymbol::kMonetaryGroupingSeparatorSymbol) : micros.symbols->getSymbol( @@ -484,20 +501,21 @@ int32_t NumberFormatterImpl::writeIntegerDigits(const MicroProps& micros, Decima // Get and append the next digit value int8_t nextDigit = quantity.getDigit(i); length += utils::insertDigitFromSymbols( - string, 0, nextDigit, *micros.symbols, UNUM_INTEGER_FIELD, status); + string, index, nextDigit, *micros.symbols, UNUM_INTEGER_FIELD, status); } return length; } int32_t NumberFormatterImpl::writeFractionDigits(const MicroProps& micros, DecimalQuantity& quantity, - NumberStringBuilder& string, UErrorCode& status) { + NumberStringBuilder& string, int32_t index, + UErrorCode& status) { int length = 0; int fractionCount = -quantity.getLowerDisplayMagnitude(); for (int i = 0; i < fractionCount; i++) { // Get and append the next digit value int8_t nextDigit = quantity.getDigit(-i - 1); length += utils::insertDigitFromSymbols( - string, string.length(), nextDigit, *micros.symbols, UNUM_FRACTION_FIELD, status); + string, length + index, nextDigit, *micros.symbols, UNUM_FRACTION_FIELD, status); } return length; } diff --git a/deps/icu-small/source/i18n/number_formatimpl.h b/deps/icu-small/source/i18n/number_formatimpl.h index 744fecec13..fda38c9284 100644 --- a/deps/icu-small/source/i18n/number_formatimpl.h +++ b/deps/icu-small/source/i18n/number_formatimpl.h @@ -29,14 +29,14 @@ class NumberFormatterImpl : public UMemory { * Builds a "safe" MicroPropsGenerator, which is thread-safe and can be used repeatedly. * The caller owns the returned NumberFormatterImpl. */ - static NumberFormatterImpl *fromMacros(const MacroProps ¯os, UErrorCode &status); + NumberFormatterImpl(const MacroProps ¯os, UErrorCode &status); /** * Builds and evaluates an "unsafe" MicroPropsGenerator, which is cheaper but can be used only once. */ - static void - applyStatic(const MacroProps ¯os, DecimalQuantity &inValue, NumberStringBuilder &outString, - UErrorCode &status); + static int32_t + formatStatic(const MacroProps ¯os, DecimalQuantity &inValue, NumberStringBuilder &outString, + UErrorCode &status); /** * Prints only the prefix and suffix; used for DecimalFormat getters. @@ -51,7 +51,12 @@ class NumberFormatterImpl : public UMemory { /** * Evaluates the "safe" MicroPropsGenerator created by "fromMacros". */ - void apply(DecimalQuantity& inValue, NumberStringBuilder& outString, UErrorCode& status) const; + int32_t format(DecimalQuantity& inValue, NumberStringBuilder& outString, UErrorCode& status) const; + + /** + * Like format(), but saves the result into an output MicroProps without additional processing. + */ + void preProcess(DecimalQuantity& inValue, MicroProps& microsOut, UErrorCode& status) const; /** * Like getPrefixSuffixStatic() but uses the safe compiled object. @@ -59,6 +64,19 @@ class NumberFormatterImpl : public UMemory { int32_t getPrefixSuffix(int8_t signum, StandardPlural::Form plural, NumberStringBuilder& outString, UErrorCode& status) const; + /** + * Synthesizes the output string from a MicroProps and DecimalQuantity. + * This method formats only the main number, not affixes. + */ + static int32_t writeNumber(const MicroProps& micros, DecimalQuantity& quantity, + NumberStringBuilder& string, int32_t index, UErrorCode& status); + + /** + * Adds the affixes. Intended to be called immediately after formatNumber. + */ + static int32_t writeAffixes(const MicroProps& micros, NumberStringBuilder& string, int32_t start, + int32_t end, UErrorCode& status); + private: // Head of the MicroPropsGenerator linked list: const MicroPropsGenerator *fMicroPropsGenerator = nullptr; @@ -85,7 +103,7 @@ class NumberFormatterImpl : public UMemory { NumberFormatterImpl(const MacroProps ¯os, bool safe, UErrorCode &status); - void applyUnsafe(DecimalQuantity &inValue, NumberStringBuilder &outString, UErrorCode &status); + MicroProps& preProcessUnsafe(DecimalQuantity &inValue, UErrorCode &status); int32_t getPrefixSuffixUnsafe(int8_t signum, StandardPlural::Form plural, NumberStringBuilder& outString, UErrorCode& status); @@ -113,31 +131,13 @@ class NumberFormatterImpl : public UMemory { const MicroPropsGenerator * macrosToMicroGenerator(const MacroProps ¯os, bool safe, UErrorCode &status); - /** - * Synthesizes the output string from a MicroProps and DecimalQuantity. - * - * @param micros - * The MicroProps after the quantity has been consumed. Will not be mutated. - * @param quantity - * The DecimalQuantity to be rendered. May be mutated. - * @param string - * The output string. Will be mutated. - */ - static int32_t - microsToString(const MicroProps µs, DecimalQuantity &quantity, NumberStringBuilder &string, - UErrorCode &status); - - static int32_t - writeNumber(const MicroProps µs, DecimalQuantity &quantity, NumberStringBuilder &string, - UErrorCode &status); - static int32_t writeIntegerDigits(const MicroProps µs, DecimalQuantity &quantity, NumberStringBuilder &string, - UErrorCode &status); + int32_t index, UErrorCode &status); static int32_t writeFractionDigits(const MicroProps µs, DecimalQuantity &quantity, NumberStringBuilder &string, - UErrorCode &status); + int32_t index, UErrorCode &status); }; } // namespace impl diff --git a/deps/icu-small/source/i18n/number_grouping.cpp b/deps/icu-small/source/i18n/number_grouping.cpp index 4a1cceb499..da32cca99a 100644 --- a/deps/icu-small/source/i18n/number_grouping.cpp +++ b/deps/icu-small/source/i18n/number_grouping.cpp @@ -80,7 +80,7 @@ void Grouper::setLocaleData(const impl::ParsedPatternInfo &patternInfo, const Lo if (fMinGrouping == -2) { fMinGrouping = getMinGroupingForLocale(locale); } else if (fMinGrouping == -3) { - fMinGrouping = uprv_max(2, getMinGroupingForLocale(locale)); + fMinGrouping = static_cast<int16_t>(uprv_max(2, getMinGroupingForLocale(locale))); } else { // leave fMinGrouping alone } diff --git a/deps/icu-small/source/i18n/number_longnames.cpp b/deps/icu-small/source/i18n/number_longnames.cpp index 26f9af4c9b..fd8e8d381a 100644 --- a/deps/icu-small/source/i18n/number_longnames.cpp +++ b/deps/icu-small/source/i18n/number_longnames.cpp @@ -39,7 +39,7 @@ static int32_t getIndex(const char* pluralKeyword, UErrorCode& status) { static UnicodeString getWithPlural( const UnicodeString* strings, - int32_t plural, + StandardPlural::Form plural, UErrorCode& status) { UnicodeString result = strings[plural]; if (result.isBogus()) { @@ -156,7 +156,7 @@ UnicodeString getPerUnitFormat(const Locale& locale, const UNumberUnitWidth &wid } // namespace -LongNameHandler +LongNameHandler* LongNameHandler::forMeasureUnit(const Locale &loc, const MeasureUnit &unitRef, const MeasureUnit &perUnit, const UNumberUnitWidth &width, const PluralRules *rules, const MicroPropsGenerator *parent, UErrorCode &status) { @@ -173,20 +173,28 @@ LongNameHandler::forMeasureUnit(const Locale &loc, const MeasureUnit &unitRef, c } } - LongNameHandler result(rules, parent); + auto* result = new LongNameHandler(rules, parent); + if (result == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + return nullptr; + } UnicodeString simpleFormats[ARRAY_LENGTH]; getMeasureData(loc, unit, width, simpleFormats, status); if (U_FAILURE(status)) { return result; } // TODO: What field to use for units? - simpleFormatsToModifiers(simpleFormats, UNUM_FIELD_COUNT, result.fModifiers, status); + result->simpleFormatsToModifiers(simpleFormats, UNUM_FIELD_COUNT, status); return result; } -LongNameHandler +LongNameHandler* LongNameHandler::forCompoundUnit(const Locale &loc, const MeasureUnit &unit, const MeasureUnit &perUnit, const UNumberUnitWidth &width, const PluralRules *rules, const MicroPropsGenerator *parent, UErrorCode &status) { - LongNameHandler result(rules, parent); + auto* result = new LongNameHandler(rules, parent); + if (result == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + return nullptr; + } UnicodeString primaryData[ARRAY_LENGTH]; getMeasureData(loc, unit, width, primaryData, status); if (U_FAILURE(status)) { return result; } @@ -213,46 +221,52 @@ LongNameHandler::forCompoundUnit(const Locale &loc, const MeasureUnit &unit, con if (U_FAILURE(status)) { return result; } } // TODO: What field to use for units? - multiSimpleFormatsToModifiers(primaryData, perUnitFormat, UNUM_FIELD_COUNT, result.fModifiers, status); + result->multiSimpleFormatsToModifiers(primaryData, perUnitFormat, UNUM_FIELD_COUNT, status); return result; } -LongNameHandler LongNameHandler::forCurrencyLongNames(const Locale &loc, const CurrencyUnit ¤cy, +LongNameHandler* LongNameHandler::forCurrencyLongNames(const Locale &loc, const CurrencyUnit ¤cy, const PluralRules *rules, const MicroPropsGenerator *parent, UErrorCode &status) { - LongNameHandler result(rules, parent); + auto* result = new LongNameHandler(rules, parent); + if (result == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + return nullptr; + } UnicodeString simpleFormats[ARRAY_LENGTH]; getCurrencyLongNameData(loc, currency, simpleFormats, status); - if (U_FAILURE(status)) { return result; } - simpleFormatsToModifiers(simpleFormats, UNUM_CURRENCY_FIELD, result.fModifiers, status); + if (U_FAILURE(status)) { return nullptr; } + result->simpleFormatsToModifiers(simpleFormats, UNUM_CURRENCY_FIELD, status); return result; } void LongNameHandler::simpleFormatsToModifiers(const UnicodeString *simpleFormats, Field field, - SimpleModifier *output, UErrorCode &status) { + UErrorCode &status) { for (int32_t i = 0; i < StandardPlural::Form::COUNT; i++) { - UnicodeString simpleFormat = getWithPlural(simpleFormats, i, status); + StandardPlural::Form plural = static_cast<StandardPlural::Form>(i); + UnicodeString simpleFormat = getWithPlural(simpleFormats, plural, status); if (U_FAILURE(status)) { return; } SimpleFormatter compiledFormatter(simpleFormat, 0, 1, status); if (U_FAILURE(status)) { return; } - output[i] = SimpleModifier(compiledFormatter, field, false); + fModifiers[i] = SimpleModifier(compiledFormatter, field, false, {this, 0, plural}); } } void LongNameHandler::multiSimpleFormatsToModifiers(const UnicodeString *leadFormats, UnicodeString trailFormat, - Field field, SimpleModifier *output, UErrorCode &status) { + Field field, UErrorCode &status) { SimpleFormatter trailCompiled(trailFormat, 1, 1, status); if (U_FAILURE(status)) { return; } for (int32_t i = 0; i < StandardPlural::Form::COUNT; i++) { - UnicodeString leadFormat = getWithPlural(leadFormats, i, status); + StandardPlural::Form plural = static_cast<StandardPlural::Form>(i); + UnicodeString leadFormat = getWithPlural(leadFormats, plural, status); if (U_FAILURE(status)) { return; } UnicodeString compoundFormat; trailCompiled.format(leadFormat, compoundFormat, status); if (U_FAILURE(status)) { return; } SimpleFormatter compoundCompiled(compoundFormat, 0, 1, status); if (U_FAILURE(status)) { return; } - output[i] = SimpleModifier(compoundCompiled, field, false); + fModifiers[i] = SimpleModifier(compoundCompiled, field, false, {this, 0, plural}); } } @@ -265,4 +279,8 @@ void LongNameHandler::processQuantity(DecimalQuantity &quantity, MicroProps &mic micros.modOuter = &fModifiers[utils::getStandardPlural(rules, copy)]; } +const Modifier* LongNameHandler::getModifier(int8_t /*signum*/, StandardPlural::Form plural) const { + return &fModifiers[plural]; +} + #endif /* #if !UCONFIG_NO_FORMATTING */ diff --git a/deps/icu-small/source/i18n/number_longnames.h b/deps/icu-small/source/i18n/number_longnames.h index 1d1e7dd3e8..a71d0caadf 100644 --- a/deps/icu-small/source/i18n/number_longnames.h +++ b/deps/icu-small/source/i18n/number_longnames.h @@ -14,13 +14,13 @@ U_NAMESPACE_BEGIN namespace number { namespace impl { -class LongNameHandler : public MicroPropsGenerator, public UMemory { +class LongNameHandler : public MicroPropsGenerator, public ModifierStore, public UMemory { public: - static LongNameHandler + static LongNameHandler* forCurrencyLongNames(const Locale &loc, const CurrencyUnit ¤cy, const PluralRules *rules, const MicroPropsGenerator *parent, UErrorCode &status); - static LongNameHandler + static LongNameHandler* forMeasureUnit(const Locale &loc, const MeasureUnit &unit, const MeasureUnit &perUnit, const UNumberUnitWidth &width, const PluralRules *rules, const MicroPropsGenerator *parent, UErrorCode &status); @@ -28,6 +28,8 @@ class LongNameHandler : public MicroPropsGenerator, public UMemory { void processQuantity(DecimalQuantity &quantity, MicroProps µs, UErrorCode &status) const U_OVERRIDE; + const Modifier* getModifier(int8_t signum, StandardPlural::Form plural) const U_OVERRIDE; + private: SimpleModifier fModifiers[StandardPlural::Form::COUNT]; const PluralRules *rules; @@ -36,15 +38,14 @@ class LongNameHandler : public MicroPropsGenerator, public UMemory { LongNameHandler(const PluralRules *rules, const MicroPropsGenerator *parent) : rules(rules), parent(parent) {} - static LongNameHandler + static LongNameHandler* forCompoundUnit(const Locale &loc, const MeasureUnit &unit, const MeasureUnit &perUnit, const UNumberUnitWidth &width, const PluralRules *rules, const MicroPropsGenerator *parent, UErrorCode &status); - static void simpleFormatsToModifiers(const UnicodeString *simpleFormats, Field field, - SimpleModifier *output, UErrorCode &status); - static void multiSimpleFormatsToModifiers(const UnicodeString *leadFormats, UnicodeString trailFormat, - Field field, SimpleModifier *output, UErrorCode &status); + void simpleFormatsToModifiers(const UnicodeString *simpleFormats, Field field, UErrorCode &status); + void multiSimpleFormatsToModifiers(const UnicodeString *leadFormats, UnicodeString trailFormat, + Field field, UErrorCode &status); }; } // namespace impl diff --git a/deps/icu-small/source/i18n/number_mapper.cpp b/deps/icu-small/source/i18n/number_mapper.cpp index d260632f93..2c9a8e5178 100644 --- a/deps/icu-small/source/i18n/number_mapper.cpp +++ b/deps/icu-small/source/i18n/number_mapper.cpp @@ -225,8 +225,8 @@ MacroProps NumberPropertyMapper::oldToNew(const DecimalFormatProperties& propert // TODO: Overriding here is a bit of a hack. Should this logic go earlier? if (macros.precision.fType == Precision::PrecisionType::RND_FRACTION) { // For the purposes of rounding, get the original min/max int/frac, since the local - // variables - // have been manipulated for display purposes. + // variables have been manipulated for display purposes. + int maxInt_ = properties.maximumIntegerDigits; int minInt_ = properties.minimumIntegerDigits; int minFrac_ = properties.minimumFractionDigits; int maxFrac_ = properties.maximumFractionDigits; @@ -237,9 +237,15 @@ MacroProps NumberPropertyMapper::oldToNew(const DecimalFormatProperties& propert // Patterns like "#.##E0" (no zeros in the mantissa), which mean round to maxFrac+1 macros.precision = Precision::constructSignificant(1, maxFrac_ + 1).withMode(roundingMode); } else { - // All other scientific patterns, which mean round to minInt+maxFrac - macros.precision = Precision::constructSignificant( - minInt_ + minFrac_, minInt_ + maxFrac_).withMode(roundingMode); + int maxSig_ = minInt_ + maxFrac_; + // Bug #20058: if maxInt_ > minInt_ > 1, then minInt_ should be 1. + if (maxInt_ > minInt_ && minInt_ > 1) { + minInt_ = 1; + } + int minSig_ = minInt_ + minFrac_; + // To avoid regression, maxSig is not reset when minInt_ set to 1. + // TODO: Reset maxSig_ = 1 + minFrac_ to follow the spec. + macros.precision = Precision::constructSignificant(minSig_, maxSig_).withMode(roundingMode); } } } diff --git a/deps/icu-small/source/i18n/number_modifiers.cpp b/deps/icu-small/source/i18n/number_modifiers.cpp index 4385499b54..d92ec63b08 100644 --- a/deps/icu-small/source/i18n/number_modifiers.cpp +++ b/deps/icu-small/source/i18n/number_modifiers.cpp @@ -53,6 +53,21 @@ void U_CALLCONV initDefaultCurrencySpacing(UErrorCode &status) { Modifier::~Modifier() = default; +Modifier::Parameters::Parameters() + : obj(nullptr) {} + +Modifier::Parameters::Parameters( + const ModifierStore* _obj, int8_t _signum, StandardPlural::Form _plural) + : obj(_obj), signum(_signum), plural(_plural) {} + +ModifierStore::~ModifierStore() = default; + +AdoptingModifierStore::~AdoptingModifierStore() { + for (const Modifier *mod : mods) { + delete mod; + } +} + int32_t ConstantAffixModifier::apply(NumberStringBuilder &output, int leftIndex, int rightIndex, UErrorCode &status) const { @@ -62,13 +77,11 @@ int32_t ConstantAffixModifier::apply(NumberStringBuilder &output, int leftIndex, return length; } -int32_t ConstantAffixModifier::getPrefixLength(UErrorCode &status) const { - (void)status; +int32_t ConstantAffixModifier::getPrefixLength() const { return fPrefix.length(); } -int32_t ConstantAffixModifier::getCodePointCount(UErrorCode &status) const { - (void)status; +int32_t ConstantAffixModifier::getCodePointCount() const { return fPrefix.countChar32() + fSuffix.countChar32(); } @@ -76,8 +89,38 @@ bool ConstantAffixModifier::isStrong() const { return fStrong; } +bool ConstantAffixModifier::containsField(UNumberFormatFields field) const { + (void)field; + // This method is not currently used. + U_ASSERT(false); + return false; +} + +void ConstantAffixModifier::getParameters(Parameters& output) const { + (void)output; + // This method is not currently used. + U_ASSERT(false); +} + +bool ConstantAffixModifier::semanticallyEquivalent(const Modifier& other) const { + auto* _other = dynamic_cast<const ConstantAffixModifier*>(&other); + if (_other == nullptr) { + return false; + } + return fPrefix == _other->fPrefix + && fSuffix == _other->fSuffix + && fField == _other->fField + && fStrong == _other->fStrong; +} + + SimpleModifier::SimpleModifier(const SimpleFormatter &simpleFormatter, Field field, bool strong) - : fCompiledPattern(simpleFormatter.compiledPattern), fField(field), fStrong(strong) { + : SimpleModifier(simpleFormatter, field, strong, {}) {} + +SimpleModifier::SimpleModifier(const SimpleFormatter &simpleFormatter, Field field, bool strong, + const Modifier::Parameters parameters) + : fCompiledPattern(simpleFormatter.compiledPattern), fField(field), fStrong(strong), + fParameters(parameters) { int32_t argLimit = SimpleFormatter::getArgumentLimit( fCompiledPattern.getBuffer(), fCompiledPattern.length()); if (argLimit == 0) { @@ -90,15 +133,19 @@ SimpleModifier::SimpleModifier(const SimpleFormatter &simpleFormatter, Field fie } else { U_ASSERT(argLimit == 1); if (fCompiledPattern.charAt(1) != 0) { + // Found prefix fPrefixLength = fCompiledPattern.charAt(1) - ARG_NUM_LIMIT; fSuffixOffset = 3 + fPrefixLength; } else { + // No prefix fPrefixLength = 0; fSuffixOffset = 2; } if (3 + fPrefixLength < fCompiledPattern.length()) { + // Found suffix fSuffixLength = fCompiledPattern.charAt(fSuffixOffset) - ARG_NUM_LIMIT; } else { + // No suffix fSuffixLength = 0; } } @@ -113,13 +160,11 @@ int32_t SimpleModifier::apply(NumberStringBuilder &output, int leftIndex, int ri return formatAsPrefixSuffix(output, leftIndex, rightIndex, fField, status); } -int32_t SimpleModifier::getPrefixLength(UErrorCode &status) const { - (void)status; +int32_t SimpleModifier::getPrefixLength() const { return fPrefixLength; } -int32_t SimpleModifier::getCodePointCount(UErrorCode &status) const { - (void)status; +int32_t SimpleModifier::getCodePointCount() const { int32_t count = 0; if (fPrefixLength > 0) { count += fCompiledPattern.countChar32(2, fPrefixLength); @@ -134,10 +179,35 @@ bool SimpleModifier::isStrong() const { return fStrong; } +bool SimpleModifier::containsField(UNumberFormatFields field) const { + (void)field; + // This method is not currently used. + U_ASSERT(false); + return false; +} + +void SimpleModifier::getParameters(Parameters& output) const { + output = fParameters; +} + +bool SimpleModifier::semanticallyEquivalent(const Modifier& other) const { + auto* _other = dynamic_cast<const SimpleModifier*>(&other); + if (_other == nullptr) { + return false; + } + if (fParameters.obj != nullptr) { + return fParameters.obj == _other->fParameters.obj; + } + return fCompiledPattern == _other->fCompiledPattern + && fField == _other->fField + && fStrong == _other->fStrong; +} + + int32_t SimpleModifier::formatAsPrefixSuffix(NumberStringBuilder &result, int32_t startIndex, int32_t endIndex, Field field, UErrorCode &status) const { - if (fSuffixOffset == -1) { + if (fSuffixOffset == -1 && fPrefixLength + fSuffixLength > 0) { // There is no argument for the inner number; overwrite the entire segment with our string. return result.splice(startIndex, endIndex, fCompiledPattern, 2, 2 + fPrefixLength, field, status); } else { @@ -157,6 +227,65 @@ SimpleModifier::formatAsPrefixSuffix(NumberStringBuilder &result, int32_t startI } } + +int32_t +SimpleModifier::formatTwoArgPattern(const SimpleFormatter& compiled, NumberStringBuilder& result, + int32_t index, int32_t* outPrefixLength, int32_t* outSuffixLength, + Field field, UErrorCode& status) { + const UnicodeString& compiledPattern = compiled.compiledPattern; + int32_t argLimit = SimpleFormatter::getArgumentLimit( + compiledPattern.getBuffer(), compiledPattern.length()); + if (argLimit != 2) { + status = U_INTERNAL_PROGRAM_ERROR; + return 0; + } + int32_t offset = 1; // offset into compiledPattern + int32_t length = 0; // chars added to result + + int32_t prefixLength = compiledPattern.charAt(offset); + offset++; + if (prefixLength < ARG_NUM_LIMIT) { + // No prefix + prefixLength = 0; + } else { + prefixLength -= ARG_NUM_LIMIT; + result.insert(index + length, compiledPattern, offset, offset + prefixLength, field, status); + offset += prefixLength; + length += prefixLength; + offset++; + } + + int32_t infixLength = compiledPattern.charAt(offset); + offset++; + if (infixLength < ARG_NUM_LIMIT) { + // No infix + infixLength = 0; + } else { + infixLength -= ARG_NUM_LIMIT; + result.insert(index + length, compiledPattern, offset, offset + infixLength, field, status); + offset += infixLength; + length += infixLength; + offset++; + } + + int32_t suffixLength; + if (offset == compiledPattern.length()) { + // No suffix + suffixLength = 0; + } else { + suffixLength = compiledPattern.charAt(offset) - ARG_NUM_LIMIT; + offset++; + result.insert(index + length, compiledPattern, offset, offset + suffixLength, field, status); + length += suffixLength; + } + + *outPrefixLength = prefixLength; + *outSuffixLength = suffixLength; + + return length; +} + + int32_t ConstantMultiFieldModifier::apply(NumberStringBuilder &output, int leftIndex, int rightIndex, UErrorCode &status) const { int32_t length = output.insert(leftIndex, fPrefix, status); @@ -171,13 +300,11 @@ int32_t ConstantMultiFieldModifier::apply(NumberStringBuilder &output, int leftI return length; } -int32_t ConstantMultiFieldModifier::getPrefixLength(UErrorCode &status) const { - (void)status; +int32_t ConstantMultiFieldModifier::getPrefixLength() const { return fPrefix.length(); } -int32_t ConstantMultiFieldModifier::getCodePointCount(UErrorCode &status) const { - (void)status; +int32_t ConstantMultiFieldModifier::getCodePointCount() const { return fPrefix.codePointCount() + fSuffix.codePointCount(); } @@ -185,6 +312,29 @@ bool ConstantMultiFieldModifier::isStrong() const { return fStrong; } +bool ConstantMultiFieldModifier::containsField(UNumberFormatFields field) const { + return fPrefix.containsField(field) || fSuffix.containsField(field); +} + +void ConstantMultiFieldModifier::getParameters(Parameters& output) const { + output = fParameters; +} + +bool ConstantMultiFieldModifier::semanticallyEquivalent(const Modifier& other) const { + auto* _other = dynamic_cast<const ConstantMultiFieldModifier*>(&other); + if (_other == nullptr) { + return false; + } + if (fParameters.obj != nullptr) { + return fParameters.obj == _other->fParameters.obj; + } + return fPrefix.contentEquals(_other->fPrefix) + && fSuffix.contentEquals(_other->fSuffix) + && fOverwrite == _other->fOverwrite + && fStrong == _other->fStrong; +} + + CurrencySpacingEnabledModifier::CurrencySpacingEnabledModifier(const NumberStringBuilder &prefix, const NumberStringBuilder &suffix, bool overwrite, diff --git a/deps/icu-small/source/i18n/number_modifiers.h b/deps/icu-small/source/i18n/number_modifiers.h index a553100cd9..65ada937d0 100644 --- a/deps/icu-small/source/i18n/number_modifiers.h +++ b/deps/icu-small/source/i18n/number_modifiers.h @@ -31,12 +31,18 @@ class U_I18N_API ConstantAffixModifier : public Modifier, public UObject { int32_t apply(NumberStringBuilder &output, int32_t leftIndex, int32_t rightIndex, UErrorCode &status) const U_OVERRIDE; - int32_t getPrefixLength(UErrorCode &status) const U_OVERRIDE; + int32_t getPrefixLength() const U_OVERRIDE; - int32_t getCodePointCount(UErrorCode &status) const U_OVERRIDE; + int32_t getCodePointCount() const U_OVERRIDE; bool isStrong() const U_OVERRIDE; + bool containsField(UNumberFormatFields field) const U_OVERRIDE; + + void getParameters(Parameters& output) const U_OVERRIDE; + + bool semanticallyEquivalent(const Modifier& other) const U_OVERRIDE; + private: UnicodeString fPrefix; UnicodeString fSuffix; @@ -52,21 +58,30 @@ class U_I18N_API SimpleModifier : public Modifier, public UMemory { public: SimpleModifier(const SimpleFormatter &simpleFormatter, Field field, bool strong); + SimpleModifier(const SimpleFormatter &simpleFormatter, Field field, bool strong, + const Modifier::Parameters parameters); + // Default constructor for LongNameHandler.h SimpleModifier(); int32_t apply(NumberStringBuilder &output, int32_t leftIndex, int32_t rightIndex, UErrorCode &status) const U_OVERRIDE; - int32_t getPrefixLength(UErrorCode &status) const U_OVERRIDE; + int32_t getPrefixLength() const U_OVERRIDE; - int32_t getCodePointCount(UErrorCode &status) const U_OVERRIDE; + int32_t getCodePointCount() const U_OVERRIDE; bool isStrong() const U_OVERRIDE; + bool containsField(UNumberFormatFields field) const U_OVERRIDE; + + void getParameters(Parameters& output) const U_OVERRIDE; + + bool semanticallyEquivalent(const Modifier& other) const U_OVERRIDE; + /** * TODO: This belongs in SimpleFormatterImpl. The only reason I haven't moved it there yet is because - * DoubleSidedStringBuilder is an internal class and SimpleFormatterImpl feels like it should not depend on it. + * NumberStringBuilder is an internal class and SimpleFormatterImpl feels like it should not depend on it. * * <p> * Formats a value that is already stored inside the StringBuilder <code>result</code> between the indices @@ -85,16 +100,33 @@ class U_I18N_API SimpleModifier : public Modifier, public UMemory { * @return The number of characters (UTF-16 code points) that were added to the StringBuilder. */ int32_t - formatAsPrefixSuffix(NumberStringBuilder &result, int32_t startIndex, int32_t endIndex, Field field, - UErrorCode &status) const; + formatAsPrefixSuffix(NumberStringBuilder& result, int32_t startIndex, int32_t endIndex, Field field, + UErrorCode& status) const; + + /** + * TODO: Like above, this belongs with the rest of the SimpleFormatterImpl code. + * I put it here so that the SimpleFormatter uses in NumberStringBuilder are near each other. + * + * <p> + * Applies the compiled two-argument pattern to the NumberStringBuilder. + * + * <p> + * This method is optimized for the case where the prefix and suffix are often empty, such as + * in the range pattern like "{0}-{1}". + */ + static int32_t + formatTwoArgPattern(const SimpleFormatter& compiled, NumberStringBuilder& result, + int32_t index, int32_t* outPrefixLength, int32_t* outSuffixLength, + Field field, UErrorCode& status); private: UnicodeString fCompiledPattern; Field fField; - bool fStrong; - int32_t fPrefixLength; - int32_t fSuffixOffset; - int32_t fSuffixLength; + bool fStrong = false; + int32_t fPrefixLength = 0; + int32_t fSuffixOffset = -1; + int32_t fSuffixLength = 0; + Modifier::Parameters fParameters; }; /** @@ -107,6 +139,18 @@ class U_I18N_API ConstantMultiFieldModifier : public Modifier, public UMemory { const NumberStringBuilder &prefix, const NumberStringBuilder &suffix, bool overwrite, + bool strong, + const Modifier::Parameters parameters) + : fPrefix(prefix), + fSuffix(suffix), + fOverwrite(overwrite), + fStrong(strong), + fParameters(parameters) {} + + ConstantMultiFieldModifier( + const NumberStringBuilder &prefix, + const NumberStringBuilder &suffix, + bool overwrite, bool strong) : fPrefix(prefix), fSuffix(suffix), @@ -116,12 +160,18 @@ class U_I18N_API ConstantMultiFieldModifier : public Modifier, public UMemory { int32_t apply(NumberStringBuilder &output, int32_t leftIndex, int32_t rightIndex, UErrorCode &status) const U_OVERRIDE; - int32_t getPrefixLength(UErrorCode &status) const U_OVERRIDE; + int32_t getPrefixLength() const U_OVERRIDE; - int32_t getCodePointCount(UErrorCode &status) const U_OVERRIDE; + int32_t getCodePointCount() const U_OVERRIDE; bool isStrong() const U_OVERRIDE; + bool containsField(UNumberFormatFields field) const U_OVERRIDE; + + void getParameters(Parameters& output) const U_OVERRIDE; + + bool semanticallyEquivalent(const Modifier& other) const U_OVERRIDE; + protected: // NOTE: In Java, these are stored as array pointers. In C++, the NumberStringBuilder is stored by // value and is treated internally as immutable. @@ -129,6 +179,7 @@ class U_I18N_API ConstantMultiFieldModifier : public Modifier, public UMemory { NumberStringBuilder fSuffix; bool fOverwrite; bool fStrong; + Modifier::Parameters fParameters; }; /** Identical to {@link ConstantMultiFieldModifier}, but supports currency spacing. */ @@ -192,13 +243,11 @@ class U_I18N_API EmptyModifier : public Modifier, public UMemory { return 0; } - int32_t getPrefixLength(UErrorCode &status) const U_OVERRIDE { - (void)status; + int32_t getPrefixLength() const U_OVERRIDE { return 0; } - int32_t getCodePointCount(UErrorCode &status) const U_OVERRIDE { - (void)status; + int32_t getCodePointCount() const U_OVERRIDE { return 0; } @@ -206,55 +255,75 @@ class U_I18N_API EmptyModifier : public Modifier, public UMemory { return fStrong; } + bool containsField(UNumberFormatFields field) const U_OVERRIDE { + (void)field; + return false; + } + + void getParameters(Parameters& output) const U_OVERRIDE { + output.obj = nullptr; + } + + bool semanticallyEquivalent(const Modifier& other) const U_OVERRIDE { + return other.getCodePointCount() == 0; + } + private: bool fStrong; }; /** - * A ParameterizedModifier by itself is NOT a Modifier. Rather, it wraps a data structure containing two or more - * Modifiers and returns the modifier appropriate for the current situation. + * This implementation of ModifierStore adopts Modifer pointers. */ -class U_I18N_API ParameterizedModifier : public UMemory { +class U_I18N_API AdoptingModifierStore : public ModifierStore, public UMemory { public: - // NOTE: mods is zero-initialized (to nullptr) - ParameterizedModifier() : mods() { - } + virtual ~AdoptingModifierStore(); - // No copying! - ParameterizedModifier(const ParameterizedModifier &other) = delete; + static constexpr StandardPlural::Form DEFAULT_STANDARD_PLURAL = StandardPlural::OTHER; - ~ParameterizedModifier() { - for (const Modifier *mod : mods) { - delete mod; - } - } + AdoptingModifierStore() = default; - void adoptPositiveNegativeModifiers( - const Modifier *positive, const Modifier *zero, const Modifier *negative) { - mods[2] = positive; - mods[1] = zero; - mods[0] = negative; - } + // No copying! + AdoptingModifierStore(const AdoptingModifierStore &other) = delete; - /** The modifier is ADOPTED. */ - void adoptSignPluralModifier(int8_t signum, StandardPlural::Form plural, const Modifier *mod) { + /** + * Sets the Modifier with the specified signum and plural form. + */ + void adoptModifier(int8_t signum, StandardPlural::Form plural, const Modifier *mod) { + U_ASSERT(mods[getModIndex(signum, plural)] == nullptr); mods[getModIndex(signum, plural)] = mod; } + /** + * Sets the Modifier with the specified signum. + * The modifier will apply to all plural forms. + */ + void adoptModifierWithoutPlural(int8_t signum, const Modifier *mod) { + U_ASSERT(mods[getModIndex(signum, DEFAULT_STANDARD_PLURAL)] == nullptr); + mods[getModIndex(signum, DEFAULT_STANDARD_PLURAL)] = mod; + } + /** Returns a reference to the modifier; no ownership change. */ - const Modifier *getModifier(int8_t signum) const { - return mods[signum + 1]; + const Modifier *getModifier(int8_t signum, StandardPlural::Form plural) const U_OVERRIDE { + const Modifier* modifier = mods[getModIndex(signum, plural)]; + if (modifier == nullptr && plural != DEFAULT_STANDARD_PLURAL) { + modifier = mods[getModIndex(signum, DEFAULT_STANDARD_PLURAL)]; + } + return modifier; } /** Returns a reference to the modifier; no ownership change. */ - const Modifier *getModifier(int8_t signum, StandardPlural::Form plural) const { - return mods[getModIndex(signum, plural)]; + const Modifier *getModifierWithoutPlural(int8_t signum) const { + return mods[getModIndex(signum, DEFAULT_STANDARD_PLURAL)]; } private: - const Modifier *mods[3 * StandardPlural::COUNT]; + // NOTE: mods is zero-initialized (to nullptr) + const Modifier *mods[3 * StandardPlural::COUNT] = {}; inline static int32_t getModIndex(int8_t signum, StandardPlural::Form plural) { + U_ASSERT(signum >= -1 && signum <= 1); + U_ASSERT(plural >= 0 && plural < StandardPlural::COUNT); return static_cast<int32_t>(plural) * 3 + (signum + 1); } }; diff --git a/deps/icu-small/source/i18n/number_multiplier.cpp b/deps/icu-small/source/i18n/number_multiplier.cpp index a27142c9bd..ecb50dd9b8 100644 --- a/deps/icu-small/source/i18n/number_multiplier.cpp +++ b/deps/icu-small/source/i18n/number_multiplier.cpp @@ -143,14 +143,14 @@ void Scale::applyReciprocalTo(impl::DecimalQuantity& quantity) const { void MultiplierFormatHandler::setAndChain(const Scale& multiplier, const MicroPropsGenerator* parent) { - this->multiplier = multiplier; - this->parent = parent; + fMultiplier = multiplier; + fParent = parent; } void MultiplierFormatHandler::processQuantity(DecimalQuantity& quantity, MicroProps& micros, UErrorCode& status) const { - parent->processQuantity(quantity, micros, status); - multiplier.applyTo(quantity); + fParent->processQuantity(quantity, micros, status); + fMultiplier.applyTo(quantity); } #endif /* #if !UCONFIG_NO_FORMATTING */ diff --git a/deps/icu-small/source/i18n/number_multiplier.h b/deps/icu-small/source/i18n/number_multiplier.h index 82c30c7842..d8235dc601 100644 --- a/deps/icu-small/source/i18n/number_multiplier.h +++ b/deps/icu-small/source/i18n/number_multiplier.h @@ -28,8 +28,8 @@ class U_I18N_API MultiplierFormatHandler : public MicroPropsGenerator, public UM UErrorCode& status) const U_OVERRIDE; private: - Scale multiplier; - const MicroPropsGenerator *parent; + Scale fMultiplier; + const MicroPropsGenerator *fParent; }; diff --git a/deps/icu-small/source/i18n/number_padding.cpp b/deps/icu-small/source/i18n/number_padding.cpp index 97e7b6014f..31684d7208 100644 --- a/deps/icu-small/source/i18n/number_padding.cpp +++ b/deps/icu-small/source/i18n/number_padding.cpp @@ -62,7 +62,7 @@ Padder Padder::forProperties(const DecimalFormatProperties& properties) { int32_t Padder::padAndApply(const Modifier &mod1, const Modifier &mod2, NumberStringBuilder &string, int32_t leftIndex, int32_t rightIndex, UErrorCode &status) const { - int32_t modLength = mod1.getCodePointCount(status) + mod2.getCodePointCount(status); + int32_t modLength = mod1.getCodePointCount() + mod2.getCodePointCount(); int32_t requiredPadding = fWidth - modLength - string.codePointCount(); U_ASSERT(leftIndex == 0 && rightIndex == string.length()); // fix the previous line to remove this assertion diff --git a/deps/icu-small/source/i18n/number_patternmodifier.cpp b/deps/icu-small/source/i18n/number_patternmodifier.cpp index 6417e14378..4c61a0d35b 100644 --- a/deps/icu-small/source/i18n/number_patternmodifier.cpp +++ b/deps/icu-small/source/i18n/number_patternmodifier.cpp @@ -24,11 +24,11 @@ MutablePatternModifier::MutablePatternModifier(bool isStrong) : fStrong(isStrong) {} void MutablePatternModifier::setPatternInfo(const AffixPatternProvider* patternInfo) { - this->patternInfo = patternInfo; + fPatternInfo = patternInfo; } void MutablePatternModifier::setPatternAttributes(UNumberSignDisplay signDisplay, bool perMille) { - this->signDisplay = signDisplay; + fSignDisplay = signDisplay; this->perMilleReplacesPercent = perMille; } @@ -36,20 +36,20 @@ void MutablePatternModifier::setSymbols(const DecimalFormatSymbols* symbols, const CurrencySymbols* currencySymbols, const UNumberUnitWidth unitWidth, const PluralRules* rules) { U_ASSERT((rules != nullptr) == needsPlurals()); - this->symbols = symbols; - this->currencySymbols = currencySymbols; - this->unitWidth = unitWidth; - this->rules = rules; + fSymbols = symbols; + fCurrencySymbols = currencySymbols; + fUnitWidth = unitWidth; + fRules = rules; } void MutablePatternModifier::setNumberProperties(int8_t signum, StandardPlural::Form plural) { - this->signum = signum; - this->plural = plural; + fSignum = signum; + fPlural = plural; } bool MutablePatternModifier::needsPlurals() const { UErrorCode statusLocal = U_ZERO_ERROR; - return patternInfo->containsSymbolType(AffixPatternType::TYPE_CURRENCY_TRIPLE, statusLocal); + return fPatternInfo->containsSymbolType(AffixPatternType::TYPE_CURRENCY_TRIPLE, statusLocal); // Silently ignore any error codes. } @@ -69,7 +69,7 @@ MutablePatternModifier::createImmutableAndChain(const MicroPropsGenerator* paren StandardPlural::Form::MANY, StandardPlural::Form::OTHER}; - auto pm = new ParameterizedModifier(); + auto pm = new AdoptingModifierStore(); if (pm == nullptr) { status = U_MEMORY_ALLOCATION_ERROR; return nullptr; @@ -79,26 +79,25 @@ MutablePatternModifier::createImmutableAndChain(const MicroPropsGenerator* paren // Slower path when we require the plural keyword. for (StandardPlural::Form plural : STANDARD_PLURAL_VALUES) { setNumberProperties(1, plural); - pm->adoptSignPluralModifier(1, plural, createConstantModifier(status)); + pm->adoptModifier(1, plural, createConstantModifier(status)); setNumberProperties(0, plural); - pm->adoptSignPluralModifier(0, plural, createConstantModifier(status)); + pm->adoptModifier(0, plural, createConstantModifier(status)); setNumberProperties(-1, plural); - pm->adoptSignPluralModifier(-1, plural, createConstantModifier(status)); + pm->adoptModifier(-1, plural, createConstantModifier(status)); } if (U_FAILURE(status)) { delete pm; return nullptr; } - return new ImmutablePatternModifier(pm, rules, parent); // adopts pm + return new ImmutablePatternModifier(pm, fRules, parent); // adopts pm } else { // Faster path when plural keyword is not needed. setNumberProperties(1, StandardPlural::Form::COUNT); - Modifier* positive = createConstantModifier(status); + pm->adoptModifierWithoutPlural(1, createConstantModifier(status)); setNumberProperties(0, StandardPlural::Form::COUNT); - Modifier* zero = createConstantModifier(status); + pm->adoptModifierWithoutPlural(0, createConstantModifier(status)); setNumberProperties(-1, StandardPlural::Form::COUNT); - Modifier* negative = createConstantModifier(status); - pm->adoptPositiveNegativeModifiers(positive, zero, negative); + pm->adoptModifierWithoutPlural(-1, createConstantModifier(status)); if (U_FAILURE(status)) { delete pm; return nullptr; @@ -112,15 +111,15 @@ ConstantMultiFieldModifier* MutablePatternModifier::createConstantModifier(UErro NumberStringBuilder b; insertPrefix(a, 0, status); insertSuffix(b, 0, status); - if (patternInfo->hasCurrencySign()) { + if (fPatternInfo->hasCurrencySign()) { return new CurrencySpacingEnabledModifier( - a, b, !patternInfo->hasBody(), fStrong, *symbols, status); + a, b, !fPatternInfo->hasBody(), fStrong, *fSymbols, status); } else { - return new ConstantMultiFieldModifier(a, b, !patternInfo->hasBody(), fStrong); + return new ConstantMultiFieldModifier(a, b, !fPatternInfo->hasBody(), fStrong); } } -ImmutablePatternModifier::ImmutablePatternModifier(ParameterizedModifier* pm, const PluralRules* rules, +ImmutablePatternModifier::ImmutablePatternModifier(AdoptingModifierStore* pm, const PluralRules* rules, const MicroPropsGenerator* parent) : pm(pm), rules(rules), parent(parent) {} @@ -132,7 +131,7 @@ void ImmutablePatternModifier::processQuantity(DecimalQuantity& quantity, MicroP void ImmutablePatternModifier::applyToMicros(MicroProps& micros, DecimalQuantity& quantity) const { if (rules == nullptr) { - micros.modMiddle = pm->getModifier(quantity.signum()); + micros.modMiddle = pm->getModifierWithoutPlural(quantity.signum()); } else { // TODO: Fix this. Avoid the copy. DecimalQuantity copy(quantity); @@ -144,7 +143,7 @@ void ImmutablePatternModifier::applyToMicros(MicroProps& micros, DecimalQuantity const Modifier* ImmutablePatternModifier::getModifier(int8_t signum, StandardPlural::Form plural) const { if (rules == nullptr) { - return pm->getModifier(signum); + return pm->getModifierWithoutPlural(signum); } else { return pm->getModifier(signum, plural); } @@ -153,13 +152,13 @@ const Modifier* ImmutablePatternModifier::getModifier(int8_t signum, StandardPlu /** Used by the unsafe code path. */ MicroPropsGenerator& MutablePatternModifier::addToChain(const MicroPropsGenerator* parent) { - this->parent = parent; + fParent = parent; return *this; } void MutablePatternModifier::processQuantity(DecimalQuantity& fq, MicroProps& micros, UErrorCode& status) const { - parent->processQuantity(fq, micros, status); + fParent->processQuantity(fq, micros, status); // The unsafe code path performs self-mutation, so we need a const_cast. // This method needs to be const because it overrides a const method in the parent class. auto nonConstThis = const_cast<MutablePatternModifier*>(this); @@ -167,7 +166,7 @@ void MutablePatternModifier::processQuantity(DecimalQuantity& fq, MicroProps& mi // TODO: Fix this. Avoid the copy. DecimalQuantity copy(fq); micros.rounder.apply(copy, status); - nonConstThis->setNumberProperties(fq.signum(), utils::getStandardPlural(rules, copy)); + nonConstThis->setNumberProperties(fq.signum(), utils::getStandardPlural(fRules, copy)); } else { nonConstThis->setNumberProperties(fq.signum(), StandardPlural::Form::COUNT); } @@ -183,7 +182,7 @@ int32_t MutablePatternModifier::apply(NumberStringBuilder& output, int32_t leftI int32_t suffixLen = nonConstThis->insertSuffix(output, rightIndex + prefixLen, status); // If the pattern had no decimal stem body (like #,##0.00), overwrite the value. int32_t overwriteLen = 0; - if (!patternInfo->hasBody()) { + if (!fPatternInfo->hasBody()) { overwriteLen = output.splice( leftIndex + prefixLen, rightIndex + prefixLen, @@ -199,28 +198,30 @@ int32_t MutablePatternModifier::apply(NumberStringBuilder& output, int32_t leftI prefixLen, rightIndex + overwriteLen + prefixLen, suffixLen, - *symbols, + *fSymbols, status); return prefixLen + overwriteLen + suffixLen; } -int32_t MutablePatternModifier::getPrefixLength(UErrorCode& status) const { +int32_t MutablePatternModifier::getPrefixLength() const { // The unsafe code path performs self-mutation, so we need a const_cast. // This method needs to be const because it overrides a const method in the parent class. auto nonConstThis = const_cast<MutablePatternModifier*>(this); // Enter and exit CharSequence Mode to get the length. + UErrorCode status = U_ZERO_ERROR; // status fails only with an iilegal argument exception nonConstThis->prepareAffix(true); int result = AffixUtils::unescapedCodePointCount(currentAffix, *this, status); // prefix length return result; } -int32_t MutablePatternModifier::getCodePointCount(UErrorCode& status) const { +int32_t MutablePatternModifier::getCodePointCount() const { // The unsafe code path performs self-mutation, so we need a const_cast. // This method needs to be const because it overrides a const method in the parent class. auto nonConstThis = const_cast<MutablePatternModifier*>(this); // Render the affixes to get the length + UErrorCode status = U_ZERO_ERROR; // status fails only with an iilegal argument exception nonConstThis->prepareAffix(true); int result = AffixUtils::unescapedCodePointCount(currentAffix, *this, status); // prefix length nonConstThis->prepareAffix(false); @@ -232,6 +233,26 @@ bool MutablePatternModifier::isStrong() const { return fStrong; } +bool MutablePatternModifier::containsField(UNumberFormatFields field) const { + (void)field; + // This method is not currently used. + U_ASSERT(false); + return false; +} + +void MutablePatternModifier::getParameters(Parameters& output) const { + (void)output; + // This method is not currently used. + U_ASSERT(false); +} + +bool MutablePatternModifier::semanticallyEquivalent(const Modifier& other) const { + (void)other; + // This method is not currently used. + U_ASSERT(false); + return false; +} + int32_t MutablePatternModifier::insertPrefix(NumberStringBuilder& sb, int position, UErrorCode& status) { prepareAffix(true); int length = AffixUtils::unescape(currentAffix, sb, position, *this, status); @@ -247,40 +268,40 @@ int32_t MutablePatternModifier::insertSuffix(NumberStringBuilder& sb, int positi /** This method contains the heart of the logic for rendering LDML affix strings. */ void MutablePatternModifier::prepareAffix(bool isPrefix) { PatternStringUtils::patternInfoToStringBuilder( - *patternInfo, isPrefix, signum, signDisplay, plural, perMilleReplacesPercent, currentAffix); + *fPatternInfo, isPrefix, fSignum, fSignDisplay, fPlural, perMilleReplacesPercent, currentAffix); } UnicodeString MutablePatternModifier::getSymbol(AffixPatternType type) const { UErrorCode localStatus = U_ZERO_ERROR; switch (type) { case AffixPatternType::TYPE_MINUS_SIGN: - return symbols->getSymbol(DecimalFormatSymbols::ENumberFormatSymbol::kMinusSignSymbol); + return fSymbols->getSymbol(DecimalFormatSymbols::ENumberFormatSymbol::kMinusSignSymbol); case AffixPatternType::TYPE_PLUS_SIGN: - return symbols->getSymbol(DecimalFormatSymbols::ENumberFormatSymbol::kPlusSignSymbol); + return fSymbols->getSymbol(DecimalFormatSymbols::ENumberFormatSymbol::kPlusSignSymbol); case AffixPatternType::TYPE_PERCENT: - return symbols->getSymbol(DecimalFormatSymbols::ENumberFormatSymbol::kPercentSymbol); + return fSymbols->getSymbol(DecimalFormatSymbols::ENumberFormatSymbol::kPercentSymbol); case AffixPatternType::TYPE_PERMILLE: - return symbols->getSymbol(DecimalFormatSymbols::ENumberFormatSymbol::kPerMillSymbol); + return fSymbols->getSymbol(DecimalFormatSymbols::ENumberFormatSymbol::kPerMillSymbol); case AffixPatternType::TYPE_CURRENCY_SINGLE: { // UnitWidth ISO and HIDDEN overrides the singular currency symbol. - if (unitWidth == UNumberUnitWidth::UNUM_UNIT_WIDTH_ISO_CODE) { - return currencySymbols->getIntlCurrencySymbol(localStatus); - } else if (unitWidth == UNumberUnitWidth::UNUM_UNIT_WIDTH_HIDDEN) { + if (fUnitWidth == UNumberUnitWidth::UNUM_UNIT_WIDTH_ISO_CODE) { + return fCurrencySymbols->getIntlCurrencySymbol(localStatus); + } else if (fUnitWidth == UNumberUnitWidth::UNUM_UNIT_WIDTH_HIDDEN) { return UnicodeString(); - } else if (unitWidth == UNumberUnitWidth::UNUM_UNIT_WIDTH_NARROW) { - return currencySymbols->getNarrowCurrencySymbol(localStatus); + } else if (fUnitWidth == UNumberUnitWidth::UNUM_UNIT_WIDTH_NARROW) { + return fCurrencySymbols->getNarrowCurrencySymbol(localStatus); } else { - return currencySymbols->getCurrencySymbol(localStatus); + return fCurrencySymbols->getCurrencySymbol(localStatus); } } case AffixPatternType::TYPE_CURRENCY_DOUBLE: - return currencySymbols->getIntlCurrencySymbol(localStatus); + return fCurrencySymbols->getIntlCurrencySymbol(localStatus); case AffixPatternType::TYPE_CURRENCY_TRIPLE: // NOTE: This is the code path only for patterns containing "¤¤¤". // Plural currencies set via the API are formatted in LongNameHandler. // This code path is used by DecimalFormat via CurrencyPluralInfo. - U_ASSERT(plural != StandardPlural::Form::COUNT); - return currencySymbols->getPluralName(plural, localStatus); + U_ASSERT(fPlural != StandardPlural::Form::COUNT); + return fCurrencySymbols->getPluralName(fPlural, localStatus); case AffixPatternType::TYPE_CURRENCY_QUAD: return UnicodeString(u"\uFFFD"); case AffixPatternType::TYPE_CURRENCY_QUINT: diff --git a/deps/icu-small/source/i18n/number_patternmodifier.h b/deps/icu-small/source/i18n/number_patternmodifier.h index f1359bd574..ea80d6305e 100644 --- a/deps/icu-small/source/i18n/number_patternmodifier.h +++ b/deps/icu-small/source/i18n/number_patternmodifier.h @@ -18,13 +18,13 @@ U_NAMESPACE_BEGIN // Export an explicit template instantiation of the LocalPointer that is used as a -// data member of ParameterizedModifier. +// data member of AdoptingModifierStore. // (When building DLLs for Windows this is required.) #if U_PF_WINDOWS <= U_PLATFORM && U_PLATFORM <= U_PF_CYGWIN // Ignore warning 4661 as LocalPointerBase does not use operator== or operator!= #pragma warning(suppress: 4661) -template class U_I18N_API LocalPointerBase<number::impl::ParameterizedModifier>; -template class U_I18N_API LocalPointer<number::impl::ParameterizedModifier>; +template class U_I18N_API LocalPointerBase<number::impl::AdoptingModifierStore>; +template class U_I18N_API LocalPointer<number::impl::AdoptingModifierStore>; #endif namespace number { @@ -45,10 +45,10 @@ class U_I18N_API ImmutablePatternModifier : public MicroPropsGenerator, public U const Modifier* getModifier(int8_t signum, StandardPlural::Form plural) const; private: - ImmutablePatternModifier(ParameterizedModifier* pm, const PluralRules* rules, + ImmutablePatternModifier(AdoptingModifierStore* pm, const PluralRules* rules, const MicroPropsGenerator* parent); - const LocalPointer<ParameterizedModifier> pm; + const LocalPointer<AdoptingModifierStore> pm; const PluralRules* rules; const MicroPropsGenerator* parent; @@ -178,12 +178,18 @@ class U_I18N_API MutablePatternModifier int32_t apply(NumberStringBuilder &output, int32_t leftIndex, int32_t rightIndex, UErrorCode &status) const U_OVERRIDE; - int32_t getPrefixLength(UErrorCode &status) const U_OVERRIDE; + int32_t getPrefixLength() const U_OVERRIDE; - int32_t getCodePointCount(UErrorCode &status) const U_OVERRIDE; + int32_t getCodePointCount() const U_OVERRIDE; bool isStrong() const U_OVERRIDE; + bool containsField(UNumberFormatFields field) const U_OVERRIDE; + + void getParameters(Parameters& output) const U_OVERRIDE; + + bool semanticallyEquivalent(const Modifier& other) const U_OVERRIDE; + /** * Returns the string that substitutes a given symbol type in a pattern. */ @@ -196,22 +202,22 @@ class U_I18N_API MutablePatternModifier const bool fStrong; // Pattern details (initialized in setPatternInfo and setPatternAttributes) - const AffixPatternProvider *patternInfo; - UNumberSignDisplay signDisplay; + const AffixPatternProvider *fPatternInfo; + UNumberSignDisplay fSignDisplay; bool perMilleReplacesPercent; // Symbol details (initialized in setSymbols) - const DecimalFormatSymbols *symbols; - UNumberUnitWidth unitWidth; - const CurrencySymbols *currencySymbols; - const PluralRules *rules; + const DecimalFormatSymbols *fSymbols; + UNumberUnitWidth fUnitWidth; + const CurrencySymbols *fCurrencySymbols; + const PluralRules *fRules; // Number details (initialized in setNumberProperties) - int8_t signum; - StandardPlural::Form plural; + int8_t fSignum; + StandardPlural::Form fPlural; // QuantityChain details (initialized in addToChain) - const MicroPropsGenerator *parent; + const MicroPropsGenerator *fParent; // Transient fields for rendering UnicodeString currentAffix; diff --git a/deps/icu-small/source/i18n/number_scientific.cpp b/deps/icu-small/source/i18n/number_scientific.cpp index 40952024e9..07c1ce9dac 100644 --- a/deps/icu-small/source/i18n/number_scientific.cpp +++ b/deps/icu-small/source/i18n/number_scientific.cpp @@ -76,17 +76,16 @@ int32_t ScientificModifier::apply(NumberStringBuilder &output, int32_t /*leftInd return i - rightIndex; } -int32_t ScientificModifier::getPrefixLength(UErrorCode &status) const { - (void)status; +int32_t ScientificModifier::getPrefixLength() const { // TODO: Localized exponent separator location. return 0; } -int32_t ScientificModifier::getCodePointCount(UErrorCode &status) const { - (void)status; - // This method is not used for strong modifiers. - U_ASSERT(false); - return 0; +int32_t ScientificModifier::getCodePointCount() const { + // NOTE: This method is only called one place, NumberRangeFormatterImpl. + // The call site only cares about != 0 and != 1. + // Return a very large value so that if this method is used elsewhere, we should notice. + return 999; } bool ScientificModifier::isStrong() const { @@ -94,6 +93,27 @@ bool ScientificModifier::isStrong() const { return true; } +bool ScientificModifier::containsField(UNumberFormatFields field) const { + (void)field; + // This method is not used for inner modifiers. + U_ASSERT(false); + return false; +} + +void ScientificModifier::getParameters(Parameters& output) const { + // Not part of any plural sets + output.obj = nullptr; +} + +bool ScientificModifier::semanticallyEquivalent(const Modifier& other) const { + auto* _other = dynamic_cast<const ScientificModifier*>(&other); + if (_other == nullptr) { + return false; + } + // TODO: Check for locale symbols and settings as well? Could be less efficient. + return fExponent == _other->fExponent; +} + // Note: Visual Studio does not compile this function without full name space. Why? icu::number::impl::ScientificHandler::ScientificHandler(const Notation *notation, const DecimalFormatSymbols *symbols, const MicroPropsGenerator *parent) : diff --git a/deps/icu-small/source/i18n/number_scientific.h b/deps/icu-small/source/i18n/number_scientific.h index 974ab3adb6..e377bd941e 100644 --- a/deps/icu-small/source/i18n/number_scientific.h +++ b/deps/icu-small/source/i18n/number_scientific.h @@ -24,12 +24,18 @@ class U_I18N_API ScientificModifier : public UMemory, public Modifier { int32_t apply(NumberStringBuilder &output, int32_t leftIndex, int32_t rightIndex, UErrorCode &status) const U_OVERRIDE; - int32_t getPrefixLength(UErrorCode &status) const U_OVERRIDE; + int32_t getPrefixLength() const U_OVERRIDE; - int32_t getCodePointCount(UErrorCode &status) const U_OVERRIDE; + int32_t getCodePointCount() const U_OVERRIDE; bool isStrong() const U_OVERRIDE; + bool containsField(UNumberFormatFields field) const U_OVERRIDE; + + void getParameters(Parameters& output) const U_OVERRIDE; + + bool semanticallyEquivalent(const Modifier& other) const U_OVERRIDE; + private: int32_t fExponent; const ScientificHandler *fHandler; diff --git a/deps/icu-small/source/i18n/number_stringbuilder.cpp b/deps/icu-small/source/i18n/number_stringbuilder.cpp index 37770d11d5..74ba33fbbc 100644 --- a/deps/icu-small/source/i18n/number_stringbuilder.cpp +++ b/deps/icu-small/source/i18n/number_stringbuilder.cpp @@ -241,6 +241,9 @@ NumberStringBuilder::insert(int32_t index, const NumberStringBuilder &other, UEr } int32_t NumberStringBuilder::prepareForInsert(int32_t index, int32_t count, UErrorCode &status) { + U_ASSERT(index >= 0); + U_ASSERT(index <= fLength); + U_ASSERT(count >= 0); if (index == 0 && fZero - count >= 0) { // Append to start fZero -= count; @@ -485,4 +488,13 @@ void NumberStringBuilder::getAllFieldPositions(FieldPositionIteratorHandler& fpi } } +bool NumberStringBuilder::containsField(Field field) const { + for (int32_t i = 0; i < fLength; i++) { + if (field == fieldAt(i)) { + return true; + } + } + return false; +} + #endif /* #if !UCONFIG_NO_FORMATTING */ diff --git a/deps/icu-small/source/i18n/number_stringbuilder.h b/deps/icu-small/source/i18n/number_stringbuilder.h index cd8ce2f805..b14ad9ede2 100644 --- a/deps/icu-small/source/i18n/number_stringbuilder.h +++ b/deps/icu-small/source/i18n/number_stringbuilder.h @@ -106,6 +106,8 @@ class U_I18N_API NumberStringBuilder : public UMemory { void getAllFieldPositions(FieldPositionIteratorHandler& fpih, UErrorCode& status) const; + bool containsField(Field field) const; + private: bool fUsingHeap = false; ValueOrHeapArray<char16_t> fChars; diff --git a/deps/icu-small/source/i18n/number_types.h b/deps/icu-small/source/i18n/number_types.h index 57da72f8aa..00a6818869 100644 --- a/deps/icu-small/source/i18n/number_types.h +++ b/deps/icu-small/source/i18n/number_types.h @@ -16,6 +16,7 @@ #include "uassert.h" #include "unicode/platform.h" #include "unicode/uniset.h" +#include "standardplural.h" U_NAMESPACE_BEGIN namespace number { namespace impl { @@ -45,6 +46,7 @@ class Modifier; class MutablePatternModifier; class DecimalQuantity; class NumberStringBuilder; +class ModifierStore; struct MicroProps; @@ -127,12 +129,13 @@ class U_I18N_API AffixPatternProvider { virtual bool hasBody() const = 0; }; + /** * A Modifier is an object that can be passed through the formatting pipeline until it is finally applied to the string * builder. A Modifier usually contains a prefix and a suffix that are applied, but it could contain something else, * like a {@link com.ibm.icu.text.SimpleFormatter} pattern. * - * A Modifier is usually immutable, except in cases such as {@link MurkyModifier}, which are mutable for performance + * A Modifier is usually immutable, except in cases such as {@link MutablePatternModifier}, which are mutable for performance * reasons. * * Exported as U_I18N_API because it is a base class for other exported types @@ -162,12 +165,12 @@ class U_I18N_API Modifier { * * @return The number of characters (UTF-16 code units) in the prefix. */ - virtual int32_t getPrefixLength(UErrorCode& status) const = 0; + virtual int32_t getPrefixLength() const = 0; /** * Returns the number of code points in the modifier, prefix plus suffix. */ - virtual int32_t getCodePointCount(UErrorCode& status) const = 0; + virtual int32_t getCodePointCount() const = 0; /** * Whether this modifier is strong. If a modifier is strong, it should always be applied immediately and not allowed @@ -177,8 +180,57 @@ class U_I18N_API Modifier { * @return Whether the modifier is strong. */ virtual bool isStrong() const = 0; + + /** + * Whether the modifier contains at least one occurrence of the given field. + */ + virtual bool containsField(UNumberFormatFields field) const = 0; + + /** + * A fill-in for getParameters(). obj will always be set; if non-null, the other + * two fields are also safe to read. + */ + struct U_I18N_API Parameters { + const ModifierStore* obj = nullptr; + int8_t signum; + StandardPlural::Form plural; + + Parameters(); + Parameters(const ModifierStore* _obj, int8_t _signum, StandardPlural::Form _plural); + }; + + /** + * Gets a set of "parameters" for this Modifier. + * + * TODO: Make this return a `const Parameters*` more like Java? + */ + virtual void getParameters(Parameters& output) const = 0; + + /** + * Returns whether this Modifier is *semantically equivalent* to the other Modifier; + * in many cases, this is the same as equal, but parameters should be ignored. + */ + virtual bool semanticallyEquivalent(const Modifier& other) const = 0; }; + +/** + * This is *not* a modifier; rather, it is an object that can return modifiers + * based on given parameters. + * + * Exported as U_I18N_API because it is a base class for other exported types. + */ +class U_I18N_API ModifierStore { + public: + virtual ~ModifierStore(); + + /** + * Returns a Modifier with the given parameters (best-effort). + */ + virtual const Modifier* getModifier(int8_t signum, StandardPlural::Form plural) const = 0; +}; + + /** * This interface is used when all number formatting settings, including the locale, are known, except for the quantity * itself. The {@link #processQuantity} method performs the final step in the number processing pipeline: it uses the diff --git a/deps/icu-small/source/i18n/numfmt.cpp b/deps/icu-small/source/i18n/numfmt.cpp index 13f23131b1..ef47e1e01b 100644 --- a/deps/icu-small/source/i18n/numfmt.cpp +++ b/deps/icu-small/source/i18n/numfmt.cpp @@ -1326,13 +1326,13 @@ NumberFormat::makeInstance(const Locale& desiredLocale, // if the locale has "@compat=host", create a host-specific NumberFormat if (U_SUCCESS(status) && count > 0 && uprv_strcmp(buffer, "host") == 0) { - Win32NumberFormat *f = NULL; UBool curr = TRUE; switch (style) { case UNUM_DECIMAL: curr = FALSE; // fall-through + U_FALLTHROUGH; case UNUM_CURRENCY: case UNUM_CURRENCY_ISO: // do not support plural formatting here @@ -1340,14 +1340,13 @@ NumberFormat::makeInstance(const Locale& desiredLocale, case UNUM_CURRENCY_ACCOUNTING: case UNUM_CASH_CURRENCY: case UNUM_CURRENCY_STANDARD: - f = new Win32NumberFormat(desiredLocale, curr, status); - + { + LocalPointer<Win32NumberFormat> f(new Win32NumberFormat(desiredLocale, curr, status), status); if (U_SUCCESS(status)) { - return f; + return f.orphan(); } - - delete f; - break; + } + break; default: break; } @@ -1417,8 +1416,7 @@ NumberFormat::makeInstance(const Locale& desiredLocale, } } - - NumberFormat *f; + LocalPointer<NumberFormat> f; if (ns->isAlgorithmic()) { UnicodeString nsDesc; UnicodeString nsRuleSetGroup; @@ -1453,7 +1451,7 @@ NumberFormat::makeInstance(const Locale& desiredLocale, return NULL; } r->setDefaultRuleSet(nsRuleSetName,status); - f = r; + f.adoptInstead(r); } else { // replace single currency sign in the pattern with double currency sign // if the style is UNUM_CURRENCY_ISO @@ -1462,9 +1460,22 @@ NumberFormat::makeInstance(const Locale& desiredLocale, UnicodeString(TRUE, gDoubleCurrencySign, 2)); } - // "new DecimalFormat()" does not adopt the symbols if its memory allocation fails. - DecimalFormatSymbols *syms = symbolsToAdopt.orphan(); - DecimalFormat* df = new DecimalFormat(pattern, syms, style, status); + // "new DecimalFormat()" does not adopt the symbols argument if its memory allocation fails. + // So we can't use adoptInsteadAndCheckErrorCode as we need to know if the 'new' failed. + DecimalFormatSymbols *syms = symbolsToAdopt.getAlias(); + LocalPointer<DecimalFormat> df(new DecimalFormat(pattern, syms, style, status)); + + if (df.isValid()) { + // if the DecimalFormat object was successfully new'ed, then it will own symbolsToAdopt, even if the status is a failure. + symbolsToAdopt.orphan(); + } + else { + status = U_MEMORY_ALLOCATION_ERROR; + } + + if (U_FAILURE(status)) { + return nullptr; + } // if it is cash currency style, setCurrencyUsage with usage if (style == UNUM_CASH_CURRENCY){ @@ -1472,25 +1483,18 @@ NumberFormat::makeInstance(const Locale& desiredLocale, } if (U_FAILURE(status)) { - delete df; - return NULL; + return nullptr; } - f = df; - if (f == NULL) { - delete syms; - status = U_MEMORY_ALLOCATION_ERROR; - return NULL; - } + f.adoptInstead(df.orphan()); } f->setLocaleIDs(ures_getLocaleByType(ownedResource.getAlias(), ULOC_VALID_LOCALE, &status), ures_getLocaleByType(ownedResource.getAlias(), ULOC_ACTUAL_LOCALE, &status)); if (U_FAILURE(status)) { - delete f; return NULL; } - return f; + return f.orphan(); } /** diff --git a/deps/icu-small/source/i18n/numparse_currency.cpp b/deps/icu-small/source/i18n/numparse_currency.cpp index ae8196ec48..598ace5653 100644 --- a/deps/icu-small/source/i18n/numparse_currency.cpp +++ b/deps/icu-small/source/i18n/numparse_currency.cpp @@ -111,7 +111,9 @@ bool CombinedCurrencyMatcher::matchCurrency(StringSegment& segment, ParsedNumber int32_t overlap2; if (!fCurrency2.isEmpty()) { - overlap2 = segment.getCaseSensitivePrefixLength(fCurrency2); + // ISO codes should be accepted case-insensitive. + // https://unicode-org.atlassian.net/browse/ICU-13696 + overlap2 = segment.getCommonPrefixLength(fCurrency2); } else { overlap2 = -1; } diff --git a/deps/icu-small/source/i18n/numparse_impl.cpp b/deps/icu-small/source/i18n/numparse_impl.cpp index 5fa52f6335..3192a39593 100644 --- a/deps/icu-small/source/i18n/numparse_impl.cpp +++ b/deps/icu-small/source/i18n/numparse_impl.cpp @@ -159,10 +159,10 @@ NumberParserImpl::createParserFromProperties(const number::impl::DecimalFormatPr // ICU-TC meeting, April 11, 2018: accept percent/permille only if it is in the pattern, // and to maintain regressive behavior, divide by 100 even if no percent sign is present. - if (affixProvider->containsSymbolType(AffixPatternType::TYPE_PERCENT, status)) { + if (!isStrict && affixProvider->containsSymbolType(AffixPatternType::TYPE_PERCENT, status)) { parser->addMatcher(parser->fLocalMatchers.percent = {symbols}); } - if (affixProvider->containsSymbolType(AffixPatternType::TYPE_PERMILLE, status)) { + if (!isStrict && affixProvider->containsSymbolType(AffixPatternType::TYPE_PERMILLE, status)) { parser->addMatcher(parser->fLocalMatchers.permille = {symbols}); } diff --git a/deps/icu-small/source/i18n/numparse_scientific.cpp b/deps/icu-small/source/i18n/numparse_scientific.cpp index 611695e57d..de38957440 100644 --- a/deps/icu-small/source/i18n/numparse_scientific.cpp +++ b/deps/icu-small/source/i18n/numparse_scientific.cpp @@ -56,6 +56,11 @@ bool ScientificMatcher::match(StringSegment& segment, ParsedNumber& result, UErr return false; } + // Only accept one exponent per string. + if (0 != (result.flags & FLAG_HAS_EXPONENT)) { + return false; + } + // First match the scientific separator, and then match another number after it. // NOTE: This is guarded by the smoke test; no need to check fExponentSeparatorString length again. int overlap1 = segment.getCommonPrefixLength(fExponentSeparatorString); diff --git a/deps/icu-small/source/i18n/numrange_fluent.cpp b/deps/icu-small/source/i18n/numrange_fluent.cpp new file mode 100644 index 0000000000..12b006c8ad --- /dev/null +++ b/deps/icu-small/source/i18n/numrange_fluent.cpp @@ -0,0 +1,472 @@ +// © 2018 and later: Unicode, Inc. and others. +// License & terms of use: http://www.unicode.org/copyright.html + +#include "unicode/utypes.h" + +#if !UCONFIG_NO_FORMATTING + +// Allow implicit conversion from char16_t* to UnicodeString for this file: +// Helpful in toString methods and elsewhere. +#define UNISTR_FROM_STRING_EXPLICIT + +#include "numrange_impl.h" +#include "util.h" +#include "number_utypes.h" + +using namespace icu; +using namespace icu::number; +using namespace icu::number::impl; + + +// This function needs to be declared in this namespace so it can be friended. +// NOTE: In Java, this logic is handled in the resolve() function. +void icu::number::impl::touchRangeLocales(RangeMacroProps& macros) { + macros.formatter1.fMacros.locale = macros.locale; + macros.formatter2.fMacros.locale = macros.locale; +} + + +template<typename Derived> +Derived NumberRangeFormatterSettings<Derived>::numberFormatterBoth(const UnlocalizedNumberFormatter& formatter) const& { + Derived copy(*this); + copy.fMacros.formatter1 = formatter; + copy.fMacros.singleFormatter = true; + touchRangeLocales(copy.fMacros); + return copy; +} + +template<typename Derived> +Derived NumberRangeFormatterSettings<Derived>::numberFormatterBoth(const UnlocalizedNumberFormatter& formatter) && { + Derived move(std::move(*this)); + move.fMacros.formatter1 = formatter; + move.fMacros.singleFormatter = true; + touchRangeLocales(move.fMacros); + return move; +} + +template<typename Derived> +Derived NumberRangeFormatterSettings<Derived>::numberFormatterBoth(UnlocalizedNumberFormatter&& formatter) const& { + Derived copy(*this); + copy.fMacros.formatter1 = std::move(formatter); + copy.fMacros.singleFormatter = true; + touchRangeLocales(copy.fMacros); + return copy; +} + +template<typename Derived> +Derived NumberRangeFormatterSettings<Derived>::numberFormatterBoth(UnlocalizedNumberFormatter&& formatter) && { + Derived move(std::move(*this)); + move.fMacros.formatter1 = std::move(formatter); + move.fMacros.singleFormatter = true; + touchRangeLocales(move.fMacros); + return move; +} + +template<typename Derived> +Derived NumberRangeFormatterSettings<Derived>::numberFormatterFirst(const UnlocalizedNumberFormatter& formatter) const& { + Derived copy(*this); + copy.fMacros.formatter1 = formatter; + copy.fMacros.singleFormatter = false; + touchRangeLocales(copy.fMacros); + return copy; +} + +template<typename Derived> +Derived NumberRangeFormatterSettings<Derived>::numberFormatterFirst(const UnlocalizedNumberFormatter& formatter) && { + Derived move(std::move(*this)); + move.fMacros.formatter1 = formatter; + move.fMacros.singleFormatter = false; + touchRangeLocales(move.fMacros); + return move; +} + +template<typename Derived> +Derived NumberRangeFormatterSettings<Derived>::numberFormatterFirst(UnlocalizedNumberFormatter&& formatter) const& { + Derived copy(*this); + copy.fMacros.formatter1 = std::move(formatter); + copy.fMacros.singleFormatter = false; + touchRangeLocales(copy.fMacros); + return copy; +} + +template<typename Derived> +Derived NumberRangeFormatterSettings<Derived>::numberFormatterFirst(UnlocalizedNumberFormatter&& formatter) && { + Derived move(std::move(*this)); + move.fMacros.formatter1 = std::move(formatter); + move.fMacros.singleFormatter = false; + touchRangeLocales(move.fMacros); + return move; +} + +template<typename Derived> +Derived NumberRangeFormatterSettings<Derived>::numberFormatterSecond(const UnlocalizedNumberFormatter& formatter) const& { + Derived copy(*this); + copy.fMacros.formatter2 = formatter; + copy.fMacros.singleFormatter = false; + touchRangeLocales(copy.fMacros); + return copy; +} + +template<typename Derived> +Derived NumberRangeFormatterSettings<Derived>::numberFormatterSecond(const UnlocalizedNumberFormatter& formatter) && { + Derived move(std::move(*this)); + move.fMacros.formatter2 = formatter; + move.fMacros.singleFormatter = false; + touchRangeLocales(move.fMacros); + return move; +} + +template<typename Derived> +Derived NumberRangeFormatterSettings<Derived>::numberFormatterSecond(UnlocalizedNumberFormatter&& formatter) const& { + Derived copy(*this); + copy.fMacros.formatter2 = std::move(formatter); + copy.fMacros.singleFormatter = false; + touchRangeLocales(copy.fMacros); + return copy; +} + +template<typename Derived> +Derived NumberRangeFormatterSettings<Derived>::numberFormatterSecond(UnlocalizedNumberFormatter&& formatter) && { + Derived move(std::move(*this)); + move.fMacros.formatter2 = std::move(formatter); + move.fMacros.singleFormatter = false; + touchRangeLocales(move.fMacros); + return move; +} + +template<typename Derived> +Derived NumberRangeFormatterSettings<Derived>::collapse(UNumberRangeCollapse collapse) const& { + Derived copy(*this); + copy.fMacros.collapse = collapse; + return copy; +} + +template<typename Derived> +Derived NumberRangeFormatterSettings<Derived>::collapse(UNumberRangeCollapse collapse) && { + Derived move(std::move(*this)); + move.fMacros.collapse = collapse; + return move; +} + +template<typename Derived> +Derived NumberRangeFormatterSettings<Derived>::identityFallback(UNumberRangeIdentityFallback identityFallback) const& { + Derived copy(*this); + copy.fMacros.identityFallback = identityFallback; + return copy; +} + +template<typename Derived> +Derived NumberRangeFormatterSettings<Derived>::identityFallback(UNumberRangeIdentityFallback identityFallback) && { + Derived move(std::move(*this)); + move.fMacros.identityFallback = identityFallback; + return move; +} + +// Declare all classes that implement NumberRangeFormatterSettings +// See https://stackoverflow.com/a/495056/1407170 +template +class icu::number::NumberRangeFormatterSettings<icu::number::UnlocalizedNumberRangeFormatter>; +template +class icu::number::NumberRangeFormatterSettings<icu::number::LocalizedNumberRangeFormatter>; + + +UnlocalizedNumberRangeFormatter NumberRangeFormatter::with() { + UnlocalizedNumberRangeFormatter result; + return result; +} + +LocalizedNumberRangeFormatter NumberRangeFormatter::withLocale(const Locale& locale) { + return with().locale(locale); +} + + +template<typename T> using NFS = NumberRangeFormatterSettings<T>; +using LNF = LocalizedNumberRangeFormatter; +using UNF = UnlocalizedNumberRangeFormatter; + +UnlocalizedNumberRangeFormatter::UnlocalizedNumberRangeFormatter(const UNF& other) + : UNF(static_cast<const NFS<UNF>&>(other)) {} + +UnlocalizedNumberRangeFormatter::UnlocalizedNumberRangeFormatter(const NFS<UNF>& other) + : NFS<UNF>(other) { + // No additional fields to assign +} + +// Make default copy constructor call the NumberRangeFormatterSettings copy constructor. +UnlocalizedNumberRangeFormatter::UnlocalizedNumberRangeFormatter(UNF&& src) U_NOEXCEPT + : UNF(static_cast<NFS<UNF>&&>(src)) {} + +UnlocalizedNumberRangeFormatter::UnlocalizedNumberRangeFormatter(NFS<UNF>&& src) U_NOEXCEPT + : NFS<UNF>(std::move(src)) { + // No additional fields to assign +} + +UnlocalizedNumberRangeFormatter& UnlocalizedNumberRangeFormatter::operator=(const UNF& other) { + NFS<UNF>::operator=(static_cast<const NFS<UNF>&>(other)); + // No additional fields to assign + return *this; +} + +UnlocalizedNumberRangeFormatter& UnlocalizedNumberRangeFormatter::operator=(UNF&& src) U_NOEXCEPT { + NFS<UNF>::operator=(static_cast<NFS<UNF>&&>(src)); + // No additional fields to assign + return *this; +} + +// Make default copy constructor call the NumberRangeFormatterSettings copy constructor. +LocalizedNumberRangeFormatter::LocalizedNumberRangeFormatter(const LNF& other) + : LNF(static_cast<const NFS<LNF>&>(other)) {} + +LocalizedNumberRangeFormatter::LocalizedNumberRangeFormatter(const NFS<LNF>& other) + : NFS<LNF>(other) { + // No additional fields to assign +} + +LocalizedNumberRangeFormatter::LocalizedNumberRangeFormatter(LocalizedNumberRangeFormatter&& src) U_NOEXCEPT + : LNF(static_cast<NFS<LNF>&&>(src)) {} + +LocalizedNumberRangeFormatter::LocalizedNumberRangeFormatter(NFS<LNF>&& src) U_NOEXCEPT + : NFS<LNF>(std::move(src)) { + // Steal the compiled formatter + LNF&& _src = static_cast<LNF&&>(src); + auto* stolen = _src.fAtomicFormatter.exchange(nullptr); + delete fAtomicFormatter.exchange(stolen); +} + +LocalizedNumberRangeFormatter& LocalizedNumberRangeFormatter::operator=(const LNF& other) { + NFS<LNF>::operator=(static_cast<const NFS<LNF>&>(other)); + // Do not steal; just clear + delete fAtomicFormatter.exchange(nullptr); + return *this; +} + +LocalizedNumberRangeFormatter& LocalizedNumberRangeFormatter::operator=(LNF&& src) U_NOEXCEPT { + NFS<LNF>::operator=(static_cast<NFS<LNF>&&>(src)); + // Steal the compiled formatter + auto* stolen = src.fAtomicFormatter.exchange(nullptr); + delete fAtomicFormatter.exchange(stolen); + return *this; +} + + +LocalizedNumberRangeFormatter::~LocalizedNumberRangeFormatter() { + delete fAtomicFormatter.exchange(nullptr); +} + +LocalizedNumberRangeFormatter::LocalizedNumberRangeFormatter(const RangeMacroProps& macros, const Locale& locale) { + fMacros = macros; + fMacros.locale = locale; + touchRangeLocales(fMacros); +} + +LocalizedNumberRangeFormatter::LocalizedNumberRangeFormatter(RangeMacroProps&& macros, const Locale& locale) { + fMacros = std::move(macros); + fMacros.locale = locale; + touchRangeLocales(fMacros); +} + +LocalizedNumberRangeFormatter UnlocalizedNumberRangeFormatter::locale(const Locale& locale) const& { + return LocalizedNumberRangeFormatter(fMacros, locale); +} + +LocalizedNumberRangeFormatter UnlocalizedNumberRangeFormatter::locale(const Locale& locale)&& { + return LocalizedNumberRangeFormatter(std::move(fMacros), locale); +} + + +FormattedNumberRange LocalizedNumberRangeFormatter::formatFormattableRange( + const Formattable& first, const Formattable& second, UErrorCode& status) const { + if (U_FAILURE(status)) { + return FormattedNumberRange(U_ILLEGAL_ARGUMENT_ERROR); + } + + auto results = new UFormattedNumberRangeData(); + if (results == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + return FormattedNumberRange(status); + } + + first.populateDecimalQuantity(results->quantity1, status); + if (U_FAILURE(status)) { + return FormattedNumberRange(status); + } + + second.populateDecimalQuantity(results->quantity2, status); + if (U_FAILURE(status)) { + return FormattedNumberRange(status); + } + + formatImpl(*results, first == second, status); + + // Do not save the results object if we encountered a failure. + if (U_SUCCESS(status)) { + return FormattedNumberRange(results); + } else { + delete results; + return FormattedNumberRange(status); + } +} + +void LocalizedNumberRangeFormatter::formatImpl( + UFormattedNumberRangeData& results, bool equalBeforeRounding, UErrorCode& status) const { + auto* impl = getFormatter(status); + if (U_FAILURE(status)) { + return; + } + if (impl == nullptr) { + status = U_INTERNAL_PROGRAM_ERROR; + return; + } + impl->format(results, equalBeforeRounding, status); +} + +const impl::NumberRangeFormatterImpl* +LocalizedNumberRangeFormatter::getFormatter(UErrorCode& status) const { + // TODO: Move this into umutex.h? (similar logic also in decimfmt.cpp) + // See ICU-20146 + + if (U_FAILURE(status)) { + return nullptr; + } + + // First try to get the pre-computed formatter + auto* ptr = fAtomicFormatter.load(); + if (ptr != nullptr) { + return ptr; + } + + // Try computing the formatter on our own + auto* temp = new NumberRangeFormatterImpl(fMacros, status); + if (U_FAILURE(status)) { + return nullptr; + } + if (temp == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + return nullptr; + } + + // Note: ptr starts as nullptr; during compare_exchange, + // it is set to what is actually stored in the atomic + // if another thread beat us to computing the formatter object. + auto* nonConstThis = const_cast<LocalizedNumberRangeFormatter*>(this); + if (!nonConstThis->fAtomicFormatter.compare_exchange_strong(ptr, temp)) { + // Another thread beat us to computing the formatter + delete temp; + return ptr; + } else { + // Our copy of the formatter got stored in the atomic + return temp; + } + +} + + +FormattedNumberRange::FormattedNumberRange(FormattedNumberRange&& src) U_NOEXCEPT + : fResults(src.fResults), fErrorCode(src.fErrorCode) { + // Disown src.fResults to prevent double-deletion + src.fResults = nullptr; + src.fErrorCode = U_INVALID_STATE_ERROR; +} + +FormattedNumberRange& FormattedNumberRange::operator=(FormattedNumberRange&& src) U_NOEXCEPT { + delete fResults; + fResults = src.fResults; + fErrorCode = src.fErrorCode; + // Disown src.fResults to prevent double-deletion + src.fResults = nullptr; + src.fErrorCode = U_INVALID_STATE_ERROR; + return *this; +} + +UnicodeString FormattedNumberRange::toString(UErrorCode& status) const { + if (U_FAILURE(status)) { + return ICU_Utility::makeBogusString(); + } + if (fResults == nullptr) { + status = fErrorCode; + return ICU_Utility::makeBogusString(); + } + return fResults->string.toUnicodeString(); +} + +Appendable& FormattedNumberRange::appendTo(Appendable& appendable, UErrorCode& status) const { + if (U_FAILURE(status)) { + return appendable; + } + if (fResults == nullptr) { + status = fErrorCode; + return appendable; + } + appendable.appendString(fResults->string.chars(), fResults->string.length()); + return appendable; +} + +UBool FormattedNumberRange::nextFieldPosition(FieldPosition& fieldPosition, UErrorCode& status) const { + if (U_FAILURE(status)) { + return FALSE; + } + if (fResults == nullptr) { + status = fErrorCode; + return FALSE; + } + // NOTE: MSVC sometimes complains when implicitly converting between bool and UBool + return fResults->string.nextFieldPosition(fieldPosition, status) ? TRUE : FALSE; +} + +void FormattedNumberRange::getAllFieldPositions(FieldPositionIterator& iterator, UErrorCode& status) const { + FieldPositionIteratorHandler fpih(&iterator, status); + getAllFieldPositionsImpl(fpih, status); +} + +void FormattedNumberRange::getAllFieldPositionsImpl( + FieldPositionIteratorHandler& fpih, UErrorCode& status) const { + if (U_FAILURE(status)) { + return; + } + if (fResults == nullptr) { + status = fErrorCode; + return; + } + fResults->string.getAllFieldPositions(fpih, status); +} + +UnicodeString FormattedNumberRange::getFirstDecimal(UErrorCode& status) const { + if (U_FAILURE(status)) { + return ICU_Utility::makeBogusString(); + } + if (fResults == nullptr) { + status = fErrorCode; + return ICU_Utility::makeBogusString(); + } + return fResults->quantity1.toScientificString(); +} + +UnicodeString FormattedNumberRange::getSecondDecimal(UErrorCode& status) const { + if (U_FAILURE(status)) { + return ICU_Utility::makeBogusString(); + } + if (fResults == nullptr) { + status = fErrorCode; + return ICU_Utility::makeBogusString(); + } + return fResults->quantity2.toScientificString(); +} + +UNumberRangeIdentityResult FormattedNumberRange::getIdentityResult(UErrorCode& status) const { + if (U_FAILURE(status)) { + return UNUM_IDENTITY_RESULT_NOT_EQUAL; + } + if (fResults == nullptr) { + status = fErrorCode; + return UNUM_IDENTITY_RESULT_NOT_EQUAL; + } + return fResults->identityResult; +} + +FormattedNumberRange::~FormattedNumberRange() { + delete fResults; +} + + + +#endif /* #if !UCONFIG_NO_FORMATTING */ diff --git a/deps/icu-small/source/i18n/numrange_impl.cpp b/deps/icu-small/source/i18n/numrange_impl.cpp new file mode 100644 index 0000000000..21365bfc59 --- /dev/null +++ b/deps/icu-small/source/i18n/numrange_impl.cpp @@ -0,0 +1,486 @@ +// © 2018 and later: Unicode, Inc. and others. +// License & terms of use: http://www.unicode.org/copyright.html + +#include "unicode/utypes.h" + +#if !UCONFIG_NO_FORMATTING + +// Allow implicit conversion from char16_t* to UnicodeString for this file: +// Helpful in toString methods and elsewhere. +#define UNISTR_FROM_STRING_EXPLICIT + +#include "unicode/numberrangeformatter.h" +#include "numrange_impl.h" +#include "patternprops.h" +#include "uresimp.h" +#include "util.h" + +using namespace icu; +using namespace icu::number; +using namespace icu::number::impl; + +namespace { + +// Helper function for 2-dimensional switch statement +constexpr int8_t identity2d(UNumberRangeIdentityFallback a, UNumberRangeIdentityResult b) { + return static_cast<int8_t>(a) | (static_cast<int8_t>(b) << 4); +} + + +struct NumberRangeData { + SimpleFormatter rangePattern; + SimpleFormatter approximatelyPattern; +}; + +class NumberRangeDataSink : public ResourceSink { + public: + NumberRangeDataSink(NumberRangeData& data) : fData(data) {} + + void put(const char* key, ResourceValue& value, UBool /*noFallback*/, UErrorCode& status) U_OVERRIDE { + ResourceTable miscTable = value.getTable(status); + if (U_FAILURE(status)) { return; } + for (int i = 0; miscTable.getKeyAndValue(i, key, value); i++) { + if (uprv_strcmp(key, "range") == 0) { + if (fData.rangePattern.getArgumentLimit() != 0) { + continue; // have already seen this pattern + } + fData.rangePattern = {value.getUnicodeString(status), status}; + } else if (uprv_strcmp(key, "approximately") == 0) { + if (fData.approximatelyPattern.getArgumentLimit() != 0) { + continue; // have already seen this pattern + } + fData.approximatelyPattern = {value.getUnicodeString(status), status}; + } + } + } + + private: + NumberRangeData& fData; +}; + +void getNumberRangeData(const char* localeName, const char* nsName, NumberRangeData& data, UErrorCode& status) { + if (U_FAILURE(status)) { return; } + LocalUResourceBundlePointer rb(ures_open(NULL, localeName, &status)); + if (U_FAILURE(status)) { return; } + NumberRangeDataSink sink(data); + + CharString dataPath; + dataPath.append("NumberElements/", -1, status); + dataPath.append(nsName, -1, status); + dataPath.append("/miscPatterns", -1, status); + ures_getAllItemsWithFallback(rb.getAlias(), dataPath.data(), sink, status); + if (U_FAILURE(status)) { return; } + + // TODO: Is it necessary to manually fall back to latn, or does the data sink take care of that? + + if (data.rangePattern.getArgumentLimit() == 0) { + // No data! + data.rangePattern = {u"{0}–{1}", status}; + } + if (data.approximatelyPattern.getArgumentLimit() == 0) { + // No data! + data.approximatelyPattern = {u"~{0}", status}; + } +} + +class PluralRangesDataSink : public ResourceSink { + public: + PluralRangesDataSink(StandardPluralRanges& output) : fOutput(output) {} + + void put(const char* /*key*/, ResourceValue& value, UBool /*noFallback*/, UErrorCode& status) U_OVERRIDE { + ResourceArray entriesArray = value.getArray(status); + if (U_FAILURE(status)) { return; } + fOutput.setCapacity(entriesArray.getSize()); + for (int i = 0; entriesArray.getValue(i, value); i++) { + ResourceArray pluralFormsArray = value.getArray(status); + if (U_FAILURE(status)) { return; } + pluralFormsArray.getValue(0, value); + StandardPlural::Form first = StandardPlural::fromString(value.getUnicodeString(status), status); + if (U_FAILURE(status)) { return; } + pluralFormsArray.getValue(1, value); + StandardPlural::Form second = StandardPlural::fromString(value.getUnicodeString(status), status); + if (U_FAILURE(status)) { return; } + pluralFormsArray.getValue(2, value); + StandardPlural::Form result = StandardPlural::fromString(value.getUnicodeString(status), status); + if (U_FAILURE(status)) { return; } + fOutput.addPluralRange(first, second, result); + } + } + + private: + StandardPluralRanges& fOutput; +}; + +void getPluralRangesData(const Locale& locale, StandardPluralRanges& output, UErrorCode& status) { + if (U_FAILURE(status)) { return; } + LocalUResourceBundlePointer rb(ures_openDirect(nullptr, "pluralRanges", &status)); + if (U_FAILURE(status)) { return; } + + CharString dataPath; + dataPath.append("locales/", -1, status); + dataPath.append(locale.getLanguage(), -1, status); + if (U_FAILURE(status)) { return; } + int32_t setLen; + // Not all languages are covered: fail gracefully + UErrorCode internalStatus = U_ZERO_ERROR; + const UChar* set = ures_getStringByKeyWithFallback(rb.getAlias(), dataPath.data(), &setLen, &internalStatus); + if (U_FAILURE(internalStatus)) { return; } + + dataPath.clear(); + dataPath.append("rules/", -1, status); + dataPath.appendInvariantChars(set, setLen, status); + if (U_FAILURE(status)) { return; } + PluralRangesDataSink sink(output); + ures_getAllItemsWithFallback(rb.getAlias(), dataPath.data(), sink, status); + if (U_FAILURE(status)) { return; } +} + +} // namespace + + +void StandardPluralRanges::initialize(const Locale& locale, UErrorCode& status) { + getPluralRangesData(locale, *this, status); +} + +void StandardPluralRanges::addPluralRange( + StandardPlural::Form first, + StandardPlural::Form second, + StandardPlural::Form result) { + U_ASSERT(fTriplesLen < fTriples.getCapacity()); + fTriples[fTriplesLen] = {first, second, result}; + fTriplesLen++; +} + +void StandardPluralRanges::setCapacity(int32_t length) { + if (length > fTriples.getCapacity()) { + fTriples.resize(length, 0); + } +} + +StandardPlural::Form +StandardPluralRanges::resolve(StandardPlural::Form first, StandardPlural::Form second) const { + for (int32_t i=0; i<fTriplesLen; i++) { + const auto& triple = fTriples[i]; + if (triple.first == first && triple.second == second) { + return triple.result; + } + } + // Default fallback + return StandardPlural::OTHER; +} + + +NumberRangeFormatterImpl::NumberRangeFormatterImpl(const RangeMacroProps& macros, UErrorCode& status) + : formatterImpl1(macros.formatter1.fMacros, status), + formatterImpl2(macros.formatter2.fMacros, status), + fSameFormatters(macros.singleFormatter), + fCollapse(macros.collapse), + fIdentityFallback(macros.identityFallback) { + + // TODO: As of this writing (ICU 63), there is no locale that has different number miscPatterns + // based on numbering system. Therefore, data is loaded only from latn. If this changes, + // this part of the code should be updated to load from the local numbering system. + // The numbering system could come from the one specified in the NumberFormatter passed to + // numberFormatterBoth() or similar. + // See ICU-20144 + + NumberRangeData data; + getNumberRangeData(macros.locale.getName(), "latn", data, status); + if (U_FAILURE(status)) { return; } + fRangeFormatter = data.rangePattern; + fApproximatelyModifier = {data.approximatelyPattern, UNUM_FIELD_COUNT, false}; + + // TODO: Get locale from PluralRules instead? + fPluralRanges.initialize(macros.locale, status); + if (U_FAILURE(status)) { return; } +} + +void NumberRangeFormatterImpl::format(UFormattedNumberRangeData& data, bool equalBeforeRounding, UErrorCode& status) const { + if (U_FAILURE(status)) { + return; + } + + MicroProps micros1; + MicroProps micros2; + formatterImpl1.preProcess(data.quantity1, micros1, status); + if (fSameFormatters) { + formatterImpl1.preProcess(data.quantity2, micros2, status); + } else { + formatterImpl2.preProcess(data.quantity2, micros2, status); + } + if (U_FAILURE(status)) { + return; + } + + // If any of the affixes are different, an identity is not possible + // and we must use formatRange(). + // TODO: Write this as MicroProps operator==() ? + // TODO: Avoid the redundancy of these equality operations with the + // ones in formatRange? + if (!micros1.modInner->semanticallyEquivalent(*micros2.modInner) + || !micros1.modMiddle->semanticallyEquivalent(*micros2.modMiddle) + || !micros1.modOuter->semanticallyEquivalent(*micros2.modOuter)) { + formatRange(data, micros1, micros2, status); + data.identityResult = UNUM_IDENTITY_RESULT_NOT_EQUAL; + return; + } + + // Check for identity + if (equalBeforeRounding) { + data.identityResult = UNUM_IDENTITY_RESULT_EQUAL_BEFORE_ROUNDING; + } else if (data.quantity1 == data.quantity2) { + data.identityResult = UNUM_IDENTITY_RESULT_EQUAL_AFTER_ROUNDING; + } else { + data.identityResult = UNUM_IDENTITY_RESULT_NOT_EQUAL; + } + + switch (identity2d(fIdentityFallback, data.identityResult)) { + case identity2d(UNUM_IDENTITY_FALLBACK_RANGE, + UNUM_IDENTITY_RESULT_NOT_EQUAL): + case identity2d(UNUM_IDENTITY_FALLBACK_RANGE, + UNUM_IDENTITY_RESULT_EQUAL_AFTER_ROUNDING): + case identity2d(UNUM_IDENTITY_FALLBACK_RANGE, + UNUM_IDENTITY_RESULT_EQUAL_BEFORE_ROUNDING): + case identity2d(UNUM_IDENTITY_FALLBACK_APPROXIMATELY, + UNUM_IDENTITY_RESULT_NOT_EQUAL): + case identity2d(UNUM_IDENTITY_FALLBACK_APPROXIMATELY_OR_SINGLE_VALUE, + UNUM_IDENTITY_RESULT_NOT_EQUAL): + case identity2d(UNUM_IDENTITY_FALLBACK_SINGLE_VALUE, + UNUM_IDENTITY_RESULT_NOT_EQUAL): + formatRange(data, micros1, micros2, status); + break; + + case identity2d(UNUM_IDENTITY_FALLBACK_APPROXIMATELY, + UNUM_IDENTITY_RESULT_EQUAL_AFTER_ROUNDING): + case identity2d(UNUM_IDENTITY_FALLBACK_APPROXIMATELY, + UNUM_IDENTITY_RESULT_EQUAL_BEFORE_ROUNDING): + case identity2d(UNUM_IDENTITY_FALLBACK_APPROXIMATELY_OR_SINGLE_VALUE, + UNUM_IDENTITY_RESULT_EQUAL_AFTER_ROUNDING): + formatApproximately(data, micros1, micros2, status); + break; + + case identity2d(UNUM_IDENTITY_FALLBACK_APPROXIMATELY_OR_SINGLE_VALUE, + UNUM_IDENTITY_RESULT_EQUAL_BEFORE_ROUNDING): + case identity2d(UNUM_IDENTITY_FALLBACK_SINGLE_VALUE, + UNUM_IDENTITY_RESULT_EQUAL_AFTER_ROUNDING): + case identity2d(UNUM_IDENTITY_FALLBACK_SINGLE_VALUE, + UNUM_IDENTITY_RESULT_EQUAL_BEFORE_ROUNDING): + formatSingleValue(data, micros1, micros2, status); + break; + + default: + U_ASSERT(false); + break; + } +} + + +void NumberRangeFormatterImpl::formatSingleValue(UFormattedNumberRangeData& data, + MicroProps& micros1, MicroProps& micros2, + UErrorCode& status) const { + if (U_FAILURE(status)) { return; } + if (fSameFormatters) { + int32_t length = NumberFormatterImpl::writeNumber(micros1, data.quantity1, data.string, 0, status); + NumberFormatterImpl::writeAffixes(micros1, data.string, 0, length, status); + } else { + formatRange(data, micros1, micros2, status); + } +} + + +void NumberRangeFormatterImpl::formatApproximately (UFormattedNumberRangeData& data, + MicroProps& micros1, MicroProps& micros2, + UErrorCode& status) const { + if (U_FAILURE(status)) { return; } + if (fSameFormatters) { + int32_t length = NumberFormatterImpl::writeNumber(micros1, data.quantity1, data.string, 0, status); + // HEURISTIC: Desired modifier order: inner, middle, approximately, outer. + length += micros1.modInner->apply(data.string, 0, length, status); + length += micros1.modMiddle->apply(data.string, 0, length, status); + length += fApproximatelyModifier.apply(data.string, 0, length, status); + micros1.modOuter->apply(data.string, 0, length, status); + } else { + formatRange(data, micros1, micros2, status); + } +} + + +void NumberRangeFormatterImpl::formatRange(UFormattedNumberRangeData& data, + MicroProps& micros1, MicroProps& micros2, + UErrorCode& status) const { + if (U_FAILURE(status)) { return; } + + // modInner is always notation (scientific); collapsable in ALL. + // modOuter is always units; collapsable in ALL, AUTO, and UNIT. + // modMiddle could be either; collapsable in ALL and sometimes AUTO and UNIT. + // Never collapse an outer mod but not an inner mod. + bool collapseOuter, collapseMiddle, collapseInner; + switch (fCollapse) { + case UNUM_RANGE_COLLAPSE_ALL: + case UNUM_RANGE_COLLAPSE_AUTO: + case UNUM_RANGE_COLLAPSE_UNIT: + { + // OUTER MODIFIER + collapseOuter = micros1.modOuter->semanticallyEquivalent(*micros2.modOuter); + + if (!collapseOuter) { + // Never collapse inner mods if outer mods are not collapsable + collapseMiddle = false; + collapseInner = false; + break; + } + + // MIDDLE MODIFIER + collapseMiddle = micros1.modMiddle->semanticallyEquivalent(*micros2.modMiddle); + + if (!collapseMiddle) { + // Never collapse inner mods if outer mods are not collapsable + collapseInner = false; + break; + } + + // MIDDLE MODIFIER HEURISTICS + // (could disable collapsing of the middle modifier) + // The modifiers are equal by this point, so we can look at just one of them. + const Modifier* mm = micros1.modMiddle; + if (fCollapse == UNUM_RANGE_COLLAPSE_UNIT) { + // Only collapse if the modifier is a unit. + // TODO: Make a better way to check for a unit? + // TODO: Handle case where the modifier has both notation and unit (compact currency)? + if (!mm->containsField(UNUM_CURRENCY_FIELD) && !mm->containsField(UNUM_PERCENT_FIELD)) { + collapseMiddle = false; + } + } else if (fCollapse == UNUM_RANGE_COLLAPSE_AUTO) { + // Heuristic as of ICU 63: collapse only if the modifier is more than one code point. + if (mm->getCodePointCount() <= 1) { + collapseMiddle = false; + } + } + + if (!collapseMiddle || fCollapse != UNUM_RANGE_COLLAPSE_ALL) { + collapseInner = false; + break; + } + + // INNER MODIFIER + collapseInner = micros1.modInner->semanticallyEquivalent(*micros2.modInner); + + // All done checking for collapsability. + break; + } + + default: + collapseOuter = false; + collapseMiddle = false; + collapseInner = false; + break; + } + + NumberStringBuilder& string = data.string; + int32_t lengthPrefix = 0; + int32_t length1 = 0; + int32_t lengthInfix = 0; + int32_t length2 = 0; + int32_t lengthSuffix = 0; + + // Use #define so that these are evaluated at the call site. + #define UPRV_INDEX_0 (lengthPrefix) + #define UPRV_INDEX_1 (lengthPrefix + length1) + #define UPRV_INDEX_2 (lengthPrefix + length1 + lengthInfix) + #define UPRV_INDEX_3 (lengthPrefix + length1 + lengthInfix + length2) + + int32_t lengthRange = SimpleModifier::formatTwoArgPattern( + fRangeFormatter, + string, + 0, + &lengthPrefix, + &lengthSuffix, + UNUM_FIELD_COUNT, + status); + if (U_FAILURE(status)) { return; } + lengthInfix = lengthRange - lengthPrefix - lengthSuffix; + U_ASSERT(lengthInfix > 0); + + // SPACING HEURISTIC + // Add spacing unless all modifiers are collapsed. + // TODO: add API to control this? + // TODO: Use a data-driven heuristic like currency spacing? + // TODO: Use Unicode [:whitespace:] instead of PatternProps whitespace? (consider speed implications) + { + bool repeatInner = !collapseInner && micros1.modInner->getCodePointCount() > 0; + bool repeatMiddle = !collapseMiddle && micros1.modMiddle->getCodePointCount() > 0; + bool repeatOuter = !collapseOuter && micros1.modOuter->getCodePointCount() > 0; + if (repeatInner || repeatMiddle || repeatOuter) { + // Add spacing if there is not already spacing + if (!PatternProps::isWhiteSpace(string.charAt(UPRV_INDEX_1))) { + lengthInfix += string.insertCodePoint(UPRV_INDEX_1, u'\u0020', UNUM_FIELD_COUNT, status); + } + if (!PatternProps::isWhiteSpace(string.charAt(UPRV_INDEX_2 - 1))) { + lengthInfix += string.insertCodePoint(UPRV_INDEX_2, u'\u0020', UNUM_FIELD_COUNT, status); + } + } + } + + length1 += NumberFormatterImpl::writeNumber(micros1, data.quantity1, string, UPRV_INDEX_0, status); + length2 += NumberFormatterImpl::writeNumber(micros2, data.quantity2, string, UPRV_INDEX_2, status); + + // TODO: Support padding? + + if (collapseInner) { + // Note: this is actually a mix of prefix and suffix, but adding to infix length works + const Modifier& mod = resolveModifierPlurals(*micros1.modInner, *micros2.modInner); + lengthInfix += mod.apply(string, UPRV_INDEX_0, UPRV_INDEX_3, status); + } else { + length1 += micros1.modInner->apply(string, UPRV_INDEX_0, UPRV_INDEX_1, status); + length2 += micros2.modInner->apply(string, UPRV_INDEX_2, UPRV_INDEX_3, status); + } + + if (collapseMiddle) { + // Note: this is actually a mix of prefix and suffix, but adding to infix length works + const Modifier& mod = resolveModifierPlurals(*micros1.modMiddle, *micros2.modMiddle); + lengthInfix += mod.apply(string, UPRV_INDEX_0, UPRV_INDEX_3, status); + } else { + length1 += micros1.modMiddle->apply(string, UPRV_INDEX_0, UPRV_INDEX_1, status); + length2 += micros2.modMiddle->apply(string, UPRV_INDEX_2, UPRV_INDEX_3, status); + } + + if (collapseOuter) { + // Note: this is actually a mix of prefix and suffix, but adding to infix length works + const Modifier& mod = resolveModifierPlurals(*micros1.modOuter, *micros2.modOuter); + lengthInfix += mod.apply(string, UPRV_INDEX_0, UPRV_INDEX_3, status); + } else { + length1 += micros1.modOuter->apply(string, UPRV_INDEX_0, UPRV_INDEX_1, status); + length2 += micros2.modOuter->apply(string, UPRV_INDEX_2, UPRV_INDEX_3, status); + } +} + + +const Modifier& +NumberRangeFormatterImpl::resolveModifierPlurals(const Modifier& first, const Modifier& second) const { + Modifier::Parameters parameters; + first.getParameters(parameters); + if (parameters.obj == nullptr) { + // No plural form; return a fallback (e.g., the first) + return first; + } + StandardPlural::Form firstPlural = parameters.plural; + + second.getParameters(parameters); + if (parameters.obj == nullptr) { + // No plural form; return a fallback (e.g., the first) + return first; + } + StandardPlural::Form secondPlural = parameters.plural; + + // Get the required plural form from data + StandardPlural::Form resultPlural = fPluralRanges.resolve(firstPlural, secondPlural); + + // Get and return the new Modifier + const Modifier* mod = parameters.obj->getModifier(parameters.signum, resultPlural); + U_ASSERT(mod != nullptr); + return *mod; +} + + + +#endif /* #if !UCONFIG_NO_FORMATTING */ diff --git a/deps/icu-small/source/i18n/numrange_impl.h b/deps/icu-small/source/i18n/numrange_impl.h new file mode 100644 index 0000000000..787fc65686 --- /dev/null +++ b/deps/icu-small/source/i18n/numrange_impl.h @@ -0,0 +1,114 @@ +// © 2018 and later: Unicode, Inc. and others. +// License & terms of use: http://www.unicode.org/copyright.html + +#include "unicode/utypes.h" + +#if !UCONFIG_NO_FORMATTING +#ifndef __SOURCE_NUMRANGE_TYPES_H__ +#define __SOURCE_NUMRANGE_TYPES_H__ + +#include "unicode/numberformatter.h" +#include "unicode/numberrangeformatter.h" +#include "unicode/simpleformatter.h" +#include "number_types.h" +#include "number_decimalquantity.h" +#include "number_formatimpl.h" +#include "number_stringbuilder.h" + +U_NAMESPACE_BEGIN namespace number { +namespace impl { + + +/** + * Class similar to UFormattedNumberData. + * + * Has incomplete magic number logic that will need to be finished + * if this is to be exposed as C API in the future. + */ +struct UFormattedNumberRangeData : public UMemory { + // The magic number to identify incoming objects. + // Reads in ASCII as "FDR" (FormatteDnumberRange with room at the end) + static constexpr int32_t kMagic = 0x46445200; + + // Data members: + int32_t fMagic = kMagic; + DecimalQuantity quantity1; + DecimalQuantity quantity2; + NumberStringBuilder string; + UNumberRangeIdentityResult identityResult = UNUM_IDENTITY_RESULT_COUNT; + + // No C conversion methods (no C API yet) +}; + + +class StandardPluralRanges : public UMemory { + public: + void initialize(const Locale& locale, UErrorCode& status); + StandardPlural::Form resolve(StandardPlural::Form first, StandardPlural::Form second) const; + + /** Used for data loading. */ + void addPluralRange( + StandardPlural::Form first, + StandardPlural::Form second, + StandardPlural::Form result); + + /** Used for data loading. */ + void setCapacity(int32_t length); + + private: + struct StandardPluralRangeTriple { + StandardPlural::Form first; + StandardPlural::Form second; + StandardPlural::Form result; + }; + + // TODO: An array is simple here, but it results in linear lookup time. + // Certain locales have 20-30 entries in this list. + // Consider changing to a smarter data structure. + typedef MaybeStackArray<StandardPluralRangeTriple, 3> PluralRangeTriples; + PluralRangeTriples fTriples; + int32_t fTriplesLen = 0; +}; + + +class NumberRangeFormatterImpl : public UMemory { + public: + NumberRangeFormatterImpl(const RangeMacroProps& macros, UErrorCode& status); + + void format(UFormattedNumberRangeData& data, bool equalBeforeRounding, UErrorCode& status) const; + + private: + NumberFormatterImpl formatterImpl1; + NumberFormatterImpl formatterImpl2; + bool fSameFormatters; + + UNumberRangeCollapse fCollapse; + UNumberRangeIdentityFallback fIdentityFallback; + + SimpleFormatter fRangeFormatter; + SimpleModifier fApproximatelyModifier; + + StandardPluralRanges fPluralRanges; + + void formatSingleValue(UFormattedNumberRangeData& data, + MicroProps& micros1, MicroProps& micros2, + UErrorCode& status) const; + + void formatApproximately(UFormattedNumberRangeData& data, + MicroProps& micros1, MicroProps& micros2, + UErrorCode& status) const; + + void formatRange(UFormattedNumberRangeData& data, + MicroProps& micros1, MicroProps& micros2, + UErrorCode& status) const; + + const Modifier& resolveModifierPlurals(const Modifier& first, const Modifier& second) const; +}; + + +} // namespace impl +} // namespace number +U_NAMESPACE_END + +#endif //__SOURCE_NUMRANGE_TYPES_H__ +#endif /* #if !UCONFIG_NO_FORMATTING */ diff --git a/deps/icu-small/source/i18n/numsys.cpp b/deps/icu-small/source/i18n/numsys.cpp index 087dd277eb..514fe05e5a 100644 --- a/deps/icu-small/source/i18n/numsys.cpp +++ b/deps/icu-small/source/i18n/numsys.cpp @@ -79,43 +79,45 @@ NumberingSystem* U_EXPORT2 NumberingSystem::createInstance(int32_t radix_in, UBool isAlgorithmic_in, const UnicodeString & desc_in, UErrorCode &status) { if (U_FAILURE(status)) { - return NULL; + return nullptr; } if ( radix_in < 2 ) { status = U_ILLEGAL_ARGUMENT_ERROR; - return NULL; + return nullptr; } if ( !isAlgorithmic_in ) { if ( desc_in.countChar32() != radix_in ) { status = U_ILLEGAL_ARGUMENT_ERROR; - return NULL; + return nullptr; } } - NumberingSystem *ns = new NumberingSystem(); + LocalPointer<NumberingSystem> ns(new NumberingSystem(), status); + if (U_FAILURE(status)) { + return nullptr; + } ns->setRadix(radix_in); ns->setDesc(desc_in); ns->setAlgorithmic(isAlgorithmic_in); - ns->setName(NULL); - return ns; + ns->setName(nullptr); + return ns.orphan(); } - NumberingSystem* U_EXPORT2 NumberingSystem::createInstance(const Locale & inLocale, UErrorCode& status) { if (U_FAILURE(status)) { - return NULL; + return nullptr; } UBool nsResolved = TRUE; UBool usingFallback = FALSE; char buffer[ULOC_KEYWORDS_CAPACITY]; - int32_t count = inLocale.getKeywordValue("numbers",buffer, sizeof(buffer),status); + int32_t count = inLocale.getKeywordValue("numbers", buffer, sizeof(buffer), status); if (U_FAILURE(status) || status == U_STRING_NOT_TERMINATED_WARNING) { // the "numbers" keyword exceeds ULOC_KEYWORDS_CAPACITY; ignore and use default. count = 0; @@ -129,20 +131,30 @@ NumberingSystem::createInstance(const Locale & inLocale, UErrorCode& status) { nsResolved = FALSE; } } else { - uprv_strcpy(buffer,gDefault); + uprv_strcpy(buffer, gDefault); nsResolved = FALSE; } if (!nsResolved) { // Resolve the numbering system ( default, native, traditional or finance ) into a "real" numbering system UErrorCode localStatus = U_ZERO_ERROR; - UResourceBundle *resource = ures_open(NULL, inLocale.getName(), &localStatus); - UResourceBundle *numberElementsRes = ures_getByKey(resource,gNumberElements,NULL,&localStatus); + LocalUResourceBundlePointer resource(ures_open(nullptr, inLocale.getName(), &localStatus)); + LocalUResourceBundlePointer numberElementsRes(ures_getByKey(resource.getAlias(), gNumberElements, nullptr, &localStatus)); + // Don't stomp on the catastrophic failure of OOM. + if (localStatus == U_MEMORY_ALLOCATION_ERROR) { + status = U_MEMORY_ALLOCATION_ERROR; + return nullptr; + } while (!nsResolved) { localStatus = U_ZERO_ERROR; count = 0; - const UChar *nsName = ures_getStringByKeyWithFallback(numberElementsRes, buffer, &count, &localStatus); + const UChar *nsName = ures_getStringByKeyWithFallback(numberElementsRes.getAlias(), buffer, &count, &localStatus); + // Don't stomp on the catastrophic failure of OOM. + if (localStatus == U_MEMORY_ALLOCATION_ERROR) { + status = U_MEMORY_ALLOCATION_ERROR; + return nullptr; + } if ( count > 0 && count < ULOC_KEYWORDS_CAPACITY ) { // numbering system found - u_UCharsToChars(nsName,buffer,count); + u_UCharsToChars(nsName, buffer, count); buffer[count] = '\0'; // Make sure it is null terminated. nsResolved = TRUE; } @@ -158,16 +170,17 @@ NumberingSystem::createInstance(const Locale & inLocale, UErrorCode& status) { } } } - ures_close(numberElementsRes); - ures_close(resource); } if (usingFallback) { status = U_USING_FALLBACK_WARNING; NumberingSystem *ns = new NumberingSystem(); + if (ns == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + } return ns; } else { - return NumberingSystem::createInstanceByName(buffer,status); + return NumberingSystem::createInstanceByName(buffer, status); } } @@ -178,36 +191,37 @@ NumberingSystem::createInstance(UErrorCode& status) { NumberingSystem* U_EXPORT2 NumberingSystem::createInstanceByName(const char *name, UErrorCode& status) { - UResourceBundle *numberingSystemsInfo = NULL; - UResourceBundle *nsTop, *nsCurrent; int32_t radix = 10; int32_t algorithmic = 0; - numberingSystemsInfo = ures_openDirect(NULL,gNumberingSystems, &status); - nsCurrent = ures_getByKey(numberingSystemsInfo,gNumberingSystems,NULL,&status); - nsTop = ures_getByKey(nsCurrent,name,NULL,&status); - UnicodeString nsd = ures_getUnicodeStringByKey(nsTop,gDesc,&status); + LocalUResourceBundlePointer numberingSystemsInfo(ures_openDirect(nullptr, gNumberingSystems, &status)); + LocalUResourceBundlePointer nsCurrent(ures_getByKey(numberingSystemsInfo.getAlias(), gNumberingSystems, nullptr, &status)); + LocalUResourceBundlePointer nsTop(ures_getByKey(nsCurrent.getAlias(), name, nullptr, &status)); - ures_getByKey(nsTop,gRadix,nsCurrent,&status); - radix = ures_getInt(nsCurrent,&status); + UnicodeString nsd = ures_getUnicodeStringByKey(nsTop.getAlias(), gDesc, &status); - ures_getByKey(nsTop,gAlgorithmic,nsCurrent,&status); - algorithmic = ures_getInt(nsCurrent,&status); + ures_getByKey(nsTop.getAlias(), gRadix, nsCurrent.getAlias(), &status); + radix = ures_getInt(nsCurrent.getAlias(), &status); - UBool isAlgorithmic = ( algorithmic == 1 ); + ures_getByKey(nsTop.getAlias(), gAlgorithmic, nsCurrent.getAlias(), &status); + algorithmic = ures_getInt(nsCurrent.getAlias(), &status); - ures_close(nsCurrent); - ures_close(nsTop); - ures_close(numberingSystemsInfo); + UBool isAlgorithmic = ( algorithmic == 1 ); if (U_FAILURE(status)) { - status = U_UNSUPPORTED_ERROR; - return NULL; + // Don't stomp on the catastrophic failure of OOM. + if (status != U_MEMORY_ALLOCATION_ERROR) { + status = U_UNSUPPORTED_ERROR; + } + return nullptr; } - NumberingSystem* ns = NumberingSystem::createInstance(radix,isAlgorithmic,nsd,status); + LocalPointer<NumberingSystem> ns(NumberingSystem::createInstance(radix, isAlgorithmic, nsd, status), status); + if (U_FAILURE(status)) { + return nullptr; + } ns->setName(name); - return ns; + return ns.orphan(); } /** @@ -241,11 +255,11 @@ void NumberingSystem::setDesc(const UnicodeString &d) { desc.setTo(d); } void NumberingSystem::setName(const char *n) { - if ( n == NULL ) { + if ( n == nullptr ) { name[0] = (char) 0; } else { uprv_strncpy(name,n,NUMSYS_NAME_CAPACITY); - name[NUMSYS_NAME_CAPACITY] = (char)0; // Make sure it is null terminated. + name[NUMSYS_NAME_CAPACITY] = '\0'; // Make sure it is null terminated. } } UBool NumberingSystem::isAlgorithmic() const { @@ -254,43 +268,57 @@ UBool NumberingSystem::isAlgorithmic() const { StringEnumeration* NumberingSystem::getAvailableNames(UErrorCode &status) { // TODO(ticket #11908): Init-once static cache, with u_cleanup() callback. - static StringEnumeration* availableNames = NULL; + static StringEnumeration* availableNames = nullptr; if (U_FAILURE(status)) { - return NULL; + return nullptr; } - if ( availableNames == NULL ) { + if ( availableNames == nullptr ) { // TODO: Simple array of UnicodeString objects, based on length of table resource? - LocalPointer<UVector> numsysNames(new UVector(uprv_deleteUObject, NULL, status), status); + LocalPointer<UVector> numsysNames(new UVector(uprv_deleteUObject, nullptr, status), status); if (U_FAILURE(status)) { - return NULL; + return nullptr; } UErrorCode rbstatus = U_ZERO_ERROR; - UResourceBundle *numberingSystemsInfo = ures_openDirect(NULL, "numberingSystems", &rbstatus); - numberingSystemsInfo = ures_getByKey(numberingSystemsInfo,"numberingSystems",numberingSystemsInfo,&rbstatus); - if(U_FAILURE(rbstatus)) { - status = U_MISSING_RESOURCE_ERROR; + UResourceBundle *numberingSystemsInfo = ures_openDirect(nullptr, "numberingSystems", &rbstatus); + numberingSystemsInfo = ures_getByKey(numberingSystemsInfo, "numberingSystems", numberingSystemsInfo, &rbstatus); + if (U_FAILURE(rbstatus)) { + // Don't stomp on the catastrophic failure of OOM. + if (rbstatus == U_MEMORY_ALLOCATION_ERROR) { + status = rbstatus; + } else { + status = U_MISSING_RESOURCE_ERROR; + } ures_close(numberingSystemsInfo); - return NULL; + return nullptr; } - while ( ures_hasNext(numberingSystemsInfo) ) { - UResourceBundle *nsCurrent = ures_getNextResource(numberingSystemsInfo,NULL,&rbstatus); - const char *nsName = ures_getKey(nsCurrent); - numsysNames->addElement(new UnicodeString(nsName, -1, US_INV),status); - ures_close(nsCurrent); + while ( ures_hasNext(numberingSystemsInfo) && U_SUCCESS(status) ) { + LocalUResourceBundlePointer nsCurrent(ures_getNextResource(numberingSystemsInfo, nullptr, &rbstatus)); + if (rbstatus == U_MEMORY_ALLOCATION_ERROR) { + status = rbstatus; // we want to report OOM failure back to the caller. + break; + } + const char *nsName = ures_getKey(nsCurrent.getAlias()); + LocalPointer<UnicodeString> newElem(new UnicodeString(nsName, -1, US_INV), status); + if (U_SUCCESS(status)) { + numsysNames->addElement(newElem.getAlias(), status); + if (U_SUCCESS(status)) { + newElem.orphan(); // on success, the numsysNames vector owns newElem. + } + } } ures_close(numberingSystemsInfo); if (U_FAILURE(status)) { - return NULL; + return nullptr; } availableNames = new NumsysNameEnumeration(numsysNames.getAlias(), status); - if (availableNames == NULL) { + if (availableNames == nullptr) { status = U_MEMORY_ALLOCATION_ERROR; - return NULL; + return nullptr; } numsysNames.orphan(); // The names got adopted. } @@ -305,10 +333,10 @@ NumsysNameEnumeration::NumsysNameEnumeration(UVector *numsysNames, UErrorCode& / const UnicodeString* NumsysNameEnumeration::snext(UErrorCode& status) { - if (U_SUCCESS(status) && pos < fNumsysNames->size()) { + if (U_SUCCESS(status) && (fNumsysNames != nullptr) && (pos < fNumsysNames->size())) { return (const UnicodeString*)fNumsysNames->elementAt(pos++); } - return NULL; + return nullptr; } void @@ -318,7 +346,7 @@ NumsysNameEnumeration::reset(UErrorCode& /*status*/) { int32_t NumsysNameEnumeration::count(UErrorCode& /*status*/) const { - return (fNumsysNames==NULL) ? 0 : fNumsysNames->size(); + return (fNumsysNames==nullptr) ? 0 : fNumsysNames->size(); } NumsysNameEnumeration::~NumsysNameEnumeration() { diff --git a/deps/icu-small/source/i18n/numsys_impl.h b/deps/icu-small/source/i18n/numsys_impl.h index 6385fa5408..733e102365 100644 --- a/deps/icu-small/source/i18n/numsys_impl.h +++ b/deps/icu-small/source/i18n/numsys_impl.h @@ -37,7 +37,7 @@ public: virtual int32_t count(UErrorCode& status) const; private: int32_t pos; - UVector *fNumsysNames; + UVector *fNumsysNames = nullptr; }; U_NAMESPACE_END diff --git a/deps/icu-small/source/i18n/olsontz.cpp b/deps/icu-small/source/i18n/olsontz.cpp index df025c0808..c350057466 100644 --- a/deps/icu-small/source/i18n/olsontz.cpp +++ b/deps/icu-small/source/i18n/olsontz.cpp @@ -140,7 +140,7 @@ OlsonTimeZone::OlsonTimeZone(const UResourceBundle* top, // Pre-32bit second transitions ures_getByKey(res, kTRANSPRE32, &r, &ec); transitionTimesPre32 = ures_getIntVector(&r, &len, &ec); - transitionCountPre32 = len >> 1; + transitionCountPre32 = static_cast<int16_t>(len >> 1); if (ec == U_MISSING_RESOURCE_ERROR) { // No pre-32bit transitions transitionTimesPre32 = NULL; @@ -153,7 +153,7 @@ OlsonTimeZone::OlsonTimeZone(const UResourceBundle* top, // 32bit second transitions ures_getByKey(res, kTRANS, &r, &ec); transitionTimes32 = ures_getIntVector(&r, &len, &ec); - transitionCount32 = len; + transitionCount32 = static_cast<int16_t>(len); if (ec == U_MISSING_RESOURCE_ERROR) { // No 32bit transitions transitionTimes32 = NULL; @@ -166,7 +166,7 @@ OlsonTimeZone::OlsonTimeZone(const UResourceBundle* top, // Post-32bit second transitions ures_getByKey(res, kTRANSPOST32, &r, &ec); transitionTimesPost32 = ures_getIntVector(&r, &len, &ec); - transitionCountPost32 = len >> 1; + transitionCountPost32 = static_cast<int16_t>(len >> 1); if (ec == U_MISSING_RESOURCE_ERROR) { // No pre-32bit transitions transitionTimesPost32 = NULL; diff --git a/deps/icu-small/source/i18n/plurrule.cpp b/deps/icu-small/source/i18n/plurrule.cpp index 9597e8eb00..3caa48a57b 100644 --- a/deps/icu-small/source/i18n/plurrule.cpp +++ b/deps/icu-small/source/i18n/plurrule.cpp @@ -65,13 +65,15 @@ UOBJECT_DEFINE_RTTI_IMPLEMENTATION(PluralKeywordEnumeration) PluralRules::PluralRules(UErrorCode& /*status*/) : UObject(), - mRules(NULL) + mRules(nullptr), + mInternalStatus(U_ZERO_ERROR) { } PluralRules::PluralRules(const PluralRules& other) : UObject(other), - mRules(NULL) + mRules(nullptr), + mInternalStatus(U_ZERO_ERROR) { *this=other; } @@ -86,54 +88,67 @@ SharedPluralRules::~SharedPluralRules() { PluralRules* PluralRules::clone() const { - return new PluralRules(*this); + PluralRules* newObj = new PluralRules(*this); + // Since clone doesn't have a 'status' parameter, the best we can do is return nullptr if + // the newly created object was not fully constructed properly (an error occurred). + if (newObj != nullptr && U_FAILURE(newObj->mInternalStatus)) { + delete newObj; + newObj = nullptr; + } + return newObj; } PluralRules& PluralRules::operator=(const PluralRules& other) { if (this != &other) { delete mRules; - if (other.mRules==NULL) { - mRules = NULL; + mRules = nullptr; + mInternalStatus = other.mInternalStatus; + if (U_FAILURE(mInternalStatus)) { + // bail out early if the object we were copying from was already 'invalid'. + return *this; } - else { + if (other.mRules != nullptr) { mRules = new RuleChain(*other.mRules); + if (mRules == nullptr) { + mInternalStatus = U_MEMORY_ALLOCATION_ERROR; + } + else if (U_FAILURE(mRules->fInternalStatus)) { + // If the RuleChain wasn't fully copied, then set our status to failure as well. + mInternalStatus = mRules->fInternalStatus; + } } } - return *this; } StringEnumeration* PluralRules::getAvailableLocales(UErrorCode &status) { - StringEnumeration *result = new PluralAvailableLocalesEnumeration(status); - if (result == NULL && U_SUCCESS(status)) { - status = U_MEMORY_ALLOCATION_ERROR; + if (U_FAILURE(status)) { + return nullptr; } + LocalPointer<StringEnumeration> result(new PluralAvailableLocalesEnumeration(status), status); if (U_FAILURE(status)) { - delete result; - result = NULL; + return nullptr; } - return result; + return result.orphan(); } PluralRules* U_EXPORT2 PluralRules::createRules(const UnicodeString& description, UErrorCode& status) { if (U_FAILURE(status)) { - return NULL; + return nullptr; } - PluralRuleParser parser; - PluralRules *newRules = new PluralRules(status); - if (U_SUCCESS(status) && newRules == NULL) { - status = U_MEMORY_ALLOCATION_ERROR; + LocalPointer<PluralRules> newRules(new PluralRules(status), status); + if (U_FAILURE(status)) { + return nullptr; } - parser.parse(description, newRules, status); + parser.parse(description, newRules.getAlias(), status); if (U_FAILURE(status)) { - delete newRules; - newRules = NULL; + newRules.adoptInstead(nullptr); } - return newRules; + return newRules.orphan(); } @@ -149,19 +164,17 @@ template<> U_I18N_API const SharedPluralRules *LocaleCacheKey<SharedPluralRules>::createObject( const void * /*unused*/, UErrorCode &status) const { const char *localeId = fLoc.getName(); - PluralRules *pr = PluralRules::internalForLocale( - localeId, UPLURAL_TYPE_CARDINAL, status); + LocalPointer<PluralRules> pr(PluralRules::internalForLocale(localeId, UPLURAL_TYPE_CARDINAL, status), status); if (U_FAILURE(status)) { - return NULL; + return nullptr; } - SharedPluralRules *result = new SharedPluralRules(pr); - if (result == NULL) { - status = U_MEMORY_ALLOCATION_ERROR; - delete pr; - return NULL; + LocalPointer<SharedPluralRules> result(new SharedPluralRules(pr.getAlias()), status); + if (U_FAILURE(status)) { + return nullptr; } + pr.orphan(); // result was successfully created so it nows pr. result->addRef(); - return result; + return result.orphan(); } /* end plural rules cache */ @@ -171,13 +184,13 @@ const SharedPluralRules* U_EXPORT2 PluralRules::createSharedInstance( const Locale& locale, UPluralType type, UErrorCode& status) { if (U_FAILURE(status)) { - return NULL; + return nullptr; } if (type != UPLURAL_TYPE_CARDINAL) { status = U_UNSUPPORTED_ERROR; - return NULL; + return nullptr; } - const SharedPluralRules *result = NULL; + const SharedPluralRules *result = nullptr; UnifiedCache::getByLocale(locale, result, status); return result; } @@ -195,11 +208,11 @@ PluralRules::forLocale(const Locale& locale, UPluralType type, UErrorCode& statu const SharedPluralRules *shared = createSharedInstance( locale, type, status); if (U_FAILURE(status)) { - return NULL; + return nullptr; } PluralRules *result = (*shared)->clone(); shared->removeRef(); - if (result == NULL) { + if (result == nullptr) { status = U_MEMORY_ALLOCATION_ERROR; } return result; @@ -208,20 +221,23 @@ PluralRules::forLocale(const Locale& locale, UPluralType type, UErrorCode& statu PluralRules* U_EXPORT2 PluralRules::internalForLocale(const Locale& locale, UPluralType type, UErrorCode& status) { if (U_FAILURE(status)) { - return NULL; + return nullptr; } if (type >= UPLURAL_TYPE_COUNT) { status = U_ILLEGAL_ARGUMENT_ERROR; - return NULL; + return nullptr; } - PluralRules *newObj = new PluralRules(status); - if (newObj==NULL || U_FAILURE(status)) { - delete newObj; - return NULL; + LocalPointer<PluralRules> newObj(new PluralRules(status), status); + if (U_FAILURE(status)) { + return nullptr; } UnicodeString locRule = newObj->getRuleFromResource(locale, type, status); - // TODO: which errors, if any, should be returned? + // TODO: which other errors, if any, should be returned? if (locRule.length() == 0) { + // If an out-of-memory error occurred, then stop and report the failure. + if (status == U_MEMORY_ALLOCATION_ERROR) { + return nullptr; + } // Locales with no specific rules (all numbers have the "other" category // will return a U_MISSING_RESOURCE_ERROR at this point. This is not // an error. @@ -229,13 +245,13 @@ PluralRules::internalForLocale(const Locale& locale, UPluralType type, UErrorCod status = U_ZERO_ERROR; } PluralRuleParser parser; - parser.parse(locRule, newObj, status); + parser.parse(locRule, newObj.getAlias(), status); // TODO: should rule parse errors be returned, or // should we silently use default rules? // Original impl used default rules. // Ask the question to ICU Core. - return newObj; + return newObj.orphan(); } UnicodeString @@ -250,7 +266,7 @@ PluralRules::select(double number) const { UnicodeString PluralRules::select(const IFixedDecimal &number) const { - if (mRules == NULL) { + if (mRules == nullptr) { return UnicodeString(TRUE, PLURAL_DEFAULT_RULE, -1); } else { @@ -262,14 +278,18 @@ PluralRules::select(const IFixedDecimal &number) const { StringEnumeration* PluralRules::getKeywords(UErrorCode& status) const { - if (U_FAILURE(status)) return NULL; - StringEnumeration* nameEnumerator = new PluralKeywordEnumeration(mRules, status); if (U_FAILURE(status)) { - delete nameEnumerator; - return NULL; + return nullptr; } - - return nameEnumerator; + if (U_FAILURE(mInternalStatus)) { + status = mInternalStatus; + return nullptr; + } + LocalPointer<StringEnumeration> nameEnumerator(new PluralKeywordEnumeration(mRules, status), status); + if (U_FAILURE(status)) { + return nullptr; + } + return nameEnumerator.orphan(); } double @@ -367,8 +387,15 @@ getSamplesFromString(const UnicodeString &samples, double *dest, int32_t PluralRules::getSamples(const UnicodeString &keyword, double *dest, int32_t destCapacity, UErrorCode& status) { + if (destCapacity == 0 || U_FAILURE(status)) { + return 0; + } + if (U_FAILURE(mInternalStatus)) { + status = mInternalStatus; + return 0; + } RuleChain *rc = rulesForKeyword(keyword); - if (rc == NULL || destCapacity == 0 || U_FAILURE(status)) { + if (rc == nullptr) { return 0; } int32_t numSamples = getSamplesFromString(rc->fIntegerSamples, dest, destCapacity, status); @@ -381,7 +408,7 @@ PluralRules::getSamples(const UnicodeString &keyword, double *dest, RuleChain *PluralRules::rulesForKeyword(const UnicodeString &keyword) const { RuleChain *rc; - for (rc = mRules; rc != NULL; rc = rc->fNext) { + for (rc = mRules; rc != nullptr; rc = rc->fNext) { if (rc->fKeyword == keyword) { break; } @@ -395,7 +422,7 @@ PluralRules::isKeyword(const UnicodeString& keyword) const { if (0 == keyword.compare(PLURAL_KEYWORD_OTHER, 5)) { return true; } - return rulesForKeyword(keyword) != NULL; + return rulesForKeyword(keyword) != nullptr; } UnicodeString @@ -421,13 +448,13 @@ PluralRules::operator==(const PluralRules& other) const { return FALSE; } myKeywordList->reset(status); - while ((ptrKeyword=myKeywordList->snext(status))!=NULL) { + while ((ptrKeyword=myKeywordList->snext(status))!=nullptr) { if (!other.isKeyword(*ptrKeyword)) { return FALSE; } } otherKeywordList->reset(status); - while ((ptrKeyword=otherKeywordList->snext(status))!=NULL) { + while ((ptrKeyword=otherKeywordList->snext(status))!=nullptr) { if (!this->isKeyword(*ptrKeyword)) { return FALSE; } @@ -460,29 +487,33 @@ PluralRuleParser::parse(const UnicodeString& ruleData, PluralRules *prules, UErr } switch (type) { case tAnd: - U_ASSERT(curAndConstraint != NULL); - curAndConstraint = curAndConstraint->add(); + U_ASSERT(curAndConstraint != nullptr); + curAndConstraint = curAndConstraint->add(status); break; case tOr: { - U_ASSERT(currentChain != NULL); + U_ASSERT(currentChain != nullptr); OrConstraint *orNode=currentChain->ruleHeader; - while (orNode->next != NULL) { + while (orNode->next != nullptr) { orNode = orNode->next; } orNode->next= new OrConstraint(); + if (orNode->next == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + break; + } orNode=orNode->next; - orNode->next=NULL; - curAndConstraint = orNode->add(); + orNode->next=nullptr; + curAndConstraint = orNode->add(status); } break; case tIs: - U_ASSERT(curAndConstraint != NULL); + U_ASSERT(curAndConstraint != nullptr); U_ASSERT(curAndConstraint->value == -1); - U_ASSERT(curAndConstraint->rangeList == NULL); + U_ASSERT(curAndConstraint->rangeList == nullptr); break; case tNot: - U_ASSERT(curAndConstraint != NULL); + U_ASSERT(curAndConstraint != nullptr); curAndConstraint->negated=TRUE; break; @@ -492,23 +523,29 @@ PluralRuleParser::parse(const UnicodeString& ruleData, PluralRules *prules, UErr case tIn: case tWithin: case tEqual: - U_ASSERT(curAndConstraint != NULL); - curAndConstraint->rangeList = new UVector32(status); - curAndConstraint->rangeList->addElement(-1, status); // range Low - curAndConstraint->rangeList->addElement(-1, status); // range Hi - rangeLowIdx = 0; - rangeHiIdx = 1; - curAndConstraint->value=PLURAL_RANGE_HIGH; - curAndConstraint->integerOnly = (type != tWithin); + { + U_ASSERT(curAndConstraint != nullptr); + LocalPointer<UVector32> newRangeList(new UVector32(status), status); + if (U_FAILURE(status)) { + break; + } + curAndConstraint->rangeList = newRangeList.orphan(); + curAndConstraint->rangeList->addElement(-1, status); // range Low + curAndConstraint->rangeList->addElement(-1, status); // range Hi + rangeLowIdx = 0; + rangeHiIdx = 1; + curAndConstraint->value=PLURAL_RANGE_HIGH; + curAndConstraint->integerOnly = (type != tWithin); + } break; case tNumber: - U_ASSERT(curAndConstraint != NULL); + U_ASSERT(curAndConstraint != nullptr); if ( (curAndConstraint->op==AndConstraint::MOD)&& (curAndConstraint->opNum == -1 ) ) { curAndConstraint->opNum=getNumberValue(token); } else { - if (curAndConstraint->rangeList == NULL) { + if (curAndConstraint->rangeList == nullptr) { // this is for an 'is' rule curAndConstraint->value = getNumberValue(token); } else { @@ -534,7 +571,7 @@ PluralRuleParser::parse(const UnicodeString& ruleData, PluralRules *prules, UErr case tComma: // TODO: rule syntax checking is inadequate, can happen with badly formed rules. // Catch cases like "n mod 10, is 1" here instead. - if (curAndConstraint == NULL || curAndConstraint->rangeList == NULL) { + if (curAndConstraint == nullptr || curAndConstraint->rangeList == nullptr) { status = U_UNEXPECTED_TOKEN; break; } @@ -545,7 +582,7 @@ PluralRuleParser::parse(const UnicodeString& ruleData, PluralRules *prules, UErr curAndConstraint->rangeList->addElement(-1, status); // range Hi break; case tMod: - U_ASSERT(curAndConstraint != NULL); + U_ASSERT(curAndConstraint != nullptr); curAndConstraint->op=AndConstraint::MOD; break; case tVariableN: @@ -553,24 +590,24 @@ PluralRuleParser::parse(const UnicodeString& ruleData, PluralRules *prules, UErr case tVariableF: case tVariableT: case tVariableV: - U_ASSERT(curAndConstraint != NULL); + U_ASSERT(curAndConstraint != nullptr); curAndConstraint->digitsType = type; break; case tKeyword: { RuleChain *newChain = new RuleChain; - if (newChain == NULL) { + if (newChain == nullptr) { status = U_MEMORY_ALLOCATION_ERROR; break; } newChain->fKeyword = token; - if (prules->mRules == NULL) { + if (prules->mRules == nullptr) { prules->mRules = newChain; } else { // The new rule chain goes at the end of the linked list of rule chains, // unless there is an "other" keyword & chain. "other" must remain last. RuleChain *insertAfter = prules->mRules; - while (insertAfter->fNext!=NULL && + while (insertAfter->fNext!=nullptr && insertAfter->fNext->fKeyword.compare(PLURAL_KEYWORD_OTHER, 5) != 0 ){ insertAfter=insertAfter->fNext; } @@ -578,8 +615,12 @@ PluralRuleParser::parse(const UnicodeString& ruleData, PluralRules *prules, UErr insertAfter->fNext = newChain; } OrConstraint *orNode = new OrConstraint(); + if (orNode == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + break; + } newChain->ruleHeader = orNode; - curAndConstraint = orNode->add(); + curAndConstraint = orNode->add(status); currentChain = newChain; } break; @@ -629,7 +670,7 @@ PluralRules::getRuleFromResource(const Locale& locale, UPluralType type, UErrorC if (U_FAILURE(errCode)) { return emptyStr; } - LocalUResourceBundlePointer rb(ures_openDirect(NULL, "plurals", &errCode)); + LocalUResourceBundlePointer rb(ures_openDirect(nullptr, "plurals", &errCode)); if(U_FAILURE(errCode)) { return emptyStr; } @@ -646,7 +687,7 @@ PluralRules::getRuleFromResource(const Locale& locale, UPluralType type, UErrorC errCode = U_ILLEGAL_ARGUMENT_ERROR; return emptyStr; } - LocalUResourceBundlePointer locRes(ures_getByKey(rb.getAlias(), typeKey, NULL, &errCode)); + LocalUResourceBundlePointer locRes(ures_getByKey(rb.getAlias(), typeKey, nullptr, &errCode)); if(U_FAILURE(errCode)) { return emptyStr; } @@ -654,25 +695,25 @@ PluralRules::getRuleFromResource(const Locale& locale, UPluralType type, UErrorC const char *curLocaleName=locale.getName(); const UChar* s = ures_getStringByKey(locRes.getAlias(), curLocaleName, &resLen, &errCode); - if (s == NULL) { + if (s == nullptr) { // Check parent locales. UErrorCode status = U_ZERO_ERROR; char parentLocaleName[ULOC_FULLNAME_CAPACITY]; - const char *curLocaleName=locale.getName(); - uprv_strcpy(parentLocaleName, curLocaleName); + const char *curLocaleName2=locale.getName(); + uprv_strcpy(parentLocaleName, curLocaleName2); while (uloc_getParent(parentLocaleName, parentLocaleName, ULOC_FULLNAME_CAPACITY, &status) > 0) { resLen=0; s = ures_getStringByKey(locRes.getAlias(), parentLocaleName, &resLen, &status); - if (s != NULL) { + if (s != nullptr) { errCode = U_ZERO_ERROR; break; } status = U_ZERO_ERROR; } } - if (s==NULL) { + if (s==nullptr) { return emptyStr; } @@ -680,18 +721,18 @@ PluralRules::getRuleFromResource(const Locale& locale, UPluralType type, UErrorC u_UCharsToChars(s, setKey, resLen + 1); // printf("\n PluralRule: %s\n", setKey); - LocalUResourceBundlePointer ruleRes(ures_getByKey(rb.getAlias(), "rules", NULL, &errCode)); + LocalUResourceBundlePointer ruleRes(ures_getByKey(rb.getAlias(), "rules", nullptr, &errCode)); if(U_FAILURE(errCode)) { return emptyStr; } - LocalUResourceBundlePointer setRes(ures_getByKey(ruleRes.getAlias(), setKey, NULL, &errCode)); + LocalUResourceBundlePointer setRes(ures_getByKey(ruleRes.getAlias(), setKey, nullptr, &errCode)); if (U_FAILURE(errCode)) { return emptyStr; } int32_t numberKeys = ures_getSize(setRes.getAlias()); UnicodeString result; - const char *key=NULL; + const char *key=nullptr; for(int32_t i=0; i<numberKeys; ++i) { // Keys are zero, one, few, ... UnicodeString rules = ures_getNextUnicodeString(setRes.getAlias(), &key, &errCode); UnicodeString uKey(key, -1, US_INV); @@ -707,54 +748,46 @@ PluralRules::getRuleFromResource(const Locale& locale, UPluralType type, UErrorC UnicodeString PluralRules::getRules() const { UnicodeString rules; - if (mRules != NULL) { + if (mRules != nullptr) { mRules->dumpRules(rules); } return rules; } - -AndConstraint::AndConstraint() { - op = AndConstraint::NONE; - opNum=-1; - value = -1; - rangeList = NULL; - negated = FALSE; - integerOnly = FALSE; - digitsType = none; - next=NULL; -} - - AndConstraint::AndConstraint(const AndConstraint& other) { + this->fInternalStatus = other.fInternalStatus; + if (U_FAILURE(fInternalStatus)) { + return; // stop early if the object we are copying from is invalid. + } this->op = other.op; this->opNum=other.opNum; this->value=other.value; - this->rangeList=NULL; - if (other.rangeList != NULL) { - UErrorCode status = U_ZERO_ERROR; - this->rangeList = new UVector32(status); - this->rangeList->assign(*other.rangeList, status); + if (other.rangeList != nullptr) { + LocalPointer<UVector32> newRangeList(new UVector32(fInternalStatus), fInternalStatus); + if (U_FAILURE(fInternalStatus)) { + return; + } + this->rangeList = newRangeList.orphan(); + this->rangeList->assign(*other.rangeList, fInternalStatus); } this->integerOnly=other.integerOnly; this->negated=other.negated; this->digitsType = other.digitsType; - if (other.next==NULL) { - this->next=NULL; - } - else { + if (other.next != nullptr) { this->next = new AndConstraint(*other.next); + if (this->next == nullptr) { + fInternalStatus = U_MEMORY_ALLOCATION_ERROR; + } } } AndConstraint::~AndConstraint() { delete rangeList; - if (next!=NULL) { - delete next; - } + rangeList = nullptr; + delete next; + next = nullptr; } - UBool AndConstraint::isFulfilled(const IFixedDecimal &number) { UBool result = TRUE; @@ -776,7 +809,7 @@ AndConstraint::isFulfilled(const IFixedDecimal &number) { if (op == MOD) { n = fmod(n, opNum); } - if (rangeList == NULL) { + if (rangeList == nullptr) { result = value == -1 || // empty rule n == value; // 'is' rule break; @@ -796,53 +829,67 @@ AndConstraint::isFulfilled(const IFixedDecimal &number) { return result; } - AndConstraint* -AndConstraint::add() -{ +AndConstraint::add(UErrorCode& status) { + if (U_FAILURE(fInternalStatus)) { + status = fInternalStatus; + return nullptr; + } this->next = new AndConstraint(); + if (this->next == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + } return this->next; } -OrConstraint::OrConstraint() { - childNode=NULL; - next=NULL; -} OrConstraint::OrConstraint(const OrConstraint& other) { - if ( other.childNode == NULL ) { - this->childNode = NULL; + this->fInternalStatus = other.fInternalStatus; + if (U_FAILURE(fInternalStatus)) { + return; // stop early if the object we are copying from is invalid. } - else { + if ( other.childNode != nullptr ) { this->childNode = new AndConstraint(*(other.childNode)); + if (this->childNode == nullptr) { + fInternalStatus = U_MEMORY_ALLOCATION_ERROR; + return; + } } - if (other.next == NULL ) { - this->next = NULL; - } - else { + if (other.next != nullptr ) { this->next = new OrConstraint(*(other.next)); + if (this->next == nullptr) { + fInternalStatus = U_MEMORY_ALLOCATION_ERROR; + return; + } + if (U_FAILURE(this->next->fInternalStatus)) { + this->fInternalStatus = this->next->fInternalStatus; + } } } OrConstraint::~OrConstraint() { - if (childNode!=NULL) { - delete childNode; - } - if (next!=NULL) { - delete next; - } + delete childNode; + childNode = nullptr; + delete next; + next = nullptr; } AndConstraint* -OrConstraint::add() -{ +OrConstraint::add(UErrorCode& status) { + if (U_FAILURE(fInternalStatus)) { + status = fInternalStatus; + return nullptr; + } OrConstraint *curOrConstraint=this; { - while (curOrConstraint->next!=NULL) { + while (curOrConstraint->next!=nullptr) { curOrConstraint = curOrConstraint->next; } - U_ASSERT(curOrConstraint->childNode == NULL); + U_ASSERT(curOrConstraint->childNode == nullptr); curOrConstraint->childNode = new AndConstraint(); + if (curOrConstraint->childNode == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + } } return curOrConstraint->childNode; } @@ -852,10 +899,10 @@ OrConstraint::isFulfilled(const IFixedDecimal &number) { OrConstraint* orRule=this; UBool result=FALSE; - while (orRule!=NULL && !result) { + while (orRule!=nullptr && !result) { result=TRUE; AndConstraint* andRule = orRule->childNode; - while (andRule!=NULL && result) { + while (andRule!=nullptr && result) { result = andRule->isFulfilled(number); andRule=andRule->next; } @@ -866,19 +913,33 @@ OrConstraint::isFulfilled(const IFixedDecimal &number) { } -RuleChain::RuleChain(): fKeyword(), fNext(NULL), ruleHeader(NULL), fDecimalSamples(), fIntegerSamples(), - fDecimalSamplesUnbounded(FALSE), fIntegerSamplesUnbounded(FALSE) { -} - RuleChain::RuleChain(const RuleChain& other) : - fKeyword(other.fKeyword), fNext(NULL), ruleHeader(NULL), fDecimalSamples(other.fDecimalSamples), + fKeyword(other.fKeyword), fDecimalSamples(other.fDecimalSamples), fIntegerSamples(other.fIntegerSamples), fDecimalSamplesUnbounded(other.fDecimalSamplesUnbounded), - fIntegerSamplesUnbounded(other.fIntegerSamplesUnbounded) { - if (other.ruleHeader != NULL) { + fIntegerSamplesUnbounded(other.fIntegerSamplesUnbounded), fInternalStatus(other.fInternalStatus) { + if (U_FAILURE(this->fInternalStatus)) { + return; // stop early if the object we are copying from is invalid. + } + if (other.ruleHeader != nullptr) { this->ruleHeader = new OrConstraint(*(other.ruleHeader)); + if (this->ruleHeader == nullptr) { + this->fInternalStatus = U_MEMORY_ALLOCATION_ERROR; + } + else if (U_FAILURE(this->ruleHeader->fInternalStatus)) { + // If the OrConstraint wasn't fully copied, then set our status to failure as well. + this->fInternalStatus = this->ruleHeader->fInternalStatus; + return; // exit early. + } } - if (other.fNext != NULL ) { + if (other.fNext != nullptr ) { this->fNext = new RuleChain(*other.fNext); + if (this->fNext == nullptr) { + this->fInternalStatus = U_MEMORY_ALLOCATION_ERROR; + } + else if (U_FAILURE(this->fNext->fInternalStatus)) { + // If the RuleChain wasn't fully copied, then set our status to failure as well. + this->fInternalStatus = this->fNext->fInternalStatus; + } } } @@ -887,11 +948,10 @@ RuleChain::~RuleChain() { delete ruleHeader; } - UnicodeString RuleChain::select(const IFixedDecimal &number) const { if (!number.isNaN() && !number.isInfinite()) { - for (const RuleChain *rules = this; rules != NULL; rules = rules->fNext) { + for (const RuleChain *rules = this; rules != nullptr; rules = rules->fNext) { if (rules->ruleHeader->isFulfilled(number)) { return rules->fKeyword; } @@ -923,17 +983,17 @@ void RuleChain::dumpRules(UnicodeString& result) { UChar digitString[16]; - if ( ruleHeader != NULL ) { + if ( ruleHeader != nullptr ) { result += fKeyword; result += COLON; result += SPACE; OrConstraint* orRule=ruleHeader; - while ( orRule != NULL ) { + while ( orRule != nullptr ) { AndConstraint* andRule=orRule->childNode; - while ( andRule != NULL ) { - if ((andRule->op==AndConstraint::NONE) && (andRule->rangeList==NULL) && (andRule->value == -1)) { + while ( andRule != nullptr ) { + if ((andRule->op==AndConstraint::NONE) && (andRule->rangeList==nullptr) && (andRule->value == -1)) { // Empty Rules. - } else if ( (andRule->op==AndConstraint::NONE) && (andRule->rangeList==NULL) ) { + } else if ( (andRule->op==AndConstraint::NONE) && (andRule->rangeList==nullptr) ) { result += tokenString(andRule->digitsType); result += UNICODE_STRING_SIMPLE(" is "); if (andRule->negated) { @@ -950,7 +1010,7 @@ RuleChain::dumpRules(UnicodeString& result) { uprv_itou(digitString,16, andRule->opNum,10,0); result += UnicodeString(digitString); } - if (andRule->rangeList==NULL) { + if (andRule->rangeList==nullptr) { if (andRule->negated) { result += UNICODE_STRING_SIMPLE(" is not "); uprv_itou(digitString,16, andRule->value,10,0); @@ -993,16 +1053,16 @@ RuleChain::dumpRules(UnicodeString& result) { } } } - if ( (andRule=andRule->next) != NULL) { + if ( (andRule=andRule->next) != nullptr) { result += UNICODE_STRING_SIMPLE(" and "); } } - if ( (orRule = orRule->next) != NULL ) { + if ( (orRule = orRule->next) != nullptr ) { result += UNICODE_STRING_SIMPLE(" or "); } } } - if ( fNext != NULL ) { + if ( fNext != nullptr ) { result += UNICODE_STRING_SIMPLE("; "); fNext->dumpRules(result); } @@ -1011,6 +1071,9 @@ RuleChain::dumpRules(UnicodeString& result) { UErrorCode RuleChain::getKeywords(int32_t capacityOfKeywords, UnicodeString* keywords, int32_t& arraySize) const { + if (U_FAILURE(fInternalStatus)) { + return fInternalStatus; + } if ( arraySize < capacityOfKeywords-1 ) { keywords[arraySize++]=fKeyword; } @@ -1018,7 +1081,7 @@ RuleChain::getKeywords(int32_t capacityOfKeywords, UnicodeString* keywords, int3 return U_BUFFER_OVERFLOW_ERROR; } - if ( fNext != NULL ) { + if ( fNext != nullptr ) { return fNext->getKeywords(capacityOfKeywords, keywords, arraySize); } else { @@ -1032,7 +1095,7 @@ RuleChain::isKeyword(const UnicodeString& keywordParam) const { return TRUE; } - if ( fNext != NULL ) { + if ( fNext != nullptr ) { return fNext->isKeyword(keywordParam); } else { @@ -1043,7 +1106,7 @@ RuleChain::isKeyword(const UnicodeString& keywordParam) const { PluralRuleParser::PluralRuleParser() : ruleIndex(0), token(), type(none), prevType(none), - curAndConstraint(NULL), currentChain(NULL), rangeLowIdx(-1), rangeHiIdx(-1) + curAndConstraint(nullptr), currentChain(nullptr), rangeLowIdx(-1), rangeHiIdx(-1) { } @@ -1341,21 +1404,36 @@ PluralKeywordEnumeration::PluralKeywordEnumeration(RuleChain *header, UErrorCode return; } fKeywordNames.setDeleter(uprv_deleteUObject); - UBool addKeywordOther=TRUE; - RuleChain *node=header; - while(node!=NULL) { - fKeywordNames.addElement(new UnicodeString(node->fKeyword), status); + UBool addKeywordOther = TRUE; + RuleChain *node = header; + while (node != nullptr) { + auto newElem = new UnicodeString(node->fKeyword); + if (newElem == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + return; + } + fKeywordNames.addElement(newElem, status); if (U_FAILURE(status)) { + delete newElem; return; } if (0 == node->fKeyword.compare(PLURAL_KEYWORD_OTHER, 5)) { - addKeywordOther= FALSE; + addKeywordOther = FALSE; } - node=node->fNext; + node = node->fNext; } if (addKeywordOther) { - fKeywordNames.addElement(new UnicodeString(PLURAL_KEYWORD_OTHER), status); + auto newElem = new UnicodeString(PLURAL_KEYWORD_OTHER); + if (newElem == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + return; + } + fKeywordNames.addElement(newElem, status); + if (U_FAILURE(status)) { + delete newElem; + return; + } } } @@ -1364,7 +1442,7 @@ PluralKeywordEnumeration::snext(UErrorCode& status) { if (U_SUCCESS(status) && pos < fKeywordNames.size()) { return (const UnicodeString*)fKeywordNames.elementAt(pos++); } - return NULL; + return nullptr; } void @@ -1374,7 +1452,7 @@ PluralKeywordEnumeration::reset(UErrorCode& /*status*/) { int32_t PluralKeywordEnumeration::count(UErrorCode& /*status*/) const { - return fKeywordNames.size(); + return fKeywordNames.size(); } PluralKeywordEnumeration::~PluralKeywordEnumeration() { @@ -1634,41 +1712,39 @@ int32_t FixedDecimal::getVisibleFractionDigitCount() const { PluralAvailableLocalesEnumeration::PluralAvailableLocalesEnumeration(UErrorCode &status) { - fLocales = NULL; - fRes = NULL; fOpenStatus = status; if (U_FAILURE(status)) { return; } - fOpenStatus = U_ZERO_ERROR; - LocalUResourceBundlePointer rb(ures_openDirect(NULL, "plurals", &fOpenStatus)); - fLocales = ures_getByKey(rb.getAlias(), "locales", NULL, &fOpenStatus); + fOpenStatus = U_ZERO_ERROR; // clear any warnings. + LocalUResourceBundlePointer rb(ures_openDirect(nullptr, "plurals", &fOpenStatus)); + fLocales = ures_getByKey(rb.getAlias(), "locales", nullptr, &fOpenStatus); } PluralAvailableLocalesEnumeration::~PluralAvailableLocalesEnumeration() { ures_close(fLocales); ures_close(fRes); - fLocales = NULL; - fRes = NULL; + fLocales = nullptr; + fRes = nullptr; } const char *PluralAvailableLocalesEnumeration::next(int32_t *resultLength, UErrorCode &status) { if (U_FAILURE(status)) { - return NULL; + return nullptr; } if (U_FAILURE(fOpenStatus)) { status = fOpenStatus; - return NULL; + return nullptr; } fRes = ures_getNextResource(fLocales, fRes, &status); - if (fRes == NULL || U_FAILURE(status)) { + if (fRes == nullptr || U_FAILURE(status)) { if (status == U_INDEX_OUTOFBOUNDS_ERROR) { status = U_ZERO_ERROR; } - return NULL; + return nullptr; } const char *result = ures_getKey(fRes); - if (resultLength != NULL) { + if (resultLength != nullptr) { *resultLength = static_cast<int32_t>(uprv_strlen(result)); } return result; diff --git a/deps/icu-small/source/i18n/plurrule_impl.h b/deps/icu-small/source/i18n/plurrule_impl.h index 3ab445d584..f23ae16856 100644 --- a/deps/icu-small/source/i18n/plurrule_impl.h +++ b/deps/icu-small/source/i18n/plurrule_impl.h @@ -181,7 +181,6 @@ private: kRangeList, kSamples }; - }; enum PluralOperand { @@ -311,32 +310,36 @@ public: NONE, MOD } RuleOp; - RuleOp op; - int32_t opNum; // for mod expressions, the right operand of the mod. - int32_t value; // valid for 'is' rules only. - UVector32 *rangeList; // for 'in', 'within' rules. Null otherwise. - UBool negated; // TRUE for negated rules. - UBool integerOnly; // TRUE for 'within' rules. - tokenType digitsType; // n | i | v | f constraint. - AndConstraint *next; - - AndConstraint(); + RuleOp op = AndConstraint::NONE; + int32_t opNum = -1; // for mod expressions, the right operand of the mod. + int32_t value = -1; // valid for 'is' rules only. + UVector32 *rangeList = nullptr; // for 'in', 'within' rules. Null otherwise. + UBool negated = FALSE; // TRUE for negated rules. + UBool integerOnly = FALSE; // TRUE for 'within' rules. + tokenType digitsType = none; // n | i | v | f constraint. + AndConstraint *next = nullptr; + // Internal error status, used for errors that occur during the copy constructor. + UErrorCode fInternalStatus = U_ZERO_ERROR; + + AndConstraint() = default; AndConstraint(const AndConstraint& other); virtual ~AndConstraint(); - AndConstraint* add(); + AndConstraint* add(UErrorCode& status); // UBool isFulfilled(double number); UBool isFulfilled(const IFixedDecimal &number); }; class OrConstraint : public UMemory { public: - AndConstraint *childNode; - OrConstraint *next; - OrConstraint(); + AndConstraint *childNode = nullptr; + OrConstraint *next = nullptr; + // Internal error status, used for errors that occur during the copy constructor. + UErrorCode fInternalStatus = U_ZERO_ERROR; + OrConstraint() = default; OrConstraint(const OrConstraint& other); virtual ~OrConstraint(); - AndConstraint* add(); + AndConstraint* add(UErrorCode& status); // UBool isFulfilled(double number); UBool isFulfilled(const IFixedDecimal &number); }; @@ -344,15 +347,16 @@ public: class RuleChain : public UMemory { public: UnicodeString fKeyword; - RuleChain *fNext; - OrConstraint *ruleHeader; + RuleChain *fNext = nullptr; + OrConstraint *ruleHeader = nullptr; UnicodeString fDecimalSamples; // Samples strings from rule source UnicodeString fIntegerSamples; // without @decimal or @integer, otherwise unprocessed. - UBool fDecimalSamplesUnbounded; - UBool fIntegerSamplesUnbounded; - + UBool fDecimalSamplesUnbounded = FALSE; + UBool fIntegerSamplesUnbounded = FALSE; + // Internal error status, used for errors that occur during the copy constructor. + UErrorCode fInternalStatus = U_ZERO_ERROR; - RuleChain(); + RuleChain() = default; RuleChain(const RuleChain& other); virtual ~RuleChain(); @@ -386,8 +390,8 @@ class U_I18N_API PluralAvailableLocalesEnumeration: public StringEnumeration { virtual int32_t count(UErrorCode& status) const; private: UErrorCode fOpenStatus; - UResourceBundle *fLocales; - UResourceBundle *fRes; + UResourceBundle *fLocales = nullptr; + UResourceBundle *fRes = nullptr; }; U_NAMESPACE_END diff --git a/deps/icu-small/source/i18n/rbnf.cpp b/deps/icu-small/source/i18n/rbnf.cpp index ab9ad15e8c..74707ccd22 100644 --- a/deps/icu-small/source/i18n/rbnf.cpp +++ b/deps/icu-small/source/i18n/rbnf.cpp @@ -680,7 +680,7 @@ StringLocalizationInfo::getDisplayName(int32_t localeIndex, int32_t ruleIndex) c RuleBasedNumberFormat::RuleBasedNumberFormat(const UnicodeString& description, const UnicodeString& locs, const Locale& alocale, UParseError& perror, UErrorCode& status) - : ruleSets(NULL) + : fRuleSets(NULL) , ruleSetDescriptions(NULL) , numRuleSets(0) , defaultRuleSet(NULL) @@ -689,7 +689,7 @@ RuleBasedNumberFormat::RuleBasedNumberFormat(const UnicodeString& description, , decimalFormatSymbols(NULL) , defaultInfinityRule(NULL) , defaultNaNRule(NULL) - , roundingMode(DecimalFormat::ERoundingMode::kRoundUnnecessary) + , fRoundingMode(DecimalFormat::ERoundingMode::kRoundUnnecessary) , lenient(FALSE) , lenientParseRules(NULL) , localizations(NULL) @@ -705,7 +705,7 @@ RuleBasedNumberFormat::RuleBasedNumberFormat(const UnicodeString& description, RuleBasedNumberFormat::RuleBasedNumberFormat(const UnicodeString& description, const UnicodeString& locs, UParseError& perror, UErrorCode& status) - : ruleSets(NULL) + : fRuleSets(NULL) , ruleSetDescriptions(NULL) , numRuleSets(0) , defaultRuleSet(NULL) @@ -714,7 +714,7 @@ RuleBasedNumberFormat::RuleBasedNumberFormat(const UnicodeString& description, , decimalFormatSymbols(NULL) , defaultInfinityRule(NULL) , defaultNaNRule(NULL) - , roundingMode(DecimalFormat::ERoundingMode::kRoundUnnecessary) + , fRoundingMode(DecimalFormat::ERoundingMode::kRoundUnnecessary) , lenient(FALSE) , lenientParseRules(NULL) , localizations(NULL) @@ -730,7 +730,7 @@ RuleBasedNumberFormat::RuleBasedNumberFormat(const UnicodeString& description, RuleBasedNumberFormat::RuleBasedNumberFormat(const UnicodeString& description, LocalizationInfo* info, const Locale& alocale, UParseError& perror, UErrorCode& status) - : ruleSets(NULL) + : fRuleSets(NULL) , ruleSetDescriptions(NULL) , numRuleSets(0) , defaultRuleSet(NULL) @@ -739,7 +739,7 @@ RuleBasedNumberFormat::RuleBasedNumberFormat(const UnicodeString& description, , decimalFormatSymbols(NULL) , defaultInfinityRule(NULL) , defaultNaNRule(NULL) - , roundingMode(DecimalFormat::ERoundingMode::kRoundUnnecessary) + , fRoundingMode(DecimalFormat::ERoundingMode::kRoundUnnecessary) , lenient(FALSE) , lenientParseRules(NULL) , localizations(NULL) @@ -754,7 +754,7 @@ RuleBasedNumberFormat::RuleBasedNumberFormat(const UnicodeString& description, RuleBasedNumberFormat::RuleBasedNumberFormat(const UnicodeString& description, UParseError& perror, UErrorCode& status) - : ruleSets(NULL) + : fRuleSets(NULL) , ruleSetDescriptions(NULL) , numRuleSets(0) , defaultRuleSet(NULL) @@ -763,7 +763,7 @@ RuleBasedNumberFormat::RuleBasedNumberFormat(const UnicodeString& description, , decimalFormatSymbols(NULL) , defaultInfinityRule(NULL) , defaultNaNRule(NULL) - , roundingMode(DecimalFormat::ERoundingMode::kRoundUnnecessary) + , fRoundingMode(DecimalFormat::ERoundingMode::kRoundUnnecessary) , lenient(FALSE) , lenientParseRules(NULL) , localizations(NULL) @@ -779,7 +779,7 @@ RuleBasedNumberFormat::RuleBasedNumberFormat(const UnicodeString& description, const Locale& aLocale, UParseError& perror, UErrorCode& status) - : ruleSets(NULL) + : fRuleSets(NULL) , ruleSetDescriptions(NULL) , numRuleSets(0) , defaultRuleSet(NULL) @@ -788,7 +788,7 @@ RuleBasedNumberFormat::RuleBasedNumberFormat(const UnicodeString& description, , decimalFormatSymbols(NULL) , defaultInfinityRule(NULL) , defaultNaNRule(NULL) - , roundingMode(DecimalFormat::ERoundingMode::kRoundUnnecessary) + , fRoundingMode(DecimalFormat::ERoundingMode::kRoundUnnecessary) , lenient(FALSE) , lenientParseRules(NULL) , localizations(NULL) @@ -801,7 +801,7 @@ RuleBasedNumberFormat::RuleBasedNumberFormat(const UnicodeString& description, } RuleBasedNumberFormat::RuleBasedNumberFormat(URBNFRuleSetTag tag, const Locale& alocale, UErrorCode& status) - : ruleSets(NULL) + : fRuleSets(NULL) , ruleSetDescriptions(NULL) , numRuleSets(0) , defaultRuleSet(NULL) @@ -810,7 +810,7 @@ RuleBasedNumberFormat::RuleBasedNumberFormat(URBNFRuleSetTag tag, const Locale& , decimalFormatSymbols(NULL) , defaultInfinityRule(NULL) , defaultNaNRule(NULL) - , roundingMode(DecimalFormat::ERoundingMode::kRoundUnnecessary) + , fRoundingMode(DecimalFormat::ERoundingMode::kRoundUnnecessary) , lenient(FALSE) , lenientParseRules(NULL) , localizations(NULL) @@ -868,7 +868,7 @@ RuleBasedNumberFormat::RuleBasedNumberFormat(URBNFRuleSetTag tag, const Locale& RuleBasedNumberFormat::RuleBasedNumberFormat(const RuleBasedNumberFormat& rhs) : NumberFormat(rhs) - , ruleSets(NULL) + , fRuleSets(NULL) , ruleSetDescriptions(NULL) , numRuleSets(0) , defaultRuleSet(NULL) @@ -877,7 +877,7 @@ RuleBasedNumberFormat::RuleBasedNumberFormat(const RuleBasedNumberFormat& rhs) , decimalFormatSymbols(NULL) , defaultInfinityRule(NULL) , defaultNaNRule(NULL) - , roundingMode(DecimalFormat::ERoundingMode::kRoundUnnecessary) + , fRoundingMode(DecimalFormat::ERoundingMode::kRoundUnnecessary) , lenient(FALSE) , lenientParseRules(NULL) , localizations(NULL) @@ -950,8 +950,8 @@ RuleBasedNumberFormat::operator==(const Format& other) const ? FALSE : *localizations == rhs.localizations))) { - NFRuleSet** p = ruleSets; - NFRuleSet** q = rhs.ruleSets; + NFRuleSet** p = fRuleSets; + NFRuleSet** q = rhs.fRuleSets; if (p == NULL) { return q == NULL; } else if (q == NULL) { @@ -972,8 +972,8 @@ UnicodeString RuleBasedNumberFormat::getRules() const { UnicodeString result; - if (ruleSets != NULL) { - for (NFRuleSet** p = ruleSets; *p; ++p) { + if (fRuleSets != NULL) { + for (NFRuleSet** p = fRuleSets; *p; ++p) { (*p)->appendRules(result); } } @@ -987,9 +987,9 @@ RuleBasedNumberFormat::getRuleSetName(int32_t index) const UnicodeString string(TRUE, localizations->getRuleSetName(index), (int32_t)-1); return string; } - else if (ruleSets) { + else if (fRuleSets) { UnicodeString result; - for (NFRuleSet** p = ruleSets; *p; ++p) { + for (NFRuleSet** p = fRuleSets; *p; ++p) { NFRuleSet* rs = *p; if (rs->isPublic()) { if (--index == -1) { @@ -1010,8 +1010,8 @@ RuleBasedNumberFormat::getNumberOfRuleSetNames() const if (localizations) { result = localizations->getNumberOfRuleSets(); } - else if (ruleSets) { - for (NFRuleSet** p = ruleSets; *p; ++p) { + else if (fRuleSets) { + for (NFRuleSet** p = fRuleSets; *p; ++p) { if ((**p).isPublic()) { ++result; } @@ -1098,8 +1098,8 @@ RuleBasedNumberFormat::getRuleSetDisplayName(const UnicodeString& ruleSetName, c NFRuleSet* RuleBasedNumberFormat::findRuleSet(const UnicodeString& name, UErrorCode& status) const { - if (U_SUCCESS(status) && ruleSets) { - for (NFRuleSet** p = ruleSets; *p; ++p) { + if (U_SUCCESS(status) && fRuleSets) { + for (NFRuleSet** p = fRuleSets; *p; ++p) { NFRuleSet* rs = *p; if (rs->isNamed(name)) { return rs; @@ -1132,11 +1132,17 @@ RuleBasedNumberFormat::format(const DecimalQuantity &number, // The DecimalFormat will provide more accurate results. // TODO this section should probably be optimized. The DecimalFormat is shared in ICU4J. - NumberFormat *decimalFormat = NumberFormat::createInstance(locale, UNUM_DECIMAL, status); + LocalPointer<NumberFormat> decimalFormat(NumberFormat::createInstance(locale, UNUM_DECIMAL, status), status); + if (decimalFormat.isNull()) { + return appendTo; + } Formattable f; - f.adoptDecimalQuantity(new DecimalQuantity(number)); + LocalPointer<DecimalQuantity> decimalQuantity(new DecimalQuantity(number), status); + if (decimalQuantity.isNull()) { + return appendTo; + } + f.adoptDecimalQuantity(decimalQuantity.orphan()); // f now owns decimalQuantity. decimalFormat->format(f, appendTo, posIter, status); - delete decimalFormat; } } return appendTo; @@ -1165,11 +1171,17 @@ RuleBasedNumberFormat::format(const DecimalQuantity &number, // The DecimalFormat will provide more accurate results. // TODO this section should probably be optimized. The DecimalFormat is shared in ICU4J. - NumberFormat *decimalFormat = NumberFormat::createInstance(locale, UNUM_DECIMAL, status); + LocalPointer<NumberFormat> decimalFormat(NumberFormat::createInstance(locale, UNUM_DECIMAL, status), status); + if (decimalFormat.isNull()) { + return appendTo; + } Formattable f; - f.adoptDecimalQuantity(new DecimalQuantity(number)); + LocalPointer<DecimalQuantity> decimalQuantity(new DecimalQuantity(number), status); + if (decimalQuantity.isNull()) { + return appendTo; + } + f.adoptDecimalQuantity(decimalQuantity.orphan()); // f now owns decimalQuantity. decimalFormat->format(f, appendTo, pos, status); - delete decimalFormat; } } return appendTo; @@ -1312,11 +1324,19 @@ RuleBasedNumberFormat::format(int64_t number, NFRuleSet *ruleSet, UnicodeString& // TODO this section should probably be optimized. The DecimalFormat is shared in ICU4J. NumberFormat *decimalFormat = NumberFormat::createInstance(locale, UNUM_DECIMAL, status); + if (decimalFormat == nullptr) { + return toAppendTo; + } Formattable f; FieldPosition pos(FieldPosition::DONT_CARE); - DecimalQuantity *digitList = new DecimalQuantity(); - digitList->setToLong(number); - f.adoptDecimalQuantity(digitList); + DecimalQuantity *decimalQuantity = new DecimalQuantity(); + if (decimalQuantity == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + delete decimalFormat; + return toAppendTo; + } + decimalQuantity->setToLong(number); + f.adoptDecimalQuantity(decimalQuantity); // f now owns decimalQuantity. decimalFormat->format(f, toAppendTo, pos, status); delete decimalFormat; } @@ -1358,7 +1378,7 @@ RuleBasedNumberFormat::parse(const UnicodeString& text, Formattable& result, ParsePosition& parsePosition) const { - if (!ruleSets) { + if (!fRuleSets) { parsePosition.setErrorIndex(0); return; } @@ -1369,7 +1389,7 @@ RuleBasedNumberFormat::parse(const UnicodeString& text, ParsePosition high_pp(0); Formattable high_result; - for (NFRuleSet** p = ruleSets; *p; ++p) { + for (NFRuleSet** p = fRuleSets; *p; ++p) { NFRuleSet *rp = *p; if (rp->isPublic() && rp->isParseable()) { ParsePosition working_pp(0); @@ -1457,7 +1477,7 @@ void RuleBasedNumberFormat::initDefaultRuleSet() { defaultRuleSet = NULL; - if (!ruleSets) { + if (!fRuleSets) { return; } @@ -1465,7 +1485,7 @@ RuleBasedNumberFormat::initDefaultRuleSet() const UnicodeString ordinal(UNICODE_STRING_SIMPLE("%digits-ordinal")); const UnicodeString duration(UNICODE_STRING_SIMPLE("%duration")); - NFRuleSet**p = &ruleSets[0]; + NFRuleSet**p = &fRuleSets[0]; while (*p) { if ((*p)->isNamed(spellout) || (*p)->isNamed(ordinal) || (*p)->isNamed(duration)) { defaultRuleSet = *p; @@ -1477,7 +1497,7 @@ RuleBasedNumberFormat::initDefaultRuleSet() defaultRuleSet = *--p; if (!defaultRuleSet->isPublic()) { - while (p != ruleSets) { + while (p != fRuleSets) { if ((*--p)->isPublic()) { defaultRuleSet = *p; break; @@ -1547,7 +1567,7 @@ RuleBasedNumberFormat::init(const UnicodeString& rules, LocalizationInfo* locali // from the description lenientParseRules = new UnicodeString(); /* test for NULL */ - if (lenientParseRules == 0) { + if (lenientParseRules == nullptr) { status = U_MEMORY_ALLOCATION_ERROR; return; } @@ -1568,15 +1588,15 @@ RuleBasedNumberFormat::init(const UnicodeString& rules, LocalizationInfo* locali ++numRuleSets; // our rule list is an array of the appropriate size - ruleSets = (NFRuleSet **)uprv_malloc((numRuleSets + 1) * sizeof(NFRuleSet *)); + fRuleSets = (NFRuleSet **)uprv_malloc((numRuleSets + 1) * sizeof(NFRuleSet *)); /* test for NULL */ - if (ruleSets == 0) { + if (fRuleSets == 0) { status = U_MEMORY_ALLOCATION_ERROR; return; } for (int i = 0; i <= numRuleSets; ++i) { - ruleSets[i] = NULL; + fRuleSets[i] = NULL; } // divide up the descriptions into individual rule-set descriptions @@ -1592,7 +1612,7 @@ RuleBasedNumberFormat::init(const UnicodeString& rules, LocalizationInfo* locali } ruleSetDescriptions = new UnicodeString[numRuleSets]; - if (ruleSetDescriptions == 0) { + if (ruleSetDescriptions == nullptr) { status = U_MEMORY_ALLOCATION_ERROR; return; } @@ -1602,8 +1622,8 @@ RuleBasedNumberFormat::init(const UnicodeString& rules, LocalizationInfo* locali int32_t start = 0; for (int32_t p = description.indexOf(gSemiPercent, 2, 0); p != -1; p = description.indexOf(gSemiPercent, 2, start)) { ruleSetDescriptions[curRuleSet].setTo(description, start, p + 1 - start); - ruleSets[curRuleSet] = new NFRuleSet(this, ruleSetDescriptions, curRuleSet, status); - if (ruleSets[curRuleSet] == 0) { + fRuleSets[curRuleSet] = new NFRuleSet(this, ruleSetDescriptions, curRuleSet, status); + if (fRuleSets[curRuleSet] == nullptr) { status = U_MEMORY_ALLOCATION_ERROR; return; } @@ -1611,8 +1631,8 @@ RuleBasedNumberFormat::init(const UnicodeString& rules, LocalizationInfo* locali start = p + 1; } ruleSetDescriptions[curRuleSet].setTo(description, start, description.length() - start); - ruleSets[curRuleSet] = new NFRuleSet(this, ruleSetDescriptions, curRuleSet, status); - if (ruleSets[curRuleSet] == 0) { + fRuleSets[curRuleSet] = new NFRuleSet(this, ruleSetDescriptions, curRuleSet, status); + if (fRuleSets[curRuleSet] == nullptr) { status = U_MEMORY_ALLOCATION_ERROR; return; } @@ -1630,11 +1650,11 @@ RuleBasedNumberFormat::init(const UnicodeString& rules, LocalizationInfo* locali initDefaultRuleSet(); // finally, we can go back through the temporary descriptions - // list and finish seting up the substructure (and we throw + // list and finish setting up the substructure (and we throw // away the temporary descriptions as we go) { for (int i = 0; i < numRuleSets; i++) { - ruleSets[i]->parseRules(ruleSetDescriptions[i], status); + fRuleSets[i]->parseRules(ruleSetDescriptions[i], status); } } @@ -1680,7 +1700,7 @@ RuleBasedNumberFormat::setContext(UDisplayContext value, UErrorCode& status) if ( capitalizationBrkIter == NULL && (value==UDISPCTX_CAPITALIZATION_FOR_BEGINNING_OF_SENTENCE || (value==UDISPCTX_CAPITALIZATION_FOR_UI_LIST_OR_MENU && capitalizationForUIListMenu) || (value==UDISPCTX_CAPITALIZATION_FOR_STANDALONE && capitalizationForStandAlone)) ) { - UErrorCode status = U_ZERO_ERROR; + status = U_ZERO_ERROR; capitalizationBrkIter = BreakIterator::createSentenceInstance(locale, status); if (U_FAILURE(status)) { delete capitalizationBrkIter; @@ -1704,8 +1724,8 @@ RuleBasedNumberFormat::initCapitalizationContextInfo(const Locale& thelocale) int32_t len = 0; const int32_t * intVector = ures_getIntVector(rb, &len, &status); if (U_SUCCESS(status) && intVector != NULL && len >= 2) { - capitalizationForUIListMenu = intVector[0]; - capitalizationForStandAlone = intVector[1]; + capitalizationForUIListMenu = static_cast<UBool>(intVector[0]); + capitalizationForStandAlone = static_cast<UBool>(intVector[1]); } } ures_close(rb); @@ -1740,7 +1760,7 @@ RuleBasedNumberFormat::stripWhitespace(UnicodeString& description) start = p + 1; } - // when we get here, we've seeked off the end of the sring, and + // when we get here, we've seeked off the end of the string, and // we terminate the loop (we continue until *start* is -1 rather // than until *p* is -1, because otherwise we'd miss the last // rule in the description) @@ -1756,12 +1776,12 @@ RuleBasedNumberFormat::stripWhitespace(UnicodeString& description) void RuleBasedNumberFormat::dispose() { - if (ruleSets) { - for (NFRuleSet** p = ruleSets; *p; ++p) { + if (fRuleSets) { + for (NFRuleSet** p = fRuleSets; *p; ++p) { delete *p; } - uprv_free(ruleSets); - ruleSets = NULL; + uprv_free(fRuleSets); + fRuleSets = NULL; } if (ruleSetDescriptions) { @@ -1811,7 +1831,7 @@ const RuleBasedCollator* RuleBasedNumberFormat::getCollator() const { #if !UCONFIG_NO_COLLATION - if (!ruleSets) { + if (!fRuleSets) { return NULL; } @@ -1820,7 +1840,7 @@ RuleBasedNumberFormat::getCollator() const // create a default collator based on the formatter's locale, // then pull out that collator's rules, append any additional // rules specified in the description, and create a _new_ - // collator based on the combinaiton of those rules + // collator based on the combination of those rules UErrorCode status = U_ZERO_ERROR; @@ -1863,13 +1883,10 @@ RuleBasedNumberFormat::initializeDecimalFormatSymbols(UErrorCode &status) // lazy-evaluate the DecimalFormatSymbols object. This object // is shared by all DecimalFormat instances belonging to this // formatter - if (decimalFormatSymbols == NULL) { - DecimalFormatSymbols* temp = new DecimalFormatSymbols(locale, status); + if (decimalFormatSymbols == nullptr) { + LocalPointer<DecimalFormatSymbols> temp(new DecimalFormatSymbols(locale, status), status); if (U_SUCCESS(status)) { - decimalFormatSymbols = temp; - } - else { - delete temp; + decimalFormatSymbols = temp.orphan(); } } return decimalFormatSymbols; @@ -1889,17 +1906,14 @@ NFRule* RuleBasedNumberFormat::initializeDefaultInfinityRule(UErrorCode &status) { if (U_FAILURE(status)) { - return NULL; + return nullptr; } if (defaultInfinityRule == NULL) { UnicodeString rule(UNICODE_STRING_SIMPLE("Inf: ")); rule.append(getDecimalFormatSymbols()->getSymbol(DecimalFormatSymbols::kInfinitySymbol)); - NFRule* temp = new NFRule(this, rule, status); + LocalPointer<NFRule> temp(new NFRule(this, rule, status), status); if (U_SUCCESS(status)) { - defaultInfinityRule = temp; - } - else { - delete temp; + defaultInfinityRule = temp.orphan(); } } return defaultInfinityRule; @@ -1915,17 +1929,14 @@ NFRule* RuleBasedNumberFormat::initializeDefaultNaNRule(UErrorCode &status) { if (U_FAILURE(status)) { - return NULL; + return nullptr; } - if (defaultNaNRule == NULL) { + if (defaultNaNRule == nullptr) { UnicodeString rule(UNICODE_STRING_SIMPLE("NaN: ")); rule.append(getDecimalFormatSymbols()->getSymbol(DecimalFormatSymbols::kNaNSymbol)); - NFRule* temp = new NFRule(this, rule, status); + LocalPointer<NFRule> temp(new NFRule(this, rule, status), status); if (U_SUCCESS(status)) { - defaultNaNRule = temp; - } - else { - delete temp; + defaultNaNRule = temp.orphan(); } } return defaultNaNRule; @@ -1963,15 +1974,15 @@ RuleBasedNumberFormat::adoptDecimalFormatSymbols(DecimalFormatSymbols* symbolsTo defaultNaNRule = NULL; initializeDefaultNaNRule(status); // Reset with the new DecimalFormatSymbols - if (ruleSets) { + if (fRuleSets) { for (int32_t i = 0; i < numRuleSets; i++) { - ruleSets[i]->setDecimalFormatSymbols(*symbolsToAdopt, status); + fRuleSets[i]->setDecimalFormatSymbols(*symbolsToAdopt, status); } } } } -// Setting the symbols is equlivalent to adopting a newly created localized symbols. +// Setting the symbols is equivalent to adopting a newly created localized symbols. void RuleBasedNumberFormat::setDecimalFormatSymbols(const DecimalFormatSymbols& symbols) { @@ -1983,7 +1994,11 @@ RuleBasedNumberFormat::createPluralFormat(UPluralType pluralType, const UnicodeString &pattern, UErrorCode& status) const { - return new PluralFormat(locale, pluralType, pattern, status); + auto *pf = new PluralFormat(locale, pluralType, pattern, status); + if (pf == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + } + return pf; } /** @@ -1991,7 +2006,7 @@ RuleBasedNumberFormat::createPluralFormat(UPluralType pluralType, * @return A rounding mode */ DecimalFormat::ERoundingMode RuleBasedNumberFormat::getRoundingMode() const { - return roundingMode; + return fRoundingMode; } /** @@ -2000,7 +2015,7 @@ DecimalFormat::ERoundingMode RuleBasedNumberFormat::getRoundingMode() const { * @param roundingMode A rounding mode */ void RuleBasedNumberFormat::setRoundingMode(DecimalFormat::ERoundingMode roundingMode) { - this->roundingMode = roundingMode; + fRoundingMode = roundingMode; } U_NAMESPACE_END diff --git a/deps/icu-small/source/i18n/rbt.h b/deps/icu-small/source/i18n/rbt.h index 005fb85384..a18452ab2e 100644 --- a/deps/icu-small/source/i18n/rbt.h +++ b/deps/icu-small/source/i18n/rbt.h @@ -29,261 +29,9 @@ class TransliterationRuleData; /** * <code>RuleBasedTransliterator</code> is a transliterator - * that reads a set of rules in order to determine how to perform - * translations. Rule sets are stored in resource bundles indexed by - * name. Rules within a rule set are separated by semicolons (';'). - * To include a literal semicolon, prefix it with a backslash ('\'). - * Whitespace, as defined by <code>Character.isWhitespace()</code>, - * is ignored. If the first non-blank character on a line is '#', - * the entire line is ignored as a comment. </p> - * - * <p>Each set of rules consists of two groups, one forward, and one - * reverse. This is a convention that is not enforced; rules for one - * direction may be omitted, with the result that translations in - * that direction will not modify the source text. In addition, - * bidirectional forward-reverse rules may be specified for - * symmetrical transformations.</p> - * - * <p><b>Rule syntax</b> </p> - * - * <p>Rule statements take one of the following forms: </p> - * - * <dl> - * <dt><code>$alefmadda=\u0622;</code></dt> - * <dd><strong>Variable definition.</strong> The name on the - * left is assigned the text on the right. In this example, - * after this statement, instances of the left hand name, - * "<code>$alefmadda</code>", will be replaced by - * the Unicode character U+0622. Variable names must begin - * with a letter and consist only of letters, digits, and - * underscores. Case is significant. Duplicate names cause - * an exception to be thrown, that is, variables cannot be - * redefined. The right hand side may contain well-formed - * text of any length, including no text at all ("<code>$empty=;</code>"). - * The right hand side may contain embedded <code>UnicodeSet</code> - * patterns, for example, "<code>$softvowel=[eiyEIY]</code>".</dd> - * <dd> </dd> - * <dt><code>ai>$alefmadda;</code></dt> - * <dd><strong>Forward translation rule.</strong> This rule - * states that the string on the left will be changed to the - * string on the right when performing forward - * transliteration.</dd> - * <dt> </dt> - * <dt><code>ai<$alefmadda;</code></dt> - * <dd><strong>Reverse translation rule.</strong> This rule - * states that the string on the right will be changed to - * the string on the left when performing reverse - * transliteration.</dd> - * </dl> - * - * <dl> - * <dt><code>ai<>$alefmadda;</code></dt> - * <dd><strong>Bidirectional translation rule.</strong> This - * rule states that the string on the right will be changed - * to the string on the left when performing forward - * transliteration, and vice versa when performing reverse - * transliteration.</dd> - * </dl> - * - * <p>Translation rules consist of a <em>match pattern</em> and an <em>output - * string</em>. The match pattern consists of literal characters, - * optionally preceded by context, and optionally followed by - * context. Context characters, like literal pattern characters, - * must be matched in the text being transliterated. However, unlike - * literal pattern characters, they are not replaced by the output - * text. For example, the pattern "<code>abc{def}</code>" - * indicates the characters "<code>def</code>" must be - * preceded by "<code>abc</code>" for a successful match. - * If there is a successful match, "<code>def</code>" will - * be replaced, but not "<code>abc</code>". The final '<code>}</code>' - * is optional, so "<code>abc{def</code>" is equivalent to - * "<code>abc{def}</code>". Another example is "<code>{123}456</code>" - * (or "<code>123}456</code>") in which the literal - * pattern "<code>123</code>" must be followed by "<code>456</code>". - * </p> - * - * <p>The output string of a forward or reverse rule consists of - * characters to replace the literal pattern characters. If the - * output string contains the character '<code>|</code>', this is - * taken to indicate the location of the <em>cursor</em> after - * replacement. The cursor is the point in the text at which the - * next replacement, if any, will be applied. The cursor is usually - * placed within the replacement text; however, it can actually be - * placed into the precending or following context by using the - * special character '<code>@</code>'. Examples:</p> - * - * <blockquote> - * <p><code>a {foo} z > | @ bar; # foo -> bar, move cursor - * before a<br> - * {foo} xyz > bar @@|; # foo -> bar, cursor between - * y and z</code></p> - * </blockquote> - * - * <p><b>UnicodeSet</b></p> - * - * <p><code>UnicodeSet</code> patterns may appear anywhere that - * makes sense. They may appear in variable definitions. - * Contrariwise, <code>UnicodeSet</code> patterns may themselves - * contain variable references, such as "<code>$a=[a-z];$not_a=[^$a]</code>", - * or "<code>$range=a-z;$ll=[$range]</code>".</p> - * - * <p><code>UnicodeSet</code> patterns may also be embedded directly - * into rule strings. Thus, the following two rules are equivalent:</p> - * - * <blockquote> - * <p><code>$vowel=[aeiou]; $vowel>'*'; # One way to do this<br> - * [aeiou]>'*'; - * # - * Another way</code></p> - * </blockquote> - * - * <p>See {@link UnicodeSet} for more documentation and examples.</p> - * - * <p><b>Segments</b></p> - * - * <p>Segments of the input string can be matched and copied to the - * output string. This makes certain sets of rules simpler and more - * general, and makes reordering possible. For example:</p> - * - * <blockquote> - * <p><code>([a-z]) > $1 $1; - * # - * double lowercase letters<br> - * ([:Lu:]) ([:Ll:]) > $2 $1; # reverse order of Lu-Ll pairs</code></p> - * </blockquote> - * - * <p>The segment of the input string to be copied is delimited by - * "<code>(</code>" and "<code>)</code>". Up to - * nine segments may be defined. Segments may not overlap. In the - * output string, "<code>$1</code>" through "<code>$9</code>" - * represent the input string segments, in left-to-right order of - * definition.</p> - * - * <p><b>Anchors</b></p> - * - * <p>Patterns can be anchored to the beginning or the end of the text. This is done with the - * special characters '<code>^</code>' and '<code>$</code>'. For example:</p> - * - * <blockquote> - * <p><code>^ a > 'BEG_A'; # match 'a' at start of text<br> - * a > 'A'; # match other instances - * of 'a'<br> - * z $ > 'END_Z'; # match 'z' at end of text<br> - * z > 'Z'; # match other instances - * of 'z'</code></p> - * </blockquote> - * - * <p>It is also possible to match the beginning or the end of the text using a <code>UnicodeSet</code>. - * This is done by including a virtual anchor character '<code>$</code>' at the end of the - * set pattern. Although this is usually the match chafacter for the end anchor, the set will - * match either the beginning or the end of the text, depending on its placement. For - * example:</p> - * - * <blockquote> - * <p><code>$x = [a-z$]; # match 'a' through 'z' OR anchor<br> - * $x 1 > 2; # match '1' after a-z or at the start<br> - * 3 $x > 4; # match '3' before a-z or at the end</code></p> - * </blockquote> - * - * <p><b>Example</b> </p> - * - * <p>The following example rules illustrate many of the features of - * the rule language. </p> - * - * <table border="0" cellpadding="4"> - * <tr> - * <td valign="top">Rule 1.</td> - * <td valign="top" nowrap><code>abc{def}>x|y</code></td> - * </tr> - * <tr> - * <td valign="top">Rule 2.</td> - * <td valign="top" nowrap><code>xyz>r</code></td> - * </tr> - * <tr> - * <td valign="top">Rule 3.</td> - * <td valign="top" nowrap><code>yz>q</code></td> - * </tr> - * </table> - * - * <p>Applying these rules to the string "<code>adefabcdefz</code>" - * yields the following results: </p> - * - * <table border="0" cellpadding="4"> - * <tr> - * <td valign="top" nowrap><code>|adefabcdefz</code></td> - * <td valign="top">Initial state, no rules match. Advance - * cursor.</td> - * </tr> - * <tr> - * <td valign="top" nowrap><code>a|defabcdefz</code></td> - * <td valign="top">Still no match. Rule 1 does not match - * because the preceding context is not present.</td> - * </tr> - * <tr> - * <td valign="top" nowrap><code>ad|efabcdefz</code></td> - * <td valign="top">Still no match. Keep advancing until - * there is a match...</td> - * </tr> - * <tr> - * <td valign="top" nowrap><code>ade|fabcdefz</code></td> - * <td valign="top">...</td> - * </tr> - * <tr> - * <td valign="top" nowrap><code>adef|abcdefz</code></td> - * <td valign="top">...</td> - * </tr> - * <tr> - * <td valign="top" nowrap><code>adefa|bcdefz</code></td> - * <td valign="top">...</td> - * </tr> - * <tr> - * <td valign="top" nowrap><code>adefab|cdefz</code></td> - * <td valign="top">...</td> - * </tr> - * <tr> - * <td valign="top" nowrap><code>adefabc|defz</code></td> - * <td valign="top">Rule 1 matches; replace "<code>def</code>" - * with "<code>xy</code>" and back up the cursor - * to before the '<code>y</code>'.</td> - * </tr> - * <tr> - * <td valign="top" nowrap><code>adefabcx|yz</code></td> - * <td valign="top">Although "<code>xyz</code>" is - * present, rule 2 does not match because the cursor is - * before the '<code>y</code>', not before the '<code>x</code>'. - * Rule 3 does match. Replace "<code>yz</code>" - * with "<code>q</code>".</td> - * </tr> - * <tr> - * <td valign="top" nowrap><code>adefabcxq|</code></td> - * <td valign="top">The cursor is at the end; - * transliteration is complete.</td> - * </tr> - * </table> - * - * <p>The order of rules is significant. If multiple rules may match - * at some point, the first matching rule is applied. </p> - * - * <p>Forward and reverse rules may have an empty output string. - * Otherwise, an empty left or right hand side of any statement is a - * syntax error. </p> - * - * <p>Single quotes are used to quote any character other than a - * digit or letter. To specify a single quote itself, inside or - * outside of quotes, use two single quotes in a row. For example, - * the rule "<code>'>'>o''clock</code>" changes the - * string "<code>></code>" to the string "<code>o'clock</code>". - * </p> - * - * <p><b>Notes</b> </p> - * - * <p>While a RuleBasedTransliterator is being built, it checks that - * the rules are added in proper order. For example, if the rule - * "a>x" is followed by the rule "ab>y", - * then the second rule will throw an exception. The reason is that - * the second rule can never be triggered, since the first rule - * always matches anything it matches. In other words, the first - * rule <em>masks</em> the second rule. </p> + * built from a set of rules as defined for + * Transliterator::createFromRules(). + * See the C++ class Transliterator documentation for the rule syntax. * * @author Alan Liu * @internal Use transliterator factory methods instead since this class will be removed in that release. diff --git a/deps/icu-small/source/i18n/rbt_pars.cpp b/deps/icu-small/source/i18n/rbt_pars.cpp index 8e49a8473a..e07cc8b63a 100644 --- a/deps/icu-small/source/i18n/rbt_pars.cpp +++ b/deps/icu-small/source/i18n/rbt_pars.cpp @@ -194,9 +194,9 @@ const UnicodeFunctor* ParseData::lookupMatcher(UChar32 ch) const { const UnicodeFunctor* set = NULL; int32_t i = ch - data->variablesBase; if (i >= 0 && i < variablesVector->size()) { - int32_t i = ch - data->variablesBase; - set = (i < variablesVector->size()) ? - (UnicodeFunctor*) variablesVector->elementAt(i) : 0; + int32_t j = ch - data->variablesBase; + set = (j < variablesVector->size()) ? + (UnicodeFunctor*) variablesVector->elementAt(j) : 0; } return set; } @@ -1108,8 +1108,8 @@ void TransliteratorParser::parseRules(const UnicodeString& rule, } data->variableNames.removeAll(); - int32_t pos = UHASH_FIRST; - const UHashElement* he = variableNames.nextElement(pos); + int32_t p = UHASH_FIRST; + const UHashElement* he = variableNames.nextElement(p); while (he != NULL) { UnicodeString* tempus = (UnicodeString*)(((UnicodeString*)(he->value.pointer))->clone()); if (tempus == NULL) { @@ -1118,7 +1118,7 @@ void TransliteratorParser::parseRules(const UnicodeString& rule, } data->variableNames.put(*((UnicodeString*)(he->key.pointer)), tempus, status); - he = variableNames.nextElement(pos); + he = variableNames.nextElement(p); } } variablesVector.removeAllElements(); // keeps them from getting deleted when we succeed diff --git a/deps/icu-small/source/i18n/regexcmp.cpp b/deps/icu-small/source/i18n/regexcmp.cpp index 410ff9513b..0c5fca6f67 100644 --- a/deps/icu-small/source/i18n/regexcmp.cpp +++ b/deps/icu-small/source/i18n/regexcmp.cpp @@ -28,6 +28,7 @@ #include "patternprops.h" #include "putilimp.h" #include "cmemory.h" +#include "cstr.h" #include "cstring.h" #include "uvectr32.h" #include "uvectr64.h" @@ -3892,7 +3893,7 @@ void RegexCompile::stripNOPs() { // //------------------------------------------------------------------------------ void RegexCompile::error(UErrorCode e) { - if (U_SUCCESS(*fStatus)) { + if (U_SUCCESS(*fStatus) || e == U_MEMORY_ALLOCATION_ERROR) { *fStatus = e; // Hmm. fParseErr (UParseError) line & offset fields are int32_t in public // API (see common/unicode/parseerr.h), while fLineNum and fCharNum are @@ -4370,209 +4371,209 @@ static inline void addIdentifierIgnorable(UnicodeSet *set, UErrorCode& ec) { // Includes trying the Java "properties" that aren't supported as // normal ICU UnicodeSet properties // -static const UChar posSetPrefix[] = {0x5b, 0x5c, 0x70, 0x7b, 0}; // "[\p{" -static const UChar negSetPrefix[] = {0x5b, 0x5c, 0x50, 0x7b, 0}; // "[\P{" UnicodeSet *RegexCompile::createSetForProperty(const UnicodeString &propName, UBool negated) { - UnicodeString setExpr; - UnicodeSet *set; - uint32_t usetFlags = 0; if (U_FAILURE(*fStatus)) { - return NULL; + return nullptr; } + LocalPointer<UnicodeSet> set; + UErrorCode status = U_ZERO_ERROR; - // - // First try the property as we received it - // - if (negated) { - setExpr.append(negSetPrefix, -1); - } else { - setExpr.append(posSetPrefix, -1); - } - setExpr.append(propName); - setExpr.append(chRBrace); - setExpr.append(chRBracket); - if (fModeFlags & UREGEX_CASE_INSENSITIVE) { - usetFlags |= USET_CASE_INSENSITIVE; - } - set = new UnicodeSet(setExpr, usetFlags, NULL, *fStatus); - if (U_SUCCESS(*fStatus)) { - return set; - } - delete set; - set = NULL; - - // - // The property as it was didn't work. - - // Do [:word:]. It is not recognized as a property by UnicodeSet. "word" not standard POSIX - // or standard Java, but many other regular expression packages do recognize it. - - if (propName.caseCompare(UNICODE_STRING_SIMPLE("word"), 0) == 0) { - *fStatus = U_ZERO_ERROR; - set = new UnicodeSet(*(fRXPat->fStaticSets[URX_ISWORD_SET])); - if (set == NULL) { - *fStatus = U_MEMORY_ALLOCATION_ERROR; - return set; + do { // non-loop, exists to allow breaks from the block. + // + // First try the property as we received it + // + UnicodeString setExpr; + uint32_t usetFlags = 0; + setExpr.append(u"[\\p{", -1); + setExpr.append(propName); + setExpr.append(u"}]", -1); + if (fModeFlags & UREGEX_CASE_INSENSITIVE) { + usetFlags |= USET_CASE_INSENSITIVE; } - if (negated) { - set->complement(); + set.adoptInsteadAndCheckErrorCode(new UnicodeSet(setExpr, usetFlags, NULL, status), status); + if (U_SUCCESS(status) || status == U_MEMORY_ALLOCATION_ERROR) { + break; } - return set; - } + // + // The incoming property wasn't directly recognized by ICU. - // Do Java fixes - - // InGreek -> InGreek or Coptic, that being the official Unicode name for that block. - // InCombiningMarksforSymbols -> InCombiningDiacriticalMarksforSymbols. - // - // Note on Spaces: either "InCombiningMarksForSymbols" or "InCombining Marks for Symbols" - // is accepted by Java. The property part of the name is compared - // case-insenstively. The spaces must be exactly as shown, either - // all there, or all omitted, with exactly one at each position - // if they are present. From checking against JDK 1.6 - // - // This code should be removed when ICU properties support the Java compatibility names - // (ICU 4.0?) - // - UnicodeString mPropName = propName; - if (mPropName.caseCompare(UNICODE_STRING_SIMPLE("InGreek"), 0) == 0) { - mPropName = UNICODE_STRING_SIMPLE("InGreek and Coptic"); - } - if (mPropName.caseCompare(UNICODE_STRING_SIMPLE("InCombining Marks for Symbols"), 0) == 0 || - mPropName.caseCompare(UNICODE_STRING_SIMPLE("InCombiningMarksforSymbols"), 0) == 0) { - mPropName = UNICODE_STRING_SIMPLE("InCombining Diacritical Marks for Symbols"); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("all")) == 0) { - mPropName = UNICODE_STRING_SIMPLE("javaValidCodePoint"); - } + // Check [:word:] and [:all:]. These are not recognized as a properties by ICU UnicodeSet. + // Java accepts 'word' with mixed case. + // Java accepts 'all' only in all lower case. - // See if the property looks like a Java "InBlockName", which - // we will recast as "Block=BlockName" - // - if (mPropName.startsWith(u"In", 2) && propName.length()>=3) { - setExpr.truncate(4); // Leaves "[\p{", or "[\P{" - setExpr.append(u"Block=", -1); - setExpr.append(UnicodeString(mPropName, 2)); // Property with the leading "In" removed. - setExpr.append(chRBrace); - setExpr.append(chRBracket); - *fStatus = U_ZERO_ERROR; - set = new UnicodeSet(setExpr, usetFlags, NULL, *fStatus); - if (U_SUCCESS(*fStatus)) { - return set; + status = U_ZERO_ERROR; + if (propName.caseCompare(u"word", -1, 0) == 0) { + set.adoptInsteadAndCheckErrorCode(new UnicodeSet(*(fRXPat->fStaticSets[URX_ISWORD_SET])), status); + break; + } + if (propName.compare(u"all", -1) == 0) { + set.adoptInsteadAndCheckErrorCode(new UnicodeSet(0, 0x10ffff), status); + break; } - delete set; - set = NULL; - } - if (propName.startsWith(UNICODE_STRING_SIMPLE("java")) || - propName.compare(UNICODE_STRING_SIMPLE("all")) == 0) - { - UErrorCode localStatus = U_ZERO_ERROR; - //setExpr.remove(); - set = new UnicodeSet(); - // - // Try the various Java specific properties. - // These all begin with "java" + + // Do Java InBlock expressions // - if (mPropName.compare(UNICODE_STRING_SIMPLE("javaDefined")) == 0) { - addCategory(set, U_GC_CN_MASK, localStatus); - set->complement(); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaDigit")) == 0) { - addCategory(set, U_GC_ND_MASK, localStatus); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaIdentifierIgnorable")) == 0) { - addIdentifierIgnorable(set, localStatus); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaISOControl")) == 0) { - set->add(0, 0x1F).add(0x7F, 0x9F); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaJavaIdentifierPart")) == 0) { - addCategory(set, U_GC_L_MASK, localStatus); - addCategory(set, U_GC_SC_MASK, localStatus); - addCategory(set, U_GC_PC_MASK, localStatus); - addCategory(set, U_GC_ND_MASK, localStatus); - addCategory(set, U_GC_NL_MASK, localStatus); - addCategory(set, U_GC_MC_MASK, localStatus); - addCategory(set, U_GC_MN_MASK, localStatus); - addIdentifierIgnorable(set, localStatus); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaJavaIdentifierStart")) == 0) { - addCategory(set, U_GC_L_MASK, localStatus); - addCategory(set, U_GC_NL_MASK, localStatus); - addCategory(set, U_GC_SC_MASK, localStatus); - addCategory(set, U_GC_PC_MASK, localStatus); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaLetter")) == 0) { - addCategory(set, U_GC_L_MASK, localStatus); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaLetterOrDigit")) == 0) { - addCategory(set, U_GC_L_MASK, localStatus); - addCategory(set, U_GC_ND_MASK, localStatus); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaLowerCase")) == 0) { - addCategory(set, U_GC_LL_MASK, localStatus); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaMirrored")) == 0) { - set->applyIntPropertyValue(UCHAR_BIDI_MIRRORED, 1, localStatus); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaSpaceChar")) == 0) { - addCategory(set, U_GC_Z_MASK, localStatus); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaSupplementaryCodePoint")) == 0) { - set->add(0x10000, UnicodeSet::MAX_VALUE); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaTitleCase")) == 0) { - addCategory(set, U_GC_LT_MASK, localStatus); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaUnicodeIdentifierStart")) == 0) { - addCategory(set, U_GC_L_MASK, localStatus); - addCategory(set, U_GC_NL_MASK, localStatus); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaUnicodeIdentifierPart")) == 0) { - addCategory(set, U_GC_L_MASK, localStatus); - addCategory(set, U_GC_PC_MASK, localStatus); - addCategory(set, U_GC_ND_MASK, localStatus); - addCategory(set, U_GC_NL_MASK, localStatus); - addCategory(set, U_GC_MC_MASK, localStatus); - addCategory(set, U_GC_MN_MASK, localStatus); - addIdentifierIgnorable(set, localStatus); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaUpperCase")) == 0) { - addCategory(set, U_GC_LU_MASK, localStatus); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaValidCodePoint")) == 0) { - set->add(0, UnicodeSet::MAX_VALUE); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("javaWhitespace")) == 0) { - addCategory(set, U_GC_Z_MASK, localStatus); - set->removeAll(UnicodeSet().add(0xa0).add(0x2007).add(0x202f)); - set->add(9, 0x0d).add(0x1c, 0x1f); - } - else if (mPropName.compare(UNICODE_STRING_SIMPLE("all")) == 0) { - set->add(0, UnicodeSet::MAX_VALUE); + UnicodeString mPropName = propName; + if (mPropName.startsWith(u"In", 2) && mPropName.length() >= 3) { + status = U_ZERO_ERROR; + set.adoptInsteadAndCheckErrorCode(new UnicodeSet(), status); + if (U_FAILURE(status)) { + break; + } + UnicodeString blockName(mPropName, 2); // Property with the leading "In" removed. + set->applyPropertyAlias(UnicodeString(u"Block"), blockName, status); + break; } - if (U_SUCCESS(localStatus) && !set->isEmpty()) { - *fStatus = U_ZERO_ERROR; - if (usetFlags & USET_CASE_INSENSITIVE) { + // Check for the Java form "IsBooleanPropertyValue", which we will recast + // as "BooleanPropertyValue". The property value can be either a + // a General Category or a Script Name. + + if (propName.startsWith(u"Is", 2) && propName.length()>=3) { + mPropName.remove(0, 2); // Strip the "Is" + if (mPropName.indexOf(u'=') >= 0) { + // Reject any "Is..." property expression containing an '=', that is, + // any non-binary property expression. + status = U_REGEX_PROPERTY_SYNTAX; + break; + } + + if (mPropName.caseCompare(u"assigned", -1, 0) == 0) { + mPropName.setTo(u"unassigned", -1); + negated = !negated; + } else if (mPropName.caseCompare(u"TitleCase", -1, 0) == 0) { + mPropName.setTo(u"Titlecase_Letter", -1); + } + + mPropName.insert(0, u"[\\p{", -1); + mPropName.append(u"}]", -1); + set.adoptInsteadAndCheckErrorCode(new UnicodeSet(mPropName, *fStatus), status); + + if (U_SUCCESS(status) && !set->isEmpty() && (usetFlags & USET_CASE_INSENSITIVE)) { set->closeOver(USET_CASE_INSENSITIVE); } - if (negated) { + break; + + } + + if (propName.startsWith(u"java", -1)) { + status = U_ZERO_ERROR; + set.adoptInsteadAndCheckErrorCode(new UnicodeSet(), status); + if (U_FAILURE(status)) { + break; + } + // + // Try the various Java specific properties. + // These all begin with "java" + // + if (propName.compare(u"javaDefined", -1) == 0) { + addCategory(set.getAlias(), U_GC_CN_MASK, status); set->complement(); } - return set; + else if (propName.compare(u"javaDigit", -1) == 0) { + addCategory(set.getAlias(), U_GC_ND_MASK, status); + } + else if (propName.compare(u"javaIdentifierIgnorable", -1) == 0) { + addIdentifierIgnorable(set.getAlias(), status); + } + else if (propName.compare(u"javaISOControl", -1) == 0) { + set->add(0, 0x1F).add(0x7F, 0x9F); + } + else if (propName.compare(u"javaJavaIdentifierPart", -1) == 0) { + addCategory(set.getAlias(), U_GC_L_MASK, status); + addCategory(set.getAlias(), U_GC_SC_MASK, status); + addCategory(set.getAlias(), U_GC_PC_MASK, status); + addCategory(set.getAlias(), U_GC_ND_MASK, status); + addCategory(set.getAlias(), U_GC_NL_MASK, status); + addCategory(set.getAlias(), U_GC_MC_MASK, status); + addCategory(set.getAlias(), U_GC_MN_MASK, status); + addIdentifierIgnorable(set.getAlias(), status); + } + else if (propName.compare(u"javaJavaIdentifierStart", -1) == 0) { + addCategory(set.getAlias(), U_GC_L_MASK, status); + addCategory(set.getAlias(), U_GC_NL_MASK, status); + addCategory(set.getAlias(), U_GC_SC_MASK, status); + addCategory(set.getAlias(), U_GC_PC_MASK, status); + } + else if (propName.compare(u"javaLetter", -1) == 0) { + addCategory(set.getAlias(), U_GC_L_MASK, status); + } + else if (propName.compare(u"javaLetterOrDigit", -1) == 0) { + addCategory(set.getAlias(), U_GC_L_MASK, status); + addCategory(set.getAlias(), U_GC_ND_MASK, status); + } + else if (propName.compare(u"javaLowerCase", -1) == 0) { + addCategory(set.getAlias(), U_GC_LL_MASK, status); + } + else if (propName.compare(u"javaMirrored", -1) == 0) { + set->applyIntPropertyValue(UCHAR_BIDI_MIRRORED, 1, status); + } + else if (propName.compare(u"javaSpaceChar", -1) == 0) { + addCategory(set.getAlias(), U_GC_Z_MASK, status); + } + else if (propName.compare(u"javaSupplementaryCodePoint", -1) == 0) { + set->add(0x10000, UnicodeSet::MAX_VALUE); + } + else if (propName.compare(u"javaTitleCase", -1) == 0) { + addCategory(set.getAlias(), U_GC_LT_MASK, status); + } + else if (propName.compare(u"javaUnicodeIdentifierStart", -1) == 0) { + addCategory(set.getAlias(), U_GC_L_MASK, status); + addCategory(set.getAlias(), U_GC_NL_MASK, status); + } + else if (propName.compare(u"javaUnicodeIdentifierPart", -1) == 0) { + addCategory(set.getAlias(), U_GC_L_MASK, status); + addCategory(set.getAlias(), U_GC_PC_MASK, status); + addCategory(set.getAlias(), U_GC_ND_MASK, status); + addCategory(set.getAlias(), U_GC_NL_MASK, status); + addCategory(set.getAlias(), U_GC_MC_MASK, status); + addCategory(set.getAlias(), U_GC_MN_MASK, status); + addIdentifierIgnorable(set.getAlias(), status); + } + else if (propName.compare(u"javaUpperCase", -1) == 0) { + addCategory(set.getAlias(), U_GC_LU_MASK, status); + } + else if (propName.compare(u"javaValidCodePoint", -1) == 0) { + set->add(0, UnicodeSet::MAX_VALUE); + } + else if (propName.compare(u"javaWhitespace", -1) == 0) { + addCategory(set.getAlias(), U_GC_Z_MASK, status); + set->removeAll(UnicodeSet().add(0xa0).add(0x2007).add(0x202f)); + set->add(9, 0x0d).add(0x1c, 0x1f); + } else { + status = U_REGEX_PROPERTY_SYNTAX; + } + + if (U_SUCCESS(status) && !set->isEmpty() && (usetFlags & USET_CASE_INSENSITIVE)) { + set->closeOver(USET_CASE_INSENSITIVE); + } + break; + } + + // Unrecognized property. ICU didn't like it as it was, and none of the Java compatibility + // extensions matched it. + status = U_REGEX_PROPERTY_SYNTAX; + } while (false); // End of do loop block. Code above breaks out of the block on success or hard failure. + + if (U_SUCCESS(status)) { + U_ASSERT(set.isValid()); + if (negated) { + set->complement(); } - delete set; - set = NULL; + return set.orphan(); + } else { + if (status == U_ILLEGAL_ARGUMENT_ERROR) { + status = U_REGEX_PROPERTY_SYNTAX; + } + error(status); + return nullptr; } - error(*fStatus); - return NULL; } - // // SetEval Part of the evaluation of [set expressions]. // Perform any pending (stacked) operations with precedence diff --git a/deps/icu-small/source/i18n/region.cpp b/deps/icu-small/source/i18n/region.cpp index 66f9ef35de..f182f61486 100644 --- a/deps/icu-small/source/i18n/region.cpp +++ b/deps/icu-small/source/i18n/region.cpp @@ -168,10 +168,18 @@ void U_CALLCONV Region::loadRegionData(UErrorCode &status) { continents->addElement(continentName,status); } + UResourceBundle *groupingBundle = nullptr; while ( ures_hasNext(groupingContainment.getAlias()) ) { - UnicodeString *groupingName = new UnicodeString(ures_getNextUnicodeString(groupingContainment.getAlias(),NULL,&status)); - groupings->addElement(groupingName,status); + groupingBundle = ures_getNextResource(groupingContainment.getAlias(), groupingBundle, &status); + if (U_FAILURE(status)) { + break; + } + UnicodeString *groupingName = new UnicodeString(ures_getKey(groupingBundle), -1, US_INV); + if (groupingName) { + groupings->addElement(groupingName,status); + } } + ures_close(groupingBundle); for ( int32_t i = 0 ; i < allRegions->size() ; i++ ) { LocalPointer<Region> r(new Region(), status); @@ -182,7 +190,7 @@ void U_CALLCONV Region::loadRegionData(UErrorCode &status) { r->idStr = *regionName; r->idStr.extract(0,r->idStr.length(),r->id,sizeof(r->id),US_INV); - r->type = URGN_TERRITORY; // Only temporary - figure out the real type later once the aliases are known. + r->fType = URGN_TERRITORY; // Only temporary - figure out the real type later once the aliases are known. Formattable result; UErrorCode ps = U_ZERO_ERROR; @@ -190,7 +198,7 @@ void U_CALLCONV Region::loadRegionData(UErrorCode &status) { if ( U_SUCCESS(ps) ) { r->code = result.getLong(); // Convert string to number uhash_iput(newNumericCodeMap.getAlias(),r->code,(void *)(r.getAlias()),&status); - r->type = URGN_SUBCONTINENT; + r->fType = URGN_SUBCONTINENT; } else { r->code = -1; } @@ -231,9 +239,9 @@ void U_CALLCONV Region::loadRegionData(UErrorCode &status) { } else { aliasFromRegion->code = -1; } - aliasFromRegion->type = URGN_DEPRECATED; + aliasFromRegion->fType = URGN_DEPRECATED; } else { - aliasFromRegion->type = URGN_DEPRECATED; + aliasFromRegion->fType = URGN_DEPRECATED; } { @@ -290,26 +298,26 @@ void U_CALLCONV Region::loadRegionData(UErrorCode &status) { UnicodeString WORLD_ID_STRING(WORLD_ID); r = (Region *) uhash_get(newRegionIDMap.getAlias(),(void *)&WORLD_ID_STRING); if ( r ) { - r->type = URGN_WORLD; + r->fType = URGN_WORLD; } UnicodeString UNKNOWN_REGION_ID_STRING(UNKNOWN_REGION_ID); r = (Region *) uhash_get(newRegionIDMap.getAlias(),(void *)&UNKNOWN_REGION_ID_STRING); if ( r ) { - r->type = URGN_UNKNOWN; + r->fType = URGN_UNKNOWN; } for ( int32_t i = 0 ; i < continents->size() ; i++ ) { r = (Region *) uhash_get(newRegionIDMap.getAlias(),(void *)continents->elementAt(i)); if ( r ) { - r->type = URGN_CONTINENT; + r->fType = URGN_CONTINENT; } } for ( int32_t i = 0 ; i < groupings->size() ; i++ ) { r = (Region *) uhash_get(newRegionIDMap.getAlias(),(void *)groupings->elementAt(i)); if ( r ) { - r->type = URGN_GROUPING; + r->fType = URGN_GROUPING; } } @@ -319,7 +327,7 @@ void U_CALLCONV Region::loadRegionData(UErrorCode &status) { UnicodeString OUTLYING_OCEANIA_REGION_ID_STRING(OUTLYING_OCEANIA_REGION_ID); r = (Region *) uhash_get(newRegionIDMap.getAlias(),(void *)&OUTLYING_OCEANIA_REGION_ID_STRING); if ( r ) { - r->type = URGN_SUBCONTINENT; + r->fType = URGN_SUBCONTINENT; } // Load territory containment info from the supplemental data. @@ -356,7 +364,7 @@ void U_CALLCONV Region::loadRegionData(UErrorCode &status) { // Set the parent region to be the containing region of the child. // Regions of type GROUPING can't be set as the parent, since another region // such as a SUBCONTINENT, CONTINENT, or WORLD must always be the parent. - if ( parentRegion->type != URGN_GROUPING) { + if ( parentRegion->fType != URGN_GROUPING) { childRegion->containingRegion = parentRegion; } } @@ -367,15 +375,15 @@ void U_CALLCONV Region::loadRegionData(UErrorCode &status) { int32_t pos = UHASH_FIRST; while ( const UHashElement* element = uhash_nextElement(newRegionIDMap.getAlias(),&pos)) { Region *ar = (Region *)element->value.pointer; - if ( availableRegions[ar->type] == NULL ) { + if ( availableRegions[ar->fType] == NULL ) { LocalPointer<UVector> newAr(new UVector(uprv_deleteUObject, uhash_compareUnicodeString, status), status); - availableRegions[ar->type] = newAr.orphan(); + availableRegions[ar->fType] = newAr.orphan(); } LocalPointer<UnicodeString> arString(new UnicodeString(ar->idStr), status); if( U_FAILURE(status) ) { return; // error out } - availableRegions[ar->type]->addElement((void *)arString.orphan(),status); + availableRegions[ar->fType]->addElement((void *)arString.orphan(),status); } ucln_i18n_registerCleanup(UCLN_I18N_REGION, region_cleanup); @@ -416,7 +424,7 @@ void Region::cleanupRegionData() { Region::Region () : code(-1), - type(URGN_UNKNOWN), + fType(URGN_UNKNOWN), containingRegion(NULL), containedRegions(NULL), preferredValues(NULL) { @@ -481,7 +489,7 @@ Region::getInstance(const char *region_code, UErrorCode &status) { return NULL; } - if ( r->type == URGN_DEPRECATED && r->preferredValues->size() == 1) { + if ( r->fType == URGN_DEPRECATED && r->preferredValues->size() == 1) { StringEnumeration *pv = r->getPreferredValues(status); pv->reset(status); const UnicodeString *ustr = pv->snext(status); @@ -529,7 +537,7 @@ Region::getInstance (int32_t code, UErrorCode &status) { return NULL; } - if ( r->type == URGN_DEPRECATED && r->preferredValues->size() == 1) { + if ( r->fType == URGN_DEPRECATED && r->preferredValues->size() == 1) { StringEnumeration *pv = r->getPreferredValues(status); pv->reset(status); const UnicodeString *ustr = pv->snext(status); @@ -580,7 +588,7 @@ Region::getContainingRegion(URegionType type) const { return NULL; } - return ( containingRegion->type == type )? containingRegion: containingRegion->getContainingRegion(type); + return ( containingRegion->fType == type)? containingRegion: containingRegion->getContainingRegion(type); } /** @@ -618,9 +626,9 @@ Region::getContainedRegions( URegionType type, UErrorCode &status ) const { StringEnumeration *cr = getContainedRegions(status); for ( int32_t i = 0 ; i < cr->count(status) ; i++ ) { - const char *id = cr->next(NULL,status); - const Region *r = Region::getInstance(id,status); - if ( r->getType() == type ) { + const char *regionId = cr->next(NULL,status); + const Region *r = Region::getInstance(regionId,status); + if ( r->getType() == type) { result->addElement((void *)&r->idStr,status); } else { StringEnumeration *children = r->getContainedRegions(type, status); @@ -672,7 +680,7 @@ Region::contains(const Region &other) const { StringEnumeration* Region::getPreferredValues(UErrorCode &status) const { umtx_initOnce(gRegionDataInitOnce, &loadRegionData, status); // returns immediately if U_FAILURE(status) - if (U_FAILURE(status) || type != URGN_DEPRECATED) { + if (U_FAILURE(status) || fType != URGN_DEPRECATED) { return NULL; } return new RegionNameEnumeration(preferredValues,status); @@ -697,7 +705,7 @@ Region::getNumericCode() const { */ URegionType Region::getType() const { - return type; + return fType; } RegionNameEnumeration::RegionNameEnumeration(UVector *fNameList, UErrorCode& status) { diff --git a/deps/icu-small/source/i18n/reldatefmt.cpp b/deps/icu-small/source/i18n/reldatefmt.cpp index 42ede7ae4d..ae251ed20b 100644 --- a/deps/icu-small/source/i18n/reldatefmt.cpp +++ b/deps/icu-small/source/i18n/reldatefmt.cpp @@ -51,13 +51,13 @@ U_NAMESPACE_BEGIN // RelativeDateTimeFormatter specific data for a single locale class RelativeDateTimeCacheData: public SharedObject { public: - RelativeDateTimeCacheData() : combinedDateAndTime(NULL) { + RelativeDateTimeCacheData() : combinedDateAndTime(nullptr) { // Initialize the cache arrays for (int32_t style = 0; style < UDAT_STYLE_COUNT; ++style) { - for (int32_t relUnit = 0; relUnit < UDAT_RELATIVE_UNIT_COUNT; ++relUnit) { + for (int32_t relUnit = 0; relUnit < UDAT_REL_UNIT_COUNT; ++relUnit) { for (int32_t pl = 0; pl < StandardPlural::COUNT; ++pl) { - relativeUnitsFormatters[style][relUnit][0][pl] = NULL; - relativeUnitsFormatters[style][relUnit][1][pl] = NULL; + relativeUnitsFormatters[style][relUnit][0][pl] = nullptr; + relativeUnitsFormatters[style][relUnit][1][pl] = nullptr; } } } @@ -74,7 +74,7 @@ public: // e.g., Next Tuesday; Yesterday; etc. For third index, 0 // means past, e.g., 5 days ago; 1 means future, e.g., in 5 days. SimpleFormatter *relativeUnitsFormatters[UDAT_STYLE_COUNT] - [UDAT_RELATIVE_UNIT_COUNT][2][StandardPlural::COUNT]; + [UDAT_REL_UNIT_COUNT][2][StandardPlural::COUNT]; const UnicodeString& getAbsoluteUnitString(int32_t fStyle, UDateAbsoluteUnit unit, @@ -83,6 +83,10 @@ public: UDateRelativeUnit unit, int32_t pastFutureIndex, int32_t pluralUnit) const; + const SimpleFormatter* getRelativeDateTimeUnitFormatter(int32_t fStyle, + URelativeDateTimeUnit unit, + int32_t pastFutureIndex, + int32_t pluralUnit) const; const UnicodeString emptyString; @@ -107,7 +111,7 @@ private: RelativeDateTimeCacheData::~RelativeDateTimeCacheData() { // clear out the cache arrays for (int32_t style = 0; style < UDAT_STYLE_COUNT; ++style) { - for (int32_t relUnit = 0; relUnit < UDAT_RELATIVE_UNIT_COUNT; ++relUnit) { + for (int32_t relUnit = 0; relUnit < UDAT_REL_UNIT_COUNT; ++relUnit) { for (int32_t pl = 0; pl < StandardPlural::COUNT; ++pl) { delete relativeUnitsFormatters[style][relUnit][0][pl]; delete relativeUnitsFormatters[style][relUnit][1][pl]; @@ -131,20 +135,41 @@ const UnicodeString& RelativeDateTimeCacheData::getAbsoluteUnitString( return emptyString; } - // Use fallback cache for SimpleFormatter relativeUnits. const SimpleFormatter* RelativeDateTimeCacheData::getRelativeUnitFormatter( int32_t fStyle, UDateRelativeUnit unit, int32_t pastFutureIndex, int32_t pluralUnit) const { + URelativeDateTimeUnit rdtunit = UDAT_REL_UNIT_COUNT; + switch (unit) { + case UDAT_RELATIVE_YEARS: rdtunit = UDAT_REL_UNIT_YEAR; break; + case UDAT_RELATIVE_MONTHS: rdtunit = UDAT_REL_UNIT_MONTH; break; + case UDAT_RELATIVE_WEEKS: rdtunit = UDAT_REL_UNIT_WEEK; break; + case UDAT_RELATIVE_DAYS: rdtunit = UDAT_REL_UNIT_DAY; break; + case UDAT_RELATIVE_HOURS: rdtunit = UDAT_REL_UNIT_HOUR; break; + case UDAT_RELATIVE_MINUTES: rdtunit = UDAT_REL_UNIT_MINUTE; break; + case UDAT_RELATIVE_SECONDS: rdtunit = UDAT_REL_UNIT_SECOND; break; + default: // a unit that the above method does not handle + return nullptr; + } + + return getRelativeDateTimeUnitFormatter(fStyle, rdtunit, pastFutureIndex, pluralUnit); + } + + // Use fallback cache for SimpleFormatter relativeUnits. + const SimpleFormatter* RelativeDateTimeCacheData::getRelativeDateTimeUnitFormatter( + int32_t fStyle, + URelativeDateTimeUnit unit, + int32_t pastFutureIndex, + int32_t pluralUnit) const { int32_t style = fStyle; do { - if (relativeUnitsFormatters[style][unit][pastFutureIndex][pluralUnit] != NULL) { + if (relativeUnitsFormatters[style][unit][pastFutureIndex][pluralUnit] != nullptr) { return relativeUnitsFormatters[style][unit][pastFutureIndex][pluralUnit]; } style = fallBackCache[style]; } while (style != -1); - return NULL; // No formatter found. + return nullptr; // No formatter found. } static UBool getStringWithFallback( @@ -217,23 +242,35 @@ struct RelDateTimeFmtDataSink : public ResourceSink { // Converts the generic units to UDAT_RELATIVE version. switch (genUnit) { case SECOND: - return UDAT_RELATIVE_SECONDS; + return UDAT_REL_UNIT_SECOND; case MINUTE: - return UDAT_RELATIVE_MINUTES; + return UDAT_REL_UNIT_MINUTE; case HOUR: - return UDAT_RELATIVE_HOURS; + return UDAT_REL_UNIT_HOUR; case DAY: - return UDAT_RELATIVE_DAYS; + return UDAT_REL_UNIT_DAY; case WEEK: - return UDAT_RELATIVE_WEEKS; + return UDAT_REL_UNIT_WEEK; case MONTH: - return UDAT_RELATIVE_MONTHS; - /* - * case QUARTER: - * return UDATE_RELATIVE_QUARTERS; - */ + return UDAT_REL_UNIT_MONTH; + case QUARTER: + return UDAT_REL_UNIT_QUARTER; case YEAR: - return UDAT_RELATIVE_YEARS; + return UDAT_REL_UNIT_YEAR; + case SUNDAY: + return UDAT_REL_UNIT_SUNDAY; + case MONDAY: + return UDAT_REL_UNIT_MONDAY; + case TUESDAY: + return UDAT_REL_UNIT_TUESDAY; + case WEDNESDAY: + return UDAT_REL_UNIT_WEDNESDAY; + case THURSDAY: + return UDAT_REL_UNIT_THURSDAY; + case FRIDAY: + return UDAT_REL_UNIT_FRIDAY; + case SATURDAY: + return UDAT_REL_UNIT_SATURDAY; default: return -1; } @@ -248,10 +285,8 @@ struct RelDateTimeFmtDataSink : public ResourceSink { return UDAT_ABSOLUTE_WEEK; case MONTH: return UDAT_ABSOLUTE_MONTH; - /* TODO: Add in QUARTER - * case QUARTER: - * return UDAT_ABSOLUTE_QUARTER; - */ + case QUARTER: + return UDAT_ABSOLUTE_QUARTER; case YEAR: return UDAT_ABSOLUTE_YEAR; case SUNDAY: @@ -312,7 +347,7 @@ struct RelDateTimeFmtDataSink : public ResourceSink { // Utility functions static UDateRelativeDateTimeFormatterStyle styleFromString(const char *s) { - int32_t len = uprv_strlen(s); + int32_t len = static_cast<int32_t>(uprv_strlen(s)); if (len >= 7 && uprv_strcmp(s + len - 7, "-narrow") == 0) { return UDAT_STYLE_NARROW; } @@ -430,7 +465,7 @@ struct RelDateTimeFmtDataSink : public ResourceSink { } int32_t relUnitIndex = relUnitFromGeneric(genericUnit); - if (relUnitIndex == UDAT_RELATIVE_SECONDS && uprv_strcmp(key, "0") == 0 && + if (relUnitIndex == UDAT_REL_UNIT_SECOND && uprv_strcmp(key, "0") == 0 && outputData.absoluteUnits[style][UDAT_ABSOLUTE_NOW][UDAT_DIRECTION_PLAIN].isEmpty()) { // Handle "NOW" outputData.absoluteUnits[style][UDAT_ABSOLUTE_NOW] @@ -463,10 +498,10 @@ struct RelDateTimeFmtDataSink : public ResourceSink { outputData.relativeUnitsFormatters[style][relUnitIndex] [pastFutureIndex]; // Only set if not already established. - if (patterns[pluralIndex] == NULL) { + if (patterns[pluralIndex] == nullptr) { patterns[pluralIndex] = new SimpleFormatter( value.getUnicodeString(errorCode), 0, 1, errorCode); - if (patterns[pluralIndex] == NULL) { + if (patterns[pluralIndex] == nullptr) { errorCode = U_MEMORY_ALLOCATION_ERROR; } } @@ -546,7 +581,7 @@ struct RelDateTimeFmtDataSink : public ResourceSink { consumeAlias(key, value, errorCode); } else { style = styleFromString(key); - int32_t unitSize = uprv_strlen(key) - styleSuffixLength(style); + int32_t unitSize = static_cast<int32_t>(uprv_strlen(key)) - styleSuffixLength(style); genericUnit = unitOrNegativeFromString(key, unitSize); if (style >= 0 && genericUnit != INVALID_UNIT) { consumeTimeUnit(key, value, errorCode); @@ -570,8 +605,14 @@ static void loadWeekdayNames(UnicodeString absoluteUnits[UDAT_STYLE_COUNT] [UDAT_ABSOLUTE_UNIT_COUNT][UDAT_DIRECTION_COUNT], const char* localeId, UErrorCode& status) { + if (U_FAILURE(status)) { + return; + } Locale locale(localeId); DateFormatSymbols dfSym(locale, status); + if (U_FAILURE(status)) { + return; + } for (int32_t style = 0; style < UDAT_STYLE_COUNT; ++style) { DateFormatSymbols::DtWidthType dtfmtWidth = styleToDateFormatSymbolWidth[style]; int32_t count; @@ -595,6 +636,9 @@ static UBool loadUnitData( RelDateTimeFmtDataSink sink(cacheData); ures_getAllItemsWithFallback(resource, "fields", sink, status); + if (U_FAILURE(status)) { + return false; + } // Get the weekday names from DateFormatSymbols. loadWeekdayNames(cacheData.absoluteUnits, localeId, status); @@ -619,7 +663,7 @@ static UBool getDateTimePattern( .append("/DateTimePatterns", status); LocalUResourceBundlePointer topLevel( ures_getByKeyWithFallback( - resource, pathBuffer.data(), NULL, &status)); + resource, pathBuffer.data(), nullptr, &status)); if (U_FAILURE(status)) { return FALSE; } @@ -636,68 +680,68 @@ static UBool getDateTimePattern( template<> U_I18N_API const RelativeDateTimeCacheData *LocaleCacheKey<RelativeDateTimeCacheData>::createObject(const void * /*unused*/, UErrorCode &status) const { const char *localeId = fLoc.getName(); - LocalUResourceBundlePointer topLevel(ures_open(NULL, localeId, &status)); + LocalUResourceBundlePointer topLevel(ures_open(nullptr, localeId, &status)); if (U_FAILURE(status)) { - return NULL; + return nullptr; } LocalPointer<RelativeDateTimeCacheData> result( new RelativeDateTimeCacheData()); if (result.isNull()) { status = U_MEMORY_ALLOCATION_ERROR; - return NULL; + return nullptr; } if (!loadUnitData( topLevel.getAlias(), *result, localeId, status)) { - return NULL; + return nullptr; } UnicodeString dateTimePattern; if (!getDateTimePattern(topLevel.getAlias(), dateTimePattern, status)) { - return NULL; + return nullptr; } result->adoptCombinedDateAndTime( new SimpleFormatter(dateTimePattern, 2, 2, status)); if (U_FAILURE(status)) { - return NULL; + return nullptr; } result->addRef(); return result.orphan(); } RelativeDateTimeFormatter::RelativeDateTimeFormatter(UErrorCode& status) : - fCache(NULL), - fNumberFormat(NULL), - fPluralRules(NULL), + fCache(nullptr), + fNumberFormat(nullptr), + fPluralRules(nullptr), fStyle(UDAT_STYLE_LONG), fContext(UDISPCTX_CAPITALIZATION_NONE), - fOptBreakIterator(NULL) { - init(NULL, NULL, status); + fOptBreakIterator(nullptr) { + init(nullptr, nullptr, status); } RelativeDateTimeFormatter::RelativeDateTimeFormatter( const Locale& locale, UErrorCode& status) : - fCache(NULL), - fNumberFormat(NULL), - fPluralRules(NULL), + fCache(nullptr), + fNumberFormat(nullptr), + fPluralRules(nullptr), fStyle(UDAT_STYLE_LONG), fContext(UDISPCTX_CAPITALIZATION_NONE), - fOptBreakIterator(NULL), + fOptBreakIterator(nullptr), fLocale(locale) { - init(NULL, NULL, status); + init(nullptr, nullptr, status); } RelativeDateTimeFormatter::RelativeDateTimeFormatter( const Locale& locale, NumberFormat *nfToAdopt, UErrorCode& status) : - fCache(NULL), - fNumberFormat(NULL), - fPluralRules(NULL), + fCache(nullptr), + fNumberFormat(nullptr), + fPluralRules(nullptr), fStyle(UDAT_STYLE_LONG), fContext(UDISPCTX_CAPITALIZATION_NONE), - fOptBreakIterator(NULL), + fOptBreakIterator(nullptr), fLocale(locale) { - init(nfToAdopt, NULL, status); + init(nfToAdopt, nullptr, status); } RelativeDateTimeFormatter::RelativeDateTimeFormatter( @@ -706,12 +750,12 @@ RelativeDateTimeFormatter::RelativeDateTimeFormatter( UDateRelativeDateTimeFormatterStyle styl, UDisplayContext capitalizationContext, UErrorCode& status) : - fCache(NULL), - fNumberFormat(NULL), - fPluralRules(NULL), + fCache(nullptr), + fNumberFormat(nullptr), + fPluralRules(nullptr), fStyle(styl), fContext(capitalizationContext), - fOptBreakIterator(NULL), + fOptBreakIterator(nullptr), fLocale(locale) { if (U_FAILURE(status)) { return; @@ -727,7 +771,7 @@ RelativeDateTimeFormatter::RelativeDateTimeFormatter( } init(nfToAdopt, bi, status); } else { - init(nfToAdopt, NULL, status); + init(nfToAdopt, nullptr, status); } } @@ -744,7 +788,7 @@ RelativeDateTimeFormatter::RelativeDateTimeFormatter( fCache->addRef(); fNumberFormat->addRef(); fPluralRules->addRef(); - if (fOptBreakIterator != NULL) { + if (fOptBreakIterator != nullptr) { fOptBreakIterator->addRef(); } } @@ -764,16 +808,16 @@ RelativeDateTimeFormatter& RelativeDateTimeFormatter::operator=( } RelativeDateTimeFormatter::~RelativeDateTimeFormatter() { - if (fCache != NULL) { + if (fCache != nullptr) { fCache->removeRef(); } - if (fNumberFormat != NULL) { + if (fNumberFormat != nullptr) { fNumberFormat->removeRef(); } - if (fPluralRules != NULL) { + if (fPluralRules != nullptr) { fPluralRules->removeRef(); } - if (fOptBreakIterator != NULL) { + if (fOptBreakIterator != nullptr) { fOptBreakIterator->removeRef(); } } @@ -812,7 +856,7 @@ UnicodeString& RelativeDateTimeFormatter::format( const SimpleFormatter* formatter = fCache->getRelativeUnitFormatter(fStyle, unit, bFuture, pluralIndex); - if (formatter == NULL) { + if (formatter == nullptr) { // TODO: WARN - look at quantity formatter's action with an error. status = U_INVALID_FORMAT_ERROR; return appendTo; @@ -828,33 +872,35 @@ UnicodeString& RelativeDateTimeFormatter::formatNumeric( if (U_FAILURE(status)) { return appendTo; } - // TODO: - // The full implementation of this depends on CLDR data that is not yet available, - // see: http://unicode.org/cldr/trac/ticket/9165 Add more relative field data. - // In the meantime do a quick bring-up by calling the old format method; this - // leaves some holes (even for data that is currently available, such as quarter). - // When the new CLDR data is available, update the data storage accordingly, - // rewrite this to use it directly, and rewrite the old format method to call this - // new one; that is covered by http://bugs.icu-project.org/trac/ticket/12171. - UDateRelativeUnit relunit = UDAT_RELATIVE_UNIT_COUNT; - switch (unit) { - case UDAT_REL_UNIT_YEAR: relunit = UDAT_RELATIVE_YEARS; break; - case UDAT_REL_UNIT_MONTH: relunit = UDAT_RELATIVE_MONTHS; break; - case UDAT_REL_UNIT_WEEK: relunit = UDAT_RELATIVE_WEEKS; break; - case UDAT_REL_UNIT_DAY: relunit = UDAT_RELATIVE_DAYS; break; - case UDAT_REL_UNIT_HOUR: relunit = UDAT_RELATIVE_HOURS; break; - case UDAT_REL_UNIT_MINUTE: relunit = UDAT_RELATIVE_MINUTES; break; - case UDAT_REL_UNIT_SECOND: relunit = UDAT_RELATIVE_SECONDS; break; - default: // a unit that the above method does not handle - status = U_UNSUPPORTED_ERROR; - return appendTo; - } UDateDirection direction = UDAT_DIRECTION_NEXT; if (std::signbit(offset)) { // needed to handle -0.0 direction = UDAT_DIRECTION_LAST; offset = -offset; } - return format(offset, direction, relunit, appendTo, status); + if (direction != UDAT_DIRECTION_LAST && direction != UDAT_DIRECTION_NEXT) { + status = U_ILLEGAL_ARGUMENT_ERROR; + return appendTo; + } + int32_t bFuture = direction == UDAT_DIRECTION_NEXT ? 1 : 0; + FieldPosition pos(FieldPosition::DONT_CARE); + + UnicodeString result; + UnicodeString formattedNumber; + + StandardPlural::Form pluralIndex = QuantityFormatter::selectPlural( + offset, **fNumberFormat, **fPluralRules, formattedNumber, pos, + status); + + const SimpleFormatter* formatter = + fCache->getRelativeDateTimeUnitFormatter(fStyle, unit, bFuture, pluralIndex); + if (formatter == nullptr) { + // TODO: WARN - look at quantity formatter's action with an error. + status = U_INVALID_FORMAT_ERROR; + return appendTo; + } + formatter->format(formattedNumber, result, status); + adjustForContext(result); + return appendTo.append(result); } UnicodeString& RelativeDateTimeFormatter::format( @@ -871,7 +917,7 @@ UnicodeString& RelativeDateTimeFormatter::format( // Get string using fallback. UnicodeString result; result.fastCopyFrom(fCache->getAbsoluteUnitString(fStyle, unit, direction)); - if (fOptBreakIterator != NULL) { + if (fOptBreakIterator != nullptr) { adjustForContext(result); } return appendTo.append(result); @@ -908,6 +954,7 @@ UnicodeString& RelativeDateTimeFormatter::format( UDateAbsoluteUnit absunit = UDAT_ABSOLUTE_UNIT_COUNT; switch (unit) { case UDAT_REL_UNIT_YEAR: absunit = UDAT_ABSOLUTE_YEAR; break; + case UDAT_REL_UNIT_QUARTER: absunit = UDAT_ABSOLUTE_QUARTER; break; case UDAT_REL_UNIT_MONTH: absunit = UDAT_ABSOLUTE_MONTH; break; case UDAT_REL_UNIT_WEEK: absunit = UDAT_ABSOLUTE_WEEK; break; case UDAT_REL_UNIT_DAY: absunit = UDAT_ABSOLUTE_DAY; break; @@ -930,7 +977,7 @@ UnicodeString& RelativeDateTimeFormatter::format( const UnicodeString &unitFormatString = fCache->getAbsoluteUnitString(fStyle, absunit, direction); if (!unitFormatString.isEmpty()) { - if (fOptBreakIterator != NULL) { + if (fOptBreakIterator != nullptr) { UnicodeString result(unitFormatString); adjustForContext(result); return appendTo.append(result); @@ -951,7 +998,7 @@ UnicodeString& RelativeDateTimeFormatter::combineDateAndTime( } void RelativeDateTimeFormatter::adjustForContext(UnicodeString &str) const { - if (fOptBreakIterator == NULL + if (fOptBreakIterator == nullptr || str.length() == 0 || !u_islower(str.char32At(0))) { return; } @@ -992,7 +1039,7 @@ void RelativeDateTimeFormatter::init( shared->removeRef(); } else { SharedNumberFormat *shared = new SharedNumberFormat(nf.getAlias()); - if (shared == NULL) { + if (shared == nullptr) { status = U_MEMORY_ALLOCATION_ERROR; return; } @@ -1003,7 +1050,7 @@ void RelativeDateTimeFormatter::init( SharedObject::clearPtr(fOptBreakIterator); } else { SharedBreakIterator *shared = new SharedBreakIterator(bi.getAlias()); - if (shared == NULL) { + if (shared == nullptr) { status = U_MEMORY_ALLOCATION_ERROR; return; } @@ -1026,13 +1073,13 @@ ureldatefmt_open( const char* locale, UErrorCode* status ) { if (U_FAILURE(*status)) { - return NULL; + return nullptr; } LocalPointer<RelativeDateTimeFormatter> formatter(new RelativeDateTimeFormatter(Locale(locale), (NumberFormat*)nfToAdopt, width, capitalizationContext, *status), *status); if (U_FAILURE(*status)) { - return NULL; + return nullptr; } return (URelativeDateTimeFormatter*)formatter.orphan(); } @@ -1054,13 +1101,13 @@ ureldatefmt_formatNumeric( const URelativeDateTimeFormatter* reldatefmt, if (U_FAILURE(*status)) { return 0; } - if (result == NULL ? resultCapacity != 0 : resultCapacity < 0) { + if (result == nullptr ? resultCapacity != 0 : resultCapacity < 0) { *status = U_ILLEGAL_ARGUMENT_ERROR; return 0; } UnicodeString res; - if (result != NULL) { - // NULL destination for pure preflighting: empty dummy string + if (result != nullptr) { + // nullptr destination for pure preflighting: empty dummy string // otherwise, alias the destination buffer (copied from udat_format) res.setTo(result, 0, resultCapacity); } @@ -1082,13 +1129,13 @@ ureldatefmt_format( const URelativeDateTimeFormatter* reldatefmt, if (U_FAILURE(*status)) { return 0; } - if (result == NULL ? resultCapacity != 0 : resultCapacity < 0) { + if (result == nullptr ? resultCapacity != 0 : resultCapacity < 0) { *status = U_ILLEGAL_ARGUMENT_ERROR; return 0; } UnicodeString res; - if (result != NULL) { - // NULL destination for pure preflighting: empty dummy string + if (result != nullptr) { + // nullptr destination for pure preflighting: empty dummy string // otherwise, alias the destination buffer (copied from udat_format) res.setTo(result, 0, resultCapacity); } @@ -1112,9 +1159,9 @@ ureldatefmt_combineDateAndTime( const URelativeDateTimeFormatter* reldatefmt, if (U_FAILURE(*status)) { return 0; } - if (result == NULL ? resultCapacity != 0 : resultCapacity < 0 || - (relativeDateString == NULL ? relativeDateStringLen != 0 : relativeDateStringLen < -1) || - (timeString == NULL ? timeStringLen != 0 : timeStringLen < -1)) { + if (result == nullptr ? resultCapacity != 0 : resultCapacity < 0 || + (relativeDateString == nullptr ? relativeDateStringLen != 0 : relativeDateStringLen < -1) || + (timeString == nullptr ? timeStringLen != 0 : timeStringLen < -1)) { *status = U_ILLEGAL_ARGUMENT_ERROR; return 0; } diff --git a/deps/icu-small/source/i18n/reldtfmt.cpp b/deps/icu-small/source/i18n/reldtfmt.cpp index d3ab45dc63..753672d905 100644 --- a/deps/icu-small/source/i18n/reldtfmt.cpp +++ b/deps/icu-small/source/i18n/reldtfmt.cpp @@ -430,7 +430,7 @@ RelativeDateFormat::setContext(UDisplayContext value, UErrorCode& status) if ( fCapitalizationBrkIter == NULL && (value==UDISPCTX_CAPITALIZATION_FOR_BEGINNING_OF_SENTENCE || (value==UDISPCTX_CAPITALIZATION_FOR_UI_LIST_OR_MENU && fCapitalizationOfRelativeUnitsForUIListMenu) || (value==UDISPCTX_CAPITALIZATION_FOR_STANDALONE && fCapitalizationOfRelativeUnitsForStandAlone)) ) { - UErrorCode status = U_ZERO_ERROR; + status = U_ZERO_ERROR; fCapitalizationBrkIter = BreakIterator::createSentenceInstance(fLocale, status); if (U_FAILURE(status)) { delete fCapitalizationBrkIter; @@ -456,8 +456,8 @@ RelativeDateFormat::initCapitalizationContextInfo(const Locale& thelocale) const int32_t * intVector = ures_getIntVector(rb.getAlias(), &len, &status); if (U_SUCCESS(status) && intVector != NULL && len >= 2) { - fCapitalizationOfRelativeUnitsForUIListMenu = intVector[0]; - fCapitalizationOfRelativeUnitsForStandAlone = intVector[1]; + fCapitalizationOfRelativeUnitsForUIListMenu = static_cast<UBool>(intVector[0]); + fCapitalizationOfRelativeUnitsForStandAlone = static_cast<UBool>(intVector[1]); } } #endif diff --git a/deps/icu-small/source/i18n/reldtfmt.h b/deps/icu-small/source/i18n/reldtfmt.h index 5063a6388f..0403da11ef 100644 --- a/deps/icu-small/source/i18n/reldtfmt.h +++ b/deps/icu-small/source/i18n/reldtfmt.h @@ -235,7 +235,6 @@ public: */ virtual const DateFormatSymbols* getDateFormatSymbols(void) const; - /* Cannot use #ifndef U_HIDE_DRAFT_API for the following draft method since it is virtual */ /** * Set a particular UDisplayContext value in the formatter, such as * UDISPCTX_CAPITALIZATION_FOR_STANDALONE. Note: For getContext, see diff --git a/deps/icu-small/source/i18n/rematch.cpp b/deps/icu-small/source/i18n/rematch.cpp index f252182207..95fd0d2240 100644 --- a/deps/icu-small/source/i18n/rematch.cpp +++ b/deps/icu-small/source/i18n/rematch.cpp @@ -801,7 +801,7 @@ UBool RegexMatcher::find(UErrorCode &status) { case START_LINE: { - UChar32 c; + UChar32 ch; if (startPos == fAnchorStart) { MatchAt(startPos, FALSE, status); if (U_FAILURE(status)) { @@ -811,17 +811,17 @@ UBool RegexMatcher::find(UErrorCode &status) { return TRUE; } UTEXT_SETNATIVEINDEX(fInputText, startPos); - c = UTEXT_NEXT32(fInputText); + ch = UTEXT_NEXT32(fInputText); startPos = UTEXT_GETNATIVEINDEX(fInputText); } else { UTEXT_SETNATIVEINDEX(fInputText, startPos); - c = UTEXT_PREVIOUS32(fInputText); + ch = UTEXT_PREVIOUS32(fInputText); UTEXT_SETNATIVEINDEX(fInputText, startPos); } if (fPattern->fFlags & UREGEX_UNIX_LINES) { for (;;) { - if (c == 0x0a) { + if (ch == 0x0a) { MatchAt(startPos, FALSE, status); if (U_FAILURE(status)) { return FALSE; @@ -836,7 +836,7 @@ UBool RegexMatcher::find(UErrorCode &status) { fHitEnd = TRUE; return FALSE; } - c = UTEXT_NEXT32(fInputText); + ch = UTEXT_NEXT32(fInputText); startPos = UTEXT_GETNATIVEINDEX(fInputText); // Note that it's perfectly OK for a pattern to have a zero-length // match at the end of a string, so we must make sure that the loop @@ -846,8 +846,8 @@ UBool RegexMatcher::find(UErrorCode &status) { } } else { for (;;) { - if (isLineTerminator(c)) { - if (c == 0x0d && startPos < fActiveLimit && UTEXT_CURRENT32(fInputText) == 0x0a) { + if (isLineTerminator(ch)) { + if (ch == 0x0d && startPos < fActiveLimit && UTEXT_CURRENT32(fInputText) == 0x0a) { (void)UTEXT_NEXT32(fInputText); startPos = UTEXT_GETNATIVEINDEX(fInputText); } @@ -865,7 +865,7 @@ UBool RegexMatcher::find(UErrorCode &status) { fHitEnd = TRUE; return FALSE; } - c = UTEXT_NEXT32(fInputText); + ch = UTEXT_NEXT32(fInputText); startPos = UTEXT_GETNATIVEINDEX(fInputText); // Note that it's perfectly OK for a pattern to have a zero-length // match at the end of a string, so we must make sure that the loop @@ -1034,7 +1034,7 @@ UBool RegexMatcher::findUsingChunk(UErrorCode &status) { return FALSE; } } - U_ASSERT(FALSE); + U_ASSERT(FALSE); case START_STRING: case START_CHAR: @@ -1067,7 +1067,7 @@ UBool RegexMatcher::findUsingChunk(UErrorCode &status) { case START_LINE: { - UChar32 c; + UChar32 ch; if (startPos == fAnchorStart) { MatchChunkAt(startPos, FALSE, status); if (U_FAILURE(status)) { @@ -1081,8 +1081,8 @@ UBool RegexMatcher::findUsingChunk(UErrorCode &status) { if (fPattern->fFlags & UREGEX_UNIX_LINES) { for (;;) { - c = inputBuf[startPos-1]; - if (c == 0x0a) { + ch = inputBuf[startPos-1]; + if (ch == 0x0a) { MatchChunkAt(startPos, FALSE, status); if (U_FAILURE(status)) { return FALSE; @@ -1105,9 +1105,9 @@ UBool RegexMatcher::findUsingChunk(UErrorCode &status) { } } else { for (;;) { - c = inputBuf[startPos-1]; - if (isLineTerminator(c)) { - if (c == 0x0d && startPos < fActiveLimit && inputBuf[startPos] == 0x0a) { + ch = inputBuf[startPos-1]; + if (isLineTerminator(ch)) { + if (ch == 0x0d && startPos < fActiveLimit && inputBuf[startPos] == 0x0a) { startPos++; } MatchChunkAt(startPos, FALSE, status); @@ -2774,7 +2774,7 @@ void RegexMatcher::MatchAt(int64_t startIdx, UBool toEnd, UErrorCode &status) { int64_t *pat = fPattern->fCompiledPat->getBuffer(); const UChar *litText = fPattern->fLiteralText.getBuffer(); - UVector *sets = fPattern->fSets; + UVector *fSets = fPattern->fSets; fFrameSize = fPattern->fFrameSize; REStackFrame *fp = resetStack(); @@ -3376,7 +3376,7 @@ GC_Done: // There is input left. Pick up one char and test it for set membership. UChar32 c = UTEXT_NEXT32(fInputText); - U_ASSERT(opValue > 0 && opValue < sets->size()); + U_ASSERT(opValue > 0 && opValue < fSets->size()); if (c<256) { Regex8BitSet *s8 = &fPattern->fSets8[opValue]; if (s8->contains(c)) { @@ -3384,7 +3384,7 @@ GC_Done: break; } } else { - UnicodeSet *s = (UnicodeSet *)sets->elementAt(opValue); + UnicodeSet *s = (UnicodeSet *)fSets->elementAt(opValue); if (s->contains(c)) { // The character is in the set. A Match. fp->fInputIdx = UTEXT_GETNATIVEINDEX(fInputText); @@ -3671,9 +3671,9 @@ GC_Done: if (newFP == (int64_t *)fp) { break; } - int32_t i; - for (i=0; i<fFrameSize; i++) { - newFP[i] = ((int64_t *)fp)[i]; + int32_t j; + for (j=0; j<fFrameSize; j++) { + newFP[j] = ((int64_t *)fp)[j]; } fp = (REStackFrame *)newFP; fStack->setSize(newStackSize); @@ -3830,9 +3830,9 @@ GC_Done: // This makes the capture groups from within the look-ahead // expression available. int64_t *newFP = fStack->getBuffer() + newStackSize - fFrameSize; - int32_t i; - for (i=0; i<fFrameSize; i++) { - newFP[i] = ((int64_t *)fp)[i]; + int32_t j; + for (j=0; j<fFrameSize; j++) { + newFP[j] = ((int64_t *)fp)[j]; } fp = (REStackFrame *)newFP; fStack->setSize(newStackSize); @@ -4123,9 +4123,9 @@ GC_Done: // This op scans through all matching input. // The following LOOP_C op emulates stack unwinding if the following pattern fails. { - U_ASSERT(opValue > 0 && opValue < sets->size()); + U_ASSERT(opValue > 0 && opValue < fSets->size()); Regex8BitSet *s8 = &fPattern->fSets8[opValue]; - UnicodeSet *s = (UnicodeSet *)sets->elementAt(opValue); + UnicodeSet *s = (UnicodeSet *)fSets->elementAt(opValue); // Loop through input, until either the input is exhausted or // we reach a character that is not a member of the set. @@ -4350,7 +4350,7 @@ void RegexMatcher::MatchChunkAt(int32_t startIdx, UBool toEnd, UErrorCode &statu int64_t *pat = fPattern->fCompiledPat->getBuffer(); const UChar *litText = fPattern->fLiteralText.getBuffer(); - UVector *sets = fPattern->fSets; + UVector *fSets = fPattern->fSets; const UChar *inputBuf = fInputText->chunkContents; @@ -4928,7 +4928,7 @@ GC_Done: break; } - U_ASSERT(opValue > 0 && opValue < sets->size()); + U_ASSERT(opValue > 0 && opValue < fSets->size()); // There is input left. Pick up one char and test it for set membership. UChar32 c; @@ -4940,7 +4940,7 @@ GC_Done: break; } } else { - UnicodeSet *s = (UnicodeSet *)sets->elementAt(opValue); + UnicodeSet *s = (UnicodeSet *)fSets->elementAt(opValue); if (s->contains(c)) { // The character is in the set. A Match. break; @@ -5214,9 +5214,9 @@ GC_Done: if (newFP == (int64_t *)fp) { break; } - int32_t i; - for (i=0; i<fFrameSize; i++) { - newFP[i] = ((int64_t *)fp)[i]; + int32_t j; + for (j=0; j<fFrameSize; j++) { + newFP[j] = ((int64_t *)fp)[j]; } fp = (REStackFrame *)newFP; fStack->setSize(newStackSize); @@ -5361,9 +5361,9 @@ GC_Done: // This makes the capture groups from within the look-ahead // expression available. int64_t *newFP = fStack->getBuffer() + newStackSize - fFrameSize; - int32_t i; - for (i=0; i<fFrameSize; i++) { - newFP[i] = ((int64_t *)fp)[i]; + int32_t j; + for (j=0; j<fFrameSize; j++) { + newFP[j] = ((int64_t *)fp)[j]; } fp = (REStackFrame *)newFP; fStack->setSize(newStackSize); @@ -5623,9 +5623,9 @@ GC_Done: // This op scans through all matching input. // The following LOOP_C op emulates stack unwinding if the following pattern fails. { - U_ASSERT(opValue > 0 && opValue < sets->size()); + U_ASSERT(opValue > 0 && opValue < fSets->size()); Regex8BitSet *s8 = &fPattern->fSets8[opValue]; - UnicodeSet *s = (UnicodeSet *)sets->elementAt(opValue); + UnicodeSet *s = (UnicodeSet *)fSets->elementAt(opValue); // Loop through input, until either the input is exhausted or // we reach a character that is not a member of the set. diff --git a/deps/icu-small/source/i18n/rulebasedcollator.cpp b/deps/icu-small/source/i18n/rulebasedcollator.cpp index ab65f10a3b..b057b6bbd5 100644 --- a/deps/icu-small/source/i18n/rulebasedcollator.cpp +++ b/deps/icu-small/source/i18n/rulebasedcollator.cpp @@ -764,9 +764,9 @@ RuleBasedCollator::internalCompareUTF8(const char *left, int32_t leftLength, // Make sure both or neither strings have a known length. // We do not optimize for mixed length/termination. if(leftLength >= 0) { - if(rightLength < 0) { rightLength = uprv_strlen(right); } + if(rightLength < 0) { rightLength = static_cast<int32_t>(uprv_strlen(right)); } } else { - if(rightLength >= 0) { leftLength = uprv_strlen(left); } + if(rightLength >= 0) { leftLength = static_cast<int32_t>(uprv_strlen(left)); } } return doCompare(reinterpret_cast<const uint8_t *>(left), leftLength, reinterpret_cast<const uint8_t *>(right), rightLength, errorCode); @@ -862,9 +862,9 @@ public: } else { str.setTo(text, (int32_t)(spanLimit - text)); { - ReorderingBuffer buffer(nfcImpl, str); - if(buffer.init(str.length(), errorCode)) { - nfcImpl.makeFCD(spanLimit, textLimit, &buffer, errorCode); + ReorderingBuffer r_buffer(nfcImpl, str); + if(r_buffer.init(str.length(), errorCode)) { + nfcImpl.makeFCD(spanLimit, textLimit, &r_buffer, errorCode); } } if(U_SUCCESS(errorCode)) { diff --git a/deps/icu-small/source/i18n/scriptset.cpp b/deps/icu-small/source/i18n/scriptset.cpp index 9358e63b9e..558d178e2f 100644 --- a/deps/icu-small/source/i18n/scriptset.cpp +++ b/deps/icu-small/source/i18n/scriptset.cpp @@ -298,7 +298,7 @@ uhash_compareScriptSet(UElement key0, UElement key1) { icu::ScriptSet *s0 = static_cast<icu::ScriptSet *>(key0.pointer); icu::ScriptSet *s1 = static_cast<icu::ScriptSet *>(key1.pointer); int32_t diff = s0->countMembers() - s1->countMembers(); - if (diff != 0) return diff; + if (diff != 0) return static_cast<UBool>(diff); int32_t i0 = s0->nextSetBit(0); int32_t i1 = s1->nextSetBit(0); while ((diff = i0-i1) == 0 && i0 > 0) { diff --git a/deps/icu-small/source/i18n/shareddateformatsymbols.h b/deps/icu-small/source/i18n/shareddateformatsymbols.h index ca9a210819..66a06ecae5 100644 --- a/deps/icu-small/source/i18n/shareddateformatsymbols.h +++ b/deps/icu-small/source/i18n/shareddateformatsymbols.h @@ -12,6 +12,9 @@ #define __SHARED_DATEFORMATSYMBOLS_H__ #include "unicode/utypes.h" + +#if !UCONFIG_NO_FORMATTING + #include "sharedobject.h" #include "unicode/dtfmtsym.h" @@ -33,4 +36,6 @@ private: U_NAMESPACE_END +#endif /* !UCONFIG_NO_FORMATTING */ + #endif diff --git a/deps/icu-small/source/i18n/smpdtfmt.cpp b/deps/icu-small/source/i18n/smpdtfmt.cpp index b1b90882fc..2bc8e49625 100644 --- a/deps/icu-small/source/i18n/smpdtfmt.cpp +++ b/deps/icu-small/source/i18n/smpdtfmt.cpp @@ -244,9 +244,9 @@ SimpleDateFormat::NSOverride::~NSOverride() { void SimpleDateFormat::NSOverride::free() { NSOverride *cur = this; while (cur) { - NSOverride *next = cur->next; + NSOverride *next_temp = cur->next; delete cur; - cur = next; + cur = next_temp; } } @@ -1304,15 +1304,15 @@ SimpleDateFormat::processOverrideString(const Locale &locale, const UnicodeStrin int32_t nsNameHash = nsName.hashCode(); // See if the numbering system is in the override list, if not, then add it. - NSOverride *cur = overrideList; + NSOverride *curr = overrideList; const SharedNumberFormat *snf = NULL; UBool found = FALSE; - while ( cur && !found ) { - if ( cur->hash == nsNameHash ) { - snf = cur->snf; + while ( curr && !found ) { + if ( curr->hash == nsNameHash ) { + snf = curr->snf; found = TRUE; } - cur = cur->next; + curr = curr->next; } if (!found) { @@ -1824,14 +1824,14 @@ SimpleDateFormat::subFormat(UnicodeString &appendTo, // Stealing am/pm value to use as our array index. // It works out: am/midnight are both 0, pm/noon are both 1, // 12 am is 12 midnight, and 12 pm is 12 noon. - int32_t value = cal.get(UCAL_AM_PM, status); + int32_t val = cal.get(UCAL_AM_PM, status); if (count <= 3) { - toAppend = &fSymbols->fAbbreviatedDayPeriods[value]; + toAppend = &fSymbols->fAbbreviatedDayPeriods[val]; } else if (count == 4 || count > 5) { - toAppend = &fSymbols->fWideDayPeriods[value]; + toAppend = &fSymbols->fWideDayPeriods[val]; } else { // count == 5 - toAppend = &fSymbols->fNarrowDayPeriods[value]; + toAppend = &fSymbols->fNarrowDayPeriods[val]; } } @@ -2281,10 +2281,10 @@ SimpleDateFormat::parse(const UnicodeString& text, Calendar& cal, ParsePosition& if (i+1 < fPattern.length()) { // move to next pattern character - UChar ch = fPattern.charAt(i+1); + UChar c = fPattern.charAt(i+1); // check for whitespace - if (PatternProps::isWhiteSpace(ch)) { + if (PatternProps::isWhiteSpace(c)) { i++; // Advance over run in pattern while ((i+1)<fPattern.length() && @@ -3162,8 +3162,8 @@ int32_t SimpleDateFormat::subParse(const UnicodeString& text, int32_t& start, UC if (!strcmp(cal.getType(),"hebrew")) { HebrewCalendar *hc = (HebrewCalendar*)&cal; if (cal.isSet(UCAL_YEAR)) { - UErrorCode status = U_ZERO_ERROR; - if (!hc->isLeapYear(hc->get(UCAL_YEAR,status)) && value >= 6) { + UErrorCode monthStatus = U_ZERO_ERROR; + if (!hc->isLeapYear(hc->get(UCAL_YEAR, monthStatus)) && value >= 6) { cal.set(UCAL_MONTH, value); } else { cal.set(UCAL_MONTH, value - 1); @@ -3571,21 +3571,21 @@ int32_t SimpleDateFormat::subParse(const UnicodeString& text, int32_t& start, UC static const UChar alt_sep = DateFormatSymbols::ALTERNATE_TIME_SEPARATOR; // Try matching a time separator. - int32_t count = 1; + int32_t count_sep = 1; UnicodeString data[3]; fSymbols->getTimeSeparatorString(data[0]); // Add the default, if different from the locale. if (data[0].compare(&def_sep, 1) != 0) { - data[count++].setTo(def_sep); + data[count_sep++].setTo(def_sep); } // If lenient, add also the alternate, if different from the locale. if (isLenient() && data[0].compare(&alt_sep, 1) != 0) { - data[count++].setTo(alt_sep); + data[count_sep++].setTo(alt_sep); } - return matchString(text, start, UCAL_FIELD_COUNT /* => nothing to set */, data, count, NULL, cal); + return matchString(text, start, UCAL_FIELD_COUNT /* => nothing to set */, data, count_sep, NULL, cal); } case UDAT_AM_PM_MIDNIGHT_NOON_FIELD: @@ -3674,7 +3674,7 @@ int32_t SimpleDateFormat::subParse(const UnicodeString& text, int32_t& start, UC } parseInt(*src, number, pos, allowNegative,currentNumberFormat); if (pos.getIndex() != parseStart) { - int32_t value = number.getLong(); + int32_t val = number.getLong(); // Don't need suffix processing here (as in number processing at the beginning of the function); // the new fields being handled as numeric values (month, weekdays, quarters) should not have suffixes. @@ -3682,7 +3682,7 @@ int32_t SimpleDateFormat::subParse(const UnicodeString& text, int32_t& start, UC if (!getBooleanAttribute(UDAT_PARSE_ALLOW_NUMERIC, status)) { // Check the range of the value int32_t bias = gFieldRangeBias[patternCharIndex]; - if (bias >= 0 && (value > cal.getMaximum(field) + bias || value < cal.getMinimum(field) + bias)) { + if (bias >= 0 && (val > cal.getMaximum(field) + bias || val < cal.getMinimum(field) + bias)) { return -start; } } @@ -3696,35 +3696,35 @@ int32_t SimpleDateFormat::subParse(const UnicodeString& text, int32_t& start, UC if (!strcmp(cal.getType(),"hebrew")) { HebrewCalendar *hc = (HebrewCalendar*)&cal; if (cal.isSet(UCAL_YEAR)) { - UErrorCode status = U_ZERO_ERROR; - if (!hc->isLeapYear(hc->get(UCAL_YEAR,status)) && value >= 6) { - cal.set(UCAL_MONTH, value); + UErrorCode monthStatus = U_ZERO_ERROR; + if (!hc->isLeapYear(hc->get(UCAL_YEAR, monthStatus)) && val >= 6) { + cal.set(UCAL_MONTH, val); } else { - cal.set(UCAL_MONTH, value - 1); + cal.set(UCAL_MONTH, val - 1); } } else { - saveHebrewMonth = value; + saveHebrewMonth = val; } } else { - cal.set(UCAL_MONTH, value - 1); + cal.set(UCAL_MONTH, val - 1); } break; case UDAT_STANDALONE_MONTH_FIELD: - cal.set(UCAL_MONTH, value - 1); + cal.set(UCAL_MONTH, val - 1); break; case UDAT_DOW_LOCAL_FIELD: case UDAT_STANDALONE_DAY_FIELD: - cal.set(UCAL_DOW_LOCAL, value); + cal.set(UCAL_DOW_LOCAL, val); break; case UDAT_QUARTER_FIELD: case UDAT_STANDALONE_QUARTER_FIELD: - cal.set(UCAL_MONTH, (value - 1) * 3); + cal.set(UCAL_MONTH, (val - 1) * 3); break; case UDAT_RELATED_YEAR_FIELD: - cal.setRelatedYear(value); + cal.setRelatedYear(val); break; default: - cal.set(field, value); + cal.set(field, val); break; } return pos.getIndex(); @@ -3971,7 +3971,7 @@ SimpleDateFormat::setContext(UDisplayContext value, UErrorCode& status) if (U_SUCCESS(status)) { if ( fCapitalizationBrkIter == NULL && (value==UDISPCTX_CAPITALIZATION_FOR_BEGINNING_OF_SENTENCE || value==UDISPCTX_CAPITALIZATION_FOR_UI_LIST_OR_MENU || value==UDISPCTX_CAPITALIZATION_FOR_STANDALONE) ) { - UErrorCode status = U_ZERO_ERROR; + status = U_ZERO_ERROR; fCapitalizationBrkIter = BreakIterator::createSentenceInstance(fLocale, status); if (U_FAILURE(status)) { delete fCapitalizationBrkIter; diff --git a/deps/icu-small/source/i18n/timezone.cpp b/deps/icu-small/source/i18n/timezone.cpp index f7f45c7d3e..dbf614469e 100644 --- a/deps/icu-small/source/i18n/timezone.cpp +++ b/deps/icu-small/source/i18n/timezone.cpp @@ -1031,8 +1031,8 @@ TimeZone::getEquivalentID(const UnicodeString& id, int32_t index) { UResourceBundle *ares = ures_getByKey(top, kNAMES, NULL, &ec); // dereference Zones section if (U_SUCCESS(ec)) { int32_t idLen = 0; - const UChar* id = ures_getStringByIndex(ares, zone, &idLen, &ec); - result.fastCopyFrom(UnicodeString(TRUE, id, idLen)); + const UChar* id2 = ures_getStringByIndex(ares, zone, &idLen, &ec); + result.fastCopyFrom(UnicodeString(TRUE, id2, idLen)); U_DEBUG_TZ_MSG(("gei(%d) -> %d, len%d, %s\n", index, zone, result.length(), u_errorName(ec))); } ures_close(ares); @@ -1199,7 +1199,7 @@ TimeZone::getDisplayName(UBool daylight, EDisplayType style, const Locale& local { UErrorCode status = U_ZERO_ERROR; UDate date = Calendar::getNow(); - UTimeZoneFormatTimeType timeType; + UTimeZoneFormatTimeType timeType = UTZFMT_TIME_TYPE_UNKNOWN; int32_t offset; if (style == GENERIC_LOCATION || style == LONG_GENERIC || style == SHORT_GENERIC) { @@ -1612,7 +1612,7 @@ TimeZone::getWindowsID(const UnicodeString& id, UnicodeString& winid, UErrorCode end = tzids + len; hasNext = FALSE; } - if (canonicalID.compare(start, end - start) == 0) { + if (canonicalID.compare(start, static_cast<int32_t>(end - start)) == 0) { winid = UnicodeString(ures_getKey(winzone), -1 , US_INV); found = TRUE; break; @@ -1673,7 +1673,7 @@ TimeZone::getIDForWindowsID(const UnicodeString& winid, const char* region, Unic if (end == NULL) { id.setTo(tzids, -1); } else { - id.setTo(tzids, end - tzids); + id.setTo(tzids, static_cast<int32_t>(end - tzids)); } gotID = TRUE; } diff --git a/deps/icu-small/source/i18n/transreg.cpp b/deps/icu-small/source/i18n/transreg.cpp index 331f4efdeb..4884773faf 100644 --- a/deps/icu-small/source/i18n/transreg.cpp +++ b/deps/icu-small/source/i18n/transreg.cpp @@ -1330,12 +1330,12 @@ Transliterator* TransliteratorRegistry::instantiateEntry(const UnicodeString& ID int32_t passNumber = 1; for (int32_t i = 0; U_SUCCESS(status) && i < entry->u.dataVector->size(); i++) { // TODO: Should passNumber be turned into a decimal-string representation (1 -> "1")? - Transliterator* t = new RuleBasedTransliterator(UnicodeString(CompoundTransliterator::PASS_STRING) + UnicodeString(passNumber++), + Transliterator* tl = new RuleBasedTransliterator(UnicodeString(CompoundTransliterator::PASS_STRING) + UnicodeString(passNumber++), (TransliterationRuleData*)(entry->u.dataVector->elementAt(i)), FALSE); - if (t == 0) + if (tl == 0) status = U_MEMORY_ALLOCATION_ERROR; else - rbts->addElement(t, status); + rbts->addElement(tl, status); } if (U_FAILURE(status)) { delete rbts; diff --git a/deps/icu-small/source/i18n/tzfmt.cpp b/deps/icu-small/source/i18n/tzfmt.cpp index 69da4667bc..df4dec1feb 100644 --- a/deps/icu-small/source/i18n/tzfmt.cpp +++ b/deps/icu-small/source/i18n/tzfmt.cpp @@ -2648,12 +2648,12 @@ TimeZoneFormat::checkAbuttingHoursAndMinutes() { UVector *items = fGMTOffsetPatternItems[type]; for (int32_t i = 0; i < items->size(); i++) { const GMTOffsetField* item = (GMTOffsetField*)items->elementAt(i); - GMTOffsetField::FieldType type = item->getType(); - if (type != GMTOffsetField::TEXT) { + GMTOffsetField::FieldType fieldType = item->getType(); + if (fieldType != GMTOffsetField::TEXT) { if (afterH) { fAbuttingOffsetHoursAndMinutes = TRUE; break; - } else if (type == GMTOffsetField::HOUR) { + } else if (fieldType == GMTOffsetField::HOUR) { afterH = TRUE; } } else if (afterH) { diff --git a/deps/icu-small/source/i18n/tzgnames.cpp b/deps/icu-small/source/i18n/tzgnames.cpp index c2e685272e..5f5b7db302 100644 --- a/deps/icu-small/source/i18n/tzgnames.cpp +++ b/deps/icu-small/source/i18n/tzgnames.cpp @@ -407,7 +407,7 @@ TZGNCore::initialize(const Locale& locale, UErrorCode& status) { // target region const char* region = fLocale.getCountry(); - int32_t regionLen = uprv_strlen(region); + int32_t regionLen = static_cast<int32_t>(uprv_strlen(region)); if (regionLen == 0) { char loc[ULOC_FULLNAME_CAPACITY]; uloc_addLikelySubtags(fLocale.getName(), loc, sizeof(loc), &status); diff --git a/deps/icu-small/source/i18n/tznames.cpp b/deps/icu-small/source/i18n/tznames.cpp index 689fdeb091..5a79c22aac 100644 --- a/deps/icu-small/source/i18n/tznames.cpp +++ b/deps/icu-small/source/i18n/tznames.cpp @@ -87,7 +87,7 @@ static void sweepCache() { const UHashElement* elem; double now = (double)uprv_getUTCtime(); - while ((elem = uhash_nextElement(gTimeZoneNamesCache, &pos))) { + while ((elem = uhash_nextElement(gTimeZoneNamesCache, &pos)) != 0) { TimeZoneNamesCacheEntry *entry = (TimeZoneNamesCacheEntry *)elem->value.pointer; if (entry->refCount <= 0 && (now - entry->lastAccess) > CACHE_EXPIRATION) { // delete this entry diff --git a/deps/icu-small/source/i18n/tznames_impl.cpp b/deps/icu-small/source/i18n/tznames_impl.cpp index ef04b31c13..6a303ea4a0 100644 --- a/deps/icu-small/source/i18n/tznames_impl.cpp +++ b/deps/icu-small/source/i18n/tznames_impl.cpp @@ -1285,7 +1285,7 @@ static void mergeTimeZoneKey(const UnicodeString& mzID, char* result) { char mzIdChar[ZID_KEY_MAX + 1]; int32_t keyLen; - int32_t prefixLen = uprv_strlen(gMZPrefix); + int32_t prefixLen = static_cast<int32_t>(uprv_strlen(gMZPrefix)); keyLen = mzID.extract(0, mzID.length(), mzIdChar, ZID_KEY_MAX + 1, US_INV); uprv_memcpy((void *)result, (void *)gMZPrefix, prefixLen); uprv_memcpy((void *)(result + prefixLen), (void *)mzIdChar, keyLen); @@ -1453,7 +1453,7 @@ struct TimeZoneNamesImpl::ZoneStringsLoader : public ResourceSink { virtual ~ZoneStringsLoader(); void* createKey(const char* key, UErrorCode& status) { - int32_t len = sizeof(char) * (uprv_strlen(key) + 1); + int32_t len = sizeof(char) * (static_cast<int32_t>(uprv_strlen(key)) + 1); char* newKey = (char*) uprv_malloc(len); if (newKey == NULL) { status = U_MEMORY_ALLOCATION_ERROR; @@ -1469,7 +1469,7 @@ struct TimeZoneNamesImpl::ZoneStringsLoader : public ResourceSink { } UnicodeString mzIDFromKey(const char* key) { - return UnicodeString(key + MZ_PREFIX_LEN, uprv_strlen(key) - MZ_PREFIX_LEN, US_INV); + return UnicodeString(key + MZ_PREFIX_LEN, static_cast<int32_t>(uprv_strlen(key)) - MZ_PREFIX_LEN, US_INV); } UnicodeString tzIDFromKey(const char* key) { @@ -1944,8 +1944,8 @@ TZDBNameSearchHandler::handleMatch(int32_t matchLength, const CharacterNode *nod // metazone mapping for "CST" is America_Central, // but if region is one of CN/MO/TW, "CST" is parsed // as metazone China (China Standard Time). - for (int32_t i = 0; i < ninfo->nRegions; i++) { - const char *region = ninfo->parseRegions[i]; + for (int32_t j = 0; j < ninfo->nRegions; j++) { + const char *region = ninfo->parseRegions[j]; if (uprv_strcmp(fRegion, region) == 0) { match = ninfo; matchRegion = TRUE; @@ -2059,7 +2059,7 @@ static void U_CALLCONV prepareFind(UErrorCode &status) { const UnicodeString *mzID; StringEnumeration *mzIDs = TimeZoneNamesImpl::_getAvailableMetaZoneIDs(status); if (U_SUCCESS(status)) { - while ((mzID = mzIDs->snext(status)) && U_SUCCESS(status)) { + while ((mzID = mzIDs->snext(status)) != 0 && U_SUCCESS(status)) { const TZDBNames *names = TZDBTimeZoneNames::getMetaZoneNames(*mzID, status); if (U_FAILURE(status)) { break; @@ -2128,7 +2128,7 @@ TZDBTimeZoneNames::TZDBTimeZoneNames(const Locale& locale) : fLocale(locale) { UBool useWorld = TRUE; const char* region = fLocale.getCountry(); - int32_t regionLen = uprv_strlen(region); + int32_t regionLen = static_cast<int32_t>(uprv_strlen(region)); if (regionLen == 0) { UErrorCode status = U_ZERO_ERROR; char loc[ULOC_FULLNAME_CAPACITY]; diff --git a/deps/icu-small/source/i18n/ucln_in.h b/deps/icu-small/source/i18n/ucln_in.h index 318eafc143..4c13b9ffcb 100644 --- a/deps/icu-small/source/i18n/ucln_in.h +++ b/deps/icu-small/source/i18n/ucln_in.h @@ -32,6 +32,7 @@ typedef enum ECleanupI18NType { UCLN_I18N_SPOOFDATA, UCLN_I18N_TRANSLITERATOR, UCLN_I18N_REGEX, + UCLN_I18N_JAPANESE_CALENDAR, UCLN_I18N_ISLAMIC_CALENDAR, UCLN_I18N_CHINESE_CALENDAR, UCLN_I18N_HEBREW_CALENDAR, @@ -58,6 +59,7 @@ typedef enum ECleanupI18NType { UCLN_I18N_GENDERINFO, UCLN_I18N_CDFINFO, UCLN_I18N_REGION, + UCLN_I18N_LIST_FORMATTER, UCLN_I18N_COUNT /* This must be last */ } ECleanupI18NType; diff --git a/deps/icu-small/source/i18n/ucol_res.cpp b/deps/icu-small/source/i18n/ucol_res.cpp index 76975ecc01..56ed5b3c19 100644 --- a/deps/icu-small/source/i18n/ucol_res.cpp +++ b/deps/icu-small/source/i18n/ucol_res.cpp @@ -400,11 +400,11 @@ CollationLoader::loadFromData(UErrorCode &errorCode) { // Try to fetch the optional rules string. { UErrorCode internalErrorCode = U_ZERO_ERROR; - int32_t length; - const UChar *s = ures_getStringByKey(data, "Sequence", &length, + int32_t len; + const UChar *s = ures_getStringByKey(data, "Sequence", &len, &internalErrorCode); if(U_SUCCESS(internalErrorCode)) { - t->rules.setTo(TRUE, s, length); + t->rules.setTo(TRUE, s, len); } } @@ -426,10 +426,10 @@ CollationLoader::loadFromData(UErrorCode &errorCode) { LocalUResourceBundlePointer def( ures_getByKeyWithFallback(actualBundle.getAlias(), "collations/default", NULL, &internalErrorCode)); - int32_t length; - const UChar *s = ures_getString(def.getAlias(), &length, &internalErrorCode); - if(U_SUCCESS(internalErrorCode) && length < UPRV_LENGTHOF(defaultType)) { - u_UCharsToChars(s, defaultType, length + 1); + int32_t len; + const UChar *s = ures_getString(def.getAlias(), &len, &internalErrorCode); + if(U_SUCCESS(internalErrorCode) && len < UPRV_LENGTHOF(defaultType)) { + u_UCharsToChars(s, defaultType, len + 1); } else { uprv_strcpy(defaultType, "standard"); } diff --git a/deps/icu-small/source/i18n/udat.cpp b/deps/icu-small/source/i18n/udat.cpp index d086067c03..b7d85cc179 100644 --- a/deps/icu-small/source/i18n/udat.cpp +++ b/deps/icu-small/source/i18n/udat.cpp @@ -603,7 +603,7 @@ udat_getSymbols(const UDateFormat *fmt, } else { return -1; } - int32_t count; + int32_t count = 0; const UnicodeString *res = NULL; switch(type) { diff --git a/deps/icu-small/source/i18n/ulistformatter.cpp b/deps/icu-small/source/i18n/ulistformatter.cpp new file mode 100644 index 0000000000..c140c784b5 --- /dev/null +++ b/deps/icu-small/source/i18n/ulistformatter.cpp @@ -0,0 +1,91 @@ +// © 2016 and later: Unicode, Inc. and others. +// License & terms of use: http://www.unicode.org/copyright.html +/* +***************************************************************************************** +* Copyright (C) 2015, International Business Machines +* Corporation and others. All Rights Reserved. +***************************************************************************************** +*/ + +#include "unicode/utypes.h" + +#if !UCONFIG_NO_FORMATTING + +#include "unicode/ulistformatter.h" +#include "unicode/listformatter.h" +#include "unicode/localpointer.h" +#include "cmemory.h" + +U_NAMESPACE_USE + +U_CAPI UListFormatter* U_EXPORT2 +ulistfmt_open(const char* locale, + UErrorCode* status) +{ + if (U_FAILURE(*status)) { + return NULL; + } + LocalPointer<ListFormatter> listfmt(ListFormatter::createInstance(Locale(locale), *status)); + if (U_FAILURE(*status)) { + return NULL; + } + return (UListFormatter*)listfmt.orphan(); +} + + +U_CAPI void U_EXPORT2 +ulistfmt_close(UListFormatter *listfmt) +{ + delete (ListFormatter*)listfmt; +} + + +U_CAPI int32_t U_EXPORT2 +ulistfmt_format(const UListFormatter* listfmt, + const UChar* const strings[], + const int32_t * stringLengths, + int32_t stringCount, + UChar* result, + int32_t resultCapacity, + UErrorCode* status) +{ + if (U_FAILURE(*status)) { + return -1; + } + if (stringCount < 0 || (strings == NULL && stringCount > 0) || ((result == NULL)? resultCapacity != 0 : resultCapacity < 0)) { + *status = U_ILLEGAL_ARGUMENT_ERROR; + return -1; + } + UnicodeString ustringsStackBuf[4]; + UnicodeString* ustrings = ustringsStackBuf; + if (stringCount > UPRV_LENGTHOF(ustringsStackBuf)) { + ustrings = new UnicodeString[stringCount]; + if (ustrings == NULL) { + *status = U_MEMORY_ALLOCATION_ERROR; + return -1; + } + } + if (stringLengths == NULL) { + for (int32_t stringIndex = 0; stringIndex < stringCount; stringIndex++) { + ustrings[stringIndex].setTo(TRUE, strings[stringIndex], -1); + } + } else { + for (int32_t stringIndex = 0; stringIndex < stringCount; stringIndex++) { + ustrings[stringIndex].setTo(stringLengths[stringIndex] < 0, strings[stringIndex], stringLengths[stringIndex]); + } + } + UnicodeString res; + if (result != NULL) { + // NULL destination for pure preflighting: empty dummy string + // otherwise, alias the destination buffer (copied from udat_format) + res.setTo(result, 0, resultCapacity); + } + ((const ListFormatter*)listfmt)->format( ustrings, stringCount, res, *status ); + if (ustrings != ustringsStackBuf) { + delete[] ustrings; + } + return res.extract(result, resultCapacity, *status); +} + + +#endif /* #if !UCONFIG_NO_FORMATTING */ diff --git a/deps/icu-small/source/i18n/ulocdata.cpp b/deps/icu-small/source/i18n/ulocdata.cpp index 551f6c64ed..f651fee6fc 100644 --- a/deps/icu-small/source/i18n/ulocdata.cpp +++ b/deps/icu-small/source/i18n/ulocdata.cpp @@ -372,7 +372,7 @@ ulocdata_getLocaleSeparator(ULocaleData *uld, p1=u_strstr(separator, sub1); if (p0!=NULL && p1!=NULL && p0<=p1) { separator = (const UChar *)p0 + subLen; - len = p1 - separator; + len = static_cast<int32_t>(p1 - separator); /* Desired separator is no longer zero-terminated; handle that if necessary */ if (len < resultCapacity) { u_strncpy(result, separator, len); diff --git a/deps/icu-small/source/i18n/unicode/alphaindex.h b/deps/icu-small/source/i18n/unicode/alphaindex.h index 54bd29ff88..4ebdf1cc56 100644 --- a/deps/icu-small/source/i18n/unicode/alphaindex.h +++ b/deps/icu-small/source/i18n/unicode/alphaindex.h @@ -266,6 +266,8 @@ public: * Use getBucket() to get the bucket's properties. * * @param name the string to be sorted into an index bucket + * @param errorCode Error code, will be set with the reason if the + * operation fails. * @return the bucket number for the name * @stable ICU 51 */ @@ -377,9 +379,10 @@ public: /** - * Get the default label used for abbreviated buckets <i>between</i> other index characters. - * For example, consider the labels when Latin and Greek are used: - * X Y Z ... Α Β Γ. + * Get the default label used for abbreviated buckets *between* other index characters. + * For example, consider the labels when Latin (X Y Z) and Greek (Α Β Γ) are used: + * + * X Y Z ... Α Β Γ. * * @return inflow label * @stable ICU 4.8 @@ -700,6 +703,7 @@ public: /** * A (name, data) pair, to be sorted by name into one of the index buckets. * The user data is not used by the index implementation. + * \cond * @internal */ struct Record: public UMemory { @@ -708,6 +712,7 @@ public: Record(const UnicodeString &name, const void *data); ~Record(); }; + /** \endcond */ #endif /* U_HIDE_INTERNAL_API */ private: diff --git a/deps/icu-small/source/i18n/unicode/calendar.h b/deps/icu-small/source/i18n/unicode/calendar.h index 48021534b4..023cf053f2 100644 --- a/deps/icu-small/source/i18n/unicode/calendar.h +++ b/deps/icu-small/source/i18n/unicode/calendar.h @@ -52,83 +52,64 @@ typedef int32_t UFieldResolutionTable[12][8]; class BasicTimeZone; /** - * <code>Calendar</code> is an abstract base class for converting between - * a <code>UDate</code> object and a set of integer fields such as - * <code>YEAR</code>, <code>MONTH</code>, <code>DAY</code>, <code>HOUR</code>, - * and so on. (A <code>UDate</code> object represents a specific instant in + * `Calendar` is an abstract base class for converting between + * a `UDate` object and a set of integer fields such as + * `YEAR`, `MONTH`, `DAY`, `HOUR`, and so on. + * (A `UDate` object represents a specific instant in * time with millisecond precision. See UDate - * for information about the <code>UDate</code> class.) + * for information about the `UDate` class.) * - * <p> - * Subclasses of <code>Calendar</code> interpret a <code>UDate</code> + * Subclasses of `Calendar` interpret a `UDate` * according to the rules of a specific calendar system. - * The most commonly used subclass of <code>Calendar</code> is - * <code>GregorianCalendar</code>. Other subclasses could represent + * The most commonly used subclass of `Calendar` is + * `GregorianCalendar`. Other subclasses could represent * the various types of lunar calendars in use in many parts of the world. * - * <p> - * <b>NOTE</b>: (ICU 2.6) The subclass interface should be considered unstable - * - it WILL change. + * **NOTE**: (ICU 2.6) The subclass interface should be considered unstable - + * it WILL change. * - * <p> - * Like other locale-sensitive classes, <code>Calendar</code> provides a - * static method, <code>createInstance</code>, for getting a generally useful - * object of this type. <code>Calendar</code>'s <code>createInstance</code> method - * returns the appropriate <code>Calendar</code> subclass whose + * Like other locale-sensitive classes, `Calendar` provides a + * static method, `createInstance`, for getting a generally useful + * object of this type. `Calendar`'s `createInstance` method + * returns the appropriate `Calendar` subclass whose * time fields have been initialized with the current date and time: - * \htmlonly<blockquote>\endhtmlonly - * <pre> - * Calendar *rightNow = Calendar::createInstance(errCode); - * </pre> - * \htmlonly</blockquote>\endhtmlonly * - * <p> - * A <code>Calendar</code> object can produce all the time field values + * Calendar *rightNow = Calendar::createInstance(errCode); + * + * A `Calendar` object can produce all the time field values * needed to implement the date-time formatting for a particular language * and calendar style (for example, Japanese-Gregorian, Japanese-Traditional). * - * <p> - * When computing a <code>UDate</code> from time fields, some special circumstances + * When computing a `UDate` from time fields, some special circumstances * may arise: there may be insufficient information to compute the - * <code>UDate</code> (such as only year and month but no day in the month), + * `UDate` (such as only year and month but no day in the month), * there may be inconsistent information (such as "Tuesday, July 15, 1996" * -- July 15, 1996 is actually a Monday), or the input time might be ambiguous * because of time zone transition. * - * <p> - * <strong>Insufficient information.</strong> The calendar will use default + * **Insufficient information.** The calendar will use default * information to specify the missing fields. This may vary by calendar; for * the Gregorian calendar, the default for a field is the same as that of the * start of the epoch: i.e., YEAR = 1970, MONTH = JANUARY, DATE = 1, etc. * - * <p> - * <strong>Inconsistent information.</strong> If fields conflict, the calendar + * **Inconsistent information.** If fields conflict, the calendar * will give preference to fields set more recently. For example, when * determining the day, the calendar will look for one of the following * combinations of fields. The most recent combination, as determined by the * most recently set single field, will be used. * - * \htmlonly<blockquote>\endhtmlonly - * <pre> - * MONTH + DAY_OF_MONTH - * MONTH + WEEK_OF_MONTH + DAY_OF_WEEK - * MONTH + DAY_OF_WEEK_IN_MONTH + DAY_OF_WEEK - * DAY_OF_YEAR - * DAY_OF_WEEK + WEEK_OF_YEAR - * </pre> - * \htmlonly</blockquote>\endhtmlonly + * MONTH + DAY_OF_MONTH + * MONTH + WEEK_OF_MONTH + DAY_OF_WEEK + * MONTH + DAY_OF_WEEK_IN_MONTH + DAY_OF_WEEK + * DAY_OF_YEAR + * DAY_OF_WEEK + WEEK_OF_YEAR * * For the time of day: * - * \htmlonly<blockquote>\endhtmlonly - * <pre> - * HOUR_OF_DAY - * AM_PM + HOUR - * </pre> - * \htmlonly</blockquote>\endhtmlonly + * HOUR_OF_DAY + * AM_PM + HOUR * - * <p> - * <strong>Ambiguous Wall Clock Time.</strong> When time offset from UTC has + * **Ambiguous Wall Clock Time.** When time offset from UTC has * changed, it produces an ambiguous time slot around the transition. For example, * many US locations observe daylight saving time. On the date switching to daylight * saving time in US, wall clock time jumps from 12:59 AM (standard) to 2:00 AM @@ -138,64 +119,66 @@ class BasicTimeZone; * In this example, 1:30 AM is interpreted as 1:30 AM standard time (non-exist), * so the final result will be 2:30 AM daylight time. * - * <p>On the date switching back to standard time, wall clock time is moved back one + * On the date switching back to standard time, wall clock time is moved back one * hour at 2:00 AM. So wall clock time from 1:00 AM to 1:59 AM occur twice. In this * case, the ICU Calendar resolves the time using the UTC offset after the transition * by default. For example, 1:30 AM on the date is resolved as 1:30 AM standard time. * - * <p>Ambiguous wall clock time resolution behaviors can be customized by Calendar APIs + * Ambiguous wall clock time resolution behaviors can be customized by Calendar APIs * {@link #setRepeatedWallTimeOption} and {@link #setSkippedWallTimeOption}. * These methods are available in ICU 49 or later versions. * - * <p> - * <strong>Note:</strong> for some non-Gregorian calendars, different + * **Note:** for some non-Gregorian calendars, different * fields may be necessary for complete disambiguation. For example, a full - * specification of the historial Arabic astronomical calendar requires year, - * month, day-of-month <em>and</em> day-of-week in some cases. + * specification of the historical Arabic astronomical calendar requires year, + * month, day-of-month *and* day-of-week in some cases. * - * <p> - * <strong>Note:</strong> There are certain possible ambiguities in + * **Note:** There are certain possible ambiguities in * interpretation of certain singular times, which are resolved in the * following ways: - * <ol> - * <li> 24:00:00 "belongs" to the following day. That is, - * 23:59 on Dec 31, 1969 < 24:00 on Jan 1, 1970 < 24:01:00 on Jan 1, 1970 * - * <li> Although historically not precise, midnight also belongs to "am", - * and noon belongs to "pm", so on the same day, - * 12:00 am (midnight) < 12:01 am, and 12:00 pm (noon) < 12:01 pm - * </ol> + * 1. 24:00:00 "belongs" to the following day. That is, + * 23:59 on Dec 31, 1969 < 24:00 on Jan 1, 1970 < 24:01:00 on Jan 1, 1970 + * 2. Although historically not precise, midnight also belongs to "am", + * and noon belongs to "pm", so on the same day, + * 12:00 am (midnight) < 12:01 am, and 12:00 pm (noon) < 12:01 pm * - * <p> * The date or time format strings are not part of the definition of a * calendar, as those must be modifiable or overridable by the user at - * runtime. Use {@link DateFormat} - * to format dates. + * runtime. Use `DateFormat` to format dates. * - * <p> - * <code>Calendar</code> provides an API for field "rolling", where fields + * `Calendar` provides an API for field "rolling", where fields * can be incremented or decremented, but wrap around. For example, rolling the - * month up in the date <code>December 12, <b>1996</b></code> results in - * <code>January 12, <b>1996</b></code>. + * month up in the date December 12, **1996** results in + * January 12, **1996**. * - * <p> - * <code>Calendar</code> also provides a date arithmetic function for + * `Calendar` also provides a date arithmetic function for * adding the specified (signed) amount of time to a particular time field. - * For example, subtracting 5 days from the date <code>September 12, 1996</code> - * results in <code>September 7, 1996</code>. + * For example, subtracting 5 days from the date `September 12, 1996` + * results in `September 7, 1996`. * - * <p><big><b>Supported range</b></big> + * ***Supported range*** * - * <p>The allowable range of <code>Calendar</code> has been - * narrowed. <code>GregorianCalendar</code> used to attempt to support - * the range of dates with millisecond values from - * <code>Long.MIN_VALUE</code> to <code>Long.MAX_VALUE</code>. - * The new <code>Calendar</code> protocol specifies the + * The allowable range of `Calendar` has been narrowed. `GregorianCalendar` used + * to attempt to support the range of dates with millisecond values from + * `Long.MIN_VALUE` to `Long.MAX_VALUE`. The new `Calendar` protocol specifies the * maximum range of supportable dates as those having Julian day numbers - * of <code>-0x7F000000</code> to <code>+0x7F000000</code>. This - * corresponds to years from ~5,800,000 BCE to ~5,800,000 CE. Programmers - * should use the protected constants in <code>Calendar</code> to - * specify an extremely early or extremely late date.</p> + * of `-0x7F000000` to `+0x7F000000`. This corresponds to years from ~5,800,000 BCE + * to ~5,800,000 CE. Programmers should use the protected constants in `Calendar` to + * specify an extremely early or extremely late date. + * + * <p> + * The Japanese calendar uses a combination of era name and year number. + * When an emperor of Japan abdicates and a new emperor ascends the throne, + * a new era is declared and year number is reset to 1. Even if the date of + * abdication is scheduled ahead of time, the new era name might not be + * announced until just before the date. In such case, ICU4C may include + * a start date of future era without actual era name, but not enabled + * by default. ICU4C users who want to test the behavior of the future era + * can enable the tentative era by: + * <ul> + * <li>Environment variable <code>ICU_ENABLE_TENTATIVE_ERA=true</code>.</li> + * </ul> * * @stable ICU 2.0 */ @@ -903,7 +886,7 @@ public: /** * Sets the behavior for handling wall time repeating multiple times * at negative time zone offset transitions. For example, 1:30 AM on - * November 6, 2011 in US Eastern time (Ameirca/New_York) occurs twice; + * November 6, 2011 in US Eastern time (America/New_York) occurs twice; * 1:30 AM EDT, then 1:30 AM EST one hour later. When <code>UCAL_WALLTIME_FIRST</code> * is used, the wall time 1:30AM in this example will be interpreted as 1:30 AM EDT * (first occurrence). When <code>UCAL_WALLTIME_LAST</code> is used, it will be @@ -1718,9 +1701,7 @@ protected: /** * Validate a single field of this calendar. Subclasses should * override this method to validate any calendar-specific fields. - * Generic fields can be handled by - * <code>Calendar::validateField()</code>. - * @see #validateField(int, int, int, int&) + * Generic fields can be handled by `Calendar::validateField()`. * @internal */ virtual void validateField(UCalendarDateFields field, UErrorCode &status); @@ -2171,7 +2152,7 @@ private: TimeZone* fZone; /** - * Option for rpeated wall time + * Option for repeated wall time * @see #setRepeatedWallTimeOption */ UCalendarWallTimeOption fRepeatedWallTime; @@ -2456,7 +2437,7 @@ private: BasicTimeZone* getBasicTimeZone() const; /** - * Find the previous zone transtion near the given time. + * Find the previous zone transition near the given time. * @param base The base time, inclusive * @param transitionTime Receives the result time * @param status The error status diff --git a/deps/icu-small/source/i18n/unicode/coll.h b/deps/icu-small/source/i18n/unicode/coll.h index d03570509e..653434f54c 100644 --- a/deps/icu-small/source/i18n/unicode/coll.h +++ b/deps/icu-small/source/i18n/unicode/coll.h @@ -235,16 +235,16 @@ public: * Returns TRUE if "other" is the same as "this". * * The base class implementation returns TRUE if "other" has the same type/class as "this": - * <code>typeid(*this) == typeid(other)</code>. + * `typeid(*this) == typeid(other)`. * * Subclass implementations should do something like the following: - * <pre> - * if (this == &other) { return TRUE; } - * if (!Collator::operator==(other)) { return FALSE; } // not the same class * - * const MyCollator &o = (const MyCollator&)other; - * (compare this vs. o's subclass fields) - * </pre> + * if (this == &other) { return TRUE; } + * if (!Collator::operator==(other)) { return FALSE; } // not the same class + * + * const MyCollator &o = (const MyCollator&)other; + * (compare this vs. o's subclass fields) + * * @param other Collator object to be compared * @return TRUE if other is the same as this. * @stable ICU 2.0 diff --git a/deps/icu-small/source/i18n/unicode/compactdecimalformat.h b/deps/icu-small/source/i18n/unicode/compactdecimalformat.h index 7dc92f6100..9c1e9183f4 100644 --- a/deps/icu-small/source/i18n/unicode/compactdecimalformat.h +++ b/deps/icu-small/source/i18n/unicode/compactdecimalformat.h @@ -30,30 +30,31 @@ U_NAMESPACE_BEGIN class PluralRules; /** - * <p><strong>IMPORTANT:</strong> New users are strongly encouraged to see if + * **IMPORTANT:** New users are strongly encouraged to see if * numberformatter.h fits their use case. Although not deprecated, this header * is provided for backwards compatibility only. - * <hr/> + * + * ----------------------------------------------------------------------------- * * The CompactDecimalFormat produces abbreviated numbers, suitable for display in * environments will limited real estate. For example, 'Hits: 1.2B' instead of * 'Hits: 1,200,000,000'. The format will be appropriate for the given language, * such as "1,2 Mrd." for German. - * <p> + * * For numbers under 1000 trillion (under 10^15, such as 123,456,789,012,345), * the result will be short for supported languages. However, the result may * sometimes exceed 7 characters, such as when there are combining marks or thin * characters. In such cases, the visual width in fonts should still be short. - * <p> + * * By default, there are 3 significant digits. After creation, if more than * three significant digits are set (with setMaximumSignificantDigits), or if a * fixed number of digits are set (with setMaximumIntegerDigits or * setMaximumFractionDigits), then result may be wider. - * <p> + * * At this time, parsing is not supported, and will produce a U_UNSUPPORTED_ERROR. * Resetting the pattern prefixes or suffixes is not supported; the method calls * are ignored. - * <p> + * * @stable ICU 51 */ class U_I18N_API CompactDecimalFormat : public DecimalFormat { @@ -61,9 +62,9 @@ public: /** * Returns a compact decimal instance for specified locale. - * <p> - * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of NumberFormat. + * + * **NOTE:** New users are strongly encouraged to use + * `number::NumberFormatter` instead of NumberFormat. * @param inLocale the given locale. * @param style whether to use short or long style. * @param status error code returned here. diff --git a/deps/icu-small/source/i18n/unicode/currpinf.h b/deps/icu-small/source/i18n/unicode/currpinf.h index 1a327c5bae..80b0462513 100644 --- a/deps/icu-small/source/i18n/unicode/currpinf.h +++ b/deps/icu-small/source/i18n/unicode/currpinf.h @@ -2,7 +2,7 @@ // License & terms of use: http://www.unicode.org/copyright.html /* ******************************************************************************* - * Copyright (C) 2009-2015, International Business Machines Corporation and * + * Copyright (C) 2009-2015, International Business Machines Corporation and * * others. All Rights Reserved. * ******************************************************************************* */ @@ -240,18 +240,27 @@ private: /* * The plural rule is used to format currency plural name, * for example: "3.00 US Dollars". - * If there are 3 currency signs in the currency patttern, + * If there are 3 currency signs in the currency pattern, * the 3 currency signs will be replaced by currency plural name. */ PluralRules* fPluralRules; // locale Locale* fLocale; + +private: + /** + * An internal status variable used to indicate that the object is in an 'invalid' state. + * Used by copy constructor, the assignment operator and the clone method. + */ + UErrorCode fInternalStatus; }; inline UBool -CurrencyPluralInfo::operator!=(const CurrencyPluralInfo& info) const { return !operator==(info); } +CurrencyPluralInfo::operator!=(const CurrencyPluralInfo& info) const { + return !operator==(info); +} U_NAMESPACE_END diff --git a/deps/icu-small/source/i18n/unicode/currunit.h b/deps/icu-small/source/i18n/unicode/currunit.h index d5bc4aa6d6..48cadc10b7 100644 --- a/deps/icu-small/source/i18n/unicode/currunit.h +++ b/deps/icu-small/source/i18n/unicode/currunit.h @@ -38,7 +38,7 @@ class U_I18N_API CurrencyUnit: public MeasureUnit { public: /** * Default constructor. Initializes currency code to "XXX" (no currency). - * @draft ICU 60 + * @stable ICU 60 */ CurrencyUnit(); @@ -59,17 +59,15 @@ class U_I18N_API CurrencyUnit: public MeasureUnit { */ CurrencyUnit(const CurrencyUnit& other); -#ifndef U_HIDE_DRAFT_API /** * Copy constructor from MeasureUnit. This constructor allows you to * restore a CurrencyUnit that was sliced to MeasureUnit. * * @param measureUnit The MeasureUnit to copy from. * @param ec Set to a failing value if the MeasureUnit is not a currency. - * @draft ICU 60 + * @stable ICU 60 */ CurrencyUnit(const MeasureUnit& measureUnit, UErrorCode &ec); -#endif /* U_HIDE_DRAFT_API */ /** * Assignment operator diff --git a/deps/icu-small/source/i18n/unicode/datefmt.h b/deps/icu-small/source/i18n/unicode/datefmt.h index c895183931..13c63d9376 100644 --- a/deps/icu-small/source/i18n/unicode/datefmt.h +++ b/deps/icu-small/source/i18n/unicode/datefmt.h @@ -43,13 +43,17 @@ U_NAMESPACE_BEGIN class TimeZone; class DateTimePatternGenerator; -// explicit template instantiation. see digitlst.h -// (When building DLLs for Windows this is required.) -#if U_PF_WINDOWS <= U_PLATFORM && U_PLATFORM <= U_PF_CYGWIN +/** + * \cond + * Export an explicit template instantiation. (See digitlst.h, datefmt.h, and others.) + * (When building DLLs for Windows this is required.) + */ +#if U_PF_WINDOWS <= U_PLATFORM && U_PLATFORM <= U_PF_CYGWIN && !defined(U_IN_DOXYGEN) template class U_I18N_API EnumSet<UDateFormatBooleanAttribute, 0, UDAT_BOOLEAN_ATTRIBUTE_COUNT>; #endif +/** \endcond */ /** * DateFormat is an abstract class for a family of classes that convert dates and diff --git a/deps/icu-small/source/i18n/unicode/dcfmtsym.h b/deps/icu-small/source/i18n/unicode/dcfmtsym.h index 2f824cec30..55e3d8a6b3 100644 --- a/deps/icu-small/source/i18n/unicode/dcfmtsym.h +++ b/deps/icu-small/source/i18n/unicode/dcfmtsym.h @@ -181,7 +181,6 @@ public: */ DecimalFormatSymbols(const Locale& locale, UErrorCode& status); -#ifndef U_HIDE_DRAFT_API /** * Creates a DecimalFormatSymbols instance for the given locale with digits and symbols * corresponding to the given NumberingSystem. @@ -196,10 +195,9 @@ public: * @param ns The numbering system. * @param status Input/output parameter, set to success or * failure code upon return. - * @draft ICU 60 + * @stable ICU 60 */ DecimalFormatSymbols(const Locale& locale, const NumberingSystem& ns, UErrorCode& status); -#endif /* U_HIDE_DRAFT_API */ /** * Create a DecimalFormatSymbols object for the default locale. @@ -406,7 +404,7 @@ public: * returning a const reference to one of the symbol strings. * The returned reference becomes invalid when the symbol is changed * or when the DecimalFormatSymbols are destroyed. - * Note: moved #ifndef U_HIDE_INTERNAL_API after this, since this is needed for inline in DecimalFormat + * Note: moved \#ifndef U_HIDE_INTERNAL_API after this, since this is needed for inline in DecimalFormat * * This is not currently stable API, but if you think it should be stable, * post a comment on the following ticket and the ICU team will take a look: @@ -531,7 +529,7 @@ inline const UnicodeString& DecimalFormatSymbols::getConstDigitSymbol(int32_t di ENumberFormatSymbol key = static_cast<ENumberFormatSymbol>(kOneDigitSymbol + digit - 1); return fSymbols[key]; } -#endif +#endif /* U_HIDE_INTERNAL_API */ // ------------------------------------- diff --git a/deps/icu-small/source/i18n/unicode/decimfmt.h b/deps/icu-small/source/i18n/unicode/decimfmt.h index 3747f510f7..b3a5cc0495 100644 --- a/deps/icu-small/source/i18n/unicode/decimfmt.h +++ b/deps/icu-small/source/i18n/unicode/decimfmt.h @@ -63,19 +63,22 @@ class NumberParserImpl; } } -// explicit template instantiation. see digitlst.h -// (When building DLLs for Windows this is required.) -#if U_PF_WINDOWS <= U_PLATFORM && U_PLATFORM <= U_PF_CYGWIN +/** + * \cond + * explicit template instantiation. see digitlst.h + * (When building DLLs for Windows this is required.) + */ +#if U_PF_WINDOWS <= U_PLATFORM && U_PLATFORM <= U_PF_CYGWIN && !defined(U_IN_DOXYGEN) template class U_I18N_API EnumSet<UNumberFormatAttribute, UNUM_MAX_NONBOOLEAN_ATTRIBUTE+1, UNUM_LIMIT_BOOLEAN_ATTRIBUTE>; #endif +/** \endcond */ /** - * <p><strong>IMPORTANT:</strong> New users are strongly encouraged to see if + * **IMPORTANT:** New users are strongly encouraged to see if * numberformatter.h fits their use case. Although not deprecated, this header * is provided for backwards compatibility only. - * <hr/> * * DecimalFormat is a concrete subclass of NumberFormat that formats decimal * numbers. It has a variety of features designed to make it possible to parse @@ -85,13 +88,13 @@ template class U_I18N_API EnumSet<UNumberFormatAttribute, * ("1.23E4"), percentages ("12%"), and currency amounts ("$123", "USD123", * "123 US dollars"). All of these flavors can be easily localized. * - * <p>To obtain a NumberFormat for a specific locale (including the default + * To obtain a NumberFormat for a specific locale (including the default * locale) call one of NumberFormat's factory methods such as * createInstance(). Do not call the DecimalFormat constructors directly, unless * you know what you are doing, since the NumberFormat factory methods may * return subclasses other than DecimalFormat. * - * <p><strong>Example Usage</strong> + * **Example Usage** * * \code * // Normally we would have a GUI with a menu for this @@ -135,11 +138,11 @@ template class U_I18N_API EnumSet<UNumberFormatAttribute, * } * } * \endcode - * <P> - * Another example use createInstance(style) - * <P> - * <pre> - * <strong>// Print out a number using the localized number, currency, + * + * **Another example use createInstance(style)** + * + * \code + * // Print out a number using the localized number, currency, * // percent, scientific, integer, iso currency, and plural currency * // format for each locale</strong> * Locale* locale = new Locale("en", "US"); @@ -150,11 +153,13 @@ template class U_I18N_API EnumSet<UNumberFormatAttribute, * for (int j=NumberFormat::kNumberStyle; * j<=NumberFormat::kPluralCurrencyStyle; * ++j) { - * NumberFormat* format = NumberFormat::createInstance(locale, j, success); + * NumberFormat* form = NumberFormat::createInstance(locale, j, success); * str.remove(); * cout << "format result " << form->format(myNumber, str) << endl; * format->parse(form->format(myNumber, str), fmtable, success); - * }</pre> + * delete form; + * } + * \endcode * * * <p><strong>Patterns</strong> @@ -690,7 +695,7 @@ class U_I18N_API DecimalFormat : public NumberFormat { * locale. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of DecimalFormat. + * #icu::number::NumberFormatter instead of DecimalFormat. * @param status Output param set to success/failure code. If the * pattern is invalid this will be set to a failure code. * @stable ICU 2.0 @@ -708,7 +713,7 @@ class U_I18N_API DecimalFormat : public NumberFormat { * locale. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of DecimalFormat. + * #icu::number::NumberFormatter instead of DecimalFormat. * @param pattern A non-localized pattern string. * @param status Output param set to success/failure code. If the * pattern is invalid this will be set to a failure code. @@ -728,7 +733,7 @@ class U_I18N_API DecimalFormat : public NumberFormat { * a NumberFormat factory method. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of DecimalFormat. + * #icu::number::NumberFormatter instead of DecimalFormat. * * @param pattern a non-localized pattern string * @param symbolsToAdopt the set of symbols to be used. The caller should not @@ -782,7 +787,7 @@ class U_I18N_API DecimalFormat : public NumberFormat { * May return U_UNSUPPORTED_ERROR if this instance does not support * the specified attribute. * @param attr the attribute to set - * @param newvalue new value + * @param newValue new value * @param status the error type * @return *this - for chaining (example: format.setAttribute(...).setAttribute(...) ) * @stable ICU 51 @@ -839,7 +844,7 @@ class U_I18N_API DecimalFormat : public NumberFormat { * a NumberFormat factory method. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of DecimalFormat. + * #icu::number::NumberFormatter instead of DecimalFormat. * * @param pattern a non-localized pattern string * @param symbolsToAdopt the set of symbols to be used. The caller should not @@ -864,7 +869,7 @@ class U_I18N_API DecimalFormat : public NumberFormat { * a NumberFormat factory method. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of DecimalFormat. + * #icu::number::NumberFormatter instead of DecimalFormat. * * @param pattern a non-localized pattern string * @param symbols the set of symbols to be used @@ -986,6 +991,7 @@ class U_I18N_API DecimalFormat : public NumberFormat { * Result is appended to existing contents. * @param pos On input: an alignment field, if desired. * On output: the offsets of the alignment field. + * @param status Output param filled with success/failure status. * @return Reference to 'appendTo' parameter. * @internal */ @@ -1031,6 +1037,7 @@ class U_I18N_API DecimalFormat : public NumberFormat { * Result is appended to existing contents. * @param pos On input: an alignment field, if desired. * On output: the offsets of the alignment field. + * @param status Output param filled with success/failure status. * @return Reference to 'appendTo' parameter. * @internal */ @@ -1726,7 +1733,7 @@ class U_I18N_API DecimalFormat : public NumberFormat { virtual void setDecimalPatternMatchRequired(UBool newValue); /** - * {@icu} Returns whether to ignore exponents when parsing. + * Returns whether to ignore exponents when parsing. * * @see #setParseNoExponent * @internal This API is a technical preview. It may change in an upcoming release. @@ -1734,7 +1741,7 @@ class U_I18N_API DecimalFormat : public NumberFormat { virtual UBool isParseNoExponent() const; /** - * {@icu} Specifies whether to stop parsing when an exponent separator is encountered. For + * Specifies whether to stop parsing when an exponent separator is encountered. For * example, parses "123E4" to 123 (with parse position 3) instead of 1230000 (with parse position * 5). * @@ -1744,7 +1751,7 @@ class U_I18N_API DecimalFormat : public NumberFormat { virtual void setParseNoExponent(UBool value); /** - * {@icu} Returns whether parsing is sensitive to case (lowercase/uppercase). + * Returns whether parsing is sensitive to case (lowercase/uppercase). * * @see #setParseCaseSensitive * @internal This API is a technical preview. It may change in an upcoming release. @@ -1752,7 +1759,7 @@ class U_I18N_API DecimalFormat : public NumberFormat { virtual UBool isParseCaseSensitive() const; /** - * {@icu} Whether to pay attention to case when parsing; default is to ignore case (perform + * Whether to pay attention to case when parsing; default is to ignore case (perform * case-folding). For example, "A" == "a" in case-insensitive but not case-sensitive mode. * * Currency symbols are never case-folded. For example, "us$1.00" will not parse in case-insensitive @@ -1763,7 +1770,7 @@ class U_I18N_API DecimalFormat : public NumberFormat { virtual void setParseCaseSensitive(UBool value); /** - * {@icu} Returns whether truncation of high-order integer digits should result in an error. + * Returns whether truncation of high-order integer digits should result in an error. * By default, setMaximumIntegerDigits truncates high-order digits silently. * * @see setFormatFailIfMoreThanMaxDigits @@ -1772,7 +1779,7 @@ class U_I18N_API DecimalFormat : public NumberFormat { virtual UBool isFormatFailIfMoreThanMaxDigits() const; /** - * {@icu} Sets whether truncation of high-order integer digits should result in an error. + * Sets whether truncation of high-order integer digits should result in an error. * By default, setMaximumIntegerDigits truncates high-order digits silently. * * @internal This API is a technical preview. It may change in an upcoming release. @@ -2017,16 +2024,17 @@ class U_I18N_API DecimalFormat : public NumberFormat { virtual void setCurrency(const char16_t* theCurrency); /** - * Sets the <tt>Currency Context</tt> object used to display currency. + * Sets the `Currency Usage` object used to display currency. * This takes effect immediately, if this format is a * currency format. - * @param currencyContext new currency context object to use. + * @param newUsage new currency usage object to use. + * @param ec input-output error code * @stable ICU 54 */ void setCurrencyUsage(UCurrencyUsage newUsage, UErrorCode* ec); /** - * Returns the <tt>Currency Context</tt> object used to display currency + * Returns the `Currency Usage` object used to display currency * @stable ICU 54 */ UCurrencyUsage getCurrencyUsage() const; @@ -2050,7 +2058,7 @@ class U_I18N_API DecimalFormat : public NumberFormat { void formatToDecimalQuantity(const Formattable& number, number::impl::DecimalQuantity& output, UErrorCode& status) const; -#endif +#endif /* U_HIDE_INTERNAL_API */ #ifndef U_HIDE_DRAFT_API /** @@ -2072,7 +2080,6 @@ class U_I18N_API DecimalFormat : public NumberFormat { * FormattedNumber result = df->toNumberFormatter().formatDouble(123, status); * </pre> * - * @param output The variable into which to store the LocalizedNumberFormatter. * @return The output variable, for chaining. * @draft ICU 62 */ diff --git a/deps/icu-small/source/i18n/unicode/dtptngen.h b/deps/icu-small/source/i18n/unicode/dtptngen.h index feb465e799..26ccc64060 100644 --- a/deps/icu-small/source/i18n/unicode/dtptngen.h +++ b/deps/icu-small/source/i18n/unicode/dtptngen.h @@ -498,27 +498,23 @@ public: private: /** * Constructor. - * @stable ICU 3.8 */ DateTimePatternGenerator(UErrorCode & status); /** * Constructor. - * @stable ICU 3.8 */ DateTimePatternGenerator(const Locale& locale, UErrorCode & status); /** * Copy constructor. * @param other DateTimePatternGenerator to copy - * @stable ICU 3.8 */ DateTimePatternGenerator(const DateTimePatternGenerator& other); /** * Default assignment operator. * @param other DateTimePatternGenerator to copy - * @stable ICU 3.8 */ DateTimePatternGenerator& operator=(const DateTimePatternGenerator& other); @@ -542,6 +538,11 @@ private: int32_t fAllowedHourFormats[7]; // Actually an array of AllowedHourFormat enum type, ending with UNKNOWN. + // Internal error code used for recording/reporting errors that occur during methods that do not + // have a UErrorCode parameter. For example: the Copy Constructor, or the ::clone() method. + // When this is set to an error the object is in an invalid state. + UErrorCode internalErrorCode; + /* internal flags masks for adjustFieldTypes etc. */ enum { kDTPGNoFlags = 0, @@ -569,11 +570,10 @@ private: #endif // U_HIDE_DRAFT_API void getAppendName(UDateTimePatternField field, UnicodeString& value); UnicodeString mapSkeletonMetacharacters(const UnicodeString& patternForm, int32_t* flags, UErrorCode& status); - int32_t getCanonicalIndex(const UnicodeString& field); - const UnicodeString* getBestRaw(DateTimeMatcher& source, int32_t includeMask, DistanceInfo* missingFields, const PtnSkeleton** specifiedSkeletonPtr = 0); + const UnicodeString* getBestRaw(DateTimeMatcher& source, int32_t includeMask, DistanceInfo* missingFields, UErrorCode& status, const PtnSkeleton** specifiedSkeletonPtr = 0); UnicodeString adjustFieldTypes(const UnicodeString& pattern, const PtnSkeleton* specifiedSkeleton, int32_t flags, UDateTimePatternMatchOptions options = UDATPG_MATCH_NO_OPTIONS); - UnicodeString getBestAppending(int32_t missingFields, int32_t flags, UDateTimePatternMatchOptions options = UDATPG_MATCH_NO_OPTIONS); - int32_t getTopBitNumber(int32_t foundMask); + UnicodeString getBestAppending(int32_t missingFields, int32_t flags, UErrorCode& status, UDateTimePatternMatchOptions options = UDATPG_MATCH_NO_OPTIONS); + int32_t getTopBitNumber(int32_t foundMask) const; void setAvailableFormat(const UnicodeString &key, UErrorCode& status); UBool isAvailableFormatSet(const UnicodeString &key) const; void copyHashtable(Hashtable *other, UErrorCode &status); diff --git a/deps/icu-small/source/i18n/unicode/fmtable.h b/deps/icu-small/source/i18n/unicode/fmtable.h index 2359b61d46..a06c23dc3b 100644 --- a/deps/icu-small/source/i18n/unicode/fmtable.h +++ b/deps/icu-small/source/i18n/unicode/fmtable.h @@ -658,7 +658,7 @@ public: /** * Adopt, and set value from, a DecimalQuantity * Internal Function, do not use. - * @param dl the DecimalQuantity to be adopted + * @param dq the DecimalQuantity to be adopted * @internal */ void adoptDecimalQuantity(number::impl::DecimalQuantity *dq); diff --git a/deps/icu-small/source/i18n/unicode/gender.h b/deps/icu-small/source/i18n/unicode/gender.h index 467b64ec5e..b7c31cb554 100644 --- a/deps/icu-small/source/i18n/unicode/gender.h +++ b/deps/icu-small/source/i18n/unicode/gender.h @@ -18,6 +18,11 @@ #ifndef _GENDER #define _GENDER +/** + * \file + * \brief C++ API: GenderInfo computes the gender of a list. + */ + #include "unicode/utypes.h" #if !UCONFIG_NO_FORMATTING @@ -30,7 +35,7 @@ class GenderInfoTest; U_NAMESPACE_BEGIN -// Forward Declaration +/** \internal Forward Declaration */ void U_CALLCONV GenderInfo_initCache(UErrorCode &status); /** diff --git a/deps/icu-small/source/i18n/unicode/listformatter.h b/deps/icu-small/source/i18n/unicode/listformatter.h new file mode 100644 index 0000000000..5e36cf71cc --- /dev/null +++ b/deps/icu-small/source/i18n/unicode/listformatter.h @@ -0,0 +1,203 @@ +// © 2016 and later: Unicode, Inc. and others. +// License & terms of use: http://www.unicode.org/copyright.html +/* +******************************************************************************* +* +* Copyright (C) 2012-2016, International Business Machines +* Corporation and others. All Rights Reserved. +* +******************************************************************************* +* file name: listformatter.h +* encoding: UTF-8 +* tab size: 8 (not used) +* indentation:4 +* +* created on: 20120426 +* created by: Umesh P. Nair +*/ + +#ifndef __LISTFORMATTER_H__ +#define __LISTFORMATTER_H__ + +#include "unicode/utypes.h" + +#include "unicode/unistr.h" +#include "unicode/locid.h" + +U_NAMESPACE_BEGIN + +class FieldPositionIterator; +class FieldPositionHandler; + +/** @internal */ +class Hashtable; + +/** @internal */ +struct ListFormatInternal; + +/* The following can't be #ifndef U_HIDE_INTERNAL_API, needed for other .h file declarations */ +/** + * @internal + * \cond + */ +struct ListFormatData : public UMemory { + UnicodeString twoPattern; + UnicodeString startPattern; + UnicodeString middlePattern; + UnicodeString endPattern; + + ListFormatData(const UnicodeString& two, const UnicodeString& start, const UnicodeString& middle, const UnicodeString& end) : + twoPattern(two), startPattern(start), middlePattern(middle), endPattern(end) {} +}; +/** \endcond */ + + +/** + * \file + * \brief C++ API: API for formatting a list. + */ + + +/** + * An immutable class for formatting a list, using data from CLDR (or supplied + * separately). + * + * Example: Input data ["Alice", "Bob", "Charlie", "Delta"] will be formatted + * as "Alice, Bob, Charlie and Delta" in English. + * + * The ListFormatter class is not intended for public subclassing. + * @stable ICU 50 + */ +class U_I18N_API ListFormatter : public UObject{ + + public: + + /** + * Copy constructor. + * @stable ICU 52 + */ + ListFormatter(const ListFormatter&); + + /** + * Assignment operator. + * @stable ICU 52 + */ + ListFormatter& operator=(const ListFormatter& other); + + /** + * Creates a ListFormatter appropriate for the default locale. + * + * @param errorCode ICU error code, set if no data available for default locale. + * @return Pointer to a ListFormatter object for the default locale, + * created from internal data derived from CLDR data. + * @stable ICU 50 + */ + static ListFormatter* createInstance(UErrorCode& errorCode); + + /** + * Creates a ListFormatter appropriate for a locale. + * + * @param locale The locale. + * @param errorCode ICU error code, set if no data available for the given locale. + * @return A ListFormatter object created from internal data derived from + * CLDR data. + * @stable ICU 50 + */ + static ListFormatter* createInstance(const Locale& locale, UErrorCode& errorCode); + +#ifndef U_HIDE_INTERNAL_API + /** + * Creates a ListFormatter appropriate for a locale and style. + * + * @param locale The locale. + * @param style the style, either "standard", "duration", or "duration-short" + * @param errorCode ICU error code, set if no data available for the given locale. + * @return A ListFormatter object created from internal data derived from + * CLDR data. + * @internal + */ + static ListFormatter* createInstance(const Locale& locale, const char* style, UErrorCode& errorCode); +#endif /* U_HIDE_INTERNAL_API */ + + /** + * Destructor. + * + * @stable ICU 50 + */ + virtual ~ListFormatter(); + + + /** + * Formats a list of strings. + * + * @param items An array of strings to be combined and formatted. + * @param n_items Length of the array items. + * @param appendTo The string to which the result should be appended to. + * @param errorCode ICU error code, set if there is an error. + * @return Formatted string combining the elements of items, appended to appendTo. + * @stable ICU 50 + */ + UnicodeString& format(const UnicodeString items[], int32_t n_items, + UnicodeString& appendTo, UErrorCode& errorCode) const; + +#ifndef U_HIDE_DRAFT_API + /** + * Format a list of strings. + * + * @param items An array of strings to be combined and formatted. + * @param n_items Length of the array items. + * @param appendTo The string to which the formatted result will be + * appended. + * @param posIter On return, can be used to iterate over positions of + * fields generated by this format call. Field values are + * defined in UListFormatterField. Can be NULL. + * @param errorCode ICU error code returned here. + * @return Formatted string combining the elements of items, + * appended to appendTo. + * @draft ICU 63 + */ + UnicodeString& format(const UnicodeString items[], int32_t n_items, + UnicodeString & appendTo, FieldPositionIterator* posIter, + UErrorCode& errorCode) const; +#endif /* U_HIDE_DRAFT_API */ + +#ifndef U_HIDE_INTERNAL_API + /** + @internal for MeasureFormat + */ + UnicodeString& format( + const UnicodeString items[], + int32_t n_items, + UnicodeString& appendTo, + int32_t index, + int32_t &offset, + UErrorCode& errorCode) const; + /** + * @internal constructor made public for testing. + */ + ListFormatter(const ListFormatData &data, UErrorCode &errorCode); + /** + * @internal constructor made public for testing. + */ + ListFormatter(const ListFormatInternal* listFormatterInternal); +#endif /* U_HIDE_INTERNAL_API */ + + private: + static void initializeHash(UErrorCode& errorCode); + static const ListFormatInternal* getListFormatInternal(const Locale& locale, const char *style, UErrorCode& errorCode); + struct ListPatternsSink; + static ListFormatInternal* loadListFormatInternal(const Locale& locale, const char* style, UErrorCode& errorCode); + + UnicodeString& format_( + const UnicodeString items[], int32_t n_items, UnicodeString& appendTo, + int32_t index, int32_t &offset, FieldPositionHandler* handler, UErrorCode& errorCode) const; + + ListFormatter(); + + ListFormatInternal* owned; + const ListFormatInternal* data; +}; + +U_NAMESPACE_END + +#endif diff --git a/deps/icu-small/source/i18n/unicode/measfmt.h b/deps/icu-small/source/i18n/unicode/measfmt.h index 14399dd59a..bbdd2364bd 100644 --- a/deps/icu-small/source/i18n/unicode/measfmt.h +++ b/deps/icu-small/source/i18n/unicode/measfmt.h @@ -104,7 +104,7 @@ class U_I18N_API MeasureFormat : public Format { * Constructor. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of NumberFormat. + * {@link icu::number::NumberFormatter} instead of NumberFormat. * @stable ICU 53 */ MeasureFormat( @@ -114,7 +114,7 @@ class U_I18N_API MeasureFormat : public Format { * Constructor. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of NumberFormat. + * {@link icu::number::NumberFormatter} instead of NumberFormat. * @stable ICU 53 */ MeasureFormat( @@ -202,7 +202,7 @@ class U_I18N_API MeasureFormat : public Format { * formatted string is 3.5 meters per second. * @param measure The measure object. In above example, 3.5 meters. * @param perUnit The per unit. In above example, it is - * *MeasureUnit::createSecond(status). + * `*%MeasureUnit::createSecond(status)`. * @param appendTo formatted string appended here. * @param pos the field position. * @param status the error. @@ -223,7 +223,7 @@ class U_I18N_API MeasureFormat : public Format { * @param unit The unit for which to get a display name. * @param status the error. * @return The display name in the locale and width specified in - * {@link MeasureFormat#getInstance}, or null if there is no display name available + * the MeasureFormat constructor, or null if there is no display name available * for the specified unit. * * @stable ICU 58 @@ -236,7 +236,7 @@ class U_I18N_API MeasureFormat : public Format { * locale. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of NumberFormat. + * {@link icu::number::NumberFormatter} instead of NumberFormat. * @param locale desired locale * @param ec input-output error code * @return a formatter object, or NULL upon error @@ -250,7 +250,7 @@ class U_I18N_API MeasureFormat : public Format { * locale. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of NumberFormat. + * {@link icu::number::NumberFormatter} instead of NumberFormat. * @param ec input-output error code * @return a formatter object, or NULL upon error * @stable ICU 3.0 @@ -348,7 +348,7 @@ class U_I18N_API MeasureFormat : public Format { const MeasureFormatCacheData *cache; const SharedNumberFormat *numberFormat; const SharedPluralRules *pluralRules; - UMeasureFormatWidth width; + UMeasureFormatWidth fWidth; // Declared outside of MeasureFormatSharedData because ListFormatter // objects are relatively cheap to copy; therefore, they don't need to be diff --git a/deps/icu-small/source/i18n/unicode/measunit.h b/deps/icu-small/source/i18n/unicode/measunit.h index f552253544..676fdeb9c8 100644 --- a/deps/icu-small/source/i18n/unicode/measunit.h +++ b/deps/icu-small/source/i18n/unicode/measunit.h @@ -368,6 +368,26 @@ class U_I18N_API MeasureUnit: public UObject { */ static MeasureUnit *createPartPerMillion(UErrorCode &status); +#ifndef U_HIDE_DRAFT_API + /** + * Returns unit of concentr: percent. + * Caller owns returned value and must free it. + * @param status ICU error code. + * @draft ICU 63 + */ + static MeasureUnit *createPercent(UErrorCode &status); +#endif /* U_HIDE_DRAFT_API */ + +#ifndef U_HIDE_DRAFT_API + /** + * Returns unit of concentr: permille. + * Caller owns returned value and must free it. + * @param status ICU error code. + * @draft ICU 63 + */ + static MeasureUnit *createPermille(UErrorCode &status); +#endif /* U_HIDE_DRAFT_API */ + /** * Returns unit of consumption: liter-per-100kilometers. * Caller owns returned value and must free it. @@ -464,6 +484,16 @@ class U_I18N_API MeasureUnit: public UObject { */ static MeasureUnit *createMegabyte(UErrorCode &status); +#ifndef U_HIDE_DRAFT_API + /** + * Returns unit of digital: petabyte. + * Caller owns returned value and must free it. + * @param status ICU error code. + * @draft ICU 63 + */ + static MeasureUnit *createPetabyte(UErrorCode &status); +#endif /* U_HIDE_DRAFT_API */ + /** * Returns unit of digital: terabit. * Caller owns returned value and must free it. @@ -984,6 +1014,16 @@ class U_I18N_API MeasureUnit: public UObject { */ static MeasureUnit *createWatt(UErrorCode &status); +#ifndef U_HIDE_DRAFT_API + /** + * Returns unit of pressure: atmosphere. + * Caller owns returned value and must free it. + * @param status ICU error code. + * @draft ICU 63 + */ + static MeasureUnit *createAtmosphere(UErrorCode &status); +#endif /* U_HIDE_DRAFT_API */ + /** * Returns unit of pressure: hectopascal. * Caller owns returned value and must free it. diff --git a/deps/icu-small/source/i18n/unicode/msgfmt.h b/deps/icu-small/source/i18n/unicode/msgfmt.h index fef8010774..074d933540 100644 --- a/deps/icu-small/source/i18n/unicode/msgfmt.h +++ b/deps/icu-small/source/i18n/unicode/msgfmt.h @@ -124,7 +124,7 @@ class NumberFormat; * argNumber = '0' | ('1'..'9' ('0'..'9')*) * * argType = "number" | "date" | "time" | "spellout" | "ordinal" | "duration" - * argStyle = "short" | "medium" | "long" | "full" | "integer" | "currency" | "percent" | argStyleText + * argStyle = "short" | "medium" | "long" | "full" | "integer" | "currency" | "percent" | argStyleText | "::" argSkeletonText * </pre> * * <ul> @@ -166,7 +166,7 @@ class NumberFormat; * <td colspan=2><i>(none)</i> * <td><code>null</code> * <tr> - * <td rowspan=5><code>number</code> + * <td rowspan=6><code>number</code> * <td><i>(none)</i> * <td><code>NumberFormat.createInstance(getLocale(), status)</code> * <tr> @@ -182,6 +182,9 @@ class NumberFormat; * <td><i>argStyleText</i> * <td><code>new DecimalFormat(argStyleText, new DecimalFormatSymbols(getLocale(), status), status)</code> * <tr> + * <td><i>argSkeletonText</i> + * <td><code>NumberFormatter::forSkeleton(argSkeletonText, status).locale(getLocale()).toFormat(status)</code> + * <tr> * <td rowspan=6><code>date</code> * <td><i>(none)</i> * <td><code>DateFormat.createDateInstance(kDefault, getLocale(), status)</code> @@ -199,7 +202,7 @@ class NumberFormat; * <td><code>DateFormat.createDateInstance(kFull, getLocale(), status)</code> * <tr> * <td><i>argStyleText</i> - * <td><code>new SimpleDateFormat(argStyleText, getLocale(), status) + * <td><code>new SimpleDateFormat(argStyleText, getLocale(), status)</code> * <tr> * <td rowspan=6><code>time</code> * <td><i>(none)</i> @@ -218,7 +221,7 @@ class NumberFormat; * <td><code>DateFormat.createTimeInstance(kFull, getLocale(), status)</code> * <tr> * <td><i>argStyleText</i> - * <td><code>new SimpleDateFormat(argStyleText, getLocale(), status) + * <td><code>new SimpleDateFormat(argStyleText, getLocale(), status)</code> * <tr> * <td><code>spellout</code> * <td><i>argStyleText (optional)</i> diff --git a/deps/icu-small/source/i18n/unicode/numberformatter.h b/deps/icu-small/source/i18n/unicode/numberformatter.h index 3ab08319f7..469949a287 100644 --- a/deps/icu-small/source/i18n/unicode/numberformatter.h +++ b/deps/icu-small/source/i18n/unicode/numberformatter.h @@ -144,11 +144,31 @@ class MultiplierFormatHandler; class CurrencySymbols; class GeneratorHelpers; class DecNum; +class NumberRangeFormatterImpl; +struct RangeMacroProps; + +/** + * Used for NumberRangeFormatter and implemented in numrange_fluent.cpp. + * Declared here so it can be friended. + * + * @internal + */ +void touchRangeLocales(impl::RangeMacroProps& macros); } // namespace impl -// Reserve extra names in case they are added as classes in the future: +/** + * Extra name reserved in case it is needed in the future. + * + * @draft ICU 63 + */ typedef Notation CompactNotation; + +/** + * Extra name reserved in case it is needed in the future. + * + * @draft ICU 63 + */ typedef Notation SimpleNotation; /** @@ -308,10 +328,15 @@ class U_I18N_API Notation : public UMemory { union NotationUnion { // For NTN_SCIENTIFIC + /** @internal */ struct ScientificSettings { + /** @internal */ int8_t fEngineeringInterval; + /** @internal */ bool fRequireMinInt; + /** @internal */ impl::digits_t fMinExponentDigits; + /** @internal */ UNumberSignDisplay fExponentSignDisplay; } scientific; @@ -407,15 +432,39 @@ class U_I18N_API ScientificNotation : public Notation { friend class impl::NumberPropertyMapper; }; -// Reserve extra names in case they are added as classes in the future: +/** + * Extra name reserved in case it is needed in the future. + * + * @draft ICU 63 + */ typedef Precision SignificantDigitsPrecision; // Typedefs for ICU 60/61 compatibility. // These will be removed in ICU 64. // See http://bugs.icu-project.org/trac/ticket/13746 + +/** + * This will be removed in ICU 64. See ICU-13746. + * @deprecated ICU 63 + */ typedef Precision Rounder; + +/** + * This will be removed in ICU 64. See ICU-13746. + * @deprecated ICU 63 + */ typedef FractionPrecision FractionRounder; + +/** + * This will be removed in ICU 64. See ICU-13746. + * @deprecated ICU 63 + */ typedef IncrementPrecision IncrementRounder; + +/** + * This will be removed in ICU 64. See ICU-13746. + * @deprecated ICU 63 + */ typedef CurrencyPrecision CurrencyRounder; /** @@ -672,16 +721,25 @@ class U_I18N_API Precision : public UMemory { } fType; union PrecisionUnion { + /** @internal */ struct FractionSignificantSettings { // For RND_FRACTION, RND_SIGNIFICANT, and RND_FRACTION_SIGNIFICANT + /** @internal */ impl::digits_t fMinFrac; + /** @internal */ impl::digits_t fMaxFrac; + /** @internal */ impl::digits_t fMinSig; + /** @internal */ impl::digits_t fMaxSig; } fracSig; + /** @internal */ struct IncrementSettings { + /** @internal */ double fIncrement; + /** @internal */ impl::digits_t fMinFrac; + /** @internal */ impl::digits_t fMaxFrac; } increment; // For RND_INCREMENT UCurrencyUsage currencyUsage; // For RND_CURRENCY @@ -1205,7 +1263,7 @@ class U_I18N_API Grouper : public UMemory { public: #ifndef U_HIDE_INTERNAL_API /** @internal */ - static Grouper forStrategy(UGroupingStrategy grouping); + static Grouper forStrategy(UNumberGroupingStrategy grouping); /** * Resolve the values in Properties to a Grouper object. @@ -1216,7 +1274,7 @@ class U_I18N_API Grouper : public UMemory { // Future: static Grouper forProperties(DecimalFormatProperties& properties); /** @internal */ - Grouper(int16_t grouping1, int16_t grouping2, int16_t minGrouping, UGroupingStrategy strategy) + Grouper(int16_t grouping1, int16_t grouping2, int16_t minGrouping, UNumberGroupingStrategy strategy) : fGrouping1(grouping1), fGrouping2(grouping2), fMinGrouping(minGrouping), @@ -1251,10 +1309,10 @@ class U_I18N_API Grouper : public UMemory { int16_t fMinGrouping; /** - * The UGroupingStrategy that was used to create this Grouper, or UNUM_GROUPING_COUNT if this - * was not created from a UGroupingStrategy. + * The UNumberGroupingStrategy that was used to create this Grouper, or UNUM_GROUPING_COUNT if this + * was not created from a UNumberGroupingStrategy. */ - UGroupingStrategy fStrategy; + UNumberGroupingStrategy fStrategy; Grouper() : fGrouping1(-3) {}; @@ -1423,7 +1481,8 @@ struct U_I18N_API MacroProps : public UMemory { /** * An abstract base class for specifying settings related to number formatting. This class is implemented by - * {@link UnlocalizedNumberFormatter} and {@link LocalizedNumberFormatter}. + * {@link UnlocalizedNumberFormatter} and {@link LocalizedNumberFormatter}. This class is not intended for + * public subclassing. */ template<typename Derived> class U_I18N_API NumberFormatterSettings { @@ -1710,7 +1769,7 @@ class U_I18N_API NumberFormatterSettings { * The exact grouping widths will be chosen based on the locale. * * <p> - * Pass this method an element from the {@link UGroupingStrategy} enum. For example: + * Pass this method an element from the {@link UNumberGroupingStrategy} enum. For example: * * <pre> * NumberFormatter::with().grouping(UNUM_GROUPING_MIN2) @@ -1724,7 +1783,7 @@ class U_I18N_API NumberFormatterSettings { * @return The fluent chain. * @draft ICU 61 */ - Derived grouping(UGroupingStrategy strategy) const &; + Derived grouping(UNumberGroupingStrategy strategy) const &; /** * Overload of grouping() for use on an rvalue reference. @@ -1733,10 +1792,9 @@ class U_I18N_API NumberFormatterSettings { * The grouping strategy to use. * @return The fluent chain. * @see #grouping - * @provisional This API might change or be removed in a future release. * @draft ICU 62 */ - Derived grouping(UGroupingStrategy strategy) &&; + Derived grouping(UNumberGroupingStrategy strategy) &&; /** * Specifies the minimum and maximum number of digits to render before the decimal mark. @@ -1748,7 +1806,7 @@ class U_I18N_API NumberFormatterSettings { * </ul> * * <p> - * Pass this method the return value of {@link IntegerWidth#zeroFillTo(int)}. For example: + * Pass this method the return value of {@link IntegerWidth#zeroFillTo}. For example: * * <pre> * NumberFormatter::with().integerWidth(IntegerWidth::zeroFillTo(2)) @@ -2099,15 +2157,18 @@ class U_I18N_API NumberFormatterSettings { // NOTE: Uses default copy and move constructors. - protected: + private: impl::MacroProps fMacros; - private: // Don't construct me directly! Use (Un)LocalizedNumberFormatter. NumberFormatterSettings() = default; friend class LocalizedNumberFormatter; friend class UnlocalizedNumberFormatter; + + // Give NumberRangeFormatter access to the MacroProps + friend void impl::touchRangeLocales(impl::RangeMacroProps& macros); + friend class impl::NumberRangeFormatterImpl; }; /** @@ -2124,13 +2185,6 @@ class U_I18N_API UnlocalizedNumberFormatter * Associate the given locale with the number formatter. The locale is used for picking the appropriate symbols, * formats, and other data for number display. * - * <p> - * To use the Java default locale, call Locale::getDefault(): - * - * <pre> - * NumberFormatter::with(). ... .locale(Locale::getDefault()) - * </pre> - * * @param locale * The locale to use when loading data for number formatting. * @return The fluent chain. @@ -2156,7 +2210,6 @@ class U_I18N_API UnlocalizedNumberFormatter */ UnlocalizedNumberFormatter() = default; - // Make default copy constructor call the NumberFormatterSettings copy constructor. /** * Returns a copy of this UnlocalizedNumberFormatter. * @draft ICU 60 @@ -2271,7 +2324,7 @@ class U_I18N_API LocalizedNumberFormatter */ int32_t getCallCount() const; -#endif +#endif /* U_HIDE_INTERNAL_API */ /** * Creates a representation of this LocalizedNumberFormat as an icu::Format, enabling the use @@ -2295,7 +2348,6 @@ class U_I18N_API LocalizedNumberFormatter */ LocalizedNumberFormatter() = default; - // Make default copy constructor call the NumberFormatterSettings copy constructor. /** * Returns a copy of this LocalizedNumberFormatter. * @draft ICU 60 @@ -2333,11 +2385,12 @@ class U_I18N_API LocalizedNumberFormatter * * @param results * The results object. This method will mutate it to save the results. + * @param status * @internal */ void formatImpl(impl::UFormattedNumberData *results, UErrorCode &status) const; -#endif +#endif /* U_HIDE_INTERNAL_API */ /** * Destruct this LocalizedNumberFormatter, cleaning up any memory it might own. @@ -2359,6 +2412,8 @@ class U_I18N_API LocalizedNumberFormatter LocalizedNumberFormatter(impl::MacroProps &¯os, const Locale &locale); + void clear(); + void lnfMoveHelper(LocalizedNumberFormatter&& src); /** @@ -2430,7 +2485,7 @@ class U_I18N_API FormattedNumber : public UMemory { * @draft ICU 62 * @see Appendable */ - Appendable &appendTo(Appendable &appendable, UErrorCode& status); + Appendable &appendTo(Appendable &appendable, UErrorCode& status) const; #ifndef U_HIDE_DEPRECATED_API /** @@ -2457,9 +2512,9 @@ class U_I18N_API FormattedNumber : public UMemory { #endif /* U_HIDE_DEPRECATED_API */ /** - * Determines the start and end indices of the next occurrence of the given <em>field</em> in the - * output string. This allows you to determine the locations of, for example, the integer part, - * fraction part, or symbols. + * Determines the start (inclusive) and end (exclusive) indices of the next occurrence of the given + * <em>field</em> in the output string. This allows you to determine the locations of, for example, + * the integer part, fraction part, or symbols. * * If a field occurs just once, calling this method will find that occurrence and return it. If a * field occurs multiple times, this method may be called repeatedly with the following pattern: @@ -2478,7 +2533,7 @@ class U_I18N_API FormattedNumber : public UMemory { * Input+output variable. On input, the "field" property determines which field to look * up, and the "beginIndex" and "endIndex" properties determine where to begin the search. * On output, the "beginIndex" is set to the beginning of the first occurrence of the - * field with either begin or end indices after the input indices, "endIndex" is set to + * field with either begin or end indices after the input indices; "endIndex" is set to * the end of that occurrence of the field (exclusive index). If a field position is not * found, the method returns FALSE and the FieldPosition may or may not be changed. * @param status @@ -2537,7 +2592,7 @@ class U_I18N_API FormattedNumber : public UMemory { */ void getAllFieldPositionsImpl(FieldPositionIteratorHandler& fpih, UErrorCode& status) const; -#endif +#endif /* U_HIDE_INTERNAL_API */ /** * Copying not supported; use move constructor instead. diff --git a/deps/icu-small/source/i18n/unicode/numberrangeformatter.h b/deps/icu-small/source/i18n/unicode/numberrangeformatter.h new file mode 100644 index 0000000000..3e6248d934 --- /dev/null +++ b/deps/icu-small/source/i18n/unicode/numberrangeformatter.h @@ -0,0 +1,866 @@ +// © 2018 and later: Unicode, Inc. and others. +// License & terms of use: http://www.unicode.org/copyright.html + +#if !UCONFIG_NO_FORMATTING +#ifndef __NUMBERRANGEFORMATTER_H__ +#define __NUMBERRANGEFORMATTER_H__ + +#include <atomic> +#include "unicode/appendable.h" +#include "unicode/fieldpos.h" +#include "unicode/fpositer.h" +#include "unicode/numberformatter.h" + +#ifndef U_HIDE_DRAFT_API + +/** + * \file + * \brief C++ API: Library for localized formatting of number, currency, and unit ranges. + * + * The main entrypoint to the formatting of ranges of numbers, including currencies and other units of measurement. + * <p> + * Usage example: + * <p> + * <pre> + * NumberRangeFormatter::with() + * .identityFallback(UNUM_IDENTITY_FALLBACK_APPROXIMATELY_OR_SINGLE_VALUE) + * .numberFormatterFirst(NumberFormatter::with().adoptUnit(MeasureUnit::createMeter())) + * .numberFormatterSecond(NumberFormatter::with().adoptUnit(MeasureUnit::createKilometer())) + * .locale("en-GB") + * .formatRange(750, 1.2, status) + * .toString(status); + * // => "750 m - 1.2 km" + * </pre> + * <p> + * Like NumberFormatter, NumberRangeFormatter instances are immutable and thread-safe. This API is based on the + * <em>fluent</em> design pattern popularized by libraries such as Google's Guava. + * + * @author Shane Carr + */ + + +/** + * Defines how to merge fields that are identical across the range sign. + * + * @draft ICU 63 + */ +typedef enum UNumberRangeCollapse { + /** + * Use locale data and heuristics to determine how much of the string to collapse. Could end up collapsing none, + * some, or all repeated pieces in a locale-sensitive way. + * + * The heuristics used for this option are subject to change over time. + * + * @draft ICU 63 + */ + UNUM_RANGE_COLLAPSE_AUTO, + + /** + * Do not collapse any part of the number. Example: "3.2 thousand kilograms – 5.3 thousand kilograms" + * + * @draft ICU 63 + */ + UNUM_RANGE_COLLAPSE_NONE, + + /** + * Collapse the unit part of the number, but not the notation, if present. Example: "3.2 thousand – 5.3 thousand + * kilograms" + * + * @draft ICU 63 + */ + UNUM_RANGE_COLLAPSE_UNIT, + + /** + * Collapse any field that is equal across the range sign. May introduce ambiguity on the magnitude of the + * number. Example: "3.2 – 5.3 thousand kilograms" + * + * @draft ICU 63 + */ + UNUM_RANGE_COLLAPSE_ALL +} UNumberRangeCollapse; + +/** + * Defines the behavior when the two numbers in the range are identical after rounding. To programmatically detect + * when the identity fallback is used, compare the lower and upper BigDecimals via FormattedNumber. + * + * @draft ICU 63 + * @see NumberRangeFormatter + */ +typedef enum UNumberRangeIdentityFallback { + /** + * Show the number as a single value rather than a range. Example: "$5" + * + * @draft ICU 63 + */ + UNUM_IDENTITY_FALLBACK_SINGLE_VALUE, + + /** + * Show the number using a locale-sensitive approximation pattern. If the numbers were the same before rounding, + * show the single value. Example: "~$5" or "$5" + * + * @draft ICU 63 + */ + UNUM_IDENTITY_FALLBACK_APPROXIMATELY_OR_SINGLE_VALUE, + + /** + * Show the number using a locale-sensitive approximation pattern. Use the range pattern always, even if the + * inputs are the same. Example: "~$5" + * + * @draft ICU 63 + */ + UNUM_IDENTITY_FALLBACK_APPROXIMATELY, + + /** + * Show the number as the range of two equal values. Use the range pattern always, even if the inputs are the + * same. Example (with RangeCollapse.NONE): "$5 – $5" + * + * @draft ICU 63 + */ + UNUM_IDENTITY_FALLBACK_RANGE +} UNumberRangeIdentityFallback; + +/** + * Used in the result class FormattedNumberRange to indicate to the user whether the numbers formatted in the range + * were equal or not, and whether or not the identity fallback was applied. + * + * @draft ICU 63 + * @see NumberRangeFormatter + */ +typedef enum UNumberRangeIdentityResult { + /** + * Used to indicate that the two numbers in the range were equal, even before any rounding rules were applied. + * + * @draft ICU 63 + * @see NumberRangeFormatter + */ + UNUM_IDENTITY_RESULT_EQUAL_BEFORE_ROUNDING, + + /** + * Used to indicate that the two numbers in the range were equal, but only after rounding rules were applied. + * + * @draft ICU 63 + * @see NumberRangeFormatter + */ + UNUM_IDENTITY_RESULT_EQUAL_AFTER_ROUNDING, + + /** + * Used to indicate that the two numbers in the range were not equal, even after rounding rules were applied. + * + * @draft ICU 63 + * @see NumberRangeFormatter + */ + UNUM_IDENTITY_RESULT_NOT_EQUAL, + +#ifndef U_HIDE_INTERNAL_API + /** + * The number of entries in this enum. + * @internal + */ + UNUM_IDENTITY_RESULT_COUNT +#endif + +} UNumberRangeIdentityResult; + +U_NAMESPACE_BEGIN + +namespace number { // icu::number + +// Forward declarations: +class UnlocalizedNumberRangeFormatter; +class LocalizedNumberRangeFormatter; +class FormattedNumberRange; + +namespace impl { + +// Forward declarations: +struct RangeMacroProps; +class DecimalQuantity; +struct UFormattedNumberRangeData; +class NumberRangeFormatterImpl; + +} // namespace impl + +/** + * \cond + * Export an explicit template instantiation. See datefmt.h + * (When building DLLs for Windows this is required.) + */ +#if U_PF_WINDOWS <= U_PLATFORM && U_PLATFORM <= U_PF_CYGWIN && !defined(U_IN_DOXYGEN) +template struct U_I18N_API std::atomic<impl::NumberRangeFormatterImpl*>; +#endif +/** \endcond */ + +// Other helper classes would go here, but there are none. + +namespace impl { // icu::number::impl + +// Do not enclose entire MacroProps with #ifndef U_HIDE_INTERNAL_API, needed for a protected field +/** @internal */ +struct U_I18N_API RangeMacroProps : public UMemory { + /** @internal */ + UnlocalizedNumberFormatter formatter1; // = NumberFormatter::with(); + + /** @internal */ + UnlocalizedNumberFormatter formatter2; // = NumberFormatter::with(); + + /** @internal */ + bool singleFormatter = true; + + /** @internal */ + UNumberRangeCollapse collapse = UNUM_RANGE_COLLAPSE_AUTO; + + /** @internal */ + UNumberRangeIdentityFallback identityFallback = UNUM_IDENTITY_FALLBACK_APPROXIMATELY; + + /** @internal */ + Locale locale; + + // NOTE: Uses default copy and move constructors. + + /** + * Check all members for errors. + * @internal + */ + bool copyErrorTo(UErrorCode &status) const { + return formatter1.copyErrorTo(status) || formatter2.copyErrorTo(status); + } +}; + +} // namespace impl + +/** + * An abstract base class for specifying settings related to number formatting. This class is implemented by + * {@link UnlocalizedNumberRangeFormatter} and {@link LocalizedNumberRangeFormatter}. This class is not intended for + * public subclassing. + */ +template<typename Derived> +class U_I18N_API NumberRangeFormatterSettings { + public: + /** + * Sets the NumberFormatter instance to use for the numbers in the range. The same formatter is applied to both + * sides of the range. + * <p> + * The NumberFormatter instances must not have a locale applied yet; the locale specified on the + * NumberRangeFormatter will be used. + * + * @param formatter + * The formatter to use for both numbers in the range. + * @return The fluent chain. + * @draft ICU 63 + */ + Derived numberFormatterBoth(const UnlocalizedNumberFormatter &formatter) const &; + + /** + * Overload of numberFormatterBoth() for use on an rvalue reference. + * + * @param formatter + * The formatter to use for both numbers in the range. + * @return The fluent chain. + * @see #numberFormatterBoth + * @draft ICU 63 + */ + Derived numberFormatterBoth(const UnlocalizedNumberFormatter &formatter) &&; + + /** + * Overload of numberFormatterBoth() for use on an rvalue reference. + * + * @param formatter + * The formatter to use for both numbers in the range. + * @return The fluent chain. + * @see #numberFormatterBoth + * @draft ICU 63 + */ + Derived numberFormatterBoth(UnlocalizedNumberFormatter &&formatter) const &; + + /** + * Overload of numberFormatterBoth() for use on an rvalue reference. + * + * @param formatter + * The formatter to use for both numbers in the range. + * @return The fluent chain. + * @see #numberFormatterBoth + * @draft ICU 63 + */ + Derived numberFormatterBoth(UnlocalizedNumberFormatter &&formatter) &&; + + /** + * Sets the NumberFormatter instance to use for the first number in the range. + * <p> + * The NumberFormatter instances must not have a locale applied yet; the locale specified on the + * NumberRangeFormatter will be used. + * + * @param formatterFirst + * The formatter to use for the first number in the range. + * @return The fluent chain. + * @draft ICU 63 + */ + Derived numberFormatterFirst(const UnlocalizedNumberFormatter &formatterFirst) const &; + + /** + * Overload of numberFormatterFirst() for use on an rvalue reference. + * + * @param formatterFirst + * The formatter to use for the first number in the range. + * @return The fluent chain. + * @see #numberFormatterFirst + * @draft ICU 63 + */ + Derived numberFormatterFirst(const UnlocalizedNumberFormatter &formatterFirst) &&; + + /** + * Overload of numberFormatterFirst() for use on an rvalue reference. + * + * @param formatterFirst + * The formatter to use for the first number in the range. + * @return The fluent chain. + * @see #numberFormatterFirst + * @draft ICU 63 + */ + Derived numberFormatterFirst(UnlocalizedNumberFormatter &&formatterFirst) const &; + + /** + * Overload of numberFormatterFirst() for use on an rvalue reference. + * + * @param formatterFirst + * The formatter to use for the first number in the range. + * @return The fluent chain. + * @see #numberFormatterFirst + * @draft ICU 63 + */ + Derived numberFormatterFirst(UnlocalizedNumberFormatter &&formatterFirst) &&; + + /** + * Sets the NumberFormatter instance to use for the second number in the range. + * <p> + * The NumberFormatter instances must not have a locale applied yet; the locale specified on the + * NumberRangeFormatter will be used. + * + * @param formatterSecond + * The formatter to use for the second number in the range. + * @return The fluent chain. + * @draft ICU 63 + */ + Derived numberFormatterSecond(const UnlocalizedNumberFormatter &formatterSecond) const &; + + /** + * Overload of numberFormatterSecond() for use on an rvalue reference. + * + * @param formatterSecond + * The formatter to use for the second number in the range. + * @return The fluent chain. + * @see #numberFormatterSecond + * @draft ICU 63 + */ + Derived numberFormatterSecond(const UnlocalizedNumberFormatter &formatterSecond) &&; + + /** + * Overload of numberFormatterSecond() for use on an rvalue reference. + * + * @param formatterSecond + * The formatter to use for the second number in the range. + * @return The fluent chain. + * @see #numberFormatterSecond + * @draft ICU 63 + */ + Derived numberFormatterSecond(UnlocalizedNumberFormatter &&formatterSecond) const &; + + /** + * Overload of numberFormatterSecond() for use on an rvalue reference. + * + * @param formatterSecond + * The formatter to use for the second number in the range. + * @return The fluent chain. + * @see #numberFormatterSecond + * @draft ICU 63 + */ + Derived numberFormatterSecond(UnlocalizedNumberFormatter &&formatterSecond) &&; + + /** + * Sets the aggressiveness of "collapsing" fields across the range separator. Possible values: + * <p> + * <ul> + * <li>ALL: "3-5K miles"</li> + * <li>UNIT: "3K - 5K miles"</li> + * <li>NONE: "3K miles - 5K miles"</li> + * <li>AUTO: usually UNIT or NONE, depending on the locale and formatter settings</li> + * </ul> + * <p> + * The default value is AUTO. + * + * @param collapse + * The collapsing strategy to use for this range. + * @return The fluent chain. + * @draft ICU 63 + */ + Derived collapse(UNumberRangeCollapse collapse) const &; + + /** + * Overload of collapse() for use on an rvalue reference. + * + * @param collapse + * The collapsing strategy to use for this range. + * @return The fluent chain. + * @see #collapse + * @draft ICU 63 + */ + Derived collapse(UNumberRangeCollapse collapse) &&; + + /** + * Sets the behavior when the two sides of the range are the same. This could happen if the same two numbers are + * passed to the formatRange function, or if different numbers are passed to the function but they become the same + * after rounding rules are applied. Possible values: + * <p> + * <ul> + * <li>SINGLE_VALUE: "5 miles"</li> + * <li>APPROXIMATELY_OR_SINGLE_VALUE: "~5 miles" or "5 miles", depending on whether the number was the same before + * rounding was applied</li> + * <li>APPROXIMATELY: "~5 miles"</li> + * <li>RANGE: "5-5 miles" (with collapse=UNIT)</li> + * </ul> + * <p> + * The default value is APPROXIMATELY. + * + * @param identityFallback + * The strategy to use when formatting two numbers that end up being the same. + * @return The fluent chain. + * @draft ICU 63 + */ + Derived identityFallback(UNumberRangeIdentityFallback identityFallback) const &; + + /** + * Overload of identityFallback() for use on an rvalue reference. + * + * @param identityFallback + * The strategy to use when formatting two numbers that end up being the same. + * @return The fluent chain. + * @see #identityFallback + * @draft ICU 63 + */ + Derived identityFallback(UNumberRangeIdentityFallback identityFallback) &&; + + /** + * Sets the UErrorCode if an error occurred in the fluent chain. + * Preserves older error codes in the outErrorCode. + * @return TRUE if U_FAILURE(outErrorCode) + * @draft ICU 63 + */ + UBool copyErrorTo(UErrorCode &outErrorCode) const { + if (U_FAILURE(outErrorCode)) { + // Do not overwrite the older error code + return TRUE; + } + fMacros.copyErrorTo(outErrorCode); + return U_FAILURE(outErrorCode); + }; + + // NOTE: Uses default copy and move constructors. + + private: + impl::RangeMacroProps fMacros; + + // Don't construct me directly! Use (Un)LocalizedNumberFormatter. + NumberRangeFormatterSettings() = default; + + friend class LocalizedNumberRangeFormatter; + friend class UnlocalizedNumberRangeFormatter; +}; + +/** + * A NumberRangeFormatter that does not yet have a locale. In order to format, a locale must be specified. + * + * @see NumberRangeFormatter + * @draft ICU 63 + */ +class U_I18N_API UnlocalizedNumberRangeFormatter + : public NumberRangeFormatterSettings<UnlocalizedNumberRangeFormatter>, public UMemory { + + public: + /** + * Associate the given locale with the number range formatter. The locale is used for picking the + * appropriate symbols, formats, and other data for number display. + * + * @param locale + * The locale to use when loading data for number formatting. + * @return The fluent chain. + * @draft ICU 63 + */ + LocalizedNumberRangeFormatter locale(const icu::Locale &locale) const &; + + /** + * Overload of locale() for use on an rvalue reference. + * + * @param locale + * The locale to use when loading data for number formatting. + * @return The fluent chain. + * @see #locale + * @draft ICU 63 + */ + LocalizedNumberRangeFormatter locale(const icu::Locale &locale) &&; + + /** + * Default constructor: puts the formatter into a valid but undefined state. + * + * @draft ICU 63 + */ + UnlocalizedNumberRangeFormatter() = default; + + /** + * Returns a copy of this UnlocalizedNumberRangeFormatter. + * @draft ICU 63 + */ + UnlocalizedNumberRangeFormatter(const UnlocalizedNumberRangeFormatter &other); + + /** + * Move constructor: + * The source UnlocalizedNumberRangeFormatter will be left in a valid but undefined state. + * @draft ICU 63 + */ + UnlocalizedNumberRangeFormatter(UnlocalizedNumberRangeFormatter&& src) U_NOEXCEPT; + + /** + * Copy assignment operator. + * @draft ICU 63 + */ + UnlocalizedNumberRangeFormatter& operator=(const UnlocalizedNumberRangeFormatter& other); + + /** + * Move assignment operator: + * The source UnlocalizedNumberRangeFormatter will be left in a valid but undefined state. + * @draft ICU 63 + */ + UnlocalizedNumberRangeFormatter& operator=(UnlocalizedNumberRangeFormatter&& src) U_NOEXCEPT; + + private: + explicit UnlocalizedNumberRangeFormatter( + const NumberRangeFormatterSettings<UnlocalizedNumberRangeFormatter>& other); + + explicit UnlocalizedNumberRangeFormatter( + NumberRangeFormatterSettings<UnlocalizedNumberRangeFormatter>&& src) U_NOEXCEPT; + + // To give the fluent setters access to this class's constructor: + friend class NumberRangeFormatterSettings<UnlocalizedNumberRangeFormatter>; + + // To give NumberRangeFormatter::with() access to this class's constructor: + friend class NumberRangeFormatter; +}; + +/** + * A NumberRangeFormatter that has a locale associated with it; this means .formatRange() methods are available. + * + * @see NumberFormatter + * @draft ICU 63 + */ +class U_I18N_API LocalizedNumberRangeFormatter + : public NumberRangeFormatterSettings<LocalizedNumberRangeFormatter>, public UMemory { + public: + /** + * Format the given Formattables to a string using the settings specified in the NumberRangeFormatter fluent setting + * chain. + * + * @param first + * The first number in the range, usually to the left in LTR locales. + * @param second + * The second number in the range, usually to the right in LTR locales. + * @param status + * Set if an error occurs while formatting. + * @return A FormattedNumberRange object; call .toString() to get the string. + * @draft ICU 63 + */ + FormattedNumberRange formatFormattableRange( + const Formattable& first, const Formattable& second, UErrorCode& status) const; + + /** + * Default constructor: puts the formatter into a valid but undefined state. + * + * @draft ICU 63 + */ + LocalizedNumberRangeFormatter() = default; + + /** + * Returns a copy of this LocalizedNumberRangeFormatter. + * @draft ICU 63 + */ + LocalizedNumberRangeFormatter(const LocalizedNumberRangeFormatter &other); + + /** + * Move constructor: + * The source LocalizedNumberRangeFormatter will be left in a valid but undefined state. + * @draft ICU 63 + */ + LocalizedNumberRangeFormatter(LocalizedNumberRangeFormatter&& src) U_NOEXCEPT; + + /** + * Copy assignment operator. + * @draft ICU 63 + */ + LocalizedNumberRangeFormatter& operator=(const LocalizedNumberRangeFormatter& other); + + /** + * Move assignment operator: + * The source LocalizedNumberRangeFormatter will be left in a valid but undefined state. + * @draft ICU 63 + */ + LocalizedNumberRangeFormatter& operator=(LocalizedNumberRangeFormatter&& src) U_NOEXCEPT; + +#ifndef U_HIDE_INTERNAL_API + + /** + * @param results + * The results object. This method will mutate it to save the results. + * @param equalBeforeRounding + * Whether the number was equal before copying it into a DecimalQuantity. + * Used for determining the identity fallback behavior. + * @param status + * Set if an error occurs while formatting. + * @internal + */ + void formatImpl(impl::UFormattedNumberRangeData& results, bool equalBeforeRounding, + UErrorCode& status) const; + +#endif + + /** + * Destruct this LocalizedNumberRangeFormatter, cleaning up any memory it might own. + * @draft ICU 63 + */ + ~LocalizedNumberRangeFormatter(); + + private: + std::atomic<impl::NumberRangeFormatterImpl*> fAtomicFormatter = {}; + + const impl::NumberRangeFormatterImpl* getFormatter(UErrorCode& stauts) const; + + explicit LocalizedNumberRangeFormatter( + const NumberRangeFormatterSettings<LocalizedNumberRangeFormatter>& other); + + explicit LocalizedNumberRangeFormatter( + NumberRangeFormatterSettings<LocalizedNumberRangeFormatter>&& src) U_NOEXCEPT; + + LocalizedNumberRangeFormatter(const impl::RangeMacroProps ¯os, const Locale &locale); + + LocalizedNumberRangeFormatter(impl::RangeMacroProps &¯os, const Locale &locale); + + void clear(); + + // To give the fluent setters access to this class's constructor: + friend class NumberRangeFormatterSettings<UnlocalizedNumberRangeFormatter>; + friend class NumberRangeFormatterSettings<LocalizedNumberRangeFormatter>; + + // To give UnlocalizedNumberRangeFormatter::locale() access to this class's constructor: + friend class UnlocalizedNumberRangeFormatter; +}; + +/** + * The result of a number range formatting operation. This class allows the result to be exported in several data types, + * including a UnicodeString and a FieldPositionIterator. + * + * @draft ICU 63 + */ +class U_I18N_API FormattedNumberRange : public UMemory { + public: + /** + * Returns a UnicodeString representation of the formatted number range. + * + * @param status + * Set if an error occurs while formatting the number to the UnicodeString. + * @return a UnicodeString containing the localized number range. + * @draft ICU 63 + */ + UnicodeString toString(UErrorCode& status) const; + + /** + * Appends the formatted number range to an Appendable. + * + * @param appendable + * The Appendable to which to append the formatted number range string. + * @param status + * Set if an error occurs while formatting the number range to the Appendable. + * @return The same Appendable, for chaining. + * @draft ICU 63 + * @see Appendable + */ + Appendable &appendTo(Appendable &appendable, UErrorCode& status) const; + + /** + * Determines the start (inclusive) and end (exclusive) indices of the next occurrence of the given + * <em>field</em> in the output string. This allows you to determine the locations of, for example, + * the integer part, fraction part, or symbols. + * + * If both sides of the range have the same field, the field will occur twice, once before the + * range separator and once after the range separator, if applicable. + * + * If a field occurs just once, calling this method will find that occurrence and return it. If a + * field occurs multiple times, this method may be called repeatedly with the following pattern: + * + * <pre> + * FieldPosition fpos(UNUM_INTEGER_FIELD); + * while (formattedNumberRange.nextFieldPosition(fpos, status)) { + * // do something with fpos. + * } + * </pre> + * + * This method is useful if you know which field to query. If you want all available field position + * information, use #getAllFieldPositions(). + * + * @param fieldPosition + * Input+output variable. See {@link FormattedNumber#nextFieldPosition}. + * @param status + * Set if an error occurs while populating the FieldPosition. + * @return TRUE if a new occurrence of the field was found; FALSE otherwise. + * @draft ICU 63 + * @see UNumberFormatFields + */ + UBool nextFieldPosition(FieldPosition& fieldPosition, UErrorCode& status) const; + + /** + * Export the formatted number range to a FieldPositionIterator. This allows you to determine which characters in + * the output string correspond to which <em>fields</em>, such as the integer part, fraction part, and sign. + * + * If information on only one field is needed, use #nextFieldPosition() instead. + * + * @param iterator + * The FieldPositionIterator to populate with all of the fields present in the formatted number. + * @param status + * Set if an error occurs while populating the FieldPositionIterator. + * @draft ICU 63 + * @see UNumberFormatFields + */ + void getAllFieldPositions(FieldPositionIterator &iterator, UErrorCode &status) const; + + /** + * Export the first formatted number as a decimal number. This endpoint + * is useful for obtaining the exact number being printed after scaling + * and rounding have been applied by the number range formatting pipeline. + * + * The syntax of the unformatted number is a "numeric string" + * as defined in the Decimal Arithmetic Specification, available at + * http://speleotrove.com/decimal + * + * @return A decimal representation of the first formatted number. + * @draft ICU 63 + * @see NumberRangeFormatter + * @see #getSecondDecimal + */ + UnicodeString getFirstDecimal(UErrorCode& status) const; + + /** + * Export the second formatted number as a decimal number. This endpoint + * is useful for obtaining the exact number being printed after scaling + * and rounding have been applied by the number range formatting pipeline. + * + * The syntax of the unformatted number is a "numeric string" + * as defined in the Decimal Arithmetic Specification, available at + * http://speleotrove.com/decimal + * + * @return A decimal representation of the second formatted number. + * @draft ICU 63 + * @see NumberRangeFormatter + * @see #getFirstDecimal + */ + UnicodeString getSecondDecimal(UErrorCode& status) const; + + /** + * Returns whether the pair of numbers was successfully formatted as a range or whether an identity fallback was + * used. For example, if the first and second number were the same either before or after rounding occurred, an + * identity fallback was used. + * + * @return An indication the resulting identity situation in the formatted number range. + * @draft ICU 63 + * @see UNumberRangeIdentityFallback + */ + UNumberRangeIdentityResult getIdentityResult(UErrorCode& status) const; + + /** + * Copying not supported; use move constructor instead. + */ + FormattedNumberRange(const FormattedNumberRange&) = delete; + + /** + * Copying not supported; use move assignment instead. + */ + FormattedNumberRange& operator=(const FormattedNumberRange&) = delete; + + /** + * Move constructor: + * Leaves the source FormattedNumberRange in an undefined state. + * @draft ICU 63 + */ + FormattedNumberRange(FormattedNumberRange&& src) U_NOEXCEPT; + + /** + * Move assignment: + * Leaves the source FormattedNumberRange in an undefined state. + * @draft ICU 63 + */ + FormattedNumberRange& operator=(FormattedNumberRange&& src) U_NOEXCEPT; + + /** + * Destruct an instance of FormattedNumberRange, cleaning up any memory it might own. + * @draft ICU 63 + */ + ~FormattedNumberRange(); + + private: + // Can't use LocalPointer because UFormattedNumberRangeData is forward-declared + const impl::UFormattedNumberRangeData *fResults; + + // Error code for the terminal methods + UErrorCode fErrorCode; + + /** + * Internal constructor from data type. Adopts the data pointer. + * @internal + */ + explicit FormattedNumberRange(impl::UFormattedNumberRangeData *results) + : fResults(results), fErrorCode(U_ZERO_ERROR) {}; + + explicit FormattedNumberRange(UErrorCode errorCode) + : fResults(nullptr), fErrorCode(errorCode) {}; + + void getAllFieldPositionsImpl(FieldPositionIteratorHandler& fpih, UErrorCode& status) const; + + // To give LocalizedNumberRangeFormatter format methods access to this class's constructor: + friend class LocalizedNumberRangeFormatter; +}; + +/** + * See the main description in numberrangeformatter.h for documentation and examples. + * + * @draft ICU 63 + */ +class U_I18N_API NumberRangeFormatter final { + public: + /** + * Call this method at the beginning of a NumberRangeFormatter fluent chain in which the locale is not currently + * known at the call site. + * + * @return An {@link UnlocalizedNumberRangeFormatter}, to be used for chaining. + * @draft ICU 63 + */ + static UnlocalizedNumberRangeFormatter with(); + + /** + * Call this method at the beginning of a NumberRangeFormatter fluent chain in which the locale is known at the call + * site. + * + * @param locale + * The locale from which to load formats and symbols for number range formatting. + * @return A {@link LocalizedNumberRangeFormatter}, to be used for chaining. + * @draft ICU 63 + */ + static LocalizedNumberRangeFormatter withLocale(const Locale &locale); + + /** + * Use factory methods instead of the constructor to create a NumberFormatter. + */ + NumberRangeFormatter() = delete; +}; + +} // namespace number +U_NAMESPACE_END + +#endif // U_HIDE_DRAFT_API + +#endif // __NUMBERRANGEFORMATTER_H__ + +#endif /* #if !UCONFIG_NO_FORMATTING */ diff --git a/deps/icu-small/source/i18n/unicode/numfmt.h b/deps/icu-small/source/i18n/unicode/numfmt.h index 572e6afc71..871fbd93e1 100644 --- a/deps/icu-small/source/i18n/unicode/numfmt.h +++ b/deps/icu-small/source/i18n/unicode/numfmt.h @@ -56,7 +56,6 @@ class StringEnumeration; * <p><strong>IMPORTANT:</strong> New users are strongly encouraged to see if * numberformatter.h fits their use case. Although not deprecated, this header * is provided for backwards compatibility only. - * <hr/> * * Abstract base class for all number formats. Provides interface for * formatting and parsing a number. Also provides methods for @@ -710,7 +709,7 @@ public: * The default formatting style is locale dependent. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of NumberFormat. + * {@link icu::number::NumberFormatter} instead of NumberFormat. * @stable ICU 2.0 */ static NumberFormat* U_EXPORT2 createInstance(UErrorCode&); @@ -721,7 +720,7 @@ public: * @param inLocale the given locale. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of NumberFormat. + * {@link icu::number::NumberFormatter} instead of NumberFormat. * @stable ICU 2.0 */ static NumberFormat* U_EXPORT2 createInstance(const Locale& inLocale, @@ -731,7 +730,7 @@ public: * Create a specific style NumberFormat for the specified locale. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of NumberFormat. + * {@link icu::number::NumberFormatter} instead of NumberFormat. * @param desiredLocale the given locale. * @param style the given style. * @param errorCode Output param filled with success/failure status. @@ -770,7 +769,7 @@ public: * Returns a currency format for the current default locale. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of NumberFormat. + * {@link icu::number::NumberFormatter} instead of NumberFormat. * @stable ICU 2.0 */ static NumberFormat* U_EXPORT2 createCurrencyInstance(UErrorCode&); @@ -779,7 +778,7 @@ public: * Returns a currency format for the specified locale. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of NumberFormat. + * {@link icu::number::NumberFormatter} instead of NumberFormat. * @param inLocale the given locale. * @stable ICU 2.0 */ @@ -790,7 +789,7 @@ public: * Returns a percentage format for the current default locale. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of NumberFormat. + * {@link icu::number::NumberFormatter} instead of NumberFormat. * @stable ICU 2.0 */ static NumberFormat* U_EXPORT2 createPercentInstance(UErrorCode&); @@ -799,7 +798,7 @@ public: * Returns a percentage format for the specified locale. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of NumberFormat. + * {@link icu::number::NumberFormatter} instead of NumberFormat. * @param inLocale the given locale. * @stable ICU 2.0 */ @@ -810,7 +809,7 @@ public: * Returns a scientific format for the current default locale. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of NumberFormat. + * {@link icu::number::NumberFormatter} instead of NumberFormat. * @stable ICU 2.0 */ static NumberFormat* U_EXPORT2 createScientificInstance(UErrorCode&); @@ -819,7 +818,7 @@ public: * Returns a scientific format for the specified locale. * <p> * <strong>NOTE:</strong> New users are strongly encouraged to use - * {@link NumberFormatter} instead of NumberFormat. + * {@link icu::number::NumberFormatter} instead of NumberFormat. * @param inLocale the given locale. * @stable ICU 2.0 */ @@ -1028,14 +1027,14 @@ public: * Get the rounding mode. This will always return NumberFormat::ERoundingMode::kRoundUnnecessary * if the subclass does not support rounding. * @return A rounding mode - * @draft ICU 60 + * @stable ICU 60 */ virtual ERoundingMode getRoundingMode(void) const; /** * Set the rounding mode. If a subclass does not support rounding, this will do nothing. * @param roundingMode A rounding mode - * @draft ICU 60 + * @stable ICU 60 */ virtual void setRoundingMode(ERoundingMode roundingMode); diff --git a/deps/icu-small/source/i18n/unicode/plurfmt.h b/deps/icu-small/source/i18n/unicode/plurfmt.h index 9a83e52550..6b757c8841 100644 --- a/deps/icu-small/source/i18n/unicode/plurfmt.h +++ b/deps/icu-small/source/i18n/unicode/plurfmt.h @@ -520,15 +520,7 @@ public: */ virtual UClassID getDynamicClassID() const; -#if (defined(__xlC__) && (__xlC__ < 0x0C00)) || (U_PLATFORM == U_PF_OS390) || (U_PLATFORM ==U_PF_OS400) -// Work around a compiler bug on xlC 11.1 on AIX 7.1 that would -// prevent PluralSelectorAdapter from implementing private PluralSelector. -// xlC error message: -// 1540-0300 (S) The "private" member "class icu_49::PluralFormat::PluralSelector" cannot be accessed. -public: -#else private: -#endif /** * @internal */ @@ -564,10 +556,6 @@ private: PluralRules* pluralRules; }; -#if defined(__xlC__) -// End of xlC bug workaround, keep remaining definitions private. -private: -#endif Locale locale; MessagePattern msgPattern; NumberFormat* numberFormat; diff --git a/deps/icu-small/source/i18n/unicode/plurrule.h b/deps/icu-small/source/i18n/unicode/plurrule.h index 03dea3f1b9..daeed52bee 100644 --- a/deps/icu-small/source/i18n/unicode/plurrule.h +++ b/deps/icu-small/source/i18n/unicode/plurrule.h @@ -118,7 +118,6 @@ class SharedPluralRules; * Examples are in the following table: * </p> * <table border='1' style="border-collapse:collapse"> - * <tbody> * <tr> * <th>n</th> * <th>i</th> @@ -155,7 +154,6 @@ class SharedPluralRules; * <td align="right">23</td> * <td>2</td> * </tr> - * </tbody> * </table> * <p> * The difference between 'in' and 'within' is that 'in' only includes integers in the specified range, while 'within' @@ -499,6 +497,12 @@ private: UnicodeString getRuleFromResource(const Locale& locale, UPluralType type, UErrorCode& status); RuleChain *rulesForKeyword(const UnicodeString &keyword) const; + /** + * An internal status variable used to indicate that the object is in an 'invalid' state. + * Used by copy constructor, the assignment operator and the clone method. + */ + UErrorCode mInternalStatus; + friend class PluralRuleParser; }; diff --git a/deps/icu-small/source/i18n/unicode/rbnf.h b/deps/icu-small/source/i18n/unicode/rbnf.h index 2d284909f8..f7cd85a322 100644 --- a/deps/icu-small/source/i18n/unicode/rbnf.h +++ b/deps/icu-small/source/i18n/unicode/rbnf.h @@ -313,7 +313,6 @@ enum URBNFRuleSetTag { * <td>The rule for an IEEE 754 NaN (not a number).</td> * </tr> * <tr> - * <tr> * <td><em>nothing</em></td> * <td>If the rule's rule descriptor is left out, the base value is one plus the * preceding rule's base value (or zero if this is the first rule in the list) in a normal @@ -1013,14 +1012,14 @@ public: /** * Get the rounding mode. * @return A rounding mode - * @draft ICU 60 + * @stable ICU 60 */ virtual ERoundingMode getRoundingMode(void) const; /** * Set the rounding mode. * @param roundingMode A rounding mode - * @draft ICU 60 + * @stable ICU 60 */ virtual void setRoundingMode(ERoundingMode roundingMode); @@ -1095,7 +1094,7 @@ private: void format(double number, NFRuleSet& rs, UnicodeString& toAppendTo, UErrorCode& status) const; private: - NFRuleSet **ruleSets; + NFRuleSet **fRuleSets; UnicodeString* ruleSetDescriptions; int32_t numRuleSets; NFRuleSet *defaultRuleSet; @@ -1104,7 +1103,7 @@ private: DecimalFormatSymbols* decimalFormatSymbols; NFRule *defaultInfinityRule; NFRule *defaultNaNRule; - ERoundingMode roundingMode; + ERoundingMode fRoundingMode; UBool lenient; UnicodeString* lenientParseRules; LocalizationInfo* localizations; diff --git a/deps/icu-small/source/i18n/unicode/region.h b/deps/icu-small/source/i18n/unicode/region.h index 667c4051f0..ccd63901a5 100644 --- a/deps/icu-small/source/i18n/unicode/region.h +++ b/deps/icu-small/source/i18n/unicode/region.h @@ -192,7 +192,7 @@ private: char id[4]; UnicodeString idStr; int32_t code; - URegionType type; + URegionType fType; Region *containingRegion; UVector *containedRegions; UVector *preferredValues; diff --git a/deps/icu-small/source/i18n/unicode/reldatefmt.h b/deps/icu-small/source/i18n/unicode/reldatefmt.h index abd43522c3..dd8bc53d55 100644 --- a/deps/icu-small/source/i18n/unicode/reldatefmt.h +++ b/deps/icu-small/source/i18n/unicode/reldatefmt.h @@ -165,12 +165,20 @@ typedef enum UDateAbsoluteUnit { */ UDAT_ABSOLUTE_NOW, +#ifndef U_HIDE_DRAFT_API + /** + * Quarter + * @draft ICU 63 + */ + UDAT_ABSOLUTE_QUARTER, +#endif // U_HIDE_DRAFT_API + #ifndef U_HIDE_DEPRECATED_API /** * One more than the highest normal UDateAbsoluteUnit value. * @deprecated ICU 58 The numeric value may change over time, see ICU ticket #12420. */ - UDAT_ABSOLUTE_UNIT_COUNT + UDAT_ABSOLUTE_UNIT_COUNT = UDAT_ABSOLUTE_NOW + 2 #endif // U_HIDE_DEPRECATED_API } UDateAbsoluteUnit; @@ -328,7 +336,7 @@ public: * @param nfToAdopt Constructed object takes ownership of this pointer. * It is an error for caller to delete this pointer or change its * contents after calling this constructor. - * @status Any error is returned here. + * @param status Any error is returned here. * @stable ICU 53 */ RelativeDateTimeFormatter( @@ -346,7 +354,7 @@ public: * @param style the format style. The UDAT_RELATIVE bit field has no effect. * @param capitalizationContext A value from UDisplayContext that pertains to * capitalization. - * @status Any error is returned here. + * @param status Any error is returned here. * @stable ICU 54 */ RelativeDateTimeFormatter( diff --git a/deps/icu-small/source/i18n/unicode/smpdtfmt.h b/deps/icu-small/source/i18n/unicode/smpdtfmt.h index 305412b8d1..929c1b4675 100644 --- a/deps/icu-small/source/i18n/unicode/smpdtfmt.h +++ b/deps/icu-small/source/i18n/unicode/smpdtfmt.h @@ -1147,7 +1147,7 @@ public: * Overrides base class method and * This method clears per field NumberFormat instances * previously set by {@see adoptNumberFormat(const UnicodeString&, NumberFormat*, UErrorCode)} - * @param adoptNF the NumbeferFormat used + * @param formatToAdopt the NumbeferFormat used * @stable ICU 54 */ void adoptNumberFormat(NumberFormat *formatToAdopt); @@ -1162,7 +1162,7 @@ public: * Per field NumberFormat can also be cleared in {@see DateFormat::setNumberFormat(const NumberFormat& newNumberFormat)} * * @param fields the fields to override(like y) - * @param adoptNF the NumbeferFormat used + * @param formatToAdopt the NumbeferFormat used * @param status Receives a status code, which will be U_ZERO_ERROR * if the operation succeeds. * @stable ICU 54 diff --git a/deps/icu-small/source/i18n/unicode/translit.h b/deps/icu-small/source/i18n/unicode/translit.h index dc31d97bc6..6b4888145f 100644 --- a/deps/icu-small/source/i18n/unicode/translit.h +++ b/deps/icu-small/source/i18n/unicode/translit.h @@ -31,7 +31,6 @@ U_NAMESPACE_BEGIN class UnicodeFilter; class UnicodeSet; -class CompoundTransliterator; class TransliteratorParser; class NormalizationTransliterator; class TransliteratorIDParser; @@ -77,8 +76,7 @@ class TransliteratorIDParser; * transliteration. For example, given a string <code>input</code> * and a transliterator <code>t</code>, the call * - * \htmlonly<blockquote>\endhtmlonly<code>String result = t.transliterate(input); - * </code>\htmlonly</blockquote>\endhtmlonly + * String result = t.transliterate(input); * * will transliterate it and return the result. Other methods allow * the client to specify a substring to be transliterated and to use @@ -98,22 +96,20 @@ class TransliteratorIDParser; * contents of the buffer may show text being modified as each new * character arrives. * - * <p>Consider the simple <code>RuleBasedTransliterator</code>: - * - * \htmlonly<blockquote>\endhtmlonly<code> - * th>{theta}<br> - * t>{tau} - * </code>\htmlonly</blockquote>\endhtmlonly + * <p>Consider the simple rule-based Transliterator: + * <pre> + * th>{theta} + * t>{tau} + * </pre> * * When the user types 't', nothing will happen, since the * transliterator is waiting to see if the next character is 'h'. To * remedy this, we introduce the notion of a cursor, marked by a '|' * in the output string: - * - * \htmlonly<blockquote>\endhtmlonly<code> - * t>|{tau}<br> - * {tau}h>{theta} - * </code>\htmlonly</blockquote>\endhtmlonly + * <pre> + * t>|{tau} + * {tau}h>{theta} + * </pre> * * Now when the user types 't', tau appears, and if the next character * is 'h', the tau changes to a theta. This is accomplished by @@ -135,7 +131,7 @@ class TransliteratorIDParser; * which the transliterator last stopped, either because it reached * the end, or because it required more characters to disambiguate * between possible inputs. The <code>CURSOR</code> can also be - * explicitly set by rules in a <code>RuleBasedTransliterator</code>. + * explicitly set by rules in a rule-based Transliterator. * Any characters before the <code>CURSOR</code> index are frozen; * future keyboard transliteration calls within this input sequence * will not change them. New text is inserted at the @@ -237,6 +233,255 @@ class TransliteratorIDParser; * if the performance of these methods can be improved over the * performance obtained by the default implementations in this class. * + * <p><b>Rule syntax</b> + * + * <p>A set of rules determines how to perform translations. + * Rules within a rule set are separated by semicolons (';'). + * To include a literal semicolon, prefix it with a backslash ('\'). + * Unicode Pattern_White_Space is ignored. + * If the first non-blank character on a line is '#', + * the entire line is ignored as a comment. + * + * <p>Each set of rules consists of two groups, one forward, and one + * reverse. This is a convention that is not enforced; rules for one + * direction may be omitted, with the result that translations in + * that direction will not modify the source text. In addition, + * bidirectional forward-reverse rules may be specified for + * symmetrical transformations. + * + * <p>Note: Another description of the Transliterator rule syntax is available in + * <a href="https://www.unicode.org/reports/tr35/tr35-general.html#Transform_Rules_Syntax">section + * Transform Rules Syntax of UTS #35: Unicode LDML</a>. + * The rules are shown there using arrow symbols ← and → and ↔. + * ICU supports both those and the equivalent ASCII symbols < and > and <>. + * + * <p>Rule statements take one of the following forms: + * + * <dl> + * <dt><code>$alefmadda=\\u0622;</code></dt> + * <dd><strong>Variable definition.</strong> The name on the + * left is assigned the text on the right. In this example, + * after this statement, instances of the left hand name, + * "<code>$alefmadda</code>", will be replaced by + * the Unicode character U+0622. Variable names must begin + * with a letter and consist only of letters, digits, and + * underscores. Case is significant. Duplicate names cause + * an exception to be thrown, that is, variables cannot be + * redefined. The right hand side may contain well-formed + * text of any length, including no text at all ("<code>$empty=;</code>"). + * The right hand side may contain embedded <code>UnicodeSet</code> + * patterns, for example, "<code>$softvowel=[eiyEIY]</code>".</dd> + * <dt><code>ai>$alefmadda;</code></dt> + * <dd><strong>Forward translation rule.</strong> This rule + * states that the string on the left will be changed to the + * string on the right when performing forward + * transliteration.</dd> + * <dt><code>ai<$alefmadda;</code></dt> + * <dd><strong>Reverse translation rule.</strong> This rule + * states that the string on the right will be changed to + * the string on the left when performing reverse + * transliteration.</dd> + * </dl> + * + * <dl> + * <dt><code>ai<>$alefmadda;</code></dt> + * <dd><strong>Bidirectional translation rule.</strong> This + * rule states that the string on the right will be changed + * to the string on the left when performing forward + * transliteration, and vice versa when performing reverse + * transliteration.</dd> + * </dl> + * + * <p>Translation rules consist of a <em>match pattern</em> and an <em>output + * string</em>. The match pattern consists of literal characters, + * optionally preceded by context, and optionally followed by + * context. Context characters, like literal pattern characters, + * must be matched in the text being transliterated. However, unlike + * literal pattern characters, they are not replaced by the output + * text. For example, the pattern "<code>abc{def}</code>" + * indicates the characters "<code>def</code>" must be + * preceded by "<code>abc</code>" for a successful match. + * If there is a successful match, "<code>def</code>" will + * be replaced, but not "<code>abc</code>". The final '<code>}</code>' + * is optional, so "<code>abc{def</code>" is equivalent to + * "<code>abc{def}</code>". Another example is "<code>{123}456</code>" + * (or "<code>123}456</code>") in which the literal + * pattern "<code>123</code>" must be followed by "<code>456</code>". + * + * <p>The output string of a forward or reverse rule consists of + * characters to replace the literal pattern characters. If the + * output string contains the character '<code>|</code>', this is + * taken to indicate the location of the <em>cursor</em> after + * replacement. The cursor is the point in the text at which the + * next replacement, if any, will be applied. The cursor is usually + * placed within the replacement text; however, it can actually be + * placed into the precending or following context by using the + * special character '@'. Examples: + * + * <pre> + * a {foo} z > | @ bar; # foo -> bar, move cursor before a + * {foo} xyz > bar @@|; # foo -> bar, cursor between y and z + * </pre> + * + * <p><b>UnicodeSet</b> + * + * <p><code>UnicodeSet</code> patterns may appear anywhere that + * makes sense. They may appear in variable definitions. + * Contrariwise, <code>UnicodeSet</code> patterns may themselves + * contain variable references, such as "<code>$a=[a-z];$not_a=[^$a]</code>", + * or "<code>$range=a-z;$ll=[$range]</code>". + * + * <p><code>UnicodeSet</code> patterns may also be embedded directly + * into rule strings. Thus, the following two rules are equivalent: + * + * <pre> + * $vowel=[aeiou]; $vowel>'*'; # One way to do this + * [aeiou]>'*'; # Another way + * </pre> + * + * <p>See {@link UnicodeSet} for more documentation and examples. + * + * <p><b>Segments</b> + * + * <p>Segments of the input string can be matched and copied to the + * output string. This makes certain sets of rules simpler and more + * general, and makes reordering possible. For example: + * + * <pre> + * ([a-z]) > $1 $1; # double lowercase letters + * ([:Lu:]) ([:Ll:]) > $2 $1; # reverse order of Lu-Ll pairs + * </pre> + * + * <p>The segment of the input string to be copied is delimited by + * "<code>(</code>" and "<code>)</code>". Up to + * nine segments may be defined. Segments may not overlap. In the + * output string, "<code>$1</code>" through "<code>$9</code>" + * represent the input string segments, in left-to-right order of + * definition. + * + * <p><b>Anchors</b> + * + * <p>Patterns can be anchored to the beginning or the end of the text. This is done with the + * special characters '<code>^</code>' and '<code>$</code>'. For example: + * + * <pre> + * ^ a > 'BEG_A'; # match 'a' at start of text + * a > 'A'; # match other instances of 'a' + * z $ > 'END_Z'; # match 'z' at end of text + * z > 'Z'; # match other instances of 'z' + * </pre> + * + * <p>It is also possible to match the beginning or the end of the text using a <code>UnicodeSet</code>. + * This is done by including a virtual anchor character '<code>$</code>' at the end of the + * set pattern. Although this is usually the match chafacter for the end anchor, the set will + * match either the beginning or the end of the text, depending on its placement. For + * example: + * + * <pre> + * $x = [a-z$]; # match 'a' through 'z' OR anchor + * $x 1 > 2; # match '1' after a-z or at the start + * 3 $x > 4; # match '3' before a-z or at the end + * </pre> + * + * <p><b>Example</b> + * + * <p>The following example rules illustrate many of the features of + * the rule language. + * + * <table border="0" cellpadding="4"> + * <tr> + * <td style="vertical-align: top;">Rule 1.</td> + * <td style="vertical-align: top; write-space: nowrap;"><code>abc{def}>x|y</code></td> + * </tr> + * <tr> + * <td style="vertical-align: top;">Rule 2.</td> + * <td style="vertical-align: top; write-space: nowrap;"><code>xyz>r</code></td> + * </tr> + * <tr> + * <td style="vertical-align: top;">Rule 3.</td> + * <td style="vertical-align: top; write-space: nowrap;"><code>yz>q</code></td> + * </tr> + * </table> + * + * <p>Applying these rules to the string "<code>adefabcdefz</code>" + * yields the following results: + * + * <table border="0" cellpadding="4"> + * <tr> + * <td style="vertical-align: top; write-space: nowrap;"><code>|adefabcdefz</code></td> + * <td style="vertical-align: top;">Initial state, no rules match. Advance + * cursor.</td> + * </tr> + * <tr> + * <td style="vertical-align: top; write-space: nowrap;"><code>a|defabcdefz</code></td> + * <td style="vertical-align: top;">Still no match. Rule 1 does not match + * because the preceding context is not present.</td> + * </tr> + * <tr> + * <td style="vertical-align: top; write-space: nowrap;"><code>ad|efabcdefz</code></td> + * <td style="vertical-align: top;">Still no match. Keep advancing until + * there is a match...</td> + * </tr> + * <tr> + * <td style="vertical-align: top; write-space: nowrap;"><code>ade|fabcdefz</code></td> + * <td style="vertical-align: top;">...</td> + * </tr> + * <tr> + * <td style="vertical-align: top; write-space: nowrap;"><code>adef|abcdefz</code></td> + * <td style="vertical-align: top;">...</td> + * </tr> + * <tr> + * <td style="vertical-align: top; write-space: nowrap;"><code>adefa|bcdefz</code></td> + * <td style="vertical-align: top;">...</td> + * </tr> + * <tr> + * <td style="vertical-align: top; write-space: nowrap;"><code>adefab|cdefz</code></td> + * <td style="vertical-align: top;">...</td> + * </tr> + * <tr> + * <td style="vertical-align: top; write-space: nowrap;"><code>adefabc|defz</code></td> + * <td style="vertical-align: top;">Rule 1 matches; replace "<code>def</code>" + * with "<code>xy</code>" and back up the cursor + * to before the '<code>y</code>'.</td> + * </tr> + * <tr> + * <td style="vertical-align: top; write-space: nowrap;"><code>adefabcx|yz</code></td> + * <td style="vertical-align: top;">Although "<code>xyz</code>" is + * present, rule 2 does not match because the cursor is + * before the '<code>y</code>', not before the '<code>x</code>'. + * Rule 3 does match. Replace "<code>yz</code>" + * with "<code>q</code>".</td> + * </tr> + * <tr> + * <td style="vertical-align: top; write-space: nowrap;"><code>adefabcxq|</code></td> + * <td style="vertical-align: top;">The cursor is at the end; + * transliteration is complete.</td> + * </tr> + * </table> + * + * <p>The order of rules is significant. If multiple rules may match + * at some point, the first matching rule is applied. + * + * <p>Forward and reverse rules may have an empty output string. + * Otherwise, an empty left or right hand side of any statement is a + * syntax error. + * + * <p>Single quotes are used to quote any character other than a + * digit or letter. To specify a single quote itself, inside or + * outside of quotes, use two single quotes in a row. For example, + * the rule "<code>'>'>o''clock</code>" changes the + * string "<code>></code>" to the string "<code>o'clock</code>". + * + * <p><b>Notes</b> + * + * <p>While a Transliterator is being built from rules, it checks that + * the rules are added in proper order. For example, if the rule + * "a>x" is followed by the rule "ab>y", + * then the second rule will throw an exception. The reason is that + * the second rule can never be triggered, since the first rule + * always matches anything it matches. In other words, the first + * rule <em>masks</em> the second rule. + * * @author Alan Liu * @stable ICU 2.0 */ @@ -499,9 +744,9 @@ public: * for details. * @param text the buffer holding transliterated and * untransliterated text - * @param index an array of three integers. See {@link #transliterate(Replaceable&, UTransPosition&, const UnicodeString*, UErrorCode&) const }. + * @param index an array of three integers. * @param status Output param to filled in with a success or an error. - * @see #transliterate(Replaceable, int[], String) + * @see #transliterate(Replaceable&, UTransPosition&, const UnicodeString&, UErrorCode &) const * @stable ICU 2.0 */ virtual void transliterate(Replaceable& text, UTransPosition& index, @@ -632,7 +877,7 @@ public: /** * Transliterate a substring of text, as specified by index, taking filters * into account. This method is for subclasses that need to delegate to - * another transliterator, such as CompoundTransliterator. + * another transliterator. * @param text the text to be transliterated * @param index the position indices * @param incremental if TRUE, then assume more characters may be inserted @@ -846,17 +1091,19 @@ public: /** * Returns a <code>Transliterator</code> object constructed from - * the given rule string. This will be a RuleBasedTransliterator, + * the given rule string. This will be a rule-based Transliterator, * if the rule string contains only rules, or a - * CompoundTransliterator, if it contains ID blocks, or a - * NullTransliterator, if it contains ID blocks which parse as + * compound Transliterator, if it contains ID blocks, or a + * null Transliterator, if it contains ID blocks which parse as * empty for the given direction. + * * @param ID the id for the transliterator. * @param rules rules, separated by ';' * @param dir either FORWARD or REVERSE. - * @param parseError Struct to recieve information on position + * @param parseError Struct to receive information on position * of error if an error is encountered * @param status Output param set to success/failure code. + * @return a newly created Transliterator * @stable ICU 2.0 */ static Transliterator* U_EXPORT2 createFromRules(const UnicodeString& ID, diff --git a/deps/icu-small/source/i18n/unicode/tzfmt.h b/deps/icu-small/source/i18n/unicode/tzfmt.h index 633cd8dc69..d2aa768b8c 100644 --- a/deps/icu-small/source/i18n/unicode/tzfmt.h +++ b/deps/icu-small/source/i18n/unicode/tzfmt.h @@ -237,10 +237,10 @@ typedef enum UTimeZoneFormatParseOption { */ UTZFMT_PARSE_OPTION_ALL_STYLES = 0x01, /** - * When parsing a time zone display name in UTZFMT_STYLE_SPECIFIC_SHORT, + * When parsing a time zone display name in \link UTZFMT_STYLE_SPECIFIC_SHORT \endlink, * look for the IANA tz database compatible zone abbreviations in addition - * to the localized names coming from the {@link TimeZoneNames} currently - * used by the {@link TimeZoneFormat}. + * to the localized names coming from the icu::TimeZoneNames currently + * used by the icu::TimeZoneFormat. * @stable ICU 54 */ UTZFMT_PARSE_OPTION_TZ_DATABASE_ABBREVIATIONS = 0x02 diff --git a/deps/icu-small/source/i18n/unicode/ucal.h b/deps/icu-small/source/i18n/unicode/ucal.h index 10d8bc5274..889a1ec51d 100644 --- a/deps/icu-small/source/i18n/unicode/ucal.h +++ b/deps/icu-small/source/i18n/unicode/ucal.h @@ -139,6 +139,19 @@ * For example, subtracting 5 days from the date <code>September 12, 1996</code> * results in <code>September 7, 1996</code>. * + * <p> + * The Japanese calendar uses a combination of era name and year number. + * When an emperor of Japan abdicates and a new emperor ascends the throne, + * a new era is declared and year number is reset to 1. Even if the date of + * abdication is scheduled ahead of time, the new era name might not be + * announced until just before the date. In such case, ICU4C may include + * a start date of future era without actual era name, but not enabled + * by default. ICU4C users who want to test the behavior of the future era + * can enable the tentative era by: + * <ul> + * <li>Environment variable <code>ICU_ENABLE_TENTATIVE_ERA=true</code>.</li> + * </ul> + * * @stable ICU 2.0 */ diff --git a/deps/icu-small/source/i18n/unicode/ucol.h b/deps/icu-small/source/i18n/unicode/ucol.h index b5bacbfcb4..f084ac61e6 100644 --- a/deps/icu-small/source/i18n/unicode/ucol.h +++ b/deps/icu-small/source/i18n/unicode/ucol.h @@ -1149,7 +1149,7 @@ ucol_getUCAVersion(const UCollator* coll, UVersionInfo info); * The recommended way to achieve "merged" sorting is by * concatenating strings with U+FFFE between them. * The concatenation has the same sort order as the merged sort keys, - * but merge(getSortKey(str1), getSortKey(str2)) may differ from getSortKey(str1 + '\uFFFE' + str2). + * but merge(getSortKey(str1), getSortKey(str2)) may differ from getSortKey(str1 + '\\uFFFE' + str2). * Using strings with U+FFFE may yield shorter sort keys. * * For details about Sort Key Features see @@ -1294,6 +1294,7 @@ U_STABLE uint32_t U_EXPORT2 ucol_getVariableTop(const UCollator *coll, UErrorCod * the top of one of the supported reordering groups, * and it must not be beyond the last of those groups. * See ucol_setMaxVariable(). + * @param coll collator to be set * @param varTop primary weight, as returned by ucol_setVariableTop or ucol_getVariableTop * @param status error code * @see ucol_getVariableTop diff --git a/deps/icu-small/source/i18n/unicode/ugender.h b/deps/icu-small/source/i18n/unicode/ugender.h index d015a2300c..903f3dd5de 100644 --- a/deps/icu-small/source/i18n/unicode/ugender.h +++ b/deps/icu-small/source/i18n/unicode/ugender.h @@ -49,11 +49,11 @@ enum UGender { */ typedef enum UGender UGender; +struct UGenderInfo; /** * Opaque UGenderInfo object for use in C programs. * @stable ICU 50 */ -struct UGenderInfo; typedef struct UGenderInfo UGenderInfo; /** @@ -77,7 +77,7 @@ ugender_getInstance(const char *locale, UErrorCode *status); * @stable ICU 50 */ U_STABLE UGender U_EXPORT2 -ugender_getListGender(const UGenderInfo* genderinfo, const UGender *genders, int32_t size, UErrorCode *status); +ugender_getListGender(const UGenderInfo* genderInfo, const UGender *genders, int32_t size, UErrorCode *status); #endif /* #if !UCONFIG_NO_FORMATTING */ diff --git a/deps/icu-small/source/i18n/unicode/ulistformatter.h b/deps/icu-small/source/i18n/unicode/ulistformatter.h new file mode 100644 index 0000000000..242d7d2164 --- /dev/null +++ b/deps/icu-small/source/i18n/unicode/ulistformatter.h @@ -0,0 +1,150 @@ +// © 2016 and later: Unicode, Inc. and others. +// License & terms of use: http://www.unicode.org/copyright.html +/* +***************************************************************************************** +* Copyright (C) 2015-2016, International Business Machines +* Corporation and others. All Rights Reserved. +***************************************************************************************** +*/ + +#ifndef ULISTFORMATTER_H +#define ULISTFORMATTER_H + +#include "unicode/utypes.h" + +#if !UCONFIG_NO_FORMATTING + +#include "unicode/localpointer.h" + +/** + * \file + * \brief C API: Format a list in a locale-appropriate way. + * + * A UListFormatter is used to format a list of items in a locale-appropriate way, + * using data from CLDR. + * Example: Input data ["Alice", "Bob", "Charlie", "Delta"] will be formatted + * as "Alice, Bob, Charlie, and Delta" in English. + */ + +/** + * Opaque UListFormatter object for use in C + * @stable ICU 55 + */ +struct UListFormatter; +typedef struct UListFormatter UListFormatter; /**< C typedef for struct UListFormatter. @stable ICU 55 */ + +#ifndef U_HIDE_DRAFT_API +/** + * FieldPosition and UFieldPosition selectors for format fields + * defined by ListFormatter. + * @draft ICU 63 + */ +typedef enum UListFormatterField { + /** + * The literal text in the result which came from the resources. + * @draft ICU 63 + */ + ULISTFMT_LITERAL_FIELD, + /** + * The element text in the result which came from the input strings. + * @draft ICU 63 + */ + ULISTFMT_ELEMENT_FIELD +} UListFormatterField; +#endif // U_HIDE_DRAFT_API + +/** + * Open a new UListFormatter object using the rules for a given locale. + * @param locale + * The locale whose rules should be used; may be NULL for + * default locale. + * @param status + * A pointer to a standard ICU UErrorCode (input/output parameter). + * Its input value must pass the U_SUCCESS() test, or else the + * function returns immediately. The caller should check its output + * value with U_FAILURE(), or use with function chaining (see User + * Guide for details). + * @return + * A pointer to a UListFormatter object for the specified locale, + * or NULL if an error occurred. + * @stable ICU 55 + */ +U_CAPI UListFormatter* U_EXPORT2 +ulistfmt_open(const char* locale, + UErrorCode* status); + +/** + * Close a UListFormatter object. Once closed it may no longer be used. + * @param listfmt + * The UListFormatter object to close. + * @stable ICU 55 + */ +U_CAPI void U_EXPORT2 +ulistfmt_close(UListFormatter *listfmt); + + +#if U_SHOW_CPLUSPLUS_API + +U_NAMESPACE_BEGIN + +/** + * \class LocalUListFormatterPointer + * "Smart pointer" class, closes a UListFormatter via ulistfmt_close(). + * For most methods see the LocalPointerBase base class. + * + * @see LocalPointerBase + * @see LocalPointer + * @stable ICU 55 + */ +U_DEFINE_LOCAL_OPEN_POINTER(LocalUListFormatterPointer, UListFormatter, ulistfmt_close); + +U_NAMESPACE_END + +#endif + +/** + * Formats a list of strings using the conventions established for the + * UListFormatter object. + * @param listfmt + * The UListFormatter object specifying the list conventions. + * @param strings + * An array of pointers to UChar strings; the array length is + * specified by stringCount. Must be non-NULL if stringCount > 0. + * @param stringLengths + * An array of string lengths corresponding to the strings[] + * parameter; any individual length value may be negative to indicate + * that the corresponding strings[] entry is 0-terminated, or + * stringLengths itself may be NULL if all of the strings are + * 0-terminated. If non-NULL, the stringLengths array must have + * stringCount entries. + * @param stringCount + * the number of entries in strings[], and the number of entries + * in the stringLengths array if it is not NULL. Must be >= 0. + * @param result + * A pointer to a buffer to receive the formatted list. + * @param resultCapacity + * The maximum size of result. + * @param status + * A pointer to a standard ICU UErrorCode (input/output parameter). + * Its input value must pass the U_SUCCESS() test, or else the + * function returns immediately. The caller should check its output + * value with U_FAILURE(), or use with function chaining (see User + * Guide for details). + * @return + * The total buffer size needed; if greater than resultLength, the + * output was truncated. May be <=0 if unable to determine the + * total buffer size needed (e.g. for illegal arguments). + * @stable ICU 55 + */ +U_CAPI int32_t U_EXPORT2 +ulistfmt_format(const UListFormatter* listfmt, + const UChar* const strings[], + const int32_t * stringLengths, + int32_t stringCount, + UChar* result, + int32_t resultCapacity, + UErrorCode* status); + +#endif /* #if !UCONFIG_NO_FORMATTING */ + +#endif diff --git a/deps/icu-small/source/i18n/unicode/unumberformatter.h b/deps/icu-small/source/i18n/unicode/unumberformatter.h index b37f80c503..d05b15cdec 100644 --- a/deps/icu-small/source/i18n/unicode/unumberformatter.h +++ b/deps/icu-small/source/i18n/unicode/unumberformatter.h @@ -91,7 +91,7 @@ * </ul> * * <p> - * This enum is similar to {@link com.ibm.icu.text.MeasureFormat.FormatWidth}. + * This enum is similar to {@link UMeasureFormatWidth}. * * @draft ICU 60 */ @@ -190,10 +190,9 @@ typedef enum UNumberUnitWidth { * Note: This enum specifies the strategy for grouping sizes. To set which character to use as the * grouping separator, use the "symbols" setter. * - * @draft ICU 61 -- TODO: This should be renamed to UNumberGroupingStrategy before promoting to stable, - * for consistency with the other enums. + * @draft ICU 63 */ -typedef enum UGroupingStrategy { +typedef enum UNumberGroupingStrategy { /** * Do not display grouping separators in any locale. * @@ -254,16 +253,28 @@ typedef enum UGroupingStrategy { * * @draft ICU 61 */ - UNUM_GROUPING_THOUSANDS, + UNUM_GROUPING_THOUSANDS +#ifndef U_HIDE_INTERNAL_API + , /** - * One more than the highest UGroupingStrategy value. + * One more than the highest UNumberGroupingStrategy value. * * @internal ICU 62: The numeric value may change over time; see ICU ticket #12420. */ UNUM_GROUPING_COUNT +#endif /* U_HIDE_INTERNAL_API */ + +} UNumberGroupingStrategy; + +#ifndef U_HIDE_DEPRECATED_API +/** + * Old name for compatibility: will be removed in ICU 64. + * @deprecated ICU 63 + */ +typedef UNumberGroupingStrategy UGroupingStrategy; +#endif /* U_HIDE_DEPRECATED_API */ -} UGroupingStrategy; #endif /* U_HIDE_DRAFT_API */ #ifndef U_HIDE_DRAFT_API @@ -398,6 +409,8 @@ typedef enum UNumberDecimalSeparatorDisplay { #endif /* U_HIDE_DRAFT_API */ #ifndef U_HIDE_DRAFT_API + +struct UNumberFormatter; /** * C-compatible version of icu::number::LocalizedNumberFormatter. * @@ -405,10 +418,9 @@ typedef enum UNumberDecimalSeparatorDisplay { * * @draft ICU 62 */ -struct UNumberFormatter; typedef struct UNumberFormatter UNumberFormatter; - +struct UFormattedNumber; /** * C-compatible version of icu::number::FormattedNumber. * @@ -416,7 +428,6 @@ typedef struct UNumberFormatter UNumberFormatter; * * @draft ICU 62 */ -struct UFormattedNumber; typedef struct UFormattedNumber UFormattedNumber; @@ -559,7 +570,8 @@ unumf_resultToString(const UFormattedNumber* uresult, UChar* buffer, int32_t buf * * NOTE: All fields of the UFieldPosition must be initialized before calling this method. * - * @param fieldPosition + * @param uresult The object containing the formatted number. + * @param ufpos * Input+output variable. On input, the "field" property determines which field to look up, * and the "endIndex" property determines where to begin the search. On output, the * "beginIndex" field is set to the beginning of the first occurrence of the field after the @@ -580,7 +592,7 @@ unumf_resultNextFieldPosition(const UFormattedNumber* uresult, UFieldPosition* u * If you need information on only one field, use unumf_resultNextFieldPosition(). * * @param uresult The object containing the formatted number. - * @param fpositer + * @param ufpositer * A pointer to a UFieldPositionIterator created by {@link #ufieldpositer_open}. Iteration * information already present in the UFieldPositionIterator is deleted, and the iterator is reset * to apply to the fields in the formatted string created by this function call. The field values diff --git a/deps/icu-small/source/i18n/unicode/usearch.h b/deps/icu-small/source/i18n/unicode/usearch.h index 600f9142b4..6b495ef001 100644 --- a/deps/icu-small/source/i18n/unicode/usearch.h +++ b/deps/icu-small/source/i18n/unicode/usearch.h @@ -257,10 +257,9 @@ typedef enum { * match an e with the same diacritic or a plain e in the searched text. * * This option is similar to "asymmetric search" as described in - * <a href="http://www.unicode.org/reports/tr10/#Asymmetric_Search"> - * UTS #10 Unicode Collation Algorithm</a, but also allows unmarked - * characters in the searched text to match marked or unmarked versions of - * that character in the pattern. + * [UTS #10 Unicode Collation Algorithm](http://www.unicode.org/reports/tr10/#Asymmetric_Search), + * but also allows unmarked characters in the searched text to match + * marked or unmarked versions of that character in the pattern. * * @stable ICU 4.4 */ diff --git a/deps/icu-small/source/i18n/unicode/uspoof.h b/deps/icu-small/source/i18n/unicode/uspoof.h index 781a88247f..d15ba4b242 100644 --- a/deps/icu-small/source/i18n/unicode/uspoof.h +++ b/deps/icu-small/source/i18n/unicode/uspoof.h @@ -94,7 +94,6 @@ * // ... * \endcode * - * <p> * UTS 39 defines two strings to be <em>confusable</em> if they map to the same <em>skeleton string</em>. A skeleton can * be thought of as a "hash code". {@link uspoof_getSkeleton} computes the skeleton for a particular string, so * the following snippet is equivalent to the example above: @@ -128,7 +127,6 @@ * free(skel2); * \endcode * - * <p> * If you need to check if a string is confusable with any string in a dictionary of many strings, rather than calling * {@link uspoof_areConfusable} many times in a loop, {@link uspoof_getSkeleton} can be used instead, as shown below: * @@ -172,14 +170,12 @@ * uspoof_close(sc); * \endcode * - * <p> * <b>Note:</b> Since the Unicode confusables mapping table is frequently updated, confusable skeletons are <em>not</em> * guaranteed to be the same between ICU releases. We therefore recommend that you always compute confusable skeletons * at runtime and do not rely on creating a permanent, or difficult to update, database of skeletons. * * <h2>Spoof Detection</h2> * - * <p> * The following snippet shows a minimal example of using <code>USpoofChecker</code> to perform spoof detection on a * string: * @@ -204,16 +200,13 @@ * uset_close(allowed); * \endcode * - * <p> * As in the case for confusability checking, it is good practice to create one <code>USpoofChecker</code> instance at * startup, and call the cheaper {@link uspoof_check} online. We specify the set of * allowed characters to be those with type RECOMMENDED or INCLUSION, according to the recommendation in UTS 39. * - * <p> * In addition to {@link uspoof_check}, the function {@link uspoof_checkUTF8} is exposed for UTF8-encoded char* strings, * and {@link uspoof_checkUnicodeString} is exposed for C++ programmers. * - * <p> * If the {@link USPOOF_AUX_INFO} check is enabled, a limited amount of information on why a string failed the checks * is available in the returned bitmask. For complete information, use the {@link uspoof_check2} class of functions * with a {@link USpoofCheckResult} parameter: @@ -274,7 +267,6 @@ * // Explicit cleanup not necessary. * \endcode * - * <p> * The return value is a bitmask of the checks that failed. In this case, there was one check that failed: * {@link USPOOF_RESTRICTION_LEVEL}, corresponding to the fifth bit (16). The possible checks are: * @@ -307,7 +299,6 @@ * uspoof_close(sc); * \endcode * - * <p> * Here is an example in C++ showing how to compute the restriction level of a string: * * \code{.cpp} @@ -334,11 +325,9 @@ * printf("Restriction level: %#010x (status: %s)\n", restrictionLevel, u_errorName(status)); * \endcode * - * <p> * The code '0x50000000' corresponds to the restriction level USPOOF_MINIMALLY_RESTRICTIVE. Since * USPOOF_MINIMALLY_RESTRICTIVE is weaker than USPOOF_MODERATELY_RESTRICTIVE, the string fails the check. * - * <p> * <b>Note:</b> The Restriction Level is the most powerful of the checks. The full logic is documented in * <a href="http://unicode.org/reports/tr39/#Restriction_Level_Detection">UTS 39</a>, but the basic idea is that strings * are restricted to contain characters from only a single script, <em>except</em> that most scripts are allowed to have @@ -352,15 +341,12 @@ * * <h2>Additional Information</h2> * - * <p> * A <code>USpoofChecker</code> instance may be used repeatedly to perform checks on any number of identifiers. * - * <p> * <b>Thread Safety:</b> The test functions for checking a single identifier, or for testing whether * two identifiers are possible confusable, are thread safe. They may called concurrently, from multiple threads, * using the same USpoofChecker instance. * - * <p> * More generally, the standard ICU thread safety rules apply: functions that take a const USpoofChecker parameter are * thread safe. Those that take a non-const USpoofChecker are not thread safe.. * @@ -1219,14 +1205,21 @@ U_NAMESPACE_BEGIN /** * \class LocalUSpoofCheckResultPointer - * "Smart pointer" class, closes a USpoofCheckResult via {@link uspoof_closeCheckResult}. + * "Smart pointer" class, closes a USpoofCheckResult via `uspoof_closeCheckResult()`. * For most methods see the LocalPointerBase base class. * * @see LocalPointerBase * @see LocalPointer * @stable ICU 58 */ + +/** + * \cond + * Note: Doxygen is giving a bogus warning on this U_DEFINE_LOCAL_OPEN_POINTER. + * For now, suppress with a Doxygen cond + */ U_DEFINE_LOCAL_OPEN_POINTER(LocalUSpoofCheckResultPointer, USpoofCheckResult, uspoof_closeCheckResult); +/** \endcond */ U_NAMESPACE_END diff --git a/deps/icu-small/source/i18n/uregex.cpp b/deps/icu-small/source/i18n/uregex.cpp index 3703843634..f504aec91b 100644 --- a/deps/icu-small/source/i18n/uregex.cpp +++ b/deps/icu-small/source/i18n/uregex.cpp @@ -1654,8 +1654,8 @@ int32_t RegexCImpl::appendTail(RegularExpression *regexp, } else if (UTEXT_USES_U16(m->fInputText)) { srcIdx = (int32_t)nativeIdx; } else { - UErrorCode status = U_ZERO_ERROR; - srcIdx = utext_extract(m->fInputText, 0, nativeIdx, NULL, 0, &status); + UErrorCode newStatus = U_ZERO_ERROR; + srcIdx = utext_extract(m->fInputText, 0, nativeIdx, NULL, 0, &newStatus); } for (;;) { diff --git a/deps/icu-small/source/i18n/usearch.cpp b/deps/icu-small/source/i18n/usearch.cpp index e1e6c28e2b..0e4cca77a1 100644 --- a/deps/icu-small/source/i18n/usearch.cpp +++ b/deps/icu-small/source/i18n/usearch.cpp @@ -498,7 +498,7 @@ inline void setShiftTable(int16_t shift[], int16_t backshift[], for (count = 0; count < cesize; count ++) { // number of ces from right of array to the count int temp = defaultforward - count - 1; - shift[hashFromCE32(cetable[count])] = temp > 1 ? temp : 1; + shift[hashFromCE32(cetable[count])] = temp > 1 ? static_cast<int16_t>(temp) : 1; } shift[hashFromCE32(cetable[cesize])] = 1; // for ignorables we just shift by one. see test examples. diff --git a/deps/icu-small/source/i18n/uspoof_impl.h b/deps/icu-small/source/i18n/uspoof_impl.h index 470a31f2c9..7d6d0f76e3 100644 --- a/deps/icu-small/source/i18n/uspoof_impl.h +++ b/deps/icu-small/source/i18n/uspoof_impl.h @@ -20,7 +20,7 @@ #include "unicode/uspoof.h" #include "unicode/uscript.h" #include "unicode/udata.h" - +#include "udataswp.h" #include "utrie2.h" #if !UCONFIG_NO_NORMALIZATION diff --git a/deps/icu-small/source/i18n/vtzone.cpp b/deps/icu-small/source/i18n/vtzone.cpp index 0c76c9b6c9..e39eada51b 100644 --- a/deps/icu-small/source/i18n/vtzone.cpp +++ b/deps/icu-small/source/i18n/vtzone.cpp @@ -135,7 +135,7 @@ static UnicodeString& appendAsciiDigits(int32_t number, uint8_t length, UnicodeS digits[i++] = number % 10; number /= 10; } while (number != 0); - length = i; + length = static_cast<uint8_t>(i); } else { // fixed digits for (i = 0; i < length; i++) { diff --git a/deps/icu-small/source/i18n/zonemeta.cpp b/deps/icu-small/source/i18n/zonemeta.cpp index 02562048a5..b7139a807b 100644 --- a/deps/icu-small/source/i18n/zonemeta.cpp +++ b/deps/icu-small/source/i18n/zonemeta.cpp @@ -319,10 +319,10 @@ ZoneMeta::getCanonicalCLDRID(const UnicodeString &tzid, UErrorCode& status) { id[len] = (char) 0; // Make sure it is null terminated. // replace '/' with ':' - char *p = id; - while (*p++) { - if (*p == '/') { - *p = ':'; + char *q = id; + while (*q++) { + if (*q == '/') { + *q = ':'; } } @@ -850,13 +850,13 @@ ZoneMeta::createCustomTimeZone(int32_t offset) { negative = TRUE; tmp = -offset; } - int32_t hour, min, sec; + uint8_t hour, min, sec; tmp /= 1000; - sec = tmp % 60; + sec = static_cast<uint8_t>(tmp % 60); tmp /= 60; - min = tmp % 60; - hour = tmp / 60; + min = static_cast<uint8_t>(tmp % 60); + hour = static_cast<uint8_t>(tmp / 60); UnicodeString zid; formatCustomID(hour, min, sec, negative, zid); |