diff options
Diffstat (limited to 'deps/icu-small/source/common/uloc_tag.cpp')
| -rw-r--r-- | deps/icu-small/source/common/uloc_tag.cpp | 490 |
1 files changed, 422 insertions, 68 deletions
diff --git a/deps/icu-small/source/common/uloc_tag.cpp b/deps/icu-small/source/common/uloc_tag.cpp index f8337ec0247..8120331c4b9 100644 --- a/deps/icu-small/source/common/uloc_tag.cpp +++ b/deps/icu-small/source/common/uloc_tag.cpp @@ -12,11 +12,13 @@ #include "unicode/putil.h" #include "unicode/uloc.h" #include "ustr_imp.h" +#include "charstr.h" #include "cmemory.h" #include "cstring.h" #include "putilimp.h" #include "uinvchar.h" #include "ulocimp.h" +#include "uvector.h" #include "uassert.h" @@ -77,19 +79,34 @@ static const char LOCALE_TYPE_YES[] = "yes"; #define LANG_UND_LEN 3 +/* + Updated on 2018-09-12 from + https://www.iana.org/assignments/language-subtag-registry/language-subtag-registry . + + This table has 2 parts. The parts for Grandfathered tags is generated by the + following scripts from the IANA language tag registry. + + curl https://www.iana.org/assignments/language-subtag-registry/language-subtag-registry |\ + egrep -A 7 'Type: grandfathered' | \ + egrep 'Tag|Prefe' | grep -B1 'Preferred' | grep -v '^--' | \ + awk -n '/Tag/ {printf(" \"%s\", ", $2);} /Preferred/ {printf("\"%s\",\n", $2);}' |\ + tr 'A-Z' 'a-z' + + + The 2nd part is made of five ICU-specific entries. They're kept for + the backward compatibility for now, even though there are no preferred + values. They may have to be removed for the strict BCP 47 compliance. + +*/ static const char* const GRANDFATHERED[] = { /* grandfathered preferred */ "art-lojban", "jbo", - "cel-gaulish", "xtg-x-cel-gaulish", - "en-GB-oed", "en-GB-x-oed", + "en-gb-oed", "en-gb-oxendict", "i-ami", "ami", "i-bnn", "bnn", - "i-default", "en-x-i-default", - "i-enochian", "und-x-i-enochian", "i-hak", "hak", "i-klingon", "tlh", "i-lux", "lb", - "i-mingo", "see-x-i-mingo", "i-navajo", "nv", "i-pwn", "pwn", "i-tao", "tao", @@ -102,17 +119,175 @@ static const char* const GRANDFATHERED[] = { "sgn-ch-de", "sgg", "zh-guoyu", "cmn", "zh-hakka", "hak", - "zh-min", "nan-x-zh-min", "zh-min-nan", "nan", "zh-xiang", "hsn", - NULL, NULL + + // Grandfathered tags with no preferred value in the IANA + // registry. Kept for now for the backward compatibility + // because ICU has mapped them this way. + "cel-gaulish", "xtg-x-cel-gaulish", + "i-default", "en-x-i-default", + "i-enochian", "und-x-i-enochian", + "i-mingo", "see-x-i-mingo", + "zh-min", "nan-x-zh-min", }; +/* + Updated on 2018-09-12 from + https://www.iana.org/assignments/language-subtag-registry/language-subtag-registry . + + The table lists redundant tags with preferred value in the IANA languate tag registry. + It's generated with the following command: + + curl https://www.iana.org/assignments/language-subtag-registry/language-subtag-registry |\ + grep 'Type: redundant' -A 5 | egrep '^(Tag:|Prefer)' | grep -B1 'Preferred' | \ + awk -n '/Tag/ {printf(" \"%s\", ", $2);} /Preferred/ {printf("\"%s\",\n", $2);}' | \ + tr 'A-Z' 'a-z' + + In addition, ja-latn-hepburn-heploc is mapped to ja-latn-alalc97 because + a variant tag 'hepburn-heploc' has the preferred subtag, 'alaic97'. +*/ + +static const char* const REDUNDANT[] = { +// redundant preferred + "sgn-br", "bzs", + "sgn-co", "csn", + "sgn-de", "gsg", + "sgn-dk", "dsl", + "sgn-es", "ssp", + "sgn-fr", "fsl", + "sgn-gb", "bfi", + "sgn-gr", "gss", + "sgn-ie", "isg", + "sgn-it", "ise", + "sgn-jp", "jsl", + "sgn-mx", "mfs", + "sgn-ni", "ncs", + "sgn-nl", "dse", + "sgn-no", "nsl", + "sgn-pt", "psr", + "sgn-se", "swl", + "sgn-us", "ase", + "sgn-za", "sfs", + "zh-cmn", "cmn", + "zh-cmn-hans", "cmn-hans", + "zh-cmn-hant", "cmn-hant", + "zh-gan", "gan", + "zh-wuu", "wuu", + "zh-yue", "yue", + + // variant tag with preferred value + "ja-latn-hepburn-heploc", "ja-latn-alalc97", +}; + +/* + Updated on 2018-09-12 from + https://www.iana.org/assignments/language-subtag-registry/language-subtag-registry . + + grep 'Type: language' -A 7 language-subtag-registry | egrep 'Subtag|Prefe' | \ + grep -B1 'Preferred' | grep -v '^--' | \ + awk -n '/Subtag/ {printf(" \"%s\", ", $2);} /Preferred/ {printf("\"%s\",\n", $2);}' + + Make sure that 2-letter language subtags come before 3-letter subtags. +*/ static const char DEPRECATEDLANGS[][4] = { /* deprecated new */ + "in", "id", "iw", "he", "ji", "yi", - "in", "id" + "jw", "jv", + "mo", "ro", + "aam", "aas", + "adp", "dz", + "aue", "ktz", + "ayx", "nun", + "bgm", "bcg", + "bjd", "drl", + "ccq", "rki", + "cjr", "mom", + "cka", "cmr", + "cmk", "xch", + "coy", "pij", + "cqu", "quh", + "drh", "khk", + "drw", "prs", + "gav", "dev", + "gfx", "vaj", + "ggn", "gvr", + "gti", "nyc", + "guv", "duz", + "hrr", "jal", + "ibi", "opa", + "ilw", "gal", + "jeg", "oyb", + "kgc", "tdf", + "kgh", "kml", + "koj", "kwv", + "krm", "bmf", + "ktr", "dtp", + "kvs", "gdj", + "kwq", "yam", + "kxe", "tvd", + "kzj", "dtp", + "kzt", "dtp", + "lii", "raq", + "lmm", "rmx", + "meg", "cir", + "mst", "mry", + "mwj", "vaj", + "myt", "mry", + "nad", "xny", + "ncp", "kdz", + "nnx", "ngv", + "nts", "pij", + "oun", "vaj", + "pcr", "adx", + "pmc", "huw", + "pmu", "phr", + "ppa", "bfy", + "ppr", "lcq", + "pry", "prt", + "puz", "pub", + "sca", "hle", + "skk", "oyb", + "tdu", "dtp", + "thc", "tpo", + "thx", "oyb", + "tie", "ras", + "tkk", "twm", + "tlw", "weo", + "tmp", "tyj", + "tne", "kak", + "tnf", "prs", + "tsf", "taj", + "uok", "ema", + "xba", "cax", + "xia", "acn", + "xkh", "waw", + "xsj", "suj", + "ybd", "rki", + "yma", "lrr", + "ymt", "mtm", + "yos", "zom", + "yuu", "yug", +}; + +/* + Updated on 2018-04-24 from + + curl https://www.iana.org/assignments/language-subtag-registry/language-subtag-registry | \ + grep 'Type: region' -A 7 | egrep 'Subtag|Prefe' | \ + grep -B1 'Preferred' | \ + awk -n '/Subtag/ {printf(" \"%s\", ", $2);} /Preferred/ {printf("\"%s\",\n", $2);}' +*/ +static const char DEPRECATEDREGIONS[][3] = { +/* deprecated new */ + "BU", "MM", + "DD", "DE", + "FX", "FR", + "TP", "TL", + "YD", "YE", + "ZR", "CD", }; /* @@ -172,6 +347,46 @@ static const char* ultag_getGrandfathered(const ULanguageTag* langtag); #endif +namespace { + +// Helper class to memory manage CharString objects. +// Only ever stack-allocated, does not need to inherit UMemory. +class CharStringPool { +public: + CharStringPool() : status(U_ZERO_ERROR), pool(&deleter, nullptr, status) {} + ~CharStringPool() = default; + + CharStringPool(const CharStringPool&) = delete; + CharStringPool& operator=(const CharStringPool&) = delete; + + icu::CharString* create() { + if (U_FAILURE(status)) { + return nullptr; + } + icu::CharString* const obj = new icu::CharString; + if (obj == nullptr) { + status = U_MEMORY_ALLOCATION_ERROR; + return nullptr; + } + pool.addElement(obj, status); + if (U_FAILURE(status)) { + delete obj; + return nullptr; + } + return obj; + } + +private: + static void U_CALLCONV deleter(void* obj) { + delete static_cast<icu::CharString*>(obj); + } + + UErrorCode status; + icu::UVector pool; +}; + +} // namespace + /* * ------------------------------------------------- * @@ -675,6 +890,11 @@ _appendLanguageToLanguageTag(const char* localeID, char* appendAt, int32_t capac } else { /* resolve deprecated */ for (i = 0; i < UPRV_LENGTHOF(DEPRECATEDLANGS); i += 2) { + // 2-letter deprecated subtags are listede before 3-letter + // ones in DEPRECATEDLANGS[]. Get out of loop on coming + // across the 1st 3-letter subtag, if the input is a 2-letter code. + // to avoid continuing to try when there's no match. + if (uprv_strlen(buf) < uprv_strlen(DEPRECATEDLANGS[i])) break; if (uprv_compareInvCharsAsAscii(buf, DEPRECATEDLANGS[i]) == 0) { uprv_strcpy(buf, DEPRECATEDLANGS[i + 1]); len = (int32_t)uprv_strlen(buf); @@ -721,7 +941,6 @@ _appendScriptToLanguageTag(const char* localeID, char* appendAt, int32_t capacit *(appendAt + reslen) = SEP; } reslen++; - if (reslen < capacity) { uprv_memcpy(appendAt + reslen, buf, uprv_min(len, capacity - reslen)); } @@ -763,6 +982,14 @@ _appendRegionToLanguageTag(const char* localeID, char* appendAt, int32_t capacit *(appendAt + reslen) = SEP; } reslen++; + /* resolve deprecated */ + for (int i = 0; i < UPRV_LENGTHOF(DEPRECATEDREGIONS); i += 2) { + if (uprv_compareInvCharsAsAscii(buf, DEPRECATEDREGIONS[i]) == 0) { + uprv_strcpy(buf, DEPRECATEDREGIONS[i + 1]); + len = (int32_t)uprv_strlen(buf); + break; + } + } if (reslen < capacity) { uprv_memcpy(appendAt + reslen, buf, uprv_min(len, capacity - reslen)); @@ -900,7 +1127,6 @@ _appendVariantsToLanguageTag(const char* localeID, char* appendAt, int32_t capac static int32_t _appendKeywordsToLanguageTag(const char* localeID, char* appendAt, int32_t capacity, UBool strict, UBool hadPosix, UErrorCode* status) { - char buf[ULOC_KEYWORD_AND_VALUES_CAPACITY]; char attrBuf[ULOC_KEYWORD_AND_VALUES_CAPACITY] = { 0 }; int32_t attrBufLength = 0; UEnumeration *keywordEnum = NULL; @@ -920,22 +1146,48 @@ _appendKeywordsToLanguageTag(const char* localeID, char* appendAt, int32_t capac AttributeListEntry *firstAttr = NULL; AttributeListEntry *attr; char *attrValue; - char extBuf[ULOC_KEYWORD_AND_VALUES_CAPACITY]; - char *pExtBuf = extBuf; - int32_t extBufCapacity = sizeof(extBuf); + CharStringPool extBufPool; const char *bcpKey=nullptr, *bcpValue=nullptr; UErrorCode tmpStatus = U_ZERO_ERROR; int32_t keylen; UBool isBcpUExt; while (TRUE) { + icu::CharString buf; key = uenum_next(keywordEnum, NULL, status); if (key == NULL) { break; } - len = uloc_getKeywordValue(localeID, key, buf, sizeof(buf), &tmpStatus); - /* buf must be null-terminated */ - if (U_FAILURE(tmpStatus) || tmpStatus == U_STRING_NOT_TERMINATED_WARNING) { + char* buffer; + int32_t resultCapacity = ULOC_KEYWORD_AND_VALUES_CAPACITY; + + for (;;) { + buffer = buf.getAppendBuffer( + /*minCapacity=*/resultCapacity, + /*desiredCapacityHint=*/resultCapacity, + resultCapacity, + tmpStatus); + + if (U_FAILURE(tmpStatus)) { + break; + } + + len = uloc_getKeywordValue( + localeID, key, buffer, resultCapacity, &tmpStatus); + + if (tmpStatus != U_BUFFER_OVERFLOW_ERROR) { + break; + } + + resultCapacity = len; + tmpStatus = U_ZERO_ERROR; + } + + if (U_FAILURE(tmpStatus)) { + if (tmpStatus == U_MEMORY_ALLOCATION_ERROR) { + *status = U_MEMORY_ALLOCATION_ERROR; + break; + } if (strict) { *status = U_ILLEGAL_ARGUMENT_ERROR; break; @@ -945,6 +1197,11 @@ _appendKeywordsToLanguageTag(const char* localeID, char* appendAt, int32_t capac continue; } + buf.append(buffer, len, tmpStatus); + if (tmpStatus == U_STRING_NOT_TERMINATED_WARNING) { + tmpStatus = U_ZERO_ERROR; // Terminators provided by CharString. + } + keylen = (int32_t)uprv_strlen(key); isBcpUExt = (keylen > 1); @@ -1007,7 +1264,7 @@ _appendKeywordsToLanguageTag(const char* localeID, char* appendAt, int32_t capac } /* we've checked buf is null-terminated above */ - bcpValue = uloc_toUnicodeLocaleType(key, buf); + bcpValue = uloc_toUnicodeLocaleType(key, buf.data()); if (bcpValue == NULL) { if (strict) { *status = U_ILLEGAL_ARGUMENT_ERROR; @@ -1015,33 +1272,44 @@ _appendKeywordsToLanguageTag(const char* localeID, char* appendAt, int32_t capac } continue; } - if (bcpValue == buf) { + if (bcpValue == buf.data()) { /* When uloc_toUnicodeLocaleType(key, buf) returns the input value as is, the value is well-formed, but has no known mapping. This implementation normalizes the - the value to lower case + value to lower case */ + icu::CharString* extBuf = extBufPool.create(); + if (extBuf == nullptr) { + *status = U_MEMORY_ALLOCATION_ERROR; + break; + } int32_t bcpValueLen = static_cast<int32_t>(uprv_strlen(bcpValue)); - if (bcpValueLen < extBufCapacity) { - uprv_strcpy(pExtBuf, bcpValue); - T_CString_toLowerCase(pExtBuf); + int32_t resultCapacity; + char* pExtBuf = extBuf->getAppendBuffer( + /*minCapacity=*/bcpValueLen, + /*desiredCapacityHint=*/bcpValueLen, + resultCapacity, + tmpStatus); + if (U_FAILURE(tmpStatus)) { + *status = tmpStatus; + break; + } - bcpValue = pExtBuf; + uprv_strcpy(pExtBuf, bcpValue); + T_CString_toLowerCase(pExtBuf); - pExtBuf += (bcpValueLen + 1); - extBufCapacity -= (bcpValueLen + 1); - } else { - if (strict) { - *status = U_ILLEGAL_ARGUMENT_ERROR; - break; - } - continue; + extBuf->append(pExtBuf, bcpValueLen, tmpStatus); + if (U_FAILURE(tmpStatus)) { + *status = tmpStatus; + break; } + + bcpValue = extBuf->data(); } } else { if (*key == PRIVATEUSE) { - if (!_isPrivateuseValueSubtags(buf, len)) { + if (!_isPrivateuseValueSubtags(buf.data(), len)) { if (strict) { *status = U_ILLEGAL_ARGUMENT_ERROR; break; @@ -1049,7 +1317,7 @@ _appendKeywordsToLanguageTag(const char* localeID, char* appendAt, int32_t capac continue; } } else { - if (!_isExtensionSingleton(key, keylen) || !_isExtensionSubtags(buf, len)) { + if (!_isExtensionSingleton(key, keylen) || !_isExtensionSubtags(buf.data(), len)) { if (strict) { *status = U_ILLEGAL_ARGUMENT_ERROR; break; @@ -1058,20 +1326,17 @@ _appendKeywordsToLanguageTag(const char* localeID, char* appendAt, int32_t capac } } bcpKey = key; - if ((len + 1) < extBufCapacity) { - uprv_memcpy(pExtBuf, buf, len); - bcpValue = pExtBuf; - - pExtBuf += len; - - *pExtBuf = 0; - pExtBuf++; - - extBufCapacity -= (len + 1); - } else { - *status = U_ILLEGAL_ARGUMENT_ERROR; + icu::CharString* extBuf = extBufPool.create(); + if (extBuf == nullptr) { + *status = U_MEMORY_ALLOCATION_ERROR; break; } + extBuf->append(buf.data(), len, tmpStatus); + if (U_FAILURE(tmpStatus)) { + *status = tmpStatus; + break; + } + bcpValue = extBuf->data(); } /* create ExtensionListEntry */ @@ -1242,6 +1507,7 @@ _appendLDMLExtensionAsKeywords(const char* ldmlext, ExtensionListEntry** appendT attrBufIdx += (len + 1); } else { *status = U_ILLEGAL_ARGUMENT_ERROR; + uprv_free(attr); goto cleanup; } @@ -1460,9 +1726,9 @@ _appendLDMLExtensionAsKeywords(const char* ldmlext, ExtensionListEntry** appendT kwd->value = pType; if (!_addExtensionToList(&kwdFirst, kwd, FALSE)) { - *status = U_ILLEGAL_ARGUMENT_ERROR; + // duplicate keyword is allowed, Only the first + // is honored. uprv_free(kwd); - goto cleanup; } } @@ -1836,7 +2102,7 @@ ultag_parse(const char* tag, int32_t tagLen, int32_t* parsedLen, UErrorCode* sta } /* check if the tag is grandfathered */ - for (i = 0; GRANDFATHERED[i] != NULL; i += 2) { + for (i = 0; i < UPRV_LENGTHOF(GRANDFATHERED); i += 2) { if (uprv_stricmp(GRANDFATHERED[i], tagBuf) == 0) { int32_t newTagLength; @@ -1858,6 +2124,37 @@ ultag_parse(const char* tag, int32_t tagLen, int32_t* parsedLen, UErrorCode* sta } } + size_t parsedLenDelta = 0; + if (grandfatheredLen == 0) { + for (i = 0; i < UPRV_LENGTHOF(REDUNDANT); i += 2) { + const char* redundantTag = REDUNDANT[i]; + size_t redundantTagLen = uprv_strlen(redundantTag); + // The preferred tag for a redundant tag is always shorter than redundant + // tag. A redundant tag may or may not be followed by other subtags. + // (i.e. "zh-yue" or "zh-yue-u-co-pinyin"). + if (uprv_strnicmp(redundantTag, tagBuf, static_cast<uint32_t>(redundantTagLen)) == 0) { + const char* redundantTagEnd = tagBuf + redundantTagLen; + if (*redundantTagEnd == '\0' || *redundantTagEnd == SEP) { + const char* preferredTag = REDUNDANT[i + 1]; + size_t preferredTagLen = uprv_strlen(preferredTag); + uprv_strncpy(t->buf, preferredTag, preferredTagLen); + if (*redundantTagEnd == SEP) { + uprv_memmove(tagBuf + preferredTagLen, + redundantTagEnd, + tagLen - redundantTagLen + 1); + } else { + tagBuf[preferredTagLen] = '\0'; + } + // parsedLen should be the length of the input + // before redundantTag is replaced by preferredTag. + // Save the delta to add it back later. + parsedLenDelta = redundantTagLen - preferredTagLen; + break; + } + } + } + } + /* * langtag = language * ["-" script] @@ -1898,10 +2195,13 @@ ultag_parse(const char* tag, int32_t tagLen, int32_t* parsedLen, UErrorCode* sta if (next & LANG) { if (_isLanguageSubtag(pSubtag, subtagLen)) { *pSep = 0; /* terminate */ + // TODO: move deprecated language code handling here. t->language = T_CString_toLowerCase(pSubtag); pLastGoodPosition = pSep; - next = EXTL | SCRT | REGN | VART | EXTS | PRIV; + next = SCRT | REGN | VART | EXTS | PRIV; + if (subtagLen <= 3) + next |= EXTL; continue; } } @@ -1942,6 +2242,7 @@ ultag_parse(const char* tag, int32_t tagLen, int32_t* parsedLen, UErrorCode* sta if (next & REGN) { if (_isRegionSubtag(pSubtag, subtagLen)) { *pSep = 0; + // TODO: move deprecated region code handling here. t->region = T_CString_toUpperCase(pSubtag); pLastGoodPosition = pSep; @@ -2035,7 +2336,7 @@ ultag_parse(const char* tag, int32_t tagLen, int32_t* parsedLen, UErrorCode* sta } } if (next & PRIV) { - if (uprv_tolower(*pSubtag) == PRIVATEUSE) { + if (uprv_tolower(*pSubtag) == PRIVATEUSE && subtagLen == 1) { char *pPrivuseVal; if (pExtension != NULL) { @@ -2138,7 +2439,8 @@ ultag_parse(const char* tag, int32_t tagLen, int32_t* parsedLen, UErrorCode* sta } if (parsedLen != NULL) { - *parsedLen = (grandfatheredLen > 0) ? grandfatheredLen : (int32_t)(pLastGoodPosition - t->buf); + *parsedLen = (grandfatheredLen > 0) ? grandfatheredLen : + (int32_t)(pLastGoodPosition - t->buf + parsedLenDelta); } return t; @@ -2335,31 +2637,66 @@ uloc_toLanguageTag(const char* localeID, int32_t langtagCapacity, UBool strict, UErrorCode* status) { - /* char canonical[ULOC_FULLNAME_CAPACITY]; */ /* See #6822 */ - char canonical[256]; - int32_t reslen = 0; + icu::CharString canonical; + int32_t reslen; UErrorCode tmpStatus = U_ZERO_ERROR; UBool hadPosix = FALSE; const char* pKeywordStart; /* Note: uloc_canonicalize returns "en_US_POSIX" for input locale ID "". See #6835 */ - canonical[0] = 0; - if (uprv_strlen(localeID) > 0) { - uloc_canonicalize(localeID, canonical, sizeof(canonical), &tmpStatus); - if (tmpStatus != U_ZERO_ERROR) { + int32_t resultCapacity = static_cast<int32_t>(uprv_strlen(localeID)); + if (resultCapacity > 0) { + char* buffer; + + for (;;) { + buffer = canonical.getAppendBuffer( + /*minCapacity=*/resultCapacity, + /*desiredCapacityHint=*/resultCapacity, + resultCapacity, + tmpStatus); + + if (U_FAILURE(tmpStatus)) { + *status = tmpStatus; + return 0; + } + + reslen = + uloc_canonicalize(localeID, buffer, resultCapacity, &tmpStatus); + + if (tmpStatus != U_BUFFER_OVERFLOW_ERROR) { + break; + } + + resultCapacity = reslen; + tmpStatus = U_ZERO_ERROR; + } + + if (U_FAILURE(tmpStatus)) { *status = U_ILLEGAL_ARGUMENT_ERROR; return 0; } + + canonical.append(buffer, reslen, tmpStatus); + if (tmpStatus == U_STRING_NOT_TERMINATED_WARNING) { + tmpStatus = U_ZERO_ERROR; // Terminators provided by CharString. + } + + if (U_FAILURE(tmpStatus)) { + *status = tmpStatus; + return 0; + } } + reslen = 0; + /* For handling special case - private use only tag */ - pKeywordStart = locale_getKeywordsStart(canonical); - if (pKeywordStart == canonical) { + pKeywordStart = locale_getKeywordsStart(canonical.data()); + if (pKeywordStart == canonical.data()) { UEnumeration *kwdEnum; int kwdCnt = 0; UBool done = FALSE; - kwdEnum = uloc_openKeywords((const char*)canonical, &tmpStatus); + kwdEnum = uloc_openKeywords(canonical.data(), &tmpStatus); if (kwdEnum != NULL) { kwdCnt = uenum_count(kwdEnum, &tmpStatus); if (kwdCnt == 1) { @@ -2397,12 +2734,12 @@ uloc_toLanguageTag(const char* localeID, } } - reslen += _appendLanguageToLanguageTag(canonical, langtag, langtagCapacity, strict, status); - reslen += _appendScriptToLanguageTag(canonical, langtag + reslen, langtagCapacity - reslen, strict, status); - reslen += _appendRegionToLanguageTag(canonical, langtag + reslen, langtagCapacity - reslen, strict, status); - reslen += _appendVariantsToLanguageTag(canonical, langtag + reslen, langtagCapacity - reslen, strict, &hadPosix, status); - reslen += _appendKeywordsToLanguageTag(canonical, langtag + reslen, langtagCapacity - reslen, strict, hadPosix, status); - reslen += _appendPrivateuseToLanguageTag(canonical, langtag + reslen, langtagCapacity - reslen, strict, hadPosix, status); + reslen += _appendLanguageToLanguageTag(canonical.data(), langtag, langtagCapacity, strict, status); + reslen += _appendScriptToLanguageTag(canonical.data(), langtag + reslen, langtagCapacity - reslen, strict, status); + reslen += _appendRegionToLanguageTag(canonical.data(), langtag + reslen, langtagCapacity - reslen, strict, status); + reslen += _appendVariantsToLanguageTag(canonical.data(), langtag + reslen, langtagCapacity - reslen, strict, &hadPosix, status); + reslen += _appendKeywordsToLanguageTag(canonical.data(), langtag + reslen, langtagCapacity - reslen, strict, hadPosix, status); + reslen += _appendPrivateuseToLanguageTag(canonical.data(), langtag + reslen, langtagCapacity - reslen, strict, hadPosix, status); return reslen; } @@ -2414,6 +2751,23 @@ uloc_forLanguageTag(const char* langtag, int32_t localeIDCapacity, int32_t* parsedLength, UErrorCode* status) { + return ulocimp_forLanguageTag( + langtag, + -1, + localeID, + localeIDCapacity, + parsedLength, + status); +} + + +U_CAPI int32_t U_EXPORT2 +ulocimp_forLanguageTag(const char* langtag, + int32_t tagLen, + char* localeID, + int32_t localeIDCapacity, + int32_t* parsedLength, + UErrorCode* status) { ULanguageTag *lt; int32_t reslen = 0; const char *subtag, *p; @@ -2421,7 +2775,7 @@ uloc_forLanguageTag(const char* langtag, int32_t i, n; UBool noRegion = TRUE; - lt = ultag_parse(langtag, -1, parsedLength, status); + lt = ultag_parse(langtag, tagLen, parsedLength, status); if (U_FAILURE(*status)) { return 0; } |