/**
* Return the script code for a given name, or -1 if not found.
*/
UScriptCode AnyTransliterator::scriptNameToCode(const UnicodeString& name) {
char buf[128];
UScriptCode code;
UErrorCode ec = U_ZERO_ERROR;
name.extract(0, 128, buf, 128, "");
if (uscript_getCode(buf, &code, 1, &ec) != 1 ||
U_FAILURE(ec)) {
code = USCRIPT_INVALID_CODE;
}
return code;
}
/**
* Return the script code for a given name, or -1 if not found.
*/
static UScriptCode scriptNameToCode(const UnicodeString& name) {
char buf[128];
UScriptCode code;
UErrorCode ec = U_ZERO_ERROR;
int32_t nameLen = name.length();
UBool isInvariant = uprv_isInvariantUString(name.getBuffer(), nameLen);
if (isInvariant) {
name.extract(0, nameLen, buf, (int32_t)sizeof(buf), US_INV);
buf[127] = 0; // Make sure that we NULL terminate the string.
}
if (!isInvariant || uscript_getCode(buf, &code, 1, &ec) != 1 || U_FAILURE(ec))
{
code = USCRIPT_INVALID_CODE;
}
return code;
}
TransliteratorSpec::TransliteratorSpec(const UnicodeString& theSpec)
: top(theSpec),
res(0)
{
UErrorCode status = U_ZERO_ERROR;
Locale topLoc("");
LocaleUtility::initLocaleFromName(theSpec, topLoc);
if (!topLoc.isBogus()) {
res = new ResourceBundle(U_ICUDATA_TRANSLIT, topLoc, status);
/* test for NULL */
if (res == 0) {
return;
}
if (U_FAILURE(status) || status == U_USING_DEFAULT_WARNING) {
delete res;
res = 0;
}
}
// Canonicalize script name -or- do locale->script mapping
status = U_ZERO_ERROR;
static const int32_t capacity = 10;
UScriptCode script[capacity]={USCRIPT_INVALID_CODE};
int32_t num = uscript_getCode(CharString().appendInvariantChars(theSpec, status).data(),
script, capacity, &status);
if (num > 0 && script[0] != USCRIPT_INVALID_CODE) {
scriptName = UnicodeString(uscript_getName(script[0]), -1, US_INV);
}
// Canonicalize top
if (res != 0) {
// Canonicalize locale name
UnicodeString locStr;
LocaleUtility::initNameFromLocale(topLoc, locStr);
if (!locStr.isBogus()) {
top = locStr;
}
} else if (scriptName.length() != 0) {
// We are a script; use canonical name
top = scriptName;
}
// assert(spec != top);
reset();
}
void SpoofImpl::addScriptChars(const char *locale, UnicodeSet *allowedChars, UErrorCode &status) {
UScriptCode scripts[30];
int32_t numScripts = uscript_getCode(locale, scripts, sizeof(scripts)/sizeof(UScriptCode), &status);
if (U_FAILURE(status)) {
return;
}
if (status == U_USING_DEFAULT_WARNING) {
status = U_ILLEGAL_ARGUMENT_ERROR;
return;
}
UnicodeSet tmpSet;
int32_t i;
for (i=0; i<numScripts; i++) {
tmpSet.applyIntPropertyValue(UCHAR_SCRIPT, scripts[i], status);
allowedChars->addAll(tmpSet);
}
}
FontMap::FontMap(const char *fileName, le_int16 pointSize, GUISupport *guiSupport, LEErrorCode &status)
: fPointSize(pointSize), fFontCount(0), fAscent(0), fDescent(0), fLeading(0), fGUISupport(guiSupport)
{
le_int32 defaultFont = -1, i, script;
le_bool haveFonts = FALSE;
for (i = 0; i < scriptCodeCount; i += 1) {
fFontIndices[i] = -1;
fFontNames[i] = NULL;
fFontInstances[i] = NULL;
}
if (LE_FAILURE(status)) {
return;
}
char *c, *scriptName, *fontName, *line, buffer[BUFFER_SIZE];
FILE *file;
file = fopen(fileName, "r");
if (file == NULL) {
sprintf(errorMessage, "Could not open the font map file: %s.", fileName);
fGUISupport->postErrorMessage(errorMessage, "Font Map Error");
status = LE_FONT_FILE_NOT_FOUND_ERROR;
return;
}
while (fgets(buffer, BUFFER_SIZE, file) != NULL) {
UScriptCode scriptCode;
UErrorCode scriptStatus = U_ZERO_ERROR;
line = strip(buffer);
if (line[0] == '#' || line[0] == 0) {
continue;
}
c = strchr(line, ':');
c[0] = 0;
fontName = strip(&c[1]);
scriptName = strip(line);
if (strcmp(scriptName, "DEFAULT") == 0) {
defaultFont = getFontIndex(fontName);
haveFonts = TRUE;
continue;
}
le_int32 fillCount = uscript_getCode(scriptName, &scriptCode, 1, &scriptStatus);
if (U_FAILURE(scriptStatus) || fillCount <= 0 ||
scriptStatus == U_USING_FALLBACK_WARNING || scriptStatus == U_USING_DEFAULT_WARNING) {
sprintf(errorMessage, "The script name %s is invalid.", line);
fGUISupport->postErrorMessage(errorMessage, "Font Map Error");
continue;
}
script = (le_int32) scriptCode;
if (fFontIndices[script] >= 0) {
// FIXME: complain that this is a duplicate entry and bail (?)
fFontIndices[script] = -1;
}
fFontIndices[script] = getFontIndex(fontName);
haveFonts = TRUE;
}
if (defaultFont >= 0) {
for (script = 0; script < scriptCodeCount; script += 1) {
if (fFontIndices[script] < 0) {
fFontIndices[script] = defaultFont;
}
}
}
if (! haveFonts) {
sprintf(errorMessage, "The font map file %s does not contain any valid scripts.", fileName);
fGUISupport->postErrorMessage(errorMessage, "Font Map Error");
status = LE_ILLEGAL_ARGUMENT_ERROR;
}
fclose(file);
}
void TestUScriptCodeAPI(){
int i =0;
int numErrors =0;
{
const char* testNames[]={
/* test locale */
"en", "en_US", "sr", "ta" , "te_IN",
"hi", "he", "ar",
/* test abbr */
"Hani", "Hang","Hebr","Hira",
"Knda","Kana","Khmr","Lao",
"Latn",/*"Latf","Latg",*/
"Mlym", "Mong",
/* test names */
"CYRILLIC","DESERET","DEVANAGARI","ETHIOPIC","GEORGIAN",
"GOTHIC", "GREEK", "GUJARATI", "COMMON", "INHERITED",
/* test lower case names */
"malayalam", "mongolian", "myanmar", "ogham", "old-italic",
"oriya", "runic", "sinhala", "syriac","tamil",
"telugu", "thaana", "thai", "tibetan",
/* test the bounds*/
"tagb", "arabic",
/* test bogus */
"asfdasd", "5464", "12235",
/* test the last index */
"zyyy", "YI",
'\0'
};
UScriptCode expected[] ={
/* locales should return */
USCRIPT_LATIN, USCRIPT_LATIN, USCRIPT_CYRILLIC, USCRIPT_TAMIL, USCRIPT_TELUGU,
USCRIPT_DEVANAGARI, USCRIPT_HEBREW, USCRIPT_ARABIC,
/* abbr should return */
USCRIPT_HAN, USCRIPT_HANGUL, USCRIPT_HEBREW, USCRIPT_HIRAGANA,
USCRIPT_KANNADA, USCRIPT_KATAKANA, USCRIPT_KHMER, USCRIPT_LAO,
USCRIPT_LATIN,/* USCRIPT_LATIN, USCRIPT_LATIN,*/
USCRIPT_MALAYALAM, USCRIPT_MONGOLIAN,
/* names should return */
USCRIPT_CYRILLIC, USCRIPT_DESERET, USCRIPT_DEVANAGARI, USCRIPT_ETHIOPIC, USCRIPT_GEORGIAN,
USCRIPT_GOTHIC, USCRIPT_GREEK, USCRIPT_GUJARATI, USCRIPT_COMMON, USCRIPT_INHERITED,
/* lower case names should return */
USCRIPT_MALAYALAM, USCRIPT_MONGOLIAN, USCRIPT_MYANMAR, USCRIPT_OGHAM, USCRIPT_OLD_ITALIC,
USCRIPT_ORIYA, USCRIPT_RUNIC, USCRIPT_SINHALA, USCRIPT_SYRIAC, USCRIPT_TAMIL,
USCRIPT_TELUGU, USCRIPT_THAANA, USCRIPT_THAI, USCRIPT_TIBETAN,
/* bounds */
USCRIPT_TAGBANWA, USCRIPT_ARABIC,
/* bogus names should return invalid code */
USCRIPT_INVALID_CODE, USCRIPT_INVALID_CODE, USCRIPT_INVALID_CODE,
USCRIPT_COMMON, USCRIPT_YI,
};
UErrorCode err = U_ZERO_ERROR;
const int32_t capacity = 10;
for( ; testNames[i]!='\0'; i++){
UScriptCode script[10]={USCRIPT_INVALID_CODE};
uscript_getCode(testNames[i],script,capacity, &err);
if( script[0] != expected[i]){
log_data_err("Error getting script code Got: %i Expected: %i for name %s (Error code does not propagate if data is not present. Are you missing data?)\n",
script[0],expected[i],testNames[i]);
numErrors++;
}
}
if(numErrors >0 ){
log_data_err("Errors uchar_getScriptCode() : %i \n",numErrors);
}
}
{
UErrorCode err = U_ZERO_ERROR;
int32_t capacity=0;
int32_t j;
UScriptCode jaCode[]={USCRIPT_KATAKANA, USCRIPT_HIRAGANA, USCRIPT_HAN };
UScriptCode script[10]={USCRIPT_INVALID_CODE};
int32_t num = uscript_getCode("ja",script,capacity, &err);
/* preflight */
if(err==U_BUFFER_OVERFLOW_ERROR){
err = U_ZERO_ERROR;
capacity = 10;
num = uscript_getCode("ja",script,capacity, &err);
if(num!=(sizeof(jaCode)/sizeof(UScriptCode))){
log_err("Errors uscript_getScriptCode() for Japanese locale: num=%d, expected %d \n",
num, (sizeof(jaCode)/sizeof(UScriptCode)));
}
for(j=0;j<sizeof(jaCode)/sizeof(UScriptCode);j++) {
if(script[j]!=jaCode[j]) {
log_err("Japanese locale: code #%d was %d (%s) but expected %d (%s)\n", j,
script[j], uscript_getName(script[j]),
jaCode[j], uscript_getName(jaCode[j]));
}
}
}else{
log_data_err("Errors in uscript_getScriptCode() expected error : %s got: %s \n",
"U_BUFFER_OVERFLOW_ERROR",
u_errorName(err));
}
}
{
UScriptCode testAbbr[]={
/* names should return */
USCRIPT_CYRILLIC, USCRIPT_DESERET, USCRIPT_DEVANAGARI, USCRIPT_ETHIOPIC, USCRIPT_GEORGIAN,
USCRIPT_GOTHIC, USCRIPT_GREEK, USCRIPT_GUJARATI,
};
const char* expectedNames[]={
/* test names */
"Cyrillic","Deseret","Devanagari","Ethiopic","Georgian",
"Gothic", "Greek", "Gujarati",
'\0'
};
i=0;
while(i<sizeof(testAbbr)/sizeof(UScriptCode)){
const char* name = uscript_getName(testAbbr[i]);
if(name == NULL) {
log_data_err("Couldn't get script name\n");
return;
}
numErrors=0;
if(strcmp(expectedNames[i],name)!=0){
log_err("Error getting abbreviations Got: %s Expected: %s\n",name,expectedNames[i]);
numErrors++;
}
if(numErrors > 0){
if(numErrors >0 ){
log_err("Errors uchar_getScriptAbbr() : %i \n",numErrors);
}
}
i++;
}
}
{
UScriptCode testAbbr[]={
/* abbr should return */
USCRIPT_HAN, USCRIPT_HANGUL, USCRIPT_HEBREW, USCRIPT_HIRAGANA,
USCRIPT_KANNADA, USCRIPT_KATAKANA, USCRIPT_KHMER, USCRIPT_LAO,
USCRIPT_LATIN,
USCRIPT_MALAYALAM, USCRIPT_MONGOLIAN,
};
const char* expectedAbbr[]={
/* test abbr */
"Hani", "Hang","Hebr","Hira",
"Knda","Kana","Khmr","Laoo",
"Latn",
"Mlym", "Mong",
'\0'
};
i=0;
while(i<sizeof(testAbbr)/sizeof(UScriptCode)){
const char* name = uscript_getShortName(testAbbr[i]);
numErrors=0;
if(strcmp(expectedAbbr[i],name)!=0){
log_err("Error getting abbreviations Got: %s Expected: %s\n",name,expectedAbbr[i]);
numErrors++;
}
if(numErrors > 0){
if(numErrors >0 ){
log_err("Errors uchar_getScriptAbbr() : %i \n",numErrors);
}
}
i++;
}
}
/* now test uscript_getScript() API */
{
uint32_t codepoints[] = {
0x0000FF9D, /* USCRIPT_KATAKANA*/
0x0000FFBE, /* USCRIPT_HANGUL*/
0x0000FFC7, /* USCRIPT_HANGUL*/
0x0000FFCF, /* USCRIPT_HANGUL*/
0x0000FFD7, /* USCRIPT_HANGUL*/
0x0000FFDC, /* USCRIPT_HANGUL*/
0x00010300, /* USCRIPT_OLD_ITALIC*/
0x00010330, /* USCRIPT_GOTHIC*/
0x0001034A, /* USCRIPT_GOTHIC*/
0x00010400, /* USCRIPT_DESERET*/
0x00010428, /* USCRIPT_DESERET*/
0x0001D167, /* USCRIPT_INHERITED*/
0x0001D17B, /* USCRIPT_INHERITED*/
0x0001D185, /* USCRIPT_INHERITED*/
0x0001D1AA, /* USCRIPT_INHERITED*/
0x00020000, /* USCRIPT_HAN*/
0x00000D02, /* USCRIPT_MALAYALAM*/
0x00000D00, /* USCRIPT_UNKNOWN (new Zzzz value in Unicode 5.0) */
0x00000000, /* USCRIPT_COMMON*/
0x0001D169, /* USCRIPT_INHERITED*/
0x0001D182, /* USCRIPT_INHERITED*/
0x0001D18B, /* USCRIPT_INHERITED*/
0x0001D1AD, /* USCRIPT_INHERITED*/
};
UScriptCode expected[] = {
USCRIPT_KATAKANA ,
USCRIPT_HANGUL ,
USCRIPT_HANGUL ,
USCRIPT_HANGUL ,
USCRIPT_HANGUL ,
USCRIPT_HANGUL ,
USCRIPT_OLD_ITALIC,
USCRIPT_GOTHIC ,
USCRIPT_GOTHIC ,
USCRIPT_DESERET ,
USCRIPT_DESERET ,
USCRIPT_INHERITED,
USCRIPT_INHERITED,
USCRIPT_INHERITED,
USCRIPT_INHERITED,
USCRIPT_HAN ,
USCRIPT_MALAYALAM,
USCRIPT_UNKNOWN,
USCRIPT_COMMON,
USCRIPT_INHERITED ,
USCRIPT_INHERITED ,
USCRIPT_INHERITED ,
USCRIPT_INHERITED ,
};
UScriptCode code = USCRIPT_INVALID_CODE;
UErrorCode status = U_ZERO_ERROR;
UBool passed = TRUE;
for(i=0; i<LENGTHOF(codepoints); ++i){
code = uscript_getScript(codepoints[i],&status);
if(U_SUCCESS(status)){
if( code != expected[i] ||
code != (UScriptCode)u_getIntPropertyValue(codepoints[i], UCHAR_SCRIPT)
) {
log_err("uscript_getScript for codepoint \\U%08X failed\n",codepoints[i]);
passed = FALSE;
}
}else{
log_err("uscript_getScript for codepoint \\U%08X failed. Error: %s\n",
codepoints[i],u_errorName(status));
break;
}
}
if(passed==FALSE){
log_err("uscript_getScript failed.\n");
}
}
{
UScriptCode code= USCRIPT_INVALID_CODE;
UErrorCode status = U_ZERO_ERROR;
code = uscript_getScript(0x001D169,&status);
if(code != USCRIPT_INHERITED){
log_err("\\U001D169 is not contained in USCRIPT_INHERITED");
}
}
{
UScriptCode code= USCRIPT_INVALID_CODE;
UErrorCode status = U_ZERO_ERROR;
int32_t err = 0;
for(i = 0; i<=0x10ffff; i++){
code = uscript_getScript(i,&status);
if(code == USCRIPT_INVALID_CODE){
err++;
log_err("uscript_getScript for codepoint \\U%08X failed.\n", i);
}
}
if(err>0){
log_err("uscript_getScript failed for %d codepoints\n", err);
}
}
{
for(i=0; (UScriptCode)i< USCRIPT_CODE_LIMIT; i++){
const char* name = uscript_getName((UScriptCode)i);
if(name==NULL || strcmp(name,"")==0){
log_err("uscript_getName failed for code %i: name is NULL or \"\"\n",i);
}
}
}
{
/*
* These script codes were originally added to ICU pre-3.6, so that ICU would
* have all ISO 15924 script codes. ICU was then based on Unicode 4.1.
* These script codes were added with only short names because we don't
* want to invent long names ourselves.
* Unicode 5 and later encode some of these scripts and give them long names.
* Whenever this happens, the long script names here need to be updated.
*/
static const char* expectedLong[] = {
"Balinese", "Batk", "Blis", "Brah", "Cham", "Cirt", "Cyrs", "Egyd", "Egyh", "Egyp",
"Geok", "Hans", "Hant", "Hmng", "Hung", "Inds", "Java", "Kayah_Li", "Latf", "Latg",
"Lepcha", "Lina", "Mand", "Maya", "Mero", "Nko", "Orkh", "Perm", "Phags_Pa", "Phoenician",
"Plrd", "Roro", "Sara", "Syre", "Syrj", "Syrn", "Teng", "Vai", "Visp", "Cuneiform",
"Zxxx", "Unknown",
"Carian", "Jpan", "Lana", "Lycian", "Lydian", "Ol_Chiki", "Rejang", "Saurashtra", "Sgnw", "Sundanese",
"Moon", "Mtei",
/* new in ICU 4.0 */
"Armi", "Avst", "Cakm", "Kore",
"Kthi", "Mani", "Phli", "Phlp", "Phlv", "Prti", "Samr", "Tavt",
"Zmth", "Zsym",
};
static const char* expectedShort[] = {
"Bali", "Batk", "Blis", "Brah", "Cham", "Cirt", "Cyrs", "Egyd", "Egyh", "Egyp",
"Geok", "Hans", "Hant", "Hmng", "Hung", "Inds", "Java", "Kali", "Latf", "Latg",
"Lepc", "Lina", "Mand", "Maya", "Mero", "Nkoo", "Orkh", "Perm", "Phag", "Phnx",
"Plrd", "Roro", "Sara", "Syre", "Syrj", "Syrn", "Teng", "Vaii", "Visp", "Xsux",
"Zxxx", "Zzzz",
"Cari", "Jpan", "Lana", "Lyci", "Lydi", "Olck", "Rjng", "Saur", "Sgnw", "Sund",
"Moon", "Mtei",
/* new in ICU 4.0 */
"Armi", "Avst", "Cakm", "Kore",
"Kthi", "Mani", "Phli", "Phlp", "Phlv", "Prti", "Samr", "Tavt",
"Zmth", "Zsym",
};
int32_t j = 0;
for(i=USCRIPT_BALINESE; (UScriptCode)i<USCRIPT_CODE_LIMIT; i++, j++){
const char* name = uscript_getName((UScriptCode)i);
if(name==NULL || strcmp(name,expectedLong[j])!=0){
log_err("uscript_getName failed for code %i: %s!=%s\n", i, name, expectedLong[j]);
}
name = uscript_getShortName((UScriptCode)i);
if(name==NULL || strcmp(name,expectedShort[j])!=0){
log_err("uscript_getShortName failed for code %i: %s!=%s\n", i, name, expectedShort[j]);
}
}
for(i=0; i<LENGTHOF(expectedLong); i++){
UScriptCode fillIn[5] = {USCRIPT_INVALID_CODE};
UErrorCode status = U_ZERO_ERROR;
int32_t len = 0;
len = uscript_getCode(expectedShort[i], fillIn, LENGTHOF(fillIn), &status);
if(U_FAILURE(status)){
log_err("uscript_getCode failed for script name %s. Error: %s\n",expectedShort[i], u_errorName(status));
}
if(len>1){
log_err("uscript_getCode did not return expected number of codes for script %s. EXPECTED: 1 GOT: %i\n", expectedShort[i], len);
}
if(fillIn[0]!= (UScriptCode)(USCRIPT_BALINESE+i)){
log_err("uscript_getCode did not return expected code for script %s. EXPECTED: %i GOT: %i\n", expectedShort[i], (USCRIPT_BALINESE+i), fillIn[0] );
}
}
}
}
void TestUScriptCodeAPI(){
int i =0;
int numErrors =0;
{
const char* testNames[]={
/* test locale */
"en", "en_US", "sr", "ta" , "te_IN",
"hi", "he", "ar",
/* test abbr */
"Hani", "Hang","Hebr","Hira",
"Knda","Kana","Khmr","Lao",
"Latn",/*"Latf","Latg",*/
"Mlym", "Mong",
/* test names */
"CYRILLIC","DESERET","DEVANAGARI","ETHIOPIC","GEORGIAN",
"GOTHIC", "GREEK", "GUJARATI", "COMMON", "INHERITED",
/* test lower case names */
"malayalam", "mongolian", "myanmar", "ogham", "old-italic",
"oriya", "runic", "sinhala", "syriac","tamil",
"telugu", "thaana", "thai", "tibetan",
/* test the bounds*/
"tagb", "arabic",
/* test bogus */
"asfdasd", "5464", "12235",
/* test the last index */
"zyyy", "YI",
NULL
};
UScriptCode expected[] ={
/* locales should return */
USCRIPT_LATIN, USCRIPT_LATIN, USCRIPT_CYRILLIC, USCRIPT_TAMIL, USCRIPT_TELUGU,
USCRIPT_DEVANAGARI, USCRIPT_HEBREW, USCRIPT_ARABIC,
/* abbr should return */
USCRIPT_HAN, USCRIPT_HANGUL, USCRIPT_HEBREW, USCRIPT_HIRAGANA,
USCRIPT_KANNADA, USCRIPT_KATAKANA, USCRIPT_KHMER, USCRIPT_LAO,
USCRIPT_LATIN,/* USCRIPT_LATIN, USCRIPT_LATIN,*/
USCRIPT_MALAYALAM, USCRIPT_MONGOLIAN,
/* names should return */
USCRIPT_CYRILLIC, USCRIPT_DESERET, USCRIPT_DEVANAGARI, USCRIPT_ETHIOPIC, USCRIPT_GEORGIAN,
USCRIPT_GOTHIC, USCRIPT_GREEK, USCRIPT_GUJARATI, USCRIPT_COMMON, USCRIPT_INHERITED,
/* lower case names should return */
USCRIPT_MALAYALAM, USCRIPT_MONGOLIAN, USCRIPT_MYANMAR, USCRIPT_OGHAM, USCRIPT_OLD_ITALIC,
USCRIPT_ORIYA, USCRIPT_RUNIC, USCRIPT_SINHALA, USCRIPT_SYRIAC, USCRIPT_TAMIL,
USCRIPT_TELUGU, USCRIPT_THAANA, USCRIPT_THAI, USCRIPT_TIBETAN,
/* bounds */
USCRIPT_TAGBANWA, USCRIPT_ARABIC,
/* bogus names should return invalid code */
USCRIPT_INVALID_CODE, USCRIPT_INVALID_CODE, USCRIPT_INVALID_CODE,
USCRIPT_COMMON, USCRIPT_YI,
};
UErrorCode err = U_ZERO_ERROR;
const int32_t capacity = 10;
for( ; testNames[i]!=NULL; i++){
UScriptCode script[10]={USCRIPT_INVALID_CODE};
uscript_getCode(testNames[i],script,capacity, &err);
if( script[0] != expected[i]){
log_data_err("Error getting script code Got: %i Expected: %i for name %s (Error code does not propagate if data is not present. Are you missing data?)\n",
script[0],expected[i],testNames[i]);
numErrors++;
}
}
if(numErrors >0 ){
log_data_err("Errors uchar_getScriptCode() : %i \n",numErrors);
}
}
{
UErrorCode err = U_ZERO_ERROR;
int32_t capacity=0;
int32_t j;
UScriptCode jaCode[]={USCRIPT_KATAKANA, USCRIPT_HIRAGANA, USCRIPT_HAN };
UScriptCode script[10]={USCRIPT_INVALID_CODE};
int32_t num = uscript_getCode("ja",script,capacity, &err);
/* preflight */
if(err==U_BUFFER_OVERFLOW_ERROR){
err = U_ZERO_ERROR;
capacity = 10;
num = uscript_getCode("ja",script,capacity, &err);
if(num!=UPRV_LENGTHOF(jaCode)){
log_err("Errors uscript_getScriptCode() for Japanese locale: num=%d, expected %d \n",
num, UPRV_LENGTHOF(jaCode));
}
for(j=0;j<UPRV_LENGTHOF(jaCode);j++) {
if(script[j]!=jaCode[j]) {
log_err("Japanese locale: code #%d was %d (%s) but expected %d (%s)\n", j,
script[j], uscript_getName(script[j]),
jaCode[j], uscript_getName(jaCode[j]));
}
}
}else{
log_data_err("Errors in uscript_getScriptCode() expected error : %s got: %s \n",
"U_BUFFER_OVERFLOW_ERROR",
u_errorName(err));
}
}
{
static const UScriptCode LATIN[1] = { USCRIPT_LATIN };
static const UScriptCode CYRILLIC[1] = { USCRIPT_CYRILLIC };
static const UScriptCode DEVANAGARI[1] = { USCRIPT_DEVANAGARI };
static const UScriptCode HAN[1] = { USCRIPT_HAN };
static const UScriptCode JAPANESE[3] = { USCRIPT_KATAKANA, USCRIPT_HIRAGANA, USCRIPT_HAN };
static const UScriptCode KOREAN[2] = { USCRIPT_HANGUL, USCRIPT_HAN };
static const UScriptCode HAN_BOPO[2] = { USCRIPT_HAN, USCRIPT_BOPOMOFO };
UScriptCode scripts[5];
UErrorCode err;
int32_t num;
// Should work regardless of whether we have locale data for the language.
err = U_ZERO_ERROR;
num = uscript_getCode("tg", scripts, UPRV_LENGTHOF(scripts), &err);
assertEqualScripts("tg script: Cyrl", CYRILLIC, 1, scripts, num, err); // Tajik
err = U_ZERO_ERROR;
num = uscript_getCode("xsr", scripts, UPRV_LENGTHOF(scripts), &err);
assertEqualScripts("xsr script: Deva", DEVANAGARI, 1, scripts, num, err); // Sherpa
// Multi-script languages.
err = U_ZERO_ERROR;
num = uscript_getCode("ja", scripts, UPRV_LENGTHOF(scripts), &err);
assertEqualScripts("ja scripts: Kana Hira Hani",
JAPANESE, UPRV_LENGTHOF(JAPANESE), scripts, num, err);
err = U_ZERO_ERROR;
num = uscript_getCode("ko", scripts, UPRV_LENGTHOF(scripts), &err);
assertEqualScripts("ko scripts: Hang Hani",
KOREAN, UPRV_LENGTHOF(KOREAN), scripts, num, err);
err = U_ZERO_ERROR;
num = uscript_getCode("zh", scripts, UPRV_LENGTHOF(scripts), &err);
assertEqualScripts("zh script: Hani", HAN, 1, scripts, num, err);
err = U_ZERO_ERROR;
num = uscript_getCode("zh-Hant", scripts, UPRV_LENGTHOF(scripts), &err);
assertEqualScripts("zh-Hant scripts: Hani Bopo", HAN_BOPO, 2, scripts, num, err);
err = U_ZERO_ERROR;
num = uscript_getCode("zh-TW", scripts, UPRV_LENGTHOF(scripts), &err);
assertEqualScripts("zh-TW scripts: Hani Bopo", HAN_BOPO, 2, scripts, num, err);
// Ambiguous API, but this probably wants to return Latin rather than Rongorongo (Roro).
err = U_ZERO_ERROR;
num = uscript_getCode("ro-RO", scripts, UPRV_LENGTHOF(scripts), &err);
assertEqualScripts("ro-RO script: Latn", LATIN, 1, scripts, num, err);
}
{
UScriptCode testAbbr[]={
/* names should return */
USCRIPT_CYRILLIC, USCRIPT_DESERET, USCRIPT_DEVANAGARI, USCRIPT_ETHIOPIC, USCRIPT_GEORGIAN,
USCRIPT_GOTHIC, USCRIPT_GREEK, USCRIPT_GUJARATI,
};
const char* expectedNames[]={
/* test names */
"Cyrillic","Deseret","Devanagari","Ethiopic","Georgian",
"Gothic", "Greek", "Gujarati",
NULL
};
i=0;
while(i<UPRV_LENGTHOF(testAbbr)){
const char* name = uscript_getName(testAbbr[i]);
if(name == NULL) {
log_data_err("Couldn't get script name\n");
return;
}
numErrors=0;
if(strcmp(expectedNames[i],name)!=0){
log_err("Error getting abbreviations Got: %s Expected: %s\n",name,expectedNames[i]);
numErrors++;
}
if(numErrors > 0){
if(numErrors >0 ){
log_err("Errors uchar_getScriptAbbr() : %i \n",numErrors);
}
}
i++;
}
}
{
UScriptCode testAbbr[]={
/* abbr should return */
USCRIPT_HAN, USCRIPT_HANGUL, USCRIPT_HEBREW, USCRIPT_HIRAGANA,
USCRIPT_KANNADA, USCRIPT_KATAKANA, USCRIPT_KHMER, USCRIPT_LAO,
USCRIPT_LATIN,
USCRIPT_MALAYALAM, USCRIPT_MONGOLIAN,
};
const char* expectedAbbr[]={
/* test abbr */
"Hani", "Hang","Hebr","Hira",
"Knda","Kana","Khmr","Laoo",
"Latn",
"Mlym", "Mong",
NULL
};
i=0;
while(i<UPRV_LENGTHOF(testAbbr)){
const char* name = uscript_getShortName(testAbbr[i]);
numErrors=0;
if(strcmp(expectedAbbr[i],name)!=0){
log_err("Error getting abbreviations Got: %s Expected: %s\n",name,expectedAbbr[i]);
numErrors++;
}
if(numErrors > 0){
if(numErrors >0 ){
log_err("Errors uchar_getScriptAbbr() : %i \n",numErrors);
}
}
i++;
}
}
/* now test uscript_getScript() API */
{
uint32_t codepoints[] = {
0x0000FF9D, /* USCRIPT_KATAKANA*/
0x0000FFBE, /* USCRIPT_HANGUL*/
0x0000FFC7, /* USCRIPT_HANGUL*/
0x0000FFCF, /* USCRIPT_HANGUL*/
0x0000FFD7, /* USCRIPT_HANGUL*/
0x0000FFDC, /* USCRIPT_HANGUL*/
0x00010300, /* USCRIPT_OLD_ITALIC*/
0x00010330, /* USCRIPT_GOTHIC*/
0x0001034A, /* USCRIPT_GOTHIC*/
0x00010400, /* USCRIPT_DESERET*/
0x00010428, /* USCRIPT_DESERET*/
0x0001D167, /* USCRIPT_INHERITED*/
0x0001D17B, /* USCRIPT_INHERITED*/
0x0001D185, /* USCRIPT_INHERITED*/
0x0001D1AA, /* USCRIPT_INHERITED*/
0x00020000, /* USCRIPT_HAN*/
0x00000D02, /* USCRIPT_MALAYALAM*/
0x00000D00, /* USCRIPT_UNKNOWN (new Zzzz value in Unicode 5.0) */
0x00000000, /* USCRIPT_COMMON*/
0x0001D169, /* USCRIPT_INHERITED*/
0x0001D182, /* USCRIPT_INHERITED*/
0x0001D18B, /* USCRIPT_INHERITED*/
0x0001D1AD, /* USCRIPT_INHERITED*/
};
UScriptCode expected[] = {
USCRIPT_KATAKANA ,
USCRIPT_HANGUL ,
USCRIPT_HANGUL ,
USCRIPT_HANGUL ,
USCRIPT_HANGUL ,
USCRIPT_HANGUL ,
USCRIPT_OLD_ITALIC,
USCRIPT_GOTHIC ,
USCRIPT_GOTHIC ,
USCRIPT_DESERET ,
USCRIPT_DESERET ,
USCRIPT_INHERITED,
USCRIPT_INHERITED,
USCRIPT_INHERITED,
USCRIPT_INHERITED,
USCRIPT_HAN ,
USCRIPT_MALAYALAM,
USCRIPT_UNKNOWN,
USCRIPT_COMMON,
USCRIPT_INHERITED ,
USCRIPT_INHERITED ,
USCRIPT_INHERITED ,
USCRIPT_INHERITED ,
};
UScriptCode code = USCRIPT_INVALID_CODE;
UErrorCode status = U_ZERO_ERROR;
UBool passed = TRUE;
for(i=0; i<UPRV_LENGTHOF(codepoints); ++i){
code = uscript_getScript(codepoints[i],&status);
if(U_SUCCESS(status)){
if( code != expected[i] ||
code != (UScriptCode)u_getIntPropertyValue(codepoints[i], UCHAR_SCRIPT)
) {
log_err("uscript_getScript for codepoint \\U%08X failed\n",codepoints[i]);
passed = FALSE;
}
}else{
log_err("uscript_getScript for codepoint \\U%08X failed. Error: %s\n",
codepoints[i],u_errorName(status));
break;
}
}
if(passed==FALSE){
log_err("uscript_getScript failed.\n");
}
}
{
UScriptCode code= USCRIPT_INVALID_CODE;
UErrorCode status = U_ZERO_ERROR;
code = uscript_getScript(0x001D169,&status);
if(code != USCRIPT_INHERITED){
log_err("\\U001D169 is not contained in USCRIPT_INHERITED");
}
}
{
UScriptCode code= USCRIPT_INVALID_CODE;
UErrorCode status = U_ZERO_ERROR;
int32_t err = 0;
for(i = 0; i<=0x10ffff; i++){
code = uscript_getScript(i,&status);
if(code == USCRIPT_INVALID_CODE){
err++;
log_err("uscript_getScript for codepoint \\U%08X failed.\n", i);
}
}
if(err>0){
log_err("uscript_getScript failed for %d codepoints\n", err);
}
}
{
for(i=0; (UScriptCode)i< USCRIPT_CODE_LIMIT; i++){
const char* name = uscript_getName((UScriptCode)i);
if(name==NULL || strcmp(name,"")==0){
log_err("uscript_getName failed for code %i: name is NULL or \"\"\n",i);
}
}
}
{
/*
* These script codes were originally added to ICU pre-3.6, so that ICU would
* have all ISO 15924 script codes. ICU was then based on Unicode 4.1.
* These script codes were added with only short names because we don't
* want to invent long names ourselves.
* Unicode 5 and later encode some of these scripts and give them long names.
* Whenever this happens, the long script names here need to be updated.
*/
static const char* expectedLong[] = {
"Balinese", "Batak", "Blis", "Brahmi", "Cham", "Cirt", "Cyrs",
"Egyd", "Egyh", "Egyptian_Hieroglyphs",
"Geok", "Hans", "Hant", "Pahawh_Hmong", "Old_Hungarian", "Inds",
"Javanese", "Kayah_Li", "Latf", "Latg",
"Lepcha", "Linear_A", "Mandaic", "Maya", "Meroitic_Hieroglyphs",
"Nko", "Old_Turkic", "Old_Permic", "Phags_Pa", "Phoenician",
"Miao", "Roro", "Sara", "Syre", "Syrj", "Syrn", "Teng", "Vai", "Visp", "Cuneiform",
"Zxxx", "Unknown",
"Carian", "Jpan", "Tai_Tham", "Lycian", "Lydian", "Ol_Chiki", "Rejang", "Saurashtra", "SignWriting", "Sundanese",
"Moon", "Meetei_Mayek",
/* new in ICU 4.0 */
"Imperial_Aramaic", "Avestan", "Chakma", "Kore",
"Kaithi", "Manichaean", "Inscriptional_Pahlavi", "Psalter_Pahlavi", "Phlv",
"Inscriptional_Parthian", "Samaritan", "Tai_Viet",
"Zmth", "Zsym",
/* new in ICU 4.4 */
"Bamum", "Lisu", "Nkgb", "Old_South_Arabian",
/* new in ICU 4.6 */
"Bassa_Vah", "Duployan", "Elbasan", "Grantha", "Kpel",
"Loma", "Mende_Kikakui", "Meroitic_Cursive",
"Old_North_Arabian", "Nabataean", "Palmyrene", "Khudawadi", "Warang_Citi",
/* new in ICU 4.8 */
"Afak", "Jurc", "Mro", "Nshu", "Sharada", "Sora_Sompeng", "Takri", "Tangut", "Wole",
/* new in ICU 49 */
"Anatolian_Hieroglyphs", "Khojki", "Tirhuta",
/* new in ICU 52 */
"Caucasian_Albanian", "Mahajani",
/* new in ICU 54 */
"Ahom", "Hatran", "Modi", "Multani", "Pau_Cin_Hau", "Siddham",
// new in ICU 58
"Adlam", "Bhaiksuki", "Marchen", "Newa", "Osage", "Hanb", "Jamo", "Zsye"
};
static const char* expectedShort[] = {
"Bali", "Batk", "Blis", "Brah", "Cham", "Cirt", "Cyrs", "Egyd", "Egyh", "Egyp",
"Geok", "Hans", "Hant", "Hmng", "Hung", "Inds", "Java", "Kali", "Latf", "Latg",
"Lepc", "Lina", "Mand", "Maya", "Mero", "Nkoo", "Orkh", "Perm", "Phag", "Phnx",
"Plrd", "Roro", "Sara", "Syre", "Syrj", "Syrn", "Teng", "Vaii", "Visp", "Xsux",
"Zxxx", "Zzzz",
"Cari", "Jpan", "Lana", "Lyci", "Lydi", "Olck", "Rjng", "Saur", "Sgnw", "Sund",
"Moon", "Mtei",
/* new in ICU 4.0 */
"Armi", "Avst", "Cakm", "Kore",
"Kthi", "Mani", "Phli", "Phlp", "Phlv", "Prti", "Samr", "Tavt",
"Zmth", "Zsym",
/* new in ICU 4.4 */
"Bamu", "Lisu", "Nkgb", "Sarb",
/* new in ICU 4.6 */
"Bass", "Dupl", "Elba", "Gran", "Kpel", "Loma", "Mend", "Merc",
"Narb", "Nbat", "Palm", "Sind", "Wara",
/* new in ICU 4.8 */
"Afak", "Jurc", "Mroo", "Nshu", "Shrd", "Sora", "Takr", "Tang", "Wole",
/* new in ICU 49 */
"Hluw", "Khoj", "Tirh",
/* new in ICU 52 */
"Aghb", "Mahj",
/* new in ICU 54 */
"Ahom", "Hatr", "Modi", "Mult", "Pauc", "Sidd",
// new in ICU 58
"Adlm", "Bhks", "Marc", "Newa", "Osge", "Hanb", "Jamo", "Zsye"
};
int32_t j = 0;
if(UPRV_LENGTHOF(expectedLong)!=(USCRIPT_CODE_LIMIT-USCRIPT_BALINESE)) {
log_err("need to add new script codes in cucdapi.c!\n");
return;
}
for(i=USCRIPT_BALINESE; (UScriptCode)i<USCRIPT_CODE_LIMIT; i++, j++){
const char* name = uscript_getName((UScriptCode)i);
if(name==NULL || strcmp(name,expectedLong[j])!=0){
log_err("uscript_getName failed for code %i: %s!=%s\n", i, name, expectedLong[j]);
}
name = uscript_getShortName((UScriptCode)i);
if(name==NULL || strcmp(name,expectedShort[j])!=0){
log_err("uscript_getShortName failed for code %i: %s!=%s\n", i, name, expectedShort[j]);
}
}
for(i=0; i<UPRV_LENGTHOF(expectedLong); i++){
UScriptCode fillIn[5] = {USCRIPT_INVALID_CODE};
UErrorCode status = U_ZERO_ERROR;
int32_t len = 0;
len = uscript_getCode(expectedShort[i], fillIn, UPRV_LENGTHOF(fillIn), &status);
if(U_FAILURE(status)){
log_err("uscript_getCode failed for script name %s. Error: %s\n",expectedShort[i], u_errorName(status));
}
if(len>1){
log_err("uscript_getCode did not return expected number of codes for script %s. EXPECTED: 1 GOT: %i\n", expectedShort[i], len);
}
if(fillIn[0]!= (UScriptCode)(USCRIPT_BALINESE+i)){
log_err("uscript_getCode did not return expected code for script %s. EXPECTED: %i GOT: %i\n", expectedShort[i], (USCRIPT_BALINESE+i), fillIn[0] );
}
}
}
{
/* test characters which have Script_Extensions */
UErrorCode errorCode=U_ZERO_ERROR;
if(!(
USCRIPT_COMMON==uscript_getScript(0x0640, &errorCode) &&
USCRIPT_INHERITED==uscript_getScript(0x0650, &errorCode) &&
USCRIPT_ARABIC==uscript_getScript(0xfdf2, &errorCode)) ||
U_FAILURE(errorCode)
) {
log_err("uscript_getScript(character with Script_Extensions) failed\n");
}
}
}
U_CDECL_BEGIN
void readTestFile(const char *testFilePath, TestCaseCallback callback)
{
#if !UCONFIG_NO_REGULAR_EXPRESSIONS
UErrorCode status = U_ZERO_ERROR;
UXMLParser *parser = UXMLParser::createParser(status);
UXMLElement *root = parser->parseFile(testFilePath, status);
if (root == NULL) {
log_err("Could not open the test data file: %s\n", testFilePath);
delete parser;
return;
}
UnicodeString test_case = UNICODE_STRING_SIMPLE("test-case");
UnicodeString test_text = UNICODE_STRING_SIMPLE("test-text");
UnicodeString test_font = UNICODE_STRING_SIMPLE("test-font");
UnicodeString result_glyphs = UNICODE_STRING_SIMPLE("result-glyphs");
UnicodeString result_indices = UNICODE_STRING_SIMPLE("result-indices");
UnicodeString result_positions = UNICODE_STRING_SIMPLE("result-positions");
// test-case attributes
UnicodeString id_attr = UNICODE_STRING_SIMPLE("id");
UnicodeString script_attr = UNICODE_STRING_SIMPLE("script");
UnicodeString lang_attr = UNICODE_STRING_SIMPLE("lang");
// test-font attributes
UnicodeString name_attr = UNICODE_STRING_SIMPLE("name");
UnicodeString ver_attr = UNICODE_STRING_SIMPLE("version");
UnicodeString cksum_attr = UNICODE_STRING_SIMPLE("checksum");
const UXMLElement *testCase;
int32_t tc = 0;
while((testCase = root->nextChildElement(tc)) != NULL) {
if (testCase->getTagName().compare(test_case) == 0) {
char *id = getCString(testCase->getAttribute(id_attr));
char *script = getCString(testCase->getAttribute(script_attr));
char *lang = getCString(testCase->getAttribute(lang_attr));
char *fontName = NULL;
char *fontVer = NULL;
char *fontCksum = NULL;
const UXMLElement *element;
int32_t ec = 0;
int32_t charCount = 0;
// int32_t typoFlags = 3; // kerning + ligatures...
UScriptCode scriptCode;
le_int32 languageCode = -1;
UnicodeString text, glyphs, indices, positions;
int32_t glyphCount = 0, indexCount = 0, positionCount = 0;
TestResult expected = {0, NULL, NULL, NULL};
uscript_getCode(script, &scriptCode, 1, &status);
if (LE_FAILURE(status)) {
log_err("invalid script name: %s.\n", script);
goto free_c_strings;
}
if (lang != NULL) {
languageCode = getLanguageCode(lang);
if (languageCode < 0) {
log_err("invalid language name: %s.\n", lang);
goto free_c_strings;
}
}
while((element = testCase->nextChildElement(ec)) != NULL) {
UnicodeString tag = element->getTagName();
// TODO: make sure that each element is only used once.
if (tag.compare(test_font) == 0) {
fontName = getCString(element->getAttribute(name_attr));
fontVer = getCString(element->getAttribute(ver_attr));
fontCksum = getCString(element->getAttribute(cksum_attr));
} else if (tag.compare(test_text) == 0) {
text = element->getText(TRUE);
charCount = text.length();
} else if (tag.compare(result_glyphs) == 0) {
glyphs = element->getText(TRUE);
} else if (tag.compare(result_indices) == 0) {
indices = element->getText(TRUE);
} else if (tag.compare(result_positions) == 0) {
positions = element->getText(TRUE);
} else {
// an unknown tag...
char *cTag = getCString(&tag);
log_info("Test %s: unknown element with tag \"%s\"\n", id, cTag);
freeCString(cTag);
}
}
expected.glyphs = (LEGlyphID *) getHexArray(glyphs, glyphCount);
expected.indices = (le_int32 *) getHexArray(indices, indexCount);
expected.positions = getFloatArray(positions, positionCount);
expected.glyphCount = glyphCount;
if (glyphCount < charCount || indexCount != glyphCount || positionCount < glyphCount * 2 + 2) {
log_err("Test %s: inconsistent input data: charCount = %d, glyphCount = %d, indexCount = %d, positionCount = %d\n",
id, charCount, glyphCount, indexCount, positionCount);
goto free_expected;
};
(*callback)(id, fontName, fontVer, fontCksum, scriptCode, languageCode, text.getBuffer(), charCount, &expected);
free_expected:
DELETE_ARRAY(expected.positions);
DELETE_ARRAY(expected.indices);
DELETE_ARRAY(expected.glyphs);
free_c_strings:
freeCString(fontCksum);
freeCString(fontVer);
freeCString(fontName);
freeCString(lang);
freeCString(script);
freeCString(id);
}
}
delete root;
delete parser;
#endif
}