const LEFontInstance *FontMap::getScriptFont(le_int32 scriptCode, LEErrorCode &status)
{
if (LE_FAILURE(status)) {
return NULL;
}
if (scriptCode <= -1 || scriptCode >= scriptCodeCount) {
status = LE_ILLEGAL_ARGUMENT_ERROR;
return NULL;
}
le_int32 fontIndex = fFontIndices[scriptCode];
if (fontIndex < 0) {
sprintf(errorMessage, "No font was set for script %s", uscript_getName((UScriptCode) scriptCode));
fGUISupport->postErrorMessage(errorMessage, "Font Map Error");
status = LE_FONT_FILE_NOT_FOUND_ERROR;
return NULL;
}
if (fFontInstances[fontIndex] == NULL) {
fFontInstances[fontIndex] = openFont(fFontNames[fontIndex], fPointSize, status);
if (LE_FAILURE(status)) {
sprintf(errorMessage, "Could not open font file %s", fFontNames[fontIndex]);
fGUISupport->postErrorMessage(errorMessage, "Font Map Error");
return NULL;
}
}
return fFontInstances[fontIndex];
}
Transliterator* AnyTransliterator::getTransliterator(UScriptCode source) const {
if (source == targetScript || source == USCRIPT_INVALID_CODE) {
return NULL;
}
Transliterator* t = (Transliterator*) uhash_iget(cache, (int32_t) source);
if (t == NULL) {
UErrorCode ec = U_ZERO_ERROR;
UnicodeString sourceName(uscript_getName(source), -1, US_INV);
UnicodeString id(sourceName);
id.append(TARGET_SEP).append(target);
t = Transliterator::createInstance(id, UTRANS_FORWARD, ec);
if (U_FAILURE(ec) || t == NULL) {
delete t;
// Try to pivot around Latin, our most common script
id = sourceName;
id.append(LATIN_PIVOT, -1).append(target);
t = Transliterator::createInstance(id, UTRANS_FORWARD, ec);
if (U_FAILURE(ec) || t == NULL) {
delete t;
t = NULL;
}
}
if (t != NULL) {
uhash_iput(cache, (int32_t) source, t, &ec);
}
}
return t;
}
U_CAPI int32_t U_EXPORT2
uldn_scriptCodeDisplayName(const ULocaleDisplayNames *ldn,
UScriptCode scriptCode,
UChar *result,
int32_t maxResultSize,
UErrorCode *pErrorCode) {
return uldn_scriptDisplayName(ldn, uscript_getName(scriptCode), result, maxResultSize, pErrorCode);
}
UnicodeString&
DefaultLocaleDisplayNames::scriptDisplayName(UScriptCode scriptCode,
UnicodeString& result) const {
const char* name = uscript_getName(scriptCode);
if (name) {
return result = UnicodeString(name, -1, US_INV);
}
return result.remove();
}
void main()
{
ScriptRun scriptRun(testChars, 0, testLength);
while (scriptRun.next()) {
int32_t start = scriptRun.getScriptStart();
int32_t end = scriptRun.getScriptEnd();
UScriptCode code = scriptRun.getScriptCode();
printf("Script '%s' from %d to %d.\n", uscript_getName(code), start, end);
}
}
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();
}
UnicodeString&
LocaleDisplayNamesImpl::scriptDisplayName(UScriptCode scriptCode,
UnicodeString& result) const {
return scriptDisplayName(uscript_getName(scriptCode), result, FALSE);
}
// Helper sets the character attribute properties and sets up the script table.
// Does not set tops and bottoms.
void SetupBasicProperties(bool report_errors, bool decompose,
UNICHARSET* unicharset) {
for (int unichar_id = 0; unichar_id < unicharset->size(); ++unichar_id) {
// Convert any custom ligatures.
const char* unichar_str = unicharset->id_to_unichar(unichar_id);
for (int i = 0; UNICHARSET::kCustomLigatures[i][0] != nullptr; ++i) {
if (!strcmp(UNICHARSET::kCustomLigatures[i][1], unichar_str)) {
unichar_str = UNICHARSET::kCustomLigatures[i][0];
break;
}
}
// Convert the unichar to UTF32 representation
std::vector<char32> uni_vector = UNICHAR::UTF8ToUTF32(unichar_str);
// Assume that if the property is true for any character in the string,
// then it holds for the whole "character".
bool unichar_isalpha = false;
bool unichar_islower = false;
bool unichar_isupper = false;
bool unichar_isdigit = false;
bool unichar_ispunct = false;
for (char32 u_ch : uni_vector) {
if (u_isalpha(u_ch)) unichar_isalpha = true;
if (u_islower(u_ch)) unichar_islower = true;
if (u_isupper(u_ch)) unichar_isupper = true;
if (u_isdigit(u_ch)) unichar_isdigit = true;
if (u_ispunct(u_ch)) unichar_ispunct = true;
}
unicharset->set_isalpha(unichar_id, unichar_isalpha);
unicharset->set_islower(unichar_id, unichar_islower);
unicharset->set_isupper(unichar_id, unichar_isupper);
unicharset->set_isdigit(unichar_id, unichar_isdigit);
unicharset->set_ispunctuation(unichar_id, unichar_ispunct);
tesseract::IcuErrorCode err;
unicharset->set_script(unichar_id, uscript_getName(
uscript_getScript(uni_vector[0], err)));
const int num_code_points = uni_vector.size();
// Obtain the lower/upper case if needed and record it in the properties.
unicharset->set_other_case(unichar_id, unichar_id);
if (unichar_islower || unichar_isupper) {
std::vector<char32> other_case(num_code_points, 0);
for (int i = 0; i < num_code_points; ++i) {
// TODO(daria): Ideally u_strToLower()/ustrToUpper() should be used.
// However since they deal with UChars (so need a conversion function
// from char32 or UTF8string) and require a meaningful locale string,
// for now u_tolower()/u_toupper() are used.
other_case[i] = unichar_islower ? u_toupper(uni_vector[i]) :
u_tolower(uni_vector[i]);
}
std::string other_case_uch = UNICHAR::UTF32ToUTF8(other_case);
UNICHAR_ID other_case_id =
unicharset->unichar_to_id(other_case_uch.c_str());
if (other_case_id != INVALID_UNICHAR_ID) {
unicharset->set_other_case(unichar_id, other_case_id);
} else if (unichar_id >= SPECIAL_UNICHAR_CODES_COUNT && report_errors) {
tprintf("Other case %s of %s is not in unicharset\n",
other_case_uch.c_str(), unichar_str);
}
}
// Set RTL property and obtain mirror unichar ID from ICU.
std::vector<char32> mirrors(num_code_points, 0);
for (int i = 0; i < num_code_points; ++i) {
mirrors[i] = u_charMirror(uni_vector[i]);
if (i == 0) { // set directionality to that of the 1st code point
unicharset->set_direction(unichar_id,
static_cast<UNICHARSET::Direction>(
u_charDirection(uni_vector[i])));
}
}
std::string mirror_uch = UNICHAR::UTF32ToUTF8(mirrors);
UNICHAR_ID mirror_uch_id = unicharset->unichar_to_id(mirror_uch.c_str());
if (mirror_uch_id != INVALID_UNICHAR_ID) {
unicharset->set_mirror(unichar_id, mirror_uch_id);
} else if (report_errors) {
tprintf("Mirror %s of %s is not in unicharset\n",
mirror_uch.c_str(), unichar_str);
}
// Record normalized version of this unichar.
std::string normed_str;
if (unichar_id != 0 &&
tesseract::NormalizeUTF8String(
decompose ? tesseract::UnicodeNormMode::kNFKD
: tesseract::UnicodeNormMode::kNFKC,
tesseract::OCRNorm::kNormalize, tesseract::GraphemeNorm::kNone,
unichar_str, &normed_str) &&
!normed_str.empty()) {
unicharset->set_normed(unichar_id, normed_str.c_str());
} else {
unicharset->set_normed(unichar_id, unichar_str);
}
ASSERT_HOST(unicharset->get_other_case(unichar_id) < unicharset->size());
}
unicharset->post_load_setup();
}
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");
}
}
}
UnicodeString&
LocaleDisplayNamesImpl::scriptDisplayName(UScriptCode scriptCode,
UnicodeString& result) const {
const char* name = uscript_getName(scriptCode);
return langData.get("Scripts", name, result);
}