U_CAPI UBool U_EXPORT2
uhash_compareIChars(const UHashTok key1, const UHashTok key2) {
const char *p1 = (const char*) key1.pointer;
const char *p2 = (const char*) key2.pointer;
if (p1 == p2) {
return TRUE;
}
if (p1 == NULL || p2 == NULL) {
return FALSE;
}
while (*p1 != 0 && uprv_tolower(*p1) == uprv_tolower(*p2)) {
++p1;
++p2;
}
return (UBool)(*p1 == *p2);
}
Value(int32_t enumValue, const char *joinedAliases)
: enumValue(enumValue), joinedAliases(joinedAliases), count(0) {
if(uprv_strlen(joinedAliases)>=JOINED_ALIASES_CAPACITY) {
fprintf(stderr,
"genprops error: pnamesbuilder.cpp Value::Value(%ld, \"%s\"): "
"joined aliases too long: "
"increase JOINED_ALIASES_CAPACITY, to at least %ld\n",
(long)enumValue, joinedAliases, uprv_strlen(joinedAliases)+1);
exit(U_BUFFER_OVERFLOW_ERROR);
}
// Copy the space-separated aliases into NUL-separated ones and count them.
// Write a normalized version of each one.
const char *j=joinedAliases;
char *a=aliasesBuffer;
char *n=normalizedBuffer;
char c;
do {
aliases[count]=a;
normalized[count++]=n;
while((c=*j++)!=' ' && c!=0) {
*a++=c;
// Ignore delimiters '-' and '_'.
if(!(c=='-' || c=='_')) {
*n++=uprv_tolower(c);
}
}
*a++=0;
*n++=0;
} while(c!=0);
}
U_CAPI int U_EXPORT2
uprv_strnicmp(const char *str1, const char *str2, uint32_t n) {
if(str1==NULL) {
if(str2==NULL) {
return 0;
} else {
return -1;
}
} else if(str2==NULL) {
return 1;
} else {
/* compare non-NULL strings lexically with lowercase */
int rc;
unsigned char c1, c2;
for(; n--;) {
c1=(unsigned char)*str1;
c2=(unsigned char)*str2;
if(c1==0) {
if(c2==0) {
return 0;
} else {
return -1;
}
} else if(c2==0) {
return 1;
} else {
/* compare non-zero characters with lowercase */
rc=(int)(unsigned char)uprv_tolower(c1)-(int)(unsigned char)uprv_tolower(c2);
if(rc!=0) {
return rc;
}
}
++str1;
++str2;
}
}
return 0;
}
U_CAPI char* U_EXPORT2
T_CString_toLowerCase(char* str)
{
char* origPtr = str;
if (str) {
do
*str = (char)uprv_tolower(*str);
while (*(str++));
}
return origPtr;
}
static const char* U_CALLCONV
_processLocaleElement(CollatorSpec *spec, uint32_t value, const char* string,
UErrorCode *status)
{
int32_t len = 0;
do {
if(value == UCOL_SIT_LANGUAGE || value == UCOL_SIT_KEYWORD || value == UCOL_SIT_PROVIDER) {
spec->locElements[value][len++] = uprv_tolower(*string);
} else {
spec->locElements[value][len++] = *string;
}
} while(*(++string) != '_' && *string && len < locElementCapacity);
if(len >= locElementCapacity) {
*status = U_BUFFER_OVERFLOW_ERROR;
return string;
}
// don't skip the underscore at the end
return string;
}
U_CAPI int32_t U_EXPORT2
uhash_hashIChars(const UHashTok key) {
STRING_HASH(uint8_t, key.pointer, uprv_strlen((char*)p), uprv_tolower(*p));
}
static void TestAPI(void)
{
int32_t intValue=0;
char src[30]="HELLO THERE", dest[30];
static const char *const abc="abcdefghijklmnopqrstuvwxyz", *const ABC="ABCDEFGHIJKLMNOPQRSTUVWXYZ";
const char *temp;
int32_t i;
log_verbose("Testing uprv_tolower() and uprv_toupper()\n");
for(i=0; i<=26; ++i) {
dest[i]=uprv_tolower(abc[i]);
}
if(0!=strcmp(abc, dest)) {
log_err("uprv_tolower(abc) failed\n");
}
for(i=0; i<=26; ++i) {
dest[i]=uprv_tolower(ABC[i]);
}
if(0!=strcmp(abc, dest)) {
log_err("uprv_tolower(ABC) failed\n");
}
for(i=0; i<=26; ++i) {
dest[i]=uprv_toupper(abc[i]);
}
if(0!=strcmp(ABC, dest)) {
log_err("uprv_toupper(abc) failed\n");
}
for(i=0; i<=26; ++i) {
dest[i]=uprv_toupper(ABC[i]);
}
if(0!=strcmp(ABC, dest)) {
log_err("uprv_toupper(ABC) failed\n");
}
log_verbose("Testing the API in cstring\n");
T_CString_toLowerCase(src);
if(uprv_strcmp(src, "hello there") != 0){
log_err("FAIL: *** T_CString_toLowerCase() failed. Expected: \"hello there\", Got: \"%s\"\n", src);
}
T_CString_toUpperCase(src);
if(uprv_strcmp(src, "HELLO THERE") != 0){
log_err("FAIL: *** T_CString_toUpperCase() failed. Expected: \"HELLO THERE\", Got: \"%s\"\n", src);
}
intValue=T_CString_stringToInteger("34556", 10);
if(intValue != 34556){
log_err("FAIL: ****T_CString_stringToInteger(\"34556\", 10) failed. Expected: 34556, Got: %d\n", intValue);
}
intValue=T_CString_stringToInteger("100", 16);
if(intValue != 256){
log_err("FAIL: ****T_CString_stringToInteger(\"100\", 16) failed. Expected: 256, Got: %d\n", intValue);
}
i = T_CString_integerToString(src, 34556, 10);
if(uprv_strcmp(src, "34556") != 0 || i != 5){
log_err("FAIL: ****integerToString(src, 34566, 10); failed. Expected: \"34556\", Got: %s\n", src);
}
i = T_CString_integerToString(src, 431, 16);
if(uprv_stricmp(src, "1AF") != 0 || i != 3){
log_err("FAIL: ****integerToString(src, 431, 16); failed. Expected: \"1AF\", Got: %s\n", src);
}
i = T_CString_int64ToString(src, U_INT64_MAX, 10);
if(uprv_strcmp(src, "9223372036854775807") != 0 || i != 19){
log_err("FAIL: ****integerToString(src, 9223372036854775807, 10); failed. Got: %s\n", src);
}
i = T_CString_int64ToString(src, U_INT64_MAX, 16);
if(uprv_stricmp(src, "7FFFFFFFFFFFFFFF") != 0 || i != 16){
log_err("FAIL: ****integerToString(src, 7FFFFFFFFFFFFFFF, 16); failed. Got: %s\n", src);
}
uprv_strcpy(src, "this is lower case");
if(T_CString_stricmp(src, "THIS is lower CASE") != 0){
log_err("FAIL: *****T_CString_stricmp() failed.");
}
if((intValue=T_CString_stricmp(NULL, "first string is null") )!= -1){
log_err("FAIL: T_CString_stricmp() where the first string is null failed. Expected: -1, returned %d\n", intValue);
}
if((intValue=T_CString_stricmp("second string is null", NULL)) != 1){
log_err("FAIL: T_CString_stricmp() where the second string is null failed. Expected: 1, returned %d\n", intValue);
}
if((intValue=T_CString_stricmp(NULL, NULL)) != 0){
log_err("FAIL: T_CString_stricmp(NULL, NULL) failed. Expected: 0, returned %d\n", intValue);
}
if((intValue=T_CString_stricmp("", "")) != 0){
log_err("FAIL: T_CString_stricmp(\"\", \"\") failed. Expected: 0, returned %d\n", intValue);
}
if((intValue=T_CString_stricmp("", "abc")) != -1){
log_err("FAIL: T_CString_stricmp(\"\", \"abc\") failed. Expected: -1, returned %d\n", intValue);
}
if((intValue=T_CString_stricmp("abc", "")) != 1){
log_err("FAIL: T_CString_stricmp(\"abc\", \"\") failed. Expected: 1, returned %d\n", intValue);
}
temp=uprv_strdup("strdup");
if(uprv_strcmp(temp, "strdup") !=0 ){
log_err("FAIL: uprv_strdup() failed. Expected: \"strdup\", Got: %s\n", temp);
}
uprv_free((char *)temp);
uprv_strcpy(src, "this is lower case");
if(T_CString_strnicmp(src, "THIS", 4 ) != 0){
log_err("FAIL: *****T_CString_strnicmp() failed.");
}
if((intValue=T_CString_strnicmp(NULL, "first string is null", 10) )!= -1){
log_err("FAIL: T_CString_strnicmp() where the first string is null failed. Expected: -1, returned %d\n", intValue);
}
if((intValue=T_CString_strnicmp("second string is null", NULL, 10)) != 1){
log_err("FAIL: T_CString_strnicmp() where the second string is null failed. Expected: 1, returned %d\n", intValue);
}
if((intValue=T_CString_strnicmp(NULL, NULL, 10)) != 0){
log_err("FAIL: T_CString_strnicmp(NULL, NULL, 10) failed. Expected: 0, returned %d\n", intValue);
}
if((intValue=T_CString_strnicmp("", "", 10)) != 0){
log_err("FAIL: T_CString_strnicmp(\"\", \"\") failed. Expected: 0, returned %d\n", intValue);
}
if((intValue=T_CString_strnicmp("", "abc", 10)) != -1){
log_err("FAIL: T_CString_stricmp(\"\", \"abc\", 10) failed. Expected: -1, returned %d\n", intValue);
}
if((intValue=T_CString_strnicmp("abc", "", 10)) != 1){
log_err("FAIL: T_CString_strnicmp(\"abc\", \"\", 10) failed. Expected: 1, returned %d\n", intValue);
}
}
/* NOTE: The caller should handle thread safety */
U_CAPI const char* U_EXPORT2
uprv_getDefaultLocaleID()
{
#if U_POSIX_LOCALE
/*
Note that: (a '!' means the ID is improper somehow)
LC_ALL ----> default_loc codepage
--------------------------------------------------------
ab.CD ab CD
ab@CD ab__CD -
[email protected] ab__CD EF
ab_CD.EF@GH ab_CD_GH EF
Some 'improper' ways to do the same as above:
! [email protected] ab_CD_GH EF
! [email protected] ab_CD_GH EF
! [email protected]@GH.IJ ab_CD_GH EF
_CD@GH _CD_GH -
_CD.EF@GH _CD_GH EF
The variant cannot have dots in it.
The 'rightmost' variant (@xxx) wins.
The leftmost codepage (.xxx) wins.
*/
char *correctedPOSIXLocale = 0;
const char* posixID = uprv_getPOSIXID();
const char *p;
const char *q;
int32_t len;
/* Format: (no spaces)
ll [ _CC ] [ . MM ] [ @ VV]
l = lang, C = ctry, M = charmap, V = variant
*/
if (gCorrectedPOSIXLocale != NULL) {
return gCorrectedPOSIXLocale;
}
if ((p = uprv_strchr(posixID, '.')) != NULL) {
/* assume new locale can't be larger than old one? */
correctedPOSIXLocale = (char*) uprv_malloc(uprv_strlen(posixID)+1);
uprv_strncpy(correctedPOSIXLocale, posixID, p-posixID);
correctedPOSIXLocale[p-posixID] = 0;
/* do not copy after the @ */
if ((p = uprv_strchr(correctedPOSIXLocale, '@')) != NULL) {
correctedPOSIXLocale[p-correctedPOSIXLocale] = 0;
}
}
/* Note that we scan the *uncorrected* ID. */
if ((p = uprv_strrchr(posixID, '@')) != NULL) {
if (correctedPOSIXLocale == NULL) {
correctedPOSIXLocale = (char*) uprv_malloc(uprv_strlen(posixID)+1);
uprv_strncpy(correctedPOSIXLocale, posixID, p-posixID);
correctedPOSIXLocale[p-posixID] = 0;
}
p++;
/* Take care of any special cases here.. */
if (!uprv_strcmp(p, "nynorsk")) {
p = "NY";
/* Don't worry about no__NY. In practice, it won't appear. */
}
if (uprv_strchr(correctedPOSIXLocale,'_') == NULL) {
uprv_strcat(correctedPOSIXLocale, "__"); /* aa@b -> aa__b */
}
else {
uprv_strcat(correctedPOSIXLocale, "_"); /* aa_CC@b -> aa_CC_b */
}
if ((q = uprv_strchr(p, '.')) != NULL) {
/* How big will the resulting string be? */
len = (int32_t)(uprv_strlen(correctedPOSIXLocale) + (q-p));
uprv_strncat(correctedPOSIXLocale, p, q-p);
correctedPOSIXLocale[len] = 0;
}
else {
/* Anything following the @ sign */
uprv_strcat(correctedPOSIXLocale, p);
}
/* Should there be a map from 'no@nynorsk' -> no_NO_NY here?
* How about 'russian' -> 'ru'?
* Many of the other locales using ISO codes will be handled by the
* canonicalization functions in uloc_getDefault.
*/
}
/* Was a correction made? */
if (correctedPOSIXLocale != NULL) {
posixID = correctedPOSIXLocale;
}
else {
/* copy it, just in case the original pointer goes away. See j2395 */
correctedPOSIXLocale = (char *)uprv_malloc(uprv_strlen(posixID) + 1);
posixID = uprv_strcpy(correctedPOSIXLocale, posixID);
}
if (gCorrectedPOSIXLocale == NULL) {
gCorrectedPOSIXLocale = correctedPOSIXLocale;
ucln_common_registerCleanup(UCLN_COMMON_PUTIL, putil_cleanup);
correctedPOSIXLocale = NULL;
}
if (correctedPOSIXLocale != NULL) { /* Was already set - clean up. */
uprv_free(correctedPOSIXLocale);
}
return posixID;
#elif defined(U_WINDOWS)
UErrorCode status = U_ZERO_ERROR;
LCID id = GetThreadLocale();
const char* locID = uprv_convertToPosix(id, &status);
if (U_FAILURE(status)) {
locID = "en_US";
}
return locID;
#elif defined(XP_MAC)
int32_t script = MAC_LC_INIT_NUMBER;
/* = IntlScript(); or GetScriptManagerVariable(smSysScript);*/
int32_t region = MAC_LC_INIT_NUMBER;
/* = GetScriptManagerVariable(smRegionCode);*/
int32_t lang = MAC_LC_INIT_NUMBER;
/* = GetScriptManagerVariable(smScriptLang);*/
int32_t date_region = MAC_LC_INIT_NUMBER;
const char* posixID = 0;
int32_t count = sizeof(mac_lc_recs) / sizeof(mac_lc_rec);
int32_t i;
Intl1Hndl ih;
ih = (Intl1Hndl) GetIntlResource(1);
if (ih)
date_region = ((uint16_t)(*ih)->intl1Vers) >> 8;
for (i = 0; i < count; i++) {
if ( ((mac_lc_recs[i].script == MAC_LC_MAGIC_NUMBER)
|| (mac_lc_recs[i].script == script))
&& ((mac_lc_recs[i].region == MAC_LC_MAGIC_NUMBER)
|| (mac_lc_recs[i].region == region))
&& ((mac_lc_recs[i].lang == MAC_LC_MAGIC_NUMBER)
|| (mac_lc_recs[i].lang == lang))
&& ((mac_lc_recs[i].date_region == MAC_LC_MAGIC_NUMBER)
|| (mac_lc_recs[i].date_region == date_region))
)
{
posixID = mac_lc_recs[i].posixID;
break;
}
}
return posixID;
#elif defined(OS400)
/* locales are process scoped and are by definition thread safe */
static char correctedLocale[64];
const char *localeID = getenv("LC_ALL");
char *p;
if (localeID == NULL)
localeID = getenv("LANG");
if (localeID == NULL)
localeID = setlocale(LC_ALL, NULL);
/* Make sure we have something... */
if (localeID == NULL)
return "en_US_POSIX";
/* Extract the locale name from the path. */
if((p = uprv_strrchr(localeID, '/')) != NULL)
{
/* Increment p to start of locale name. */
p++;
localeID = p;
}
/* Copy to work location. */
uprv_strcpy(correctedLocale, localeID);
/* Strip off the '.locale' extension. */
if((p = uprv_strchr(correctedLocale, '.')) != NULL) {
*p = 0;
}
/* Upper case the locale name. */
T_CString_toUpperCase(correctedLocale);
/* See if we are using the POSIX locale. Any of the
* following are equivalent and use the same QLGPGCMA
* (POSIX) locale.
* QLGPGCMA2 means UCS2
* QLGPGCMA_4 means UTF-32
* QLGPGCMA_8 means UTF-8
*/
if ((uprv_strcmp("C", correctedLocale) == 0) ||
(uprv_strcmp("POSIX", correctedLocale) == 0) ||
(uprv_strncmp("QLGPGCMA", correctedLocale, 8) == 0))
{
uprv_strcpy(correctedLocale, "en_US_POSIX");
}
else
{
int16_t LocaleLen;
/* Lower case the lang portion. */
for(p = correctedLocale; *p != 0 && *p != '_'; p++)
{
*p = uprv_tolower(*p);
}
/* Adjust for Euro. After '_E' add 'URO'. */
LocaleLen = uprv_strlen(correctedLocale);
if (correctedLocale[LocaleLen - 2] == '_' &&
correctedLocale[LocaleLen - 1] == 'E')
{
uprv_strcat(correctedLocale, "URO");
}
/* If using Lotus-based locale then convert to
* equivalent non Lotus.
*/
else if (correctedLocale[LocaleLen - 2] == '_' &&
correctedLocale[LocaleLen - 1] == 'L')
{
correctedLocale[LocaleLen - 2] = 0;
}
/* There are separate simplified and traditional
* locales called zh_HK_S and zh_HK_T.
*/
else if (uprv_strncmp(correctedLocale, "zh_HK", 5) == 0)
{
uprv_strcpy(correctedLocale, "zh_HK");
}
/* A special zh_CN_GBK locale...
*/
else if (uprv_strcmp(correctedLocale, "zh_CN_GBK") == 0)
{
uprv_strcpy(correctedLocale, "zh_CN");
}
}
return correctedLocale;
#endif
}