static uint16_t
getTagNumber(const char *tag, uint16_t tagLen) {
char *atag;
uint16_t t;
UBool preferredName = ((tagLen > 0) ? (tag[tagLen - 1] == '*') : (FALSE));
if (tagCount >= MAX_TAG_COUNT) {
fprintf(stderr, "%s:%d: too many tags\n", path, lineNum);
exit(U_BUFFER_OVERFLOW_ERROR);
}
if (preferredName) {
/* puts(tag);*/
tagLen--;
}
for (t = 0; t < tagCount; ++t) {
const char *currTag = GET_TAG_STR(tags[t].tag);
if (uprv_strlen(currTag) == tagLen && !uprv_strnicmp(currTag, tag, tagLen)) {
return t;
}
}
/* we need to add this tag */
if (tagCount >= MAX_TAG_COUNT) {
fprintf(stderr, "%s:%d: error: too many tags\n", path, lineNum);
exit(U_BUFFER_OVERFLOW_ERROR);
}
/* allocate a new entry in the tag table */
atag = allocString(&tagBlock, tag, tagLen);
if (standardTagsUsed) {
fprintf(stderr, "%s:%d: error: Tag \"%s\" is not declared at the beginning of the alias table.\n",
path, lineNum, atag);
exit(1);
}
else if (tagLen > 0 && strcmp(tag, ALL_TAG_STR) != 0) {
fprintf(stderr, "%s:%d: warning: Tag \"%s\" was added to the list of standards because it was not declared at beginning of the alias table.\n",
path, lineNum, atag);
}
/* add the tag to the tag table */
tags[tagCount].tag = GET_TAG_NUM(atag);
/* The aliasList should be set to 0's already */
return tagCount++;
}
static int32_t
_uloc_minimizeSubtags(const char* localeID,
char* minimizedLocaleID,
int32_t minimizedLocaleIDCapacity,
UErrorCode* err)
{
/**
* ULOC_FULLNAME_CAPACITY will provide enough capacity
* that we can build a string that contains the language,
* script and region code without worrying about overrunning
* the user-supplied buffer.
**/
char maximizedTagBuffer[ULOC_FULLNAME_CAPACITY];
int32_t maximizedTagBufferLength = sizeof(maximizedTagBuffer);
char lang[ULOC_LANG_CAPACITY];
int32_t langLength = sizeof(lang);
char script[ULOC_SCRIPT_CAPACITY];
int32_t scriptLength = sizeof(script);
char region[ULOC_COUNTRY_CAPACITY];
int32_t regionLength = sizeof(region);
const char* trailing = "";
int32_t trailingLength = 0;
int32_t trailingIndex = 0;
if(U_FAILURE(*err)) {
goto error;
}
else if (localeID == NULL ||
minimizedLocaleID == NULL ||
minimizedLocaleIDCapacity <= 0) {
goto error;
}
trailingIndex =
parseTagString(
localeID,
lang,
&langLength,
script,
&scriptLength,
region,
®ionLength,
err);
if(U_FAILURE(*err)) {
/* Overflow indicates an illegal argument error */
if (*err == U_BUFFER_OVERFLOW_ERROR) {
*err = U_ILLEGAL_ARGUMENT_ERROR;
}
goto error;
}
/* Find the spot where the variants or the keywords begin, if any. */
while (_isIDSeparator(localeID[trailingIndex])) {
trailingIndex++;
}
trailing = &localeID[trailingIndex];
trailingLength = (int32_t)uprv_strlen(trailing);
CHECK_TRAILING_VARIANT_SIZE(trailing, trailingLength);
createTagString(
lang,
langLength,
script,
scriptLength,
region,
regionLength,
NULL,
0,
maximizedTagBuffer,
maximizedTagBufferLength,
err);
if(U_FAILURE(*err)) {
goto error;
}
/**
* First, we need to first get the maximization
* from AddLikelySubtags.
**/
maximizedTagBufferLength =
uloc_addLikelySubtags(
maximizedTagBuffer,
maximizedTagBuffer,
maximizedTagBufferLength,
err);
if(U_FAILURE(*err)) {
goto error;
}
/**
* Start first with just the language.
**/
{
char tagBuffer[ULOC_FULLNAME_CAPACITY];
const int32_t tagBufferLength =
createLikelySubtagsString(
lang,
langLength,
NULL,
0,
NULL,
0,
NULL,
0,
tagBuffer,
sizeof(tagBuffer),
err);
if(U_FAILURE(*err)) {
goto error;
}
else if (uprv_strnicmp(
maximizedTagBuffer,
tagBuffer,
tagBufferLength) == 0) {
return createTagString(
lang,
langLength,
NULL,
0,
NULL,
0,
trailing,
trailingLength,
minimizedLocaleID,
minimizedLocaleIDCapacity,
err);
}
}
/**
* Next, try the language and region.
**/
if (regionLength > 0) {
char tagBuffer[ULOC_FULLNAME_CAPACITY];
const int32_t tagBufferLength =
createLikelySubtagsString(
lang,
langLength,
NULL,
0,
region,
regionLength,
NULL,
0,
tagBuffer,
sizeof(tagBuffer),
err);
if(U_FAILURE(*err)) {
goto error;
}
else if (uprv_strnicmp(
maximizedTagBuffer,
tagBuffer,
tagBufferLength) == 0) {
return createTagString(
lang,
langLength,
NULL,
0,
region,
regionLength,
trailing,
trailingLength,
minimizedLocaleID,
minimizedLocaleIDCapacity,
err);
}
}
/**
* Finally, try the language and script. This is our last chance,
* since trying with all three subtags would only yield the
* maximal version that we already have.
**/
if (scriptLength > 0 && regionLength > 0) {
char tagBuffer[ULOC_FULLNAME_CAPACITY];
const int32_t tagBufferLength =
createLikelySubtagsString(
lang,
langLength,
script,
scriptLength,
NULL,
0,
NULL,
0,
tagBuffer,
sizeof(tagBuffer),
err);
if(U_FAILURE(*err)) {
goto error;
}
else if (uprv_strnicmp(
maximizedTagBuffer,
tagBuffer,
tagBufferLength) == 0) {
return createTagString(
lang,
langLength,
script,
scriptLength,
NULL,
0,
trailing,
trailingLength,
minimizedLocaleID,
minimizedLocaleIDCapacity,
err);
}
}
{
/**
* If we got here, return the locale ID parameter.
**/
const int32_t localeIDLength = (int32_t)uprv_strlen(localeID);
uprv_memcpy(
minimizedLocaleID,
localeID,
localeIDLength <= minimizedLocaleIDCapacity ?
localeIDLength : minimizedLocaleIDCapacity);
return u_terminateChars(
minimizedLocaleID,
minimizedLocaleIDCapacity,
localeIDLength,
err);
}
error:
if (!U_FAILURE(*err)) {
*err = U_ILLEGAL_ARGUMENT_ERROR;
}
return -1;
}
/**
* Parse the language, script, and region subtags from a tag string, and copy the
* results into the corresponding output parameters. The buffers are null-terminated,
* unless overflow occurs.
*
* The langLength, scriptLength, and regionLength parameters are input/output
* parameters, and must contain the capacity of their corresponding buffers on
* input. On output, they will contain the actual length of the buffers, not
* including the null terminator.
*
* If the length of any of the output subtags exceeds the capacity of the corresponding
* buffer, the function copies as many bytes to the output buffer as it can, and returns
* the error U_BUFFER_OVERFLOW_ERROR. It will not parse any more subtags once overflow
* occurs.
*
* If an illegal argument is provided, the function returns the error
* U_ILLEGAL_ARGUMENT_ERROR.
*
* @param localeID The locale ID to parse.
* @param lang The language tag buffer.
* @param langLength The length of the language tag.
* @param script The script tag buffer.
* @param scriptLength The length of the script tag.
* @param region The region tag buffer.
* @param regionLength The length of the region tag.
* @param err A pointer to a UErrorCode for error reporting.
* @return The number of chars of the localeID parameter consumed.
**/
static int32_t U_CALLCONV
parseTagString(
const char* localeID,
char* lang,
int32_t* langLength,
char* script,
int32_t* scriptLength,
char* region,
int32_t* regionLength,
UErrorCode* err)
{
const char* position = localeID;
int32_t subtagLength = 0;
if(U_FAILURE(*err) ||
localeID == NULL ||
lang == NULL ||
langLength == NULL ||
script == NULL ||
scriptLength == NULL ||
region == NULL ||
regionLength == NULL) {
goto error;
}
subtagLength = ulocimp_getLanguage(position, lang, *langLength, &position);
u_terminateChars(lang, *langLength, subtagLength, err);
/*
* Note that we explicit consider U_STRING_NOT_TERMINATED_WARNING
* to be an error, because it indicates the user-supplied tag is
* not well-formed.
*/
if(U_FAILURE(*err)) {
goto error;
}
*langLength = subtagLength;
/*
* If no language was present, use the value of unknownLanguage
* instead. Otherwise, move past any separator.
*/
if (*langLength == 0) {
uprv_strcpy(
lang,
unknownLanguage);
*langLength = (int32_t)uprv_strlen(lang);
}
else if (_isIDSeparator(*position)) {
++position;
}
subtagLength = ulocimp_getScript(position, script, *scriptLength, &position);
u_terminateChars(script, *scriptLength, subtagLength, err);
if(U_FAILURE(*err)) {
goto error;
}
*scriptLength = subtagLength;
if (*scriptLength > 0) {
if (uprv_strnicmp(script, unknownScript, *scriptLength) == 0) {
/**
* If the script part is the "unknown" script, then don't return it.
**/
*scriptLength = 0;
}
/*
* Move past any separator.
*/
if (_isIDSeparator(*position)) {
++position;
}
}
subtagLength = ulocimp_getCountry(position, region, *regionLength, &position);
u_terminateChars(region, *regionLength, subtagLength, err);
if(U_FAILURE(*err)) {
goto error;
}
*regionLength = subtagLength;
if (*regionLength > 0) {
if (uprv_strnicmp(region, unknownRegion, *regionLength) == 0) {
/**
* If the region part is the "unknown" region, then don't return it.
**/
*regionLength = 0;
}
} else if (*position != 0 && *position != '@') {
/* back up over consumed trailing separator */
--position;
}
exit:
return (int32_t)(position - localeID);
error:
/**
* If we get here, we have no explicit error, it's the result of an
* illegal argument.
**/
if (!U_FAILURE(*err)) {
*err = U_ILLEGAL_ARGUMENT_ERROR;
}
goto exit;
}
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(uprv_stricmp(src, "THIS is lower CASE") != 0){
log_err("FAIL: *****uprv_stricmp() failed.");
}
if((intValue=uprv_stricmp(NULL, "first string is null") )!= -1){
log_err("FAIL: uprv_stricmp() where the first string is null failed. Expected: -1, returned %d\n", intValue);
}
if((intValue=uprv_stricmp("second string is null", NULL)) != 1){
log_err("FAIL: uprv_stricmp() where the second string is null failed. Expected: 1, returned %d\n", intValue);
}
if((intValue=uprv_stricmp(NULL, NULL)) != 0){
log_err("FAIL: uprv_stricmp(NULL, NULL) failed. Expected: 0, returned %d\n", intValue);;
}
if((intValue=uprv_stricmp("", "")) != 0){
log_err("FAIL: uprv_stricmp(\"\", \"\") failed. Expected: 0, returned %d\n", intValue);;
}
if((intValue=uprv_stricmp("", "abc")) != -1){
log_err("FAIL: uprv_stricmp(\"\", \"abc\") failed. Expected: -1, returned %d\n", intValue);
}
if((intValue=uprv_stricmp("abc", "")) != 1){
log_err("FAIL: uprv_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(uprv_strnicmp(src, "THIS", 4 ) != 0){
log_err("FAIL: *****uprv_strnicmp() failed.");
}
if((intValue=uprv_strnicmp(NULL, "first string is null", 10) )!= -1){
log_err("FAIL: uprv_strnicmp() where the first string is null failed. Expected: -1, returned %d\n", intValue);
}
if((intValue=uprv_strnicmp("second string is null", NULL, 10)) != 1){
log_err("FAIL: uprv_strnicmp() where the second string is null failed. Expected: 1, returned %d\n", intValue);
}
if((intValue=uprv_strnicmp(NULL, NULL, 10)) != 0){
log_err("FAIL: uprv_strnicmp(NULL, NULL, 10) failed. Expected: 0, returned %d\n", intValue);;
}
if((intValue=uprv_strnicmp("", "", 10)) != 0){
log_err("FAIL: uprv_strnicmp(\"\", \"\") failed. Expected: 0, returned %d\n", intValue);;
}
if((intValue=uprv_strnicmp("", "abc", 10)) != -1){
log_err("FAIL: uprv_stricmp(\"\", \"abc\", 10) failed. Expected: -1, returned %d\n", intValue);
}
if((intValue=uprv_strnicmp("abc", "", 10)) != 1){
log_err("FAIL: uprv_strnicmp(\"abc\", \"\", 10) failed. Expected: 1, returned %d\n", intValue);
}
}
