static void TestBreakIteratorRuleError() {
/*
* TestBreakIteratorRuleError - Try to create a BI from rules with syntax errors,
* check that the error is reported correctly.
*/
char rules[] = " # This is a rule comment on line 1\n"
"[:L:]; # this rule is OK.\n"
"abcdefg); # Error, mismatched parens\n";
UChar *uRules;
void *freeHook = NULL;
UErrorCode status = U_ZERO_ERROR;
UParseError parseErr;
UBreakIterator *bi;
uRules = toUChar(rules, &freeHook);
bi = ubrk_openRules(uRules, -1, /* The rules */
NULL, -1, /* The text to be iterated over. */
&parseErr, &status);
if (U_SUCCESS(status)) {
log_err("FAIL: construction of break iterator succeeded when it should have failed.\n");
ubrk_close(bi);
} else {
if (parseErr.line != 3 || parseErr.offset != 8) {
log_data_err("FAIL: incorrect error position reported. Got line %d, char %d, expected line 3, char 7 (Are you missing data?)\n",
parseErr.line, parseErr.offset);
}
}
freeToUCharStrings(&freeHook);
}
static TextBreakIterator* initializeIteratorWithRules(const char* breakRules)
{
UParseError parseStatus;
UErrorCode openStatus = U_ZERO_ERROR;
String rules(breakRules);
TextBreakIterator* iterator = reinterpret_cast<TextBreakIterator*>(ubrk_openRules(rules.deprecatedCharacters(), rules.length(), 0, 0, &parseStatus, &openStatus));
ASSERT_WITH_MESSAGE(U_SUCCESS(openStatus), "ICU could not open a break iterator: %s (%d)", u_errorName(openStatus), openStatus);
return iterator;
}
static TextBreakIterator* initializeIteratorWithRules(const char* breakRules)
{
UParseError parseStatus;
UErrorCode openStatus = U_ZERO_ERROR;
unsigned length = strlen(breakRules);
auto upconvertedCharacters = StringView(reinterpret_cast<const LChar*>(breakRules), length).upconvertedCharacters();
TextBreakIterator* iterator = reinterpret_cast<TextBreakIterator*>(ubrk_openRules(upconvertedCharacters, length, 0, 0, &parseStatus, &openStatus));
ASSERT_WITH_MESSAGE(U_SUCCESS(openStatus), "ICU could not open a break iterator: %s (%d)", u_errorName(openStatus), openStatus);
return iterator;
}
/*
* TestsBreakIteratorStatusVals() Test the ubrk_getRuleStatusVec() funciton
*/
static void TestBreakIteratorStatusVec() {
#define RULE_STRING_LENGTH 200
UChar rules[RULE_STRING_LENGTH];
#define TEST_STRING_LENGTH 25
UChar testString[TEST_STRING_LENGTH];
UBreakIterator *bi = NULL;
int32_t pos = 0;
int32_t vals[10];
int32_t numVals;
UErrorCode status = U_ZERO_ERROR;
u_uastrncpy(rules, "[A-N]{100}; \n"
"[a-w]{200}; \n"
"[\\p{L}]{300}; \n"
"[\\p{N}]{400}; \n"
"[0-5]{500}; \n"
"!.*;\n", RULE_STRING_LENGTH);
u_uastrncpy(testString, "ABC", TEST_STRING_LENGTH);
bi = ubrk_openRules(rules, -1, testString, -1, NULL, &status);
TEST_ASSERT_SUCCESS(status);
TEST_ASSERT(bi != NULL);
/* The TEST_ASSERT above should change too... */
if (bi != NULL) {
pos = ubrk_next(bi);
TEST_ASSERT(pos == 1);
memset(vals, -1, sizeof(vals));
numVals = ubrk_getRuleStatusVec(bi, vals, 10, &status);
TEST_ASSERT_SUCCESS(status);
TEST_ASSERT(numVals == 2);
TEST_ASSERT(vals[0] == 100);
TEST_ASSERT(vals[1] == 300);
TEST_ASSERT(vals[2] == -1);
numVals = ubrk_getRuleStatusVec(bi, vals, 0, &status);
TEST_ASSERT(status == U_BUFFER_OVERFLOW_ERROR);
TEST_ASSERT(numVals == 2);
}
ubrk_close(bi);
}
/*
// Open a break iterator from char * rules. Take care of conversion
// of the rules and error checking.
*/
static UBreakIterator * testOpenRules(char *rules) {
UErrorCode status = U_ZERO_ERROR;
UChar *ruleSourceU = NULL;
void *strCleanUp = NULL;
UParseError parseErr;
UBreakIterator *bi;
ruleSourceU = toUChar(rules, &strCleanUp);
bi = ubrk_openRules(ruleSourceU, -1, /* The rules */
NULL, -1, /* The text to be iterated over. */
&parseErr, &status);
if (U_FAILURE(status)) {
log_data_err("FAIL: ubrk_openRules: ICU Error \"%s\" (Are you missing data?)\n", u_errorName(status));
bi = 0;
};
freeToUCharStrings(&strCleanUp);
return bi;
}
static TextBreakIterator* setUpIteratorWithRules(bool& createdIterator, TextBreakIterator*& iterator,
const char* breakRules, const UChar* string, int length)
{
if (!string)
return 0;
if (!createdIterator) {
UParseError parseStatus;
UErrorCode openStatus = U_ZERO_ERROR;
String rules(breakRules);
iterator = reinterpret_cast<TextBreakIterator*>(ubrk_openRules(rules.characters(), rules.length(), 0, 0, &parseStatus, &openStatus));
createdIterator = true;
ASSERT_WITH_MESSAGE(U_SUCCESS(openStatus), "ICU could not open a break iterator: %s (%d)", u_errorName(openStatus), openStatus);
}
if (!iterator)
return 0;
UErrorCode setTextStatus = U_ZERO_ERROR;
ubrk_setText(reinterpret_cast<UBreakIterator*>(iterator), string, length, &setTextStatus);
if (U_FAILURE(setTextStatus))
return 0;
return iterator;
}
void
StringCaseTest::TestCasing() {
UErrorCode status = U_ZERO_ERROR;
#if !UCONFIG_NO_BREAK_ITERATION
LocalUBreakIteratorPointer iter;
#endif
char cLocaleID[100];
UnicodeString locale, input, output, optionsString, result;
uint32_t options;
int32_t whichCase, type;
LocalPointer<TestDataModule> driver(TestDataModule::getTestDataModule("casing", *this, status));
if(U_SUCCESS(status)) {
for(whichCase=0; whichCase<TEST_COUNT; ++whichCase) {
#if UCONFIG_NO_BREAK_ITERATION
if(whichCase==TEST_TITLE) {
continue;
}
#endif
LocalPointer<TestData> casingTest(driver->createTestData(dataNames[whichCase], status));
if(U_FAILURE(status)) {
errln("TestCasing failed to createTestData(%s) - %s", dataNames[whichCase], u_errorName(status));
break;
}
const DataMap *myCase = NULL;
while(casingTest->nextCase(myCase, status)) {
input = myCase->getString("Input", status);
output = myCase->getString("Output", status);
if(whichCase!=TEST_FOLD) {
locale = myCase->getString("Locale", status);
}
locale.extract(0, 0x7fffffff, cLocaleID, sizeof(cLocaleID), "");
#if !UCONFIG_NO_BREAK_ITERATION
if(whichCase==TEST_TITLE) {
type = myCase->getInt("Type", status);
if(type>=0) {
iter.adoptInstead(ubrk_open((UBreakIteratorType)type, cLocaleID, NULL, 0, &status));
} else if(type==-2) {
// Open a trivial break iterator that only delivers { 0, length }
// or even just { 0 } as boundaries.
static const UChar rules[] = { 0x2e, 0x2a, 0x3b }; // ".*;"
UParseError parseError;
iter.adoptInstead(ubrk_openRules(rules, LENGTHOF(rules), NULL, 0, &parseError, &status));
}
}
#endif
options = 0;
if(whichCase==TEST_TITLE || whichCase==TEST_FOLD) {
optionsString = myCase->getString("Options", status);
if(optionsString.indexOf((UChar)0x54)>=0) { // T
options|=U_FOLD_CASE_EXCLUDE_SPECIAL_I;
}
if(optionsString.indexOf((UChar)0x4c)>=0) { // L
options|=U_TITLECASE_NO_LOWERCASE;
}
if(optionsString.indexOf((UChar)0x41)>=0) { // A
options|=U_TITLECASE_NO_BREAK_ADJUSTMENT;
}
}
if(U_FAILURE(status)) {
dataerrln("error: TestCasing() setup failed for %s test case from casing.res: %s", dataNames[whichCase], u_errorName(status));
status = U_ZERO_ERROR;
} else {
#if UCONFIG_NO_BREAK_ITERATION
LocalPointer<UMemory> iter;
#endif
TestCasingImpl(input, output, whichCase, iter.getAlias(), cLocaleID, options);
}
#if !UCONFIG_NO_BREAK_ITERATION
iter.adoptInstead(NULL);
#endif
}
}
}
#if !UCONFIG_NO_BREAK_ITERATION
// more tests for API coverage
status=U_ZERO_ERROR;
input=UNICODE_STRING_SIMPLE("sTrA\\u00dfE").unescape();
(result=input).toTitle(NULL);
if(result!=UNICODE_STRING_SIMPLE("Stra\\u00dfe").unescape()) {
dataerrln("UnicodeString::toTitle(NULL) failed.");
}
#endif
}
UBreakIterator*
get_rules(const char *ruleFileName, UErrorCode status)
{
/* Read in the rule source file */
long result;
long ruleFileSize;
FILE *file;
OFILE *ufile;
UBreakIterator *return_me;
file = fopen(ruleFileName, "rb");
if( file == 0 ) {
fprintf(stderr, "Could not open file \"%s\"\n", ruleFileName);
exit(-1);
}
fseek(file, 0, SEEK_END);
ruleFileSize = ftell(file);
fseek(file, 0, SEEK_SET);
char *ruleBufferC = (char *) omalloc (ruleFileSize + 1);
ruleBufferC[ruleFileSize] = '\0';
result = (long)fread(ruleBufferC, 1, ruleFileSize, file);
if (result != ruleFileSize) {
fprintf(stderr, "Error reading file \"%s\"\n", ruleFileName);
exit (-1);
}
/* Look for a Unicode Signature (BOM) on the rule file */
int32_t signatureLength;
const char * ruleSourceC = ruleBufferC;
const char* encoding = ucnv_detectUnicodeSignature(
ruleSourceC, ruleFileSize, &signatureLength, &status);
/* fprintf(stderr, "DetectUnicodeSig: \"%s\"\n", encoding); */
if (U_FAILURE(status))
{
fprintf(stderr, "\nCan not initialize ICU. status = %s\n",
u_errorName(status));
exit(1);
}
if(encoding!=NULL )
{
ruleSourceC += signatureLength;
ruleFileSize -= signatureLength;
}
/* fprintf(stderr, "encoding: \"%s\"\n", encoding); */
/* Open a converter to take the rule file to UTF-16 */
UConverter* conv;
conv = ucnv_open(encoding, &status);
if (U_FAILURE(status)) {
fprintf(stderr, "ucnv_open: ICU Error \"%s\"\n", u_errorName(status));
exit(1);
}
ufile = u_finit(file, NULL, NULL);
u_frewind(ufile);
UChar *ruleSourceU = (UChar *) omalloc ((ruleFileSize*sizeof(UChar))+1);
long charsRead = u_file_read(ruleSourceU, ruleFileSize, ufile);
/* u_fprintf(u_stderr, "Chars read: \"%i\", File size: \"%i\"\n", charsRead, ruleFileSize); */
ruleSourceU[charsRead] = 0;
/* u_fprintf(u_stderr, "RulesourceU POST: \"%S\"\n", ruleSourceU); */
ucnv_close(conv);
u_fclose(ufile);
/* Create the break iterator from the rules */
/* This will compile the rules. */
UParseError parseError;
parseError.line = 0;
parseError.offset = 0;
return_me = ubrk_openRules(ruleSourceU, ruleFileSize, NULL, 0, &parseError, &status);
if (U_FAILURE(status)) {
fprintf(stderr, "createRuleBasedBreakIterator: ICU Error \"%s\" at line %d, column %d\n",
u_errorName(status), (int)parseError.line, (int)parseError.offset);
exit(1);
};
return return_me;
}
