+// © 2016 and later: Unicode, Inc. and others.
+// License & terms of use: http://www.unicode.org/copyright.html
/********************************************************************
- * COPYRIGHT:
- * Copyright (c) 1999-2006, International Business Machines Corporation and
- * others. All Rights Reserved.
+ * Copyright (c) 1999-2016, International Business Machines
+ * Corporation and others. All Rights Reserved.
+ ********************************************************************
+ * Date Name Description
+ * 12/14/99 Madhu Creation.
+ * 01/12/2000 Madhu updated for changed API
********************************************************************/
-/************************************************************************
-* Date Name Description
-* 12/14/99 Madhu Creation.
-* 01/12/2000 Madhu updated for changed API
-************************************************************************/
#include "unicode/utypes.h"
#include "rbbidata.h"
#include "cstring.h"
#include "ubrkimpl.h"
+#include "unicode/locid.h"
#include "unicode/ustring.h"
#include "unicode/utext.h"
-
+#include "cmemory.h"
+#if !UCONFIG_NO_BREAK_ITERATION
+#include "unicode/filteredbrk.h"
+#include <stdio.h> // for sprintf
+#endif
/**
* API Test the RuleBasedBreakIterator class
*/
#define TEST_ASSERT_SUCCESS(status) {if (U_FAILURE(status)) {\
-errln("Failure at file %s, line %d, error = %s", __FILE__, __LINE__, u_errorName(status));}}
+dataerrln("Failure at file %s, line %d, error = %s", __FILE__, __LINE__, u_errorName(status));}}
-#define TEST_ASSERT(expr) {if ((expr)==FALSE) { \
-errln("Test Failure at file %s, line %d", __FILE__, __LINE__);}}
+#define TEST_ASSERT(expr) {if ((expr) == FALSE) { \
+ errln("Test Failure at file %s, line %d: \"%s\" is false.\n", __FILE__, __LINE__, #expr);};}
void RBBIAPITest::TestCloneEquals()
{
RuleBasedBreakIterator* bi3 = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createCharacterInstance(Locale::getDefault(), status);
RuleBasedBreakIterator* bi2 = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createWordInstance(Locale::getDefault(), status);
if(U_FAILURE(status)){
- errln((UnicodeString)"FAIL : in construction");
+ errcheckln(status, "Fail : in construction - %s", u_errorName(status));
return;
}
errln((UnicodeString)"ERROR:2 RBBI's == and != operator failed.");
- // Quick test of RulesBasedBreakIterator assignment -
+ // Quick test of RulesBasedBreakIterator assignment -
// Check that
// two different iterators are !=
// they are == after assignment
logln("Testing assignment");
RuleBasedBreakIterator *bix = (RuleBasedBreakIterator *)BreakIterator::createLineInstance(Locale::getDefault(), status);
if(U_FAILURE(status)){
- errln((UnicodeString)"FAIL : in construction");
+ errcheckln(status, "Fail : in construction - %s", u_errorName(status));
return;
}
RuleBasedBreakIterator* bi1clone=(RuleBasedBreakIterator*)bi1->clone();
RuleBasedBreakIterator* bi2clone=(RuleBasedBreakIterator*)bi2->clone();
- if(*bi1clone != *bi1 || *bi1clone != *biequal ||
+ if(*bi1clone != *bi1 || *bi1clone != *biequal ||
*bi1clone == *bi3 || *bi1clone == *bi2)
errln((UnicodeString)"ERROR:1 RBBI's clone() method failed");
- if(*bi2clone == *bi1 || *bi2clone == *biequal ||
+ if(*bi2clone == *bi1 || *bi2clone == *biequal ||
*bi2clone == *bi3 || *bi2clone != *bi2)
errln((UnicodeString)"ERROR:2 RBBI's clone() method failed");
if(bi1->getText() != bi1clone->getText() ||
- bi2clone->getText() != bi2->getText() ||
+ bi2clone->getText() != bi2->getText() ||
*bi2clone == *bi1clone )
errln((UnicodeString)"ERROR: RBBI's clone() method failed");
BreakIterator* a = BreakIterator::createWordInstance(Locale("hi"), status);
BreakIterator* b = BreakIterator::createWordInstance(Locale("hi_IN"),status);
if (U_FAILURE(status)) {
- errln("Creation of break iterator failed %s", u_errorName(status));
+ errcheckln(status, "Creation of break iterator failed %s", u_errorName(status));
return;
}
if(*a!=*b){
errln("Failed: boilerplate method operator!= does not return correct results");
}
- BreakIterator* c = BreakIterator::createWordInstance(Locale("ja"),status);
- if(a && c){
- if(*c==*a){
- errln("Failed: boilerplate method opertator== does not return correct results");
+ // Japanese word break iterators are identical to root with
+ // a dictionary-based break iterator
+ BreakIterator* c = BreakIterator::createCharacterInstance(Locale("ja"),status);
+ BreakIterator* d = BreakIterator::createCharacterInstance(Locale("root"),status);
+ if(c && d){
+ if(*c!=*d){
+ errln("Failed: boilerplate method operator== does not return correct results");
}
}else{
errln("creation of break iterator failed");
delete a;
delete b;
delete c;
+ delete d;
}
void RBBIAPITest::TestgetRules()
RuleBasedBreakIterator* bi1=(RuleBasedBreakIterator*)RuleBasedBreakIterator::createCharacterInstance(Locale::getDefault(), status);
RuleBasedBreakIterator* bi2=(RuleBasedBreakIterator*)RuleBasedBreakIterator::createWordInstance(Locale::getDefault(), status);
if(U_FAILURE(status)){
- errln((UnicodeString)"FAIL: in construction");
+ errcheckln(status, "FAIL: in construction - %s", u_errorName(status));
delete bi1;
delete bi2;
return;
RuleBasedBreakIterator* bi3 = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createCharacterInstance(Locale::getDefault(), status);
RuleBasedBreakIterator* bi2 = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createWordInstance(Locale::getDefault(), status);
if(U_FAILURE(status)){
- errln((UnicodeString)"FAIL : in construction");
+ errcheckln(status, "Fail : in construction - %s", u_errorName(status));
delete bi1;
delete bi2;
delete bi3;
errln((UnicodeString)"ERROR: different objects have same hashcodes");
delete bi1clone;
- delete bi2clone;
+ delete bi2clone;
delete bi1;
delete bi2;
delete bi3;
void RBBIAPITest::TestGetSetAdoptText()
{
logln((UnicodeString)"Testing getText setText ");
- UErrorCode status=U_ZERO_ERROR;
+ IcuTestErrorCode status(*this, "TestGetSetAdoptText");
UnicodeString str1="first string.";
UnicodeString str2="Second string.";
- RuleBasedBreakIterator* charIter1 = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createCharacterInstance(Locale::getDefault(), status);
- RuleBasedBreakIterator* wordIter1 = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createWordInstance(Locale::getDefault(), status);
- if(U_FAILURE(status)){
- errln((UnicodeString)"FAIL : in construction");
+ LocalPointer<RuleBasedBreakIterator> charIter1((RuleBasedBreakIterator*)RuleBasedBreakIterator::createCharacterInstance(Locale::getDefault(), status));
+ LocalPointer<RuleBasedBreakIterator> wordIter1((RuleBasedBreakIterator*)RuleBasedBreakIterator::createWordInstance(Locale::getDefault(), status));
+ if(status.isFailure()){
+ errcheckln(status, "Fail : in construction - %s", status.errorName());
return;
}
CharacterIterator* text1Clone = text1->clone();
CharacterIterator* text2= new StringCharacterIterator(str2);
CharacterIterator* text3= new StringCharacterIterator(str2, 3, 10, 3); // "ond str"
-
+
wordIter1->setText(str1);
CharacterIterator *tci = &wordIter1->getText();
UnicodeString tstr;
TEST_ASSERT(tstr == str1);
- RuleBasedBreakIterator* rb=(RuleBasedBreakIterator*)wordIter1->clone();
+ LocalPointer<RuleBasedBreakIterator> rb((RuleBasedBreakIterator*)wordIter1->clone());
rb->adoptText(text1);
if(rb->getText() != *text1)
errln((UnicodeString)"ERROR:1 error in adoptText ");
const char *s2 = "\x73\x65\x65\x20\x79\x61"; /* "see ya" in UTF-8 */
// 012345678901
- status = U_ZERO_ERROR;
- UText *ut = utext_openUTF8(NULL, s1, -1, &status);
- wordIter1->setText(ut, status);
+ status.reset();
+ LocalUTextPointer ut(utext_openUTF8(NULL, s1, -1, status));
+ wordIter1->setText(ut.getAlias(), status);
TEST_ASSERT_SUCCESS(status);
int32_t pos;
pos = wordIter1->next();
TEST_ASSERT(pos==UBRK_DONE);
- status = U_ZERO_ERROR;
- UText *ut2 = utext_openUTF8(NULL, s2, -1, &status);
+ status.reset();
+ LocalUTextPointer ut2(utext_openUTF8(NULL, s2, -1, status));
TEST_ASSERT_SUCCESS(status);
- wordIter1->setText(ut2, status);
+ wordIter1->setText(ut2.getAlias(), status);
TEST_ASSERT_SUCCESS(status);
pos = wordIter1->first();
pos = wordIter1->previous();
TEST_ASSERT(pos==UBRK_DONE);
- status = U_ZERO_ERROR;
+ status.reset();
UnicodeString sEmpty;
- UText *gut2 = utext_openUnicodeString(NULL, &sEmpty, &status);
- wordIter1->getUText(gut2, status);
+ LocalUTextPointer gut2(utext_openUnicodeString(NULL, &sEmpty, status));
+ wordIter1->getUText(gut2.getAlias(), status);
TEST_ASSERT_SUCCESS(status);
- utext_close(gut2);
-
- utext_close(ut);
- utext_close(ut2);
-
- delete wordIter1;
- delete charIter1;
- delete rb;
+ status.reset();
+}
- }
-
void RBBIAPITest::TestIteration()
{
// This test just verifies that the API is present.
UErrorCode status=U_ZERO_ERROR;
RuleBasedBreakIterator* bi = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createCharacterInstance(Locale::getDefault(), status);
if (U_FAILURE(status) || bi == NULL) {
- errln("Failure creating character break iterator. Status = %s", u_errorName(status));
+ errcheckln(status, "Failure creating character break iterator. Status = %s", u_errorName(status));
}
delete bi;
status=U_ZERO_ERROR;
bi = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createWordInstance(Locale::getDefault(), status);
if (U_FAILURE(status) || bi == NULL) {
- errln("Failure creating Word break iterator. Status = %s", u_errorName(status));
+ errcheckln(status, "Failure creating Word break iterator. Status = %s", u_errorName(status));
}
delete bi;
status=U_ZERO_ERROR;
bi = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createLineInstance(Locale::getDefault(), status);
if (U_FAILURE(status) || bi == NULL) {
- errln("Failure creating Line break iterator. Status = %s", u_errorName(status));
+ errcheckln(status, "Failure creating Line break iterator. Status = %s", u_errorName(status));
}
delete bi;
status=U_ZERO_ERROR;
bi = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createSentenceInstance(Locale::getDefault(), status);
if (U_FAILURE(status) || bi == NULL) {
- errln("Failure creating Sentence break iterator. Status = %s", u_errorName(status));
+ errcheckln(status, "Failure creating Sentence break iterator. Status = %s", u_errorName(status));
}
delete bi;
status=U_ZERO_ERROR;
bi = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createTitleInstance(Locale::getDefault(), status);
if (U_FAILURE(status) || bi == NULL) {
- errln("Failure creating Title break iterator. Status = %s", u_errorName(status));
+ errcheckln(status, "Failure creating Title break iterator. Status = %s", u_errorName(status));
}
delete bi;
status=U_ZERO_ERROR;
bi = (RuleBasedBreakIterator*)RuleBasedBreakIterator::createCharacterInstance(Locale::getDefault(), status);
if (U_FAILURE(status) || bi == NULL) {
- errln("Failure creating character break iterator. Status = %s", u_errorName(status));
+ errcheckln(status, "Failure creating character break iterator. Status = %s", u_errorName(status));
return; // Skip the rest of these tests.
}
int32_t bounds1[] = {0, 3, 6, 7, 8, 11};
UErrorCode status=U_ZERO_ERROR;
UParseError parseError;
-
+
RuleBasedBreakIterator *bi = new RuleBasedBreakIterator(rulesString1, parseError, status);
if(U_FAILURE(status)) {
- errln("FAIL : in construction");
+ dataerrln("Fail : in construction - %s", u_errorName(status));
} else {
bi->setText(testString1);
doBoundaryTest(*bi, testString1, bounds1);
int32_t bounds1[] = {0, 6, 7, 10, 11, 12};
UErrorCode status=U_ZERO_ERROR;
UParseError parseError;
-
+
RuleBasedBreakIterator *bi = new RuleBasedBreakIterator(rulesString1, parseError, status);
if(U_FAILURE(status)) {
- errln("FAIL : in construction");
+ dataerrln("Fail : in construction - %s", u_errorName(status));
} else {
bi->setText(testString1);
doBoundaryTest(*bi, testString1, bounds1);
// Test word break rule status constants.
//
void RBBIAPITest::TestRuleStatus() {
- UChar str[30];
- u_unescape("plain word 123.45 \\u9160\\u9161 \\u30a1\\u30a2 \\u3041\\u3094",
- // 012345678901234567 8 9 0 1 2 3 4 5 6
- // Ideographic Katakana Hiragana
+ UChar str[30];
+ //no longer test Han or hiragana breaking here: ruleStatusVec would return nothing
+ // changed UBRK_WORD_KANA to UBRK_WORD_IDEO
+ u_unescape("plain word 123.45 \\u30a1\\u30a2 ",
+ // 012345678901234567 8 9 0
+ // Katakana
str, 30);
UnicodeString testString1(str);
- int32_t bounds1[] = {0, 5, 6, 10, 11, 17, 18, 19, 20, 21, 23, 24, 25, 26};
+ int32_t bounds1[] = {0, 5, 6, 10, 11, 17, 18, 20, 21};
int32_t tag_lo[] = {UBRK_WORD_NONE, UBRK_WORD_LETTER, UBRK_WORD_NONE, UBRK_WORD_LETTER,
UBRK_WORD_NONE, UBRK_WORD_NUMBER, UBRK_WORD_NONE,
- UBRK_WORD_IDEO, UBRK_WORD_IDEO, UBRK_WORD_NONE,
- UBRK_WORD_KANA, UBRK_WORD_NONE, UBRK_WORD_KANA, UBRK_WORD_KANA};
+ UBRK_WORD_IDEO, UBRK_WORD_NONE};
int32_t tag_hi[] = {UBRK_WORD_NONE_LIMIT, UBRK_WORD_LETTER_LIMIT, UBRK_WORD_NONE_LIMIT, UBRK_WORD_LETTER_LIMIT,
UBRK_WORD_NONE_LIMIT, UBRK_WORD_NUMBER_LIMIT, UBRK_WORD_NONE_LIMIT,
- UBRK_WORD_IDEO_LIMIT, UBRK_WORD_IDEO_LIMIT, UBRK_WORD_NONE_LIMIT,
- UBRK_WORD_KANA_LIMIT, UBRK_WORD_NONE_LIMIT, UBRK_WORD_KANA_LIMIT, UBRK_WORD_KANA_LIMIT};
+ UBRK_WORD_IDEO_LIMIT, UBRK_WORD_NONE_LIMIT};
UErrorCode status=U_ZERO_ERROR;
-
- RuleBasedBreakIterator *bi = (RuleBasedBreakIterator *)BreakIterator::createWordInstance(Locale::getEnglish(), status);
+
+ BreakIterator *bi = BreakIterator::createWordInstance(Locale::getEnglish(), status);
if(U_FAILURE(status)) {
- errln("FAIL : in construction");
+ errcheckln(status, "Fail : in construction - %s", u_errorName(status));
} else {
bi->setText(testString1);
// First test that the breaks are in the right spots.
errln("FAIL: incorrect tag value %d at position %d", tag, pos);
break;
}
-
+
// Check that we get the same tag values from getRuleStatusVec()
int32_t vec[10];
int t = bi->getRuleStatusVec(vec, 10, status);
testString1 = "test line. \n";
// break type s s h
- bi = (RuleBasedBreakIterator *)
- BreakIterator::createLineInstance(Locale::getEnglish(), status);
+ bi = BreakIterator::createLineInstance(Locale::getEnglish(), status);
if(U_FAILURE(status)) {
- errln("failed to create word break iterator.");
+ errcheckln(status, "failed to create word break iterator. - %s", u_errorName(status));
} else {
int32_t i = 0;
int32_t pos, tag;
}
if (UBRK_LINE_SOFT >= UBRK_LINE_SOFT_LIMIT ||
UBRK_LINE_HARD >= UBRK_LINE_HARD_LIMIT ||
- UBRK_LINE_HARD > UBRK_LINE_SOFT && UBRK_LINE_HARD < UBRK_LINE_SOFT_LIMIT ) {
+ (UBRK_LINE_HARD > UBRK_LINE_SOFT && UBRK_LINE_HARD < UBRK_LINE_SOFT_LIMIT)) {
errln("UBRK_LINE_* constants from header are inconsistent.");
}
}
// Test the vector form of break rule status.
//
void RBBIAPITest::TestRuleStatusVec() {
- UnicodeString rulesString = "[A-N]{100}; \n"
+ UnicodeString rulesString( "[A-N]{100}; \n"
"[a-w]{200}; \n"
"[\\p{L}]{300}; \n"
"[\\p{N}]{400}; \n"
"[0-5]{500}; \n"
- "!.*;\n";
+ "!.*;\n", -1, US_INV);
UnicodeString testString1 = "Aapz5?";
int32_t statusVals[10];
int32_t numStatuses;
UErrorCode status=U_ZERO_ERROR;
UParseError parseError;
-
+
RuleBasedBreakIterator *bi = new RuleBasedBreakIterator(rulesString, parseError, status);
- TEST_ASSERT_SUCCESS(status);
- if (U_SUCCESS(status)) {
+ if (U_FAILURE(status)) {
+ dataerrln("Failure at file %s, line %d, error = %s", __FILE__, __LINE__, u_errorName(status));
+ } else {
bi->setText(testString1);
// A
TEST_ASSERT(statusVals[0] == 0);
//
- // Check buffer overflow error handling. Char == A
+ // Check buffer overflow error handling. Char == A
//
bi->first();
pos = bi->next();
int32_t bounds1[] = {0, 4, 8};
UErrorCode status=U_ZERO_ERROR;
UParseError parseError;
-
+
RuleBasedBreakIterator *bi = new RuleBasedBreakIterator(rulesString1, parseError, status);
if(U_FAILURE(status)) {
- errln("FAIL : in construction");
+ dataerrln("Fail : in construction - %s", u_errorName(status));
} else {
bi->setText(testString1);
doBoundaryTest(*bi, testString1, bounds1);
#if !UCONFIG_NO_SERVICE
UErrorCode status = U_ZERO_ERROR;
BreakIterator* ja_word = BreakIterator::createWordInstance("ja_JP", status);
-
// ok to not delete these if we exit because of error?
BreakIterator* ja_char = BreakIterator::createCharacterInstance("ja_JP", status);
BreakIterator* root_word = BreakIterator::createWordInstance("", status);
BreakIterator* root_char = BreakIterator::createCharacterInstance("", status);
-
+
+ if (status == U_MISSING_RESOURCE_ERROR || status == U_FILE_ACCESS_ERROR) {
+ dataerrln("Error creating instances of break interactors - %s", u_errorName(status));
+
+ delete ja_word;
+ delete ja_char;
+ delete root_word;
+ delete root_char;
+
+ return;
+ }
+
URegistryKey key = BreakIterator::registerInstance(ja_word, "xx", UBRK_WORD, status);
{
+#if 0 // With a dictionary based word breaking, ja_word is identical to root.
if (ja_word && *ja_word == *root_word) {
errln("japan not different from root");
}
+#endif
}
-
+
{
BreakIterator* result = BreakIterator::createWordInstance("xx_XX", status);
UBool fail = TRUE;
errln("bad result for xx_XX/word");
}
}
-
+
{
BreakIterator* result = BreakIterator::createCharacterInstance("ja_JP", status);
UBool fail = TRUE;
errln("bad result for ja_JP/char");
}
}
-
+
{
BreakIterator* result = BreakIterator::createCharacterInstance("xx_XX", status);
UBool fail = TRUE;
errln("bad result for xx_XX/char");
}
}
-
+
{
StringEnumeration* avail = BreakIterator::getAvailableLocales();
UBool found = FALSE;
errln("did not find test locale");
}
}
-
+
{
UBool unreg = BreakIterator::unregister(key, status);
if (!unreg) {
errln("unable to unregister");
}
}
-
+
{
BreakIterator* result = BreakIterator::createWordInstance("en_US", status);
BreakIterator* root = BreakIterator::createWordInstance("", status);
errln("did not get root break");
}
}
-
+
{
StringEnumeration* avail = BreakIterator::getAvailableLocales();
UBool found = FALSE;
errln("found test locale");
}
}
-
+
{
int32_t count;
UBool foundLocale = FALSE;
errln("BreakIterator::getAvailableLocales(&count), failed to find EN.");
}
}
-
-
+
+
// ja_word was adopted by factory
delete ja_char;
delete root_word;
UParseError parseError;
parseError.line = 0;
parseError.offset = 0;
- UDataMemory *data = udata_open(U_ICUDATA_BRKITR, "brk", dataFile, &status);
+ LocalUDataMemoryPointer data(udata_open(U_ICUDATA_BRKITR, "brk", dataFile, &status));
uint32_t length;
const UChar *builtSource;
const uint8_t *rbbiRules;
const uint8_t *builtRules;
if (U_FAILURE(status)) {
- errln("Can't open \"%s\"", dataFile);
+ errcheckln(status, "%s:%d Can't open \"%s\" - %s", __FILE__, __LINE__, dataFile, u_errorName(status));
return;
}
- builtRules = (const uint8_t *)udata_getMemory(data);
+ builtRules = (const uint8_t *)udata_getMemory(data.getAlias());
builtSource = (const UChar *)(builtRules + ((RBBIDataHeader*)builtRules)->fRuleSource);
RuleBasedBreakIterator *brkItr = new RuleBasedBreakIterator(builtSource, parseError, status);
if (U_FAILURE(status)) {
- errln("createRuleBasedBreakIterator: ICU Error \"%s\" at line %d, column %d\n",
- u_errorName(status), parseError.line, parseError.offset);
+ errln("%s:%d createRuleBasedBreakIterator: ICU Error \"%s\" at line %d, column %d\n",
+ __FILE__, __LINE__, u_errorName(status), parseError.line, parseError.offset);
+ errln(UnicodeString(builtSource));
return;
};
rbbiRules = brkItr->getBinaryRules(length);
logln("Comparing \"%s\" len=%d", dataFile, length);
if (memcmp(builtRules, rbbiRules, (int32_t)length) != 0) {
- errln("Built rules and rebuilt rules are different %s", dataFile);
+ errln("%s:%d Built rules and rebuilt rules are different %s", __FILE__, __LINE__, dataFile);
return;
}
delete brkItr;
- udata_close(data);
}
void RBBIAPITest::TestRoundtripRules() {
RoundtripRule("line");
RoundtripRule("char");
if (!quick) {
- RoundtripRule("word_ja");
RoundtripRule("word_POSIX");
}
}
+
+// Check getBinaryRules() and construction of a break iterator from those rules.
+
+void RBBIAPITest::TestGetBinaryRules() {
+ UErrorCode status=U_ZERO_ERROR;
+ LocalPointer<BreakIterator> bi(BreakIterator::createLineInstance(Locale::getEnglish(), status));
+ if (U_FAILURE(status)) {
+ dataerrln("FAIL: BreakIterator::createLineInstance for Locale::getEnglish(): %s", u_errorName(status));
+ return;
+ }
+ RuleBasedBreakIterator *rbbi = dynamic_cast<RuleBasedBreakIterator *>(bi.getAlias());
+ if (rbbi == NULL) {
+ dataerrln("FAIL: RuleBasedBreakIterator is NULL");
+ return;
+ }
+
+ // Check that the new line break iterator is nominally functional.
+ UnicodeString helloWorld("Hello, World!");
+ rbbi->setText(helloWorld);
+ int n = 0;
+ while (bi->next() != UBRK_DONE) {
+ ++n;
+ }
+ TEST_ASSERT(n == 2);
+
+ // Extract the binary rules as a uint8_t blob.
+ uint32_t ruleLength;
+ const uint8_t *binRules = rbbi->getBinaryRules(ruleLength);
+ TEST_ASSERT(ruleLength > 0);
+ TEST_ASSERT(binRules != NULL);
+
+ // Clone the binary rules, and create a break iterator from that.
+ // The break iterator does not adopt the rules; we must delete when we are finished with the iterator.
+ uint8_t *clonedRules = new uint8_t[ruleLength];
+ memcpy(clonedRules, binRules, ruleLength);
+ RuleBasedBreakIterator clonedBI(clonedRules, ruleLength, status);
+ TEST_ASSERT_SUCCESS(status);
+
+ // Check that the cloned line break iterator is nominally alive.
+ clonedBI.setText(helloWorld);
+ n = 0;
+ while (clonedBI.next() != UBRK_DONE) {
+ ++n;
+ }
+ TEST_ASSERT(n == 2);
+
+ delete[] clonedRules;
+}
+
+
+void RBBIAPITest::TestRefreshInputText() {
+ /*
+ * RefreshInput changes out the input of a Break Iterator without
+ * changing anything else in the iterator's state. Used with Java JNI,
+ * when Java moves the underlying string storage. This test
+ * runs BreakIterator::next() repeatedly, moving the text in the middle of the sequence.
+ * The right set of boundaries should still be found.
+ */
+ UChar testStr[] = {0x20, 0x41, 0x20, 0x42, 0x20, 0x43, 0x20, 0x44, 0x0}; /* = " A B C D" */
+ UChar movedStr[] = {0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0};
+ UErrorCode status = U_ZERO_ERROR;
+ UText ut1 = UTEXT_INITIALIZER;
+ UText ut2 = UTEXT_INITIALIZER;
+ RuleBasedBreakIterator *bi = (RuleBasedBreakIterator *)BreakIterator::createLineInstance(Locale::getEnglish(), status);
+ TEST_ASSERT_SUCCESS(status);
+
+ utext_openUChars(&ut1, testStr, -1, &status);
+ TEST_ASSERT_SUCCESS(status);
+
+ if (U_SUCCESS(status)) {
+ bi->setText(&ut1, status);
+ TEST_ASSERT_SUCCESS(status);
+
+ /* Line boundaries will occur before each letter in the original string */
+ TEST_ASSERT(1 == bi->next());
+ TEST_ASSERT(3 == bi->next());
+
+ /* Move the string, kill the original string. */
+ u_strcpy(movedStr, testStr);
+ u_memset(testStr, 0x20, u_strlen(testStr));
+ utext_openUChars(&ut2, movedStr, -1, &status);
+ TEST_ASSERT_SUCCESS(status);
+ RuleBasedBreakIterator *returnedBI = &bi->refreshInputText(&ut2, status);
+ TEST_ASSERT_SUCCESS(status);
+ TEST_ASSERT(bi == returnedBI);
+
+ /* Find the following matches, now working in the moved string. */
+ TEST_ASSERT(5 == bi->next());
+ TEST_ASSERT(7 == bi->next());
+ TEST_ASSERT(8 == bi->next());
+ TEST_ASSERT(UBRK_DONE == bi->next());
+
+ utext_close(&ut1);
+ utext_close(&ut2);
+ }
+ delete bi;
+
+}
+
+#if !UCONFIG_NO_BREAK_ITERATION && !UCONFIG_NO_FILTERED_BREAK_ITERATION
+static void prtbrks(BreakIterator* brk, const UnicodeString &ustr, IntlTest &it) {
+ static const UChar PILCROW=0x00B6, CHSTR=0x3010, CHEND=0x3011; // lenticular brackets
+ it.logln(UnicodeString("String:'")+ustr+UnicodeString("'"));
+
+ int32_t *pos = new int32_t[ustr.length()];
+ int32_t posCount = 0;
+
+ // calculate breaks up front, so we can print out
+ // sans any debugging
+ for(int32_t n = 0; (n=brk->next())!=UBRK_DONE; ) {
+ pos[posCount++] = n;
+ if(posCount>=ustr.length()) {
+ it.errln("brk count exceeds string length!");
+ return;
+ }
+ }
+ UnicodeString out;
+ out.append((UChar)CHSTR);
+ int32_t prev = 0;
+ for(int32_t i=0;i<posCount;i++) {
+ int32_t n=pos[i];
+ out.append(ustr.tempSubString(prev,n-prev));
+ out.append((UChar)PILCROW);
+ prev=n;
+ }
+ out.append(ustr.tempSubString(prev,ustr.length()-prev));
+ out.append((UChar)CHEND);
+ it.logln(out);
+
+ out.remove();
+ for(int32_t i=0;i<posCount;i++) {
+ char tmp[100];
+ sprintf(tmp,"%d ",pos[i]);
+ out.append(UnicodeString(tmp));
+ }
+ it.logln(out);
+ delete [] pos;
+}
+#endif
+
+void RBBIAPITest::TestFilteredBreakIteratorBuilder() {
+#if !UCONFIG_NO_BREAK_ITERATION && !UCONFIG_NO_FILTERED_BREAK_ITERATION
+ UErrorCode status = U_ZERO_ERROR;
+ LocalPointer<FilteredBreakIteratorBuilder> builder;
+ LocalPointer<BreakIterator> baseBI;
+ LocalPointer<BreakIterator> filteredBI;
+ LocalPointer<BreakIterator> frenchBI;
+
+ const UnicodeString text("In the meantime Mr. Weston arrived with his small ship, which he had now recovered. Capt. Gorges, who informed the Sgt. here that one purpose of his going east was to meet with Mr. Weston, took this opportunity to call him to account for some abuses he had to lay to his charge."); // (William Bradford, public domain. http://catalog.hathitrust.org/Record/008651224 ) - edited.
+ const UnicodeString ABBR_MR("Mr.");
+ const UnicodeString ABBR_CAPT("Capt.");
+
+ {
+ logln("Constructing empty builder\n");
+ builder.adoptInstead(FilteredBreakIteratorBuilder::createInstance(status));
+ TEST_ASSERT_SUCCESS(status);
+
+ logln("Constructing base BI\n");
+ baseBI.adoptInstead(BreakIterator::createSentenceInstance(Locale::getEnglish(), status));
+ TEST_ASSERT_SUCCESS(status);
+
+ logln("Building new BI\n");
+ filteredBI.adoptInstead(builder->build(baseBI.orphan(), status));
+ TEST_ASSERT_SUCCESS(status);
+
+ if (U_SUCCESS(status)) {
+ logln("Testing:");
+ filteredBI->setText(text);
+ TEST_ASSERT(20 == filteredBI->next()); // Mr.
+ TEST_ASSERT(84 == filteredBI->next()); // recovered.
+ TEST_ASSERT(90 == filteredBI->next()); // Capt.
+ TEST_ASSERT(181 == filteredBI->next()); // Mr.
+ TEST_ASSERT(278 == filteredBI->next()); // charge.
+ filteredBI->first();
+ prtbrks(filteredBI.getAlias(), text, *this);
+ }
+ }
+
+ {
+ logln("Constructing empty builder\n");
+ builder.adoptInstead(FilteredBreakIteratorBuilder::createInstance(status));
+ TEST_ASSERT_SUCCESS(status);
+
+ if (U_SUCCESS(status)) {
+ logln("Adding Mr. as an exception\n");
+ TEST_ASSERT(TRUE == builder->suppressBreakAfter(ABBR_MR, status));
+ TEST_ASSERT(FALSE == builder->suppressBreakAfter(ABBR_MR, status)); // already have it
+ TEST_ASSERT(TRUE == builder->unsuppressBreakAfter(ABBR_MR, status));
+ TEST_ASSERT(FALSE == builder->unsuppressBreakAfter(ABBR_MR, status)); // already removed it
+ TEST_ASSERT(TRUE == builder->suppressBreakAfter(ABBR_MR, status));
+ TEST_ASSERT_SUCCESS(status);
+
+ logln("Constructing base BI\n");
+ baseBI.adoptInstead(BreakIterator::createSentenceInstance(Locale::getEnglish(), status));
+ TEST_ASSERT_SUCCESS(status);
+
+ logln("Building new BI\n");
+ filteredBI.adoptInstead(builder->build(baseBI.orphan(), status));
+ TEST_ASSERT_SUCCESS(status);
+
+ logln("Testing:");
+ filteredBI->setText(text);
+ TEST_ASSERT(84 == filteredBI->next());
+ TEST_ASSERT(90 == filteredBI->next());// Capt.
+ TEST_ASSERT(278 == filteredBI->next());
+ filteredBI->first();
+ prtbrks(filteredBI.getAlias(), text, *this);
+ }
+ }
+
+
+ {
+ logln("Constructing empty builder\n");
+ builder.adoptInstead(FilteredBreakIteratorBuilder::createInstance(status));
+ TEST_ASSERT_SUCCESS(status);
+
+ if (U_SUCCESS(status)) {
+ logln("Adding Mr. and Capt as an exception\n");
+ TEST_ASSERT(TRUE == builder->suppressBreakAfter(ABBR_MR, status));
+ TEST_ASSERT(TRUE == builder->suppressBreakAfter(ABBR_CAPT, status));
+ TEST_ASSERT_SUCCESS(status);
+
+ logln("Constructing base BI\n");
+ baseBI.adoptInstead(BreakIterator::createSentenceInstance(Locale::getEnglish(), status));
+ TEST_ASSERT_SUCCESS(status);
+
+ logln("Building new BI\n");
+ filteredBI.adoptInstead(builder->build(baseBI.orphan(), status));
+ TEST_ASSERT_SUCCESS(status);
+
+ logln("Testing:");
+ filteredBI->setText(text);
+ TEST_ASSERT(84 == filteredBI->next());
+ TEST_ASSERT(278 == filteredBI->next());
+ filteredBI->first();
+ prtbrks(filteredBI.getAlias(), text, *this);
+ }
+ }
+
+
+ {
+ logln("Constructing English builder\n");
+ builder.adoptInstead(FilteredBreakIteratorBuilder::createInstance(Locale::getEnglish(), status));
+ TEST_ASSERT_SUCCESS(status);
+
+ logln("Constructing base BI\n");
+ baseBI.adoptInstead(BreakIterator::createSentenceInstance(Locale::getEnglish(), status));
+ TEST_ASSERT_SUCCESS(status);
+
+ if (U_SUCCESS(status)) {
+ logln("unsuppressing 'Capt'");
+ TEST_ASSERT(TRUE == builder->unsuppressBreakAfter(ABBR_CAPT, status));
+
+ logln("Building new BI\n");
+ filteredBI.adoptInstead(builder->build(baseBI.orphan(), status));
+ TEST_ASSERT_SUCCESS(status);
+
+ if(filteredBI.isValid()) {
+ logln("Testing:");
+ filteredBI->setText(text);
+ TEST_ASSERT(84 == filteredBI->next());
+ TEST_ASSERT(90 == filteredBI->next());
+ TEST_ASSERT(278 == filteredBI->next());
+ filteredBI->first();
+ prtbrks(filteredBI.getAlias(), text, *this);
+ }
+ }
+ }
+
+
+ {
+ logln("Constructing English builder\n");
+ builder.adoptInstead(FilteredBreakIteratorBuilder::createInstance(Locale::getEnglish(), status));
+ TEST_ASSERT_SUCCESS(status);
+
+ logln("Constructing base BI\n");
+ baseBI.adoptInstead(BreakIterator::createSentenceInstance(Locale::getEnglish(), status));
+ TEST_ASSERT_SUCCESS(status);
+
+ if (U_SUCCESS(status)) {
+ logln("Building new BI\n");
+ filteredBI.adoptInstead(builder->build(baseBI.orphan(), status));
+ TEST_ASSERT_SUCCESS(status);
+
+ if(filteredBI.isValid()) {
+ logln("Testing:");
+ filteredBI->setText(text);
+ TEST_ASSERT(84 == filteredBI->next());
+ TEST_ASSERT(278 == filteredBI->next());
+ filteredBI->first();
+ prtbrks(filteredBI.getAlias(), text, *this);
+ }
+ }
+ }
+
+ // reenable once french is in
+ {
+ logln("Constructing French builder");
+ builder.adoptInstead(FilteredBreakIteratorBuilder::createInstance(Locale::getFrench(), status));
+ TEST_ASSERT_SUCCESS(status);
+
+ logln("Constructing base BI\n");
+ baseBI.adoptInstead(BreakIterator::createSentenceInstance(Locale::getFrench(), status));
+ TEST_ASSERT_SUCCESS(status);
+
+ if (U_SUCCESS(status)) {
+ logln("Building new BI\n");
+ frenchBI.adoptInstead(builder->build(baseBI.orphan(), status));
+ TEST_ASSERT_SUCCESS(status);
+ }
+
+ if(frenchBI.isValid()) {
+ logln("Testing:");
+ UnicodeString frText("C'est MM. Duval.");
+ frenchBI->setText(frText);
+ TEST_ASSERT(16 == frenchBI->next());
+ TEST_ASSERT(BreakIterator::DONE == frenchBI->next());
+ frenchBI->first();
+ prtbrks(frenchBI.getAlias(), frText, *this);
+ logln("Testing against English:");
+ filteredBI->setText(frText);
+ TEST_ASSERT(10 == filteredBI->next()); // wrong for french, but filterBI is english.
+ TEST_ASSERT(16 == filteredBI->next());
+ TEST_ASSERT(BreakIterator::DONE == filteredBI->next());
+ filteredBI->first();
+ prtbrks(filteredBI.getAlias(), frText, *this);
+
+ // Verify ==
+ TEST_ASSERT_TRUE(*frenchBI == *frenchBI);
+ TEST_ASSERT_TRUE(*filteredBI != *frenchBI);
+ TEST_ASSERT_TRUE(*frenchBI != *filteredBI);
+ } else {
+ dataerrln("French BI: not valid.");
+ }
+ }
+
+#else
+ logln("Skipped- not: !UCONFIG_NO_BREAK_ITERATION && !UCONFIG_NO_FILTERED_BREAK_ITERATION");
+#endif
+}
+
//---------------------------------------------
// runIndexedTest
//---------------------------------------------
void RBBIAPITest::runIndexedTest( int32_t index, UBool exec, const char* &name, char* /*par*/ )
{
if (exec) logln((UnicodeString)"TestSuite RuleBasedBreakIterator API ");
- switch (index) {
- // case 0: name = "TestConstruction"; if (exec) TestConstruction(); break;
- case 0: name = "TestCloneEquals"; if (exec) TestCloneEquals(); break;
- case 1: name = "TestgetRules"; if (exec) TestgetRules(); break;
- case 2: name = "TestHashCode"; if (exec) TestHashCode(); break;
- case 3: name = "TestGetSetAdoptText"; if (exec) TestGetSetAdoptText(); break;
- case 4: name = "TestIteration"; if (exec) TestIteration(); break;
- case 5: name = "TestBuilder"; if (exec) TestBuilder(); break;
- case 6: name = "TestQuoteGrouping"; if (exec) TestQuoteGrouping(); break;
- case 7: name = "TestRuleStatus"; if (exec) TestRuleStatus(); break;
- case 8: name = "TestRuleStatusVec"; if (exec) TestRuleStatusVec(); break;
- case 9: name = "TestBug2190"; if (exec) TestBug2190(); break;
- case 10: name = "TestRegistration"; if (exec) TestRegistration(); break;
- case 11: name = "TestBoilerPlate"; if (exec) TestBoilerPlate(); break;
- case 12: name = "TestRoundtripRules"; if (exec) TestRoundtripRules(); break;
-
- default: name = ""; break; // needed to end loop
- }
+ TESTCASE_AUTO_BEGIN;
+#if !UCONFIG_NO_FILE_IO
+ TESTCASE_AUTO(TestCloneEquals);
+ TESTCASE_AUTO(TestgetRules);
+ TESTCASE_AUTO(TestHashCode);
+ TESTCASE_AUTO(TestGetSetAdoptText);
+ TESTCASE_AUTO(TestIteration);
+#endif
+ TESTCASE_AUTO(TestBuilder);
+ TESTCASE_AUTO(TestQuoteGrouping);
+ TESTCASE_AUTO(TestRuleStatusVec);
+ TESTCASE_AUTO(TestBug2190);
+#if !UCONFIG_NO_FILE_IO
+ TESTCASE_AUTO(TestRegistration);
+ TESTCASE_AUTO(TestBoilerPlate);
+ TESTCASE_AUTO(TestRuleStatus);
+ TESTCASE_AUTO(TestRoundtripRules);
+ TESTCASE_AUTO(TestGetBinaryRules);
+#endif
+ TESTCASE_AUTO(TestRefreshInputText);
+#if !UCONFIG_NO_BREAK_ITERATION
+ TESTCASE_AUTO(TestFilteredBreakIteratorBuilder);
+#endif
+ TESTCASE_AUTO_END;
}
+
//---------------------------------------------
//Internal subroutines
//---------------------------------------------
-void RBBIAPITest::doBoundaryTest(RuleBasedBreakIterator& bi, UnicodeString& text, int32_t *boundaries){
+void RBBIAPITest::doBoundaryTest(BreakIterator& bi, UnicodeString& text, int32_t *boundaries){
logln((UnicodeString)"testIsBoundary():");
int32_t p = 0;
UBool isB;
if(gotoffset != expectedOffset)
errln((UnicodeString)"ERROR:****returned #" + gotoffset + (UnicodeString)" instead of #" + expectedOffset);
if(start <= gotoffset){
- testString.extractBetween(start, gotoffset, selected);
+ testString.extractBetween(start, gotoffset, selected);
}
else{
testString.extractBetween(gotoffset, start, selected);
projects / apple / icu.git / blobdiff