void StringSearch::setText(CharacterIterator &text, UErrorCode &status)
{
if (U_SUCCESS(status)) {
text.getText(m_text_);
usearch_setText(m_strsrch_, m_text_.getBuffer(), m_text_.length(), &status);
}
}
void SearchIterator::setText(CharacterIterator &text, UErrorCode &status)
{
if (U_SUCCESS(status)) {
text.getText(m_text_);
setText(m_text_, status);
}
}
void printTextRange( BreakIterator& iterator,
int32_t start, int32_t end )
{
CharacterIterator *strIter = iterator.getText().clone();
UnicodeString s;
strIter->getText(s);
printf(" %ld %ld\t", (long)start, (long)end);
printUnicodeString(UnicodeString(s, 0, start));
printf("|");
printUnicodeString(UnicodeString(s, start, end-start));
printf("|");
printUnicodeString(UnicodeString(s, end));
puts("");
delete strIter;
}
Normalizer::Normalizer(const CharacterIterator& iter, UNormalizationMode mode) :
UObject(), fFilteredNorm2(NULL), fNorm2(NULL), fUMode(mode), fOptions(0),
text(iter.clone()),
currentIndex(0), nextIndex(0),
buffer(), bufferPos(0)
{
init();
}
// Sets the source to the new character iterator.
void CollationElementIterator::setText(CharacterIterator& source,
UErrorCode& status)
{
if (U_FAILURE(status))
return;
source.getText(string_);
setText(string_, status);
}
// Sets the source to the new character iterator.
void CollationElementIterator::setText(CharacterIterator& source,
UErrorCode& status)
{
if (U_FAILURE(status))
return;
int32_t length = source.getLength();
UChar *buffer = NULL;
if (length == 0) {
buffer = (UChar *)uprv_malloc(U_SIZEOF_UCHAR);
/* test for NULL */
if (buffer == NULL) {
status = U_MEMORY_ALLOCATION_ERROR;
return;
}
*buffer = 0;
}
else {
buffer = (UChar *)uprv_malloc(U_SIZEOF_UCHAR * length);
/* test for NULL */
if (buffer == NULL) {
status = U_MEMORY_ALLOCATION_ERROR;
return;
}
/*
Using this constructor will prevent buffer from being removed when
string gets removed
*/
UnicodeString string;
source.getText(string);
u_memcpy(buffer, string.getBuffer(), length);
}
if (m_data_->isWritable && m_data_->iteratordata_.string != NULL) {
uprv_free((UChar *)m_data_->iteratordata_.string);
}
m_data_->isWritable = TRUE;
/* Free offsetBuffer before initializing it. */
ucol_freeOffsetBuffer(&(m_data_->iteratordata_));
uprv_init_collIterate(m_data_->iteratordata_.coll, buffer, length,
&m_data_->iteratordata_, &status);
m_data_->reset_ = TRUE;
}
/**
* Set the input text over which this <tt>Normalizer</tt> will iterate.
* The iteration position is set to the beginning of the string.
*/
void
Normalizer::setText(const CharacterIterator& newText,
UErrorCode &status)
{
if (U_FAILURE(status)) {
return;
}
CharacterIterator *newIter = newText.clone();
if (newIter == NULL) {
status = U_MEMORY_ALLOCATION_ERROR;
return;
}
delete text;
text = newIter;
reset();
}
SearchIterator::SearchIterator(CharacterIterator &text,
BreakIterator *breakiter) :
m_breakiterator_(breakiter)
{
m_search_ = (USearch *)uprv_malloc(sizeof(USearch));
m_search_->breakIter = NULL;
m_search_->isOverlap = FALSE;
m_search_->isCanonicalMatch = FALSE;
m_search_->isForwardSearching = TRUE;
m_search_->reset = TRUE;
m_search_->matchedIndex = USEARCH_DONE;
m_search_->matchedLength = 0;
text.getText(m_text_);
m_search_->text = m_text_.getBuffer();
m_search_->textLength = m_text_.length();
m_breakiterator_ = breakiter;
}
/**
* This is the "real" constructor for this class; it constructs an iterator over
* the source text using the specified collator
*/
CollationElementIterator::CollationElementIterator(
const CharacterIterator& sourceText,
const RuleBasedCollator* order,
UErrorCode& status)
: isDataOwned_(TRUE)
{
if (U_FAILURE(status))
return;
// **** should I just drop this test? ****
/*
if ( sourceText.endIndex() != 0 )
{
// A CollationElementIterator is really a two-layered beast.
// Internally it uses a Normalizer to munge the source text into a form
// where all "composed" Unicode characters (such as \u00FC) are split into a
// normal character and a combining accent character.
// Afterward, CollationElementIterator does its own processing to handle
// expanding and contracting collation sequences, ignorables, and so on.
Normalizer::EMode decomp = order->getStrength() == Collator::IDENTICAL
? Normalizer::NO_OP : order->getDecomposition();
text = new Normalizer(sourceText, decomp);
if (text == NULL)
status = U_MEMORY_ALLOCATION_ERROR;
}
*/
int32_t length = sourceText.getLength();
UChar *buffer;
if (length > 0) {
buffer = (UChar *)uprv_malloc(U_SIZEOF_UCHAR * length);
/* test for NULL */
if (buffer == NULL) {
status = U_MEMORY_ALLOCATION_ERROR;
return;
}
/*
Using this constructor will prevent buffer from being removed when
string gets removed
*/
UnicodeString string(buffer, length, length);
((CharacterIterator &)sourceText).getText(string);
const UChar *temp = string.getBuffer();
u_memcpy(buffer, temp, length);
}
else {
buffer = (UChar *)uprv_malloc(U_SIZEOF_UCHAR);
/* test for NULL */
if (buffer == NULL) {
status = U_MEMORY_ALLOCATION_ERROR;
return;
}
*buffer = 0;
}
m_data_ = ucol_openElements(order->ucollator, buffer, length, &status);
/* Test for buffer overflows */
if (U_FAILURE(status)) {
return;
}
m_data_->isWritable = TRUE;
}
void CharIterTest::TestUCharIterator(UCharIterator *iter, CharacterIterator &ci,
const char *moves, const char *which) {
int32_t m;
UChar32 c, c2;
UBool h, h2;
for(m=0;; ++m) {
// move both iter and s[index]
switch(moves[m]) {
case '0':
h=iter->hasNext(iter);
h2=ci.hasNext();
c=iter->current(iter);
c2=ci.current();
break;
case '|':
h=iter->hasNext(iter);
h2=ci.hasNext();
c=uiter_current32(iter);
c2=ci.current32();
break;
case '+':
h=iter->hasNext(iter);
h2=ci.hasNext();
c=iter->next(iter);
c2=ci.nextPostInc();
break;
case '>':
h=iter->hasNext(iter);
h2=ci.hasNext();
c=uiter_next32(iter);
c2=ci.next32PostInc();
break;
case '-':
h=iter->hasPrevious(iter);
h2=ci.hasPrevious();
c=iter->previous(iter);
c2=ci.previous();
break;
case '<':
h=iter->hasPrevious(iter);
h2=ci.hasPrevious();
c=uiter_previous32(iter);
c2=ci.previous32();
break;
case '2':
h=h2=FALSE;
c=(UChar32)iter->move(iter, 2, UITER_CURRENT);
c2=(UChar32)ci.move(2, CharacterIterator::kCurrent);
break;
case '8':
h=h2=FALSE;
c=(UChar32)iter->move(iter, -2, UITER_CURRENT);
c2=(UChar32)ci.move(-2, CharacterIterator::kCurrent);
break;
case 0:
return;
default:
errln("error: unexpected move character '%c' in \"%s\"", moves[m], moves);
return;
}
// compare results
if(c2==0xffff) {
c2=(UChar32)-1;
}
if(c!=c2 || h!=h2 || ci.getIndex()!=iter->getIndex(iter, UITER_CURRENT)) {
errln("error: UCharIterator(%s) misbehaving at \"%s\"[%d]='%c'", which, moves, m, moves[m]);
}
}
}
void RBBIAPITest::TestGetSetAdoptText()
{
logln((UnicodeString)"Testing getText setText ");
IcuTestErrorCode status(*this, "TestGetSetAdoptText");
UnicodeString str1="first string.";
UnicodeString str2="Second string.";
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* text1= new StringCharacterIterator(str1);
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;
tci->getText(tstr);
TEST_ASSERT(tstr == str1);
if(wordIter1->current() != 0)
errln((UnicodeString)"ERROR:1 setText did not set the iteration position to the beginning of the text, it is" + wordIter1->current() + (UnicodeString)"\n");
wordIter1->next(2);
wordIter1->setText(str2);
if(wordIter1->current() != 0)
errln((UnicodeString)"ERROR:2 setText did not reset the iteration position to the beginning of the text, it is" + wordIter1->current() + (UnicodeString)"\n");
charIter1->adoptText(text1Clone);
TEST_ASSERT(wordIter1->getText() != charIter1->getText());
tci = &wordIter1->getText();
tci->getText(tstr);
TEST_ASSERT(tstr == str2);
tci = &charIter1->getText();
tci->getText(tstr);
TEST_ASSERT(tstr == str1);
LocalPointer<RuleBasedBreakIterator> rb((RuleBasedBreakIterator*)wordIter1->clone());
rb->adoptText(text1);
if(rb->getText() != *text1)
errln((UnicodeString)"ERROR:1 error in adoptText ");
rb->adoptText(text2);
if(rb->getText() != *text2)
errln((UnicodeString)"ERROR:2 error in adoptText ");
// Adopt where iterator range is less than the entire orignal source string.
// (With the change of the break engine to working with UText internally,
// CharacterIterators starting at positions other than zero are not supported)
rb->adoptText(text3);
TEST_ASSERT(rb->preceding(2) == 0);
TEST_ASSERT(rb->following(11) == BreakIterator::DONE);
//if(rb->preceding(2) != 3) {
// errln((UnicodeString)"ERROR:3 error in adoptText ");
//}
//if(rb->following(11) != BreakIterator::DONE) {
// errln((UnicodeString)"ERROR:4 error in adoptText ");
//}
// UText API
//
// Quick test to see if UText is working at all.
//
const char *s1 = "\x68\x65\x6C\x6C\x6F\x20\x77\x6F\x72\x6C\x64"; /* "hello world" in UTF-8 */
const char *s2 = "\x73\x65\x65\x20\x79\x61"; /* "see ya" in UTF-8 */
// 012345678901
status.reset();
LocalUTextPointer ut(utext_openUTF8(NULL, s1, -1, status));
wordIter1->setText(ut.getAlias(), status);
TEST_ASSERT_SUCCESS(status);
int32_t pos;
pos = wordIter1->first();
TEST_ASSERT(pos==0);
pos = wordIter1->next();
TEST_ASSERT(pos==5);
pos = wordIter1->next();
TEST_ASSERT(pos==6);
pos = wordIter1->next();
TEST_ASSERT(pos==11);
pos = wordIter1->next();
TEST_ASSERT(pos==UBRK_DONE);
status.reset();
LocalUTextPointer ut2(utext_openUTF8(NULL, s2, -1, status));
TEST_ASSERT_SUCCESS(status);
wordIter1->setText(ut2.getAlias(), status);
TEST_ASSERT_SUCCESS(status);
pos = wordIter1->first();
TEST_ASSERT(pos==0);
pos = wordIter1->next();
TEST_ASSERT(pos==3);
pos = wordIter1->next();
TEST_ASSERT(pos==4);
pos = wordIter1->last();
TEST_ASSERT(pos==6);
pos = wordIter1->previous();
TEST_ASSERT(pos==4);
pos = wordIter1->previous();
TEST_ASSERT(pos==3);
pos = wordIter1->previous();
TEST_ASSERT(pos==0);
pos = wordIter1->previous();
TEST_ASSERT(pos==UBRK_DONE);
status.reset();
UnicodeString sEmpty;
LocalUTextPointer gut2(utext_openUnicodeString(NULL, &sEmpty, status));
wordIter1->getUText(gut2.getAlias(), status);
TEST_ASSERT_SUCCESS(status);
status.reset();
}