void testIncludeLowerTrue(CuTest* tc)
{
WhitespaceAnalyzer a;
RAMDirectory* index = _CLNEW RAMDirectory();
IndexWriter* writer = _CLNEW IndexWriter(index,
&a, true);
Document doc;
doc.add(*_CLNEW Field(_T("Category"), _T("a 1"), Field::STORE_YES | Field::INDEX_TOKENIZED));
writer->addDocument(&doc); doc.clear();
doc.add(*_CLNEW Field(_T("Category"), _T("a 2"), Field::STORE_YES | Field::INDEX_TOKENIZED));
writer->addDocument(&doc); doc.clear();
doc.add(*_CLNEW Field(_T("Category"), _T("a 3"), Field::STORE_YES | Field::INDEX_TOKENIZED));
writer->addDocument(&doc); doc.clear();
writer->close();
_CLLDELETE(writer);
IndexSearcher* s = _CLNEW IndexSearcher(index);
Filter* f = _CLNEW RangeFilter(_T("Category"), _T("3"), _T("3"), true, true);
Term* t = _CLNEW Term(_T("Category"), _T("a"));
Query* q1 = _CLNEW TermQuery(t);
_CLLDECDELETE(t);
t = _CLNEW Term(_T("Category"), _T("3"));
Query* q2 = _CLNEW TermQuery(t);
_CLLDECDELETE(t);
Hits* h = s->search(q1);
assertTrue(h->length() == 3);
_CLLDELETE(h);
h = s->search(q2);
assertTrue(h->length() == 1);
_CLLDELETE(h);
h = s->search(q1, f);
assertTrue(h->length() == 1);
_CLLDELETE(h);
s->close();
_CLLDELETE(s);
_CLLDELETE(q1);
_CLLDELETE(q2);
_CLLDELETE(f);
index->close();
_CLLDECDELETE(index);
}
void TestBasics::testBoolean2()
{
Term * term1 = _CLNEW Term( _T( "field" ), _T( "sevento" ));
Term * term2 = _CLNEW Term( _T( "field" ), _T( "sevenly" ));
BooleanQuery * query = _CLNEW BooleanQuery();
query->add( _CLNEW TermQuery( term1 ), true, BooleanClause::MUST );
query->add( _CLNEW TermQuery( term2 ), true, BooleanClause::MUST );
_CLLDECDELETE( term1 );
_CLLDECDELETE( term2 );
checkHits( query, NULL, 0 );
_CLLDELETE( query );
}
void TestBasics::testBoolean()
{
int32_t expectedDocs[] = {77, 777, 177, 277, 377, 477, 577, 677, 770, 771, 772, 773, 774, 775, 776, 778, 779, 877, 977};
Term * term1 = _CLNEW Term( _T( "field" ), _T( "seventy" ));
Term * term2 = _CLNEW Term( _T( "field" ), _T( "seven" ));
BooleanQuery * query = _CLNEW BooleanQuery();
query->add( _CLNEW TermQuery( term1 ), true, BooleanClause::MUST );
query->add( _CLNEW TermQuery( term2 ), true, BooleanClause::MUST );
_CLLDECDELETE( term1 );
_CLLDECDELETE( term2 );
checkHits( query, expectedDocs, sizeof( expectedDocs ) / sizeof( expectedDocs[ 0 ] ));
_CLLDELETE( query );
}
/// TestBooleanScorer.java, ported 5/9/2009
void testBooleanScorer(CuTest *tc) {
const TCHAR* FIELD = _T("category");
RAMDirectory directory;
TCHAR* values[] = { _T("1"), _T("2"), _T("3"), _T("4"), NULL};
try {
WhitespaceAnalyzer a;
IndexWriter* writer = _CLNEW IndexWriter(&directory, &a, true);
for (size_t i = 0; values[i]!=NULL; i++) {
Document* doc = _CLNEW Document();
doc->add(*_CLNEW Field(FIELD, values[i], Field::STORE_YES | Field::INDEX_TOKENIZED));
writer->addDocument(doc);
_CLLDELETE(doc);
}
writer->close();
_CLLDELETE(writer);
BooleanQuery* booleanQuery1 = _CLNEW BooleanQuery();
Term *t = _CLNEW Term(FIELD, _T("1"));
booleanQuery1->add(_CLNEW TermQuery(t), true, BooleanClause::SHOULD);
_CLDECDELETE(t);
t = _CLNEW Term(FIELD, _T("2"));
booleanQuery1->add(_CLNEW TermQuery(t), true, BooleanClause::SHOULD);
_CLDECDELETE(t);
BooleanQuery* query = _CLNEW BooleanQuery();
query->add(booleanQuery1, true, BooleanClause::MUST);
t = _CLNEW Term(FIELD, _T("9"));
query->add(_CLNEW TermQuery(t), true, BooleanClause::MUST_NOT);
_CLDECDELETE(t);
IndexSearcher *indexSearcher = _CLNEW IndexSearcher(&directory);
Hits *hits = indexSearcher->search(query);
CLUCENE_ASSERT(2 == hits->length()); // Number of matched documents
_CLLDELETE(hits);
_CLLDELETE(indexSearcher);
_CLLDELETE(query);
}
catch (CLuceneError& e) {
CuFail(tc, e.twhat());
}
}
void testExtractFromTermQuery( CuTest * tc )
{
Directory * pIndex = setUpIndex();
IndexReader * pReader = IndexReader::open( pIndex );
TermSet termSet;
Term * t1 = _CLNEW Term( _T("data"), _T("aaaaa") );
Term * t2 = _CLNEW Term( _T("data"), _T("bbbbb") );
Query * q1 = _CLNEW TermQuery( t1 );
Query * q2 = _CLNEW TermQuery( t2 );
Query * rewrite1 = q1->rewrite( pReader );
Query * rewrite2 = q2->rewrite( pReader );
rewrite1->extractTerms( &termSet );
assertEqualsMsg( _T( "wrong number of terms" ), 1, termSet.size() );
assertEqualsMsg( _T( "wrong term" ), 0, t1->compareTo( *(termSet.begin())) );
clearTermSet( termSet );
rewrite2->extractTerms( &termSet );
assertEqualsMsg( _T( "wrong number of terms" ), 1, termSet.size() );
assertEqualsMsg( _T( "wrong term" ), 0, t2->compareTo( *(termSet.begin())) );
clearTermSet( termSet );
_CLLDECDELETE( t1 );
_CLLDECDELETE( t2 );
if( q1 != rewrite1 )
_CLDELETE( rewrite1 );
_CLDELETE( q1 );
if( q2 != rewrite2 )
_CLDELETE( rewrite2 );
_CLDELETE( q2 );
pReader->close();
_CLDELETE( pReader );
closeIndex( pIndex );
pIndex = NULL;
}
void testExtractFromBooleanQuery( CuTest * tc )
{
Directory * pIndex = setUpIndex();
IndexReader * pReader = IndexReader::open( pIndex );
TermSet termSet;
Term * t1 = _CLNEW Term( _T("data"), _T("aaaab") );
Term * t2 = _CLNEW Term( _T("data"), _T("aaabb") );
Term * t3 = _CLNEW Term( _T("data"), _T("aaabb") );
BooleanQuery * bq = _CLNEW BooleanQuery();
bq->add( _CLNEW TermQuery( t1 ), true, BooleanClause::SHOULD );
bq->add( _CLNEW TermQuery( t2 ), true, BooleanClause::SHOULD );
bq->add( _CLNEW TermQuery( t3 ), true, BooleanClause::SHOULD );
Query * rewrite = bq->rewrite( pReader );
rewrite->extractTerms( &termSet );
assertEqualsMsg( _T( "wrong number of terms" ), 2, termSet.size() );
for( TermSet::iterator itTerms = termSet.begin(); itTerms != termSet.end(); itTerms++ )
{
Term * pTerm = *itTerms;
assertTrueMsg( _T( "wrong term" ), ( 0 == t1->compareTo( pTerm ) || 0 == t2->compareTo( pTerm )));
}
clearTermSet( termSet );
_CLLDECDELETE( t1 );
_CLLDECDELETE( t2 );
_CLLDECDELETE( t3 );
if( rewrite != bq )
_CLDELETE( rewrite );
_CLDELETE( bq );
pReader->close();
_CLDELETE( pReader );
closeIndex( pIndex );
pIndex = NULL;
}
/**
* FIXME: Describe <code>rewrite</code> method here.
*
* @param reader an <code>IndexReader</code> value
* @return a <code>Query</code> value
* @exception IOException if an error occurs
*/
Query* RangeQuery::rewrite(IndexReader* reader){
BooleanQuery* query = _CLNEW BooleanQuery;
TermEnum* enumerator = reader->terms(lowerTerm);
Term* lastTerm = NULL;
try {
bool checkLower = false;
if (!inclusive) // make adjustments to set to exclusive
checkLower = true;
const TCHAR* testField = getField();
do {
lastTerm = enumerator->term();
if (lastTerm != NULL && lastTerm->field() == testField ) {
if (!checkLower || _tcscmp(lastTerm->text(),lowerTerm->text()) > 0) {
checkLower = false;
if (upperTerm != NULL) {
int compare = _tcscmp(upperTerm->text(),lastTerm->text());
/* if beyond the upper term, or is exclusive and
* this is equal to the upper term, break out */
if ((compare < 0) || (!inclusive && compare == 0))
break;
}
TermQuery* tq = _CLNEW TermQuery(lastTerm); // found a match
tq->setBoost(getBoost()); // set the boost
query->add(tq, true, false, false); // add to query
}
}else {
break;
}
_CLDECDELETE(lastTerm);
}
while (enumerator->next());
}catch(...){
_CLDECDELETE(lastTerm); //always need to delete this
_CLDELETE(query); //in case of error, delete the query
enumerator->close();
_CLDELETE(enumerator);
throw; //rethrow
}
_CLDECDELETE(lastTerm); //always need to delete this
enumerator->close();
_CLDELETE(enumerator);
return query;
}
Query* PrefixQuery::rewrite(IndexReader* reader){
BooleanQuery* query = _CLNEW BooleanQuery();
TermEnum* enumerator = reader->terms(prefix);
Term* lastTerm = NULL;
try {
const TCHAR* prefixText = prefix->text();
const TCHAR* prefixField = prefix->field();
const TCHAR* tmp;
size_t i;
int32_t prefixLen = prefix->textLength();
do {
lastTerm = enumerator->term();
if (lastTerm != NULL && lastTerm->field() == prefixField ){
//now see if term->text() starts with prefixText
int32_t termLen = lastTerm->textLength();
if ( prefixLen>termLen )
break; //the prefix is longer than the term, can't be matched
tmp = lastTerm->text();
//check for prefix match in reverse, since most change will be at the end
for ( i=prefixLen-1;i!=-1;--i ){
if ( tmp[i] != prefixText[i] ){
tmp=NULL;//signals inequality
break;
}
}
if ( tmp == NULL )
break;
TermQuery* tq = _CLNEW TermQuery(lastTerm); // found a match
tq->setBoost(getBoost()); // set the boost
query->add(tq,true,false, false); // add to query
} else
break;
_CLDECDELETE(lastTerm);
} while (enumerator->next());
}_CLFINALLY(
enumerator->close();
_CLDELETE(enumerator);
_CLDECDELETE(lastTerm);
);
/// TestBooleanQuery.java, ported 5/9/2009
void testEquality(CuTest *tc) {
BooleanQuery* bq1 = _CLNEW BooleanQuery();
Term* t = _CLNEW Term(_T("field"), _T("value1"));
bq1->add(_CLNEW TermQuery(t), true, BooleanClause::SHOULD);
_CLDECDELETE(t);
t = _CLNEW Term(_T("field"), _T("value2"));
bq1->add(_CLNEW TermQuery(t), true, BooleanClause::SHOULD);
_CLDECDELETE(t);
BooleanQuery* nested1 = _CLNEW BooleanQuery();
t = _CLNEW Term(_T("field"), _T("nestedvalue1"));
nested1->add(_CLNEW TermQuery(t), true, BooleanClause::SHOULD);
_CLDECDELETE(t);
t = _CLNEW Term(_T("field"), _T("nestedvalue2"));
nested1->add(_CLNEW TermQuery(t), true, BooleanClause::SHOULD);
_CLDECDELETE(t);
bq1->add(nested1, true, BooleanClause::SHOULD);
BooleanQuery* bq2 = _CLNEW BooleanQuery();
t = _CLNEW Term(_T("field"), _T("value1"));
bq2->add(_CLNEW TermQuery(t), true, BooleanClause::SHOULD);
_CLDECDELETE(t);
t = _CLNEW Term(_T("field"), _T("value2"));
bq2->add(_CLNEW TermQuery(t), true, BooleanClause::SHOULD);
_CLDECDELETE(t);
BooleanQuery* nested2 = _CLNEW BooleanQuery();
t = _CLNEW Term(_T("field"), _T("nestedvalue1"));
nested2->add(_CLNEW TermQuery(t), true, BooleanClause::SHOULD);
_CLDECDELETE(t);
t = _CLNEW Term(_T("field"), _T("nestedvalue2"));
nested2->add(_CLNEW TermQuery(t), true, BooleanClause::SHOULD);
_CLDECDELETE(t);
bq2->add(nested2, true, BooleanClause::SHOULD);
CLUCENE_ASSERT(bq1->equals(bq2));
_CLLDELETE(bq1);
_CLLDELETE(bq2);
}
void testRangeFilterId() {
IndexReader* reader = IndexReader::open(index);
IndexSearcher* search = new IndexSearcher(reader);
int medId = ((maxId - minId) / 2);
std::tstring minIPstr = pad(minId);
const TCHAR* minIP = minIPstr.c_str();
std::tstring maxIPstr = pad(maxId);
const TCHAR* maxIP = maxIPstr.c_str();
std::tstring medIPstr = pad(medId);
const TCHAR* medIP = medIPstr.c_str();
size_t numDocs = static_cast<size_t>(reader->numDocs());
assertEqualsMsg(_T("num of docs"), numDocs, static_cast<size_t>(1+ maxId - minId));
Hits* result;
Term* term = _CLNEW Term(_T("body"),_T("body"));
Query* q = _CLNEW TermQuery(term);
_CLDECDELETE(term);
// test id, bounded on both ends
Filter* f = _CLNEW RangeFilter(_T("id"),minIP,maxIP,T,T);
result = search->search(q, f);
assertEqualsMsg(_T("find all"), numDocs, result->length());
_CLLDELETE(result);
_CLLDELETE(f);
f=_CLNEW RangeFilter(_T("id"),minIP,maxIP,T,F);
result = search->search(q,f);
assertEqualsMsg(_T("all but last"), numDocs-1, result->length());
_CLLDELETE(result);
_CLLDELETE(f);
f =_CLNEW RangeFilter(_T("id"),minIP,maxIP,F,T);
result = search->search(q,f);
assertEqualsMsg(_T("all but first"), numDocs-1, result->length());
_CLLDELETE(result);
_CLLDELETE(f);
f=_CLNEW RangeFilter(_T("id"),minIP,maxIP,F,F);
result = search->search(q,f);
assertEqualsMsg(_T("all but ends"), numDocs-2, result->length());
_CLLDELETE(result);
_CLLDELETE(f);
f=_CLNEW RangeFilter(_T("id"),medIP,maxIP,T,T);
result = search->search(q,f);
assertEqualsMsg(_T("med and up"), 1+ maxId-medId, result->length());
_CLLDELETE(result);
_CLLDELETE(f);
f=_CLNEW RangeFilter(_T("id"),minIP,medIP,T,T);
result = search->search(q,f);
assertEqualsMsg(_T("up to med"), 1+ medId-minId, result->length());
_CLLDELETE(result);
_CLLDELETE(f);
// unbounded id
f=_CLNEW RangeFilter(_T("id"),minIP,NULL,T,F);
result = search->search(q,f);
assertEqualsMsg(_T("min and up"), numDocs, result->length());
_CLLDELETE(result);
_CLLDELETE(f);
f=_CLNEW RangeFilter(_T("id"),NULL,maxIP,F,T);
result = search->search(q,f);
assertEqualsMsg(_T("max and down"), numDocs, result->length());
_CLLDELETE(result);
_CLLDELETE(f);
f=_CLNEW RangeFilter(_T("id"),minIP,NULL,F,F);
result = search->search(q,f);
assertEqualsMsg(_T("not min, but up"), numDocs-1, result->length());
_CLLDELETE(result);
_CLLDELETE(f);
f=_CLNEW RangeFilter(_T("id"),NULL,maxIP,F,F);
result = search->search(q,f);
assertEqualsMsg(_T("not max, but down"), numDocs-1, result->length());
_CLLDELETE(result);
_CLLDELETE(f);
f=_CLNEW RangeFilter(_T("id"),medIP,maxIP,T,F);
result = search->search(q,f);
assertEqualsMsg(_T("med and up, not max"), maxId-medId, result->length());
_CLLDELETE(result);
_CLLDELETE(f);
f=_CLNEW RangeFilter(_T("id"),minIP,medIP,F,T);
result = search->search(q,f);
assertEqualsMsg(_T("not min, up to med"), medId-minId, result->length());
_CLLDELETE(result);
_CLLDELETE(f);
// very small sets
f=_CLNEW RangeFilter(_T("id"),minIP,minIP,F,F);
result = search->search(q,f);
assertEqualsMsg(_T("min,min,F,F"), 0, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("id"),medIP,medIP,F,F);
result = search->search(q,f);
assertEqualsMsg(_T("med,med,F,F"), 0, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("id"),maxIP,maxIP,F,F);
result = search->search(q,f);
assertEqualsMsg(_T("max,max,F,F"), 0, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("id"),minIP,minIP,T,T);
result = search->search(q,f);
assertEqualsMsg(_T("min,min,T,T"), 1, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("id"),NULL,minIP,F,T);
result = search->search(q,f);
assertEqualsMsg(_T("nul,min,F,T"), 1, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("id"),maxIP,maxIP,T,T);
result = search->search(q,f);
assertEqualsMsg(_T("max,max,T,T"), 1, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("id"),maxIP,NULL,T,F);
result = search->search(q,f);
assertEqualsMsg(_T("max,nul,T,T"), 1, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("id"),medIP,medIP,T,T);
result = search->search(q,f);
assertEqualsMsg(_T("med,med,T,T"), 1, result->length());
_CLLDELETE(result); _CLLDELETE(f);
search->close();
_CLLDELETE(search);
reader->close();
_CLLDELETE(reader);
_CLLDELETE(q);
}
void testRangeFilterRand()
{
IndexReader* reader = IndexReader::open(index);
IndexSearcher* search = _CLNEW IndexSearcher(reader);
std::tstring minRPstr = pad(minR);
const TCHAR* minRP = minRPstr.c_str();
std::tstring maxRPstr = pad(maxR);
const TCHAR* maxRP = maxRPstr.c_str();
size_t numDocs = static_cast<size_t>(reader->numDocs());
assertEqualsMsg(_T("num of docs"), numDocs, 1+ maxId - minId);
Hits* result;
Term* term = _CLNEW Term(_T("body"),_T("body"));
Query* q = _CLNEW TermQuery(term);
_CLDECDELETE(term);
// test extremes, bounded on both ends
Filter* f = _CLNEW RangeFilter(_T("rand"),minRP,maxRP,T,T);
result = search->search(q,f);
assertEqualsMsg(_T("find all"), numDocs, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("rand"),minRP,maxRP,T,F);
result = search->search(q,f);
assertEqualsMsg(_T("all but biggest"), numDocs-1, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("rand"),minRP,maxRP,F,T);
result = search->search(q,f);
assertEqualsMsg(_T("all but smallest"), numDocs-1, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("rand"),minRP,maxRP,F,F);
result = search->search(q,f);
assertEqualsMsg(_T("all but extremes"), numDocs-2, result->length());
_CLLDELETE(result); _CLLDELETE(f);
// unbounded
f=_CLNEW RangeFilter(_T("rand"),minRP,NULL,T,F);
result = search->search(q,f);
assertEqualsMsg(_T("smallest and up"), numDocs, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("rand"),NULL,maxRP,F,T);
result = search->search(q,f);
assertEqualsMsg(_T("biggest and down"), numDocs, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("rand"),minRP,NULL,F,F);
result = search->search(q,f);
assertEqualsMsg(_T("not smallest, but up"), numDocs-1, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("rand"),NULL,maxRP,F,F);
result = search->search(q,f);
assertEqualsMsg(_T("not biggest, but down"), numDocs-1, result->length());
_CLLDELETE(result); _CLLDELETE(f);
// very small sets
f=_CLNEW RangeFilter(_T("rand"),minRP,minRP,F,F);
result = search->search(q,f);
assertEqualsMsg(_T("min,min,F,F"), 0, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("rand"),maxRP,maxRP,F,F);
result = search->search(q,f);
assertEqualsMsg(_T("max,max,F,F"), 0, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("rand"),minRP,minRP,T,T);
result = search->search(q,f);
assertEqualsMsg(_T("min,min,T,T"), 1, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("rand"),NULL,minRP,F,T);
result = search->search(q,f);
assertEqualsMsg(_T("nul,min,F,T"), 1, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("rand"),maxRP,maxRP,T,T);
result = search->search(q,f);
assertEqualsMsg(_T("max,max,T,T"), 1, result->length());
_CLLDELETE(result); _CLLDELETE(f);
f=_CLNEW RangeFilter(_T("rand"),maxRP,NULL,T,F);
result = search->search(q,f);
assertEqualsMsg(_T("max,nul,T,T"), 1, result->length());
_CLLDELETE(result); _CLLDELETE(f);
search->close();
_CLLDELETE(search);
reader->close();
_CLLDELETE(reader);
_CLLDELETE(q);
}
