use of org.apache.lucene.queryparser.flexible.core.nodes.QuotedFieldQueryNode in project lucene-solr by apache.
the class StandardSyntaxParser method Term.
public final QueryNode Term(CharSequence field) throws ParseException {
Token term, boost = null, fuzzySlop = null, goop1, goop2;
boolean fuzzy = false;
boolean regexp = false;
boolean startInc = false;
boolean endInc = false;
QueryNode q = null;
FieldQueryNode qLower, qUpper;
float defaultMinSimilarity = org.apache.lucene.search.FuzzyQuery.defaultMinSimilarity;
switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
case TERM:
case REGEXPTERM:
case NUMBER:
switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
case TERM:
term = jj_consume_token(TERM);
q = new FieldQueryNode(field, EscapeQuerySyntaxImpl.discardEscapeChar(term.image), term.beginColumn, term.endColumn);
break;
case REGEXPTERM:
term = jj_consume_token(REGEXPTERM);
regexp = true;
break;
case NUMBER:
term = jj_consume_token(NUMBER);
break;
default:
jj_la1[13] = jj_gen;
jj_consume_token(-1);
throw new ParseException();
}
switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
case FUZZY_SLOP:
fuzzySlop = jj_consume_token(FUZZY_SLOP);
fuzzy = true;
break;
default:
jj_la1[14] = jj_gen;
;
}
switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
case CARAT:
jj_consume_token(CARAT);
boost = jj_consume_token(NUMBER);
switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
case FUZZY_SLOP:
fuzzySlop = jj_consume_token(FUZZY_SLOP);
fuzzy = true;
break;
default:
jj_la1[15] = jj_gen;
;
}
break;
default:
jj_la1[16] = jj_gen;
;
}
if (fuzzy) {
float fms = defaultMinSimilarity;
try {
fms = Float.parseFloat(fuzzySlop.image.substring(1));
} catch (Exception ignored) {
}
if (fms < 0.0f) {
{
if (true)
throw new ParseException(new MessageImpl(QueryParserMessages.INVALID_SYNTAX_FUZZY_LIMITS));
}
} else if (fms >= 1.0f && fms != (int) fms) {
{
if (true)
throw new ParseException(new MessageImpl(QueryParserMessages.INVALID_SYNTAX_FUZZY_EDITS));
}
}
q = new FuzzyQueryNode(field, EscapeQuerySyntaxImpl.discardEscapeChar(term.image), fms, term.beginColumn, term.endColumn);
} else if (regexp) {
String re = term.image.substring(1, term.image.length() - 1);
q = new RegexpQueryNode(field, re, 0, re.length());
}
break;
case RANGEIN_START:
case RANGEEX_START:
switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
case RANGEIN_START:
jj_consume_token(RANGEIN_START);
startInc = true;
break;
case RANGEEX_START:
jj_consume_token(RANGEEX_START);
break;
default:
jj_la1[17] = jj_gen;
jj_consume_token(-1);
throw new ParseException();
}
switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
case RANGE_GOOP:
goop1 = jj_consume_token(RANGE_GOOP);
break;
case RANGE_QUOTED:
goop1 = jj_consume_token(RANGE_QUOTED);
break;
case RANGE_TO:
goop1 = jj_consume_token(RANGE_TO);
break;
default:
jj_la1[18] = jj_gen;
jj_consume_token(-1);
throw new ParseException();
}
jj_consume_token(RANGE_TO);
switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
case RANGE_GOOP:
goop2 = jj_consume_token(RANGE_GOOP);
break;
case RANGE_QUOTED:
goop2 = jj_consume_token(RANGE_QUOTED);
break;
case RANGE_TO:
goop2 = jj_consume_token(RANGE_TO);
break;
default:
jj_la1[19] = jj_gen;
jj_consume_token(-1);
throw new ParseException();
}
switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
case RANGEIN_END:
jj_consume_token(RANGEIN_END);
endInc = true;
break;
case RANGEEX_END:
jj_consume_token(RANGEEX_END);
break;
default:
jj_la1[20] = jj_gen;
jj_consume_token(-1);
throw new ParseException();
}
switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
case CARAT:
jj_consume_token(CARAT);
boost = jj_consume_token(NUMBER);
break;
default:
jj_la1[21] = jj_gen;
;
}
if (goop1.kind == RANGE_QUOTED) {
goop1.image = goop1.image.substring(1, goop1.image.length() - 1);
}
if (goop2.kind == RANGE_QUOTED) {
goop2.image = goop2.image.substring(1, goop2.image.length() - 1);
}
qLower = new FieldQueryNode(field, EscapeQuerySyntaxImpl.discardEscapeChar(goop1.image), goop1.beginColumn, goop1.endColumn);
qUpper = new FieldQueryNode(field, EscapeQuerySyntaxImpl.discardEscapeChar(goop2.image), goop2.beginColumn, goop2.endColumn);
q = new TermRangeQueryNode(qLower, qUpper, startInc ? true : false, endInc ? true : false);
break;
case QUOTED:
term = jj_consume_token(QUOTED);
q = new QuotedFieldQueryNode(field, EscapeQuerySyntaxImpl.discardEscapeChar(term.image.substring(1, term.image.length() - 1)), term.beginColumn + 1, term.endColumn - 1);
switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
case FUZZY_SLOP:
fuzzySlop = jj_consume_token(FUZZY_SLOP);
break;
default:
jj_la1[22] = jj_gen;
;
}
switch((jj_ntk == -1) ? jj_ntk() : jj_ntk) {
case CARAT:
jj_consume_token(CARAT);
boost = jj_consume_token(NUMBER);
break;
default:
jj_la1[23] = jj_gen;
;
}
int phraseSlop = 0;
if (fuzzySlop != null) {
try {
phraseSlop = (int) Float.parseFloat(fuzzySlop.image.substring(1));
q = new SlopQueryNode(q, phraseSlop);
} catch (Exception ignored) {
/* Should this be handled somehow? (defaults to "no PhraseSlop", if
* slop number is invalid)
*/
}
}
break;
default:
jj_la1[24] = jj_gen;
jj_consume_token(-1);
throw new ParseException();
}
if (boost != null) {
float f = (float) 1.0;
try {
f = Float.parseFloat(boost.image);
// avoid boosting null queries, such as those caused by stop words
if (q != null) {
q = new BoostQueryNode(q, f);
}
} catch (Exception ignored) {
/* Should this be handled somehow? (defaults to "no boost", if
* boost number is invalid)
*/
}
}
{
if (true)
return q;
}
throw new Error("Missing return statement in function");
}
use of org.apache.lucene.queryparser.flexible.core.nodes.QuotedFieldQueryNode in project lucene-solr by apache.
the class WildcardQueryNodeProcessor method postProcessNode.
@Override
protected QueryNode postProcessNode(QueryNode node) throws QueryNodeException {
// we do the same here, also ignore empty terms
if (node instanceof FieldQueryNode || node instanceof FuzzyQueryNode) {
FieldQueryNode fqn = (FieldQueryNode) node;
CharSequence text = fqn.getText();
// QuotedFieldQueryNode to reproduce the old parser behavior
if (fqn.getParent() instanceof TermRangeQueryNode || fqn instanceof QuotedFieldQueryNode || text.length() <= 0) {
// Ignore empty terms
return node;
}
if (isWildcard(text)) {
Analyzer analyzer = getQueryConfigHandler().get(ConfigurationKeys.ANALYZER);
if (analyzer != null) {
text = analyzeWildcard(analyzer, fqn.getFieldAsString(), text.toString());
}
if (isPrefixWildcard(text)) {
return new PrefixWildcardQueryNode(fqn.getField(), text, fqn.getBegin(), fqn.getEnd());
} else {
return new WildcardQueryNode(fqn.getField(), text, fqn.getBegin(), fqn.getEnd());
}
}
}
return node;
}
use of org.apache.lucene.queryparser.flexible.core.nodes.QuotedFieldQueryNode in project lucene-solr by apache.
the class AnalyzerQueryNodeProcessor method postProcessNode.
@Override
protected QueryNode postProcessNode(QueryNode node) throws QueryNodeException {
if (node instanceof TextableQueryNode && !(node instanceof WildcardQueryNode) && !(node instanceof FuzzyQueryNode) && !(node instanceof RegexpQueryNode) && !(node.getParent() instanceof RangeQueryNode)) {
FieldQueryNode fieldNode = ((FieldQueryNode) node);
String text = fieldNode.getTextAsString();
String field = fieldNode.getFieldAsString();
CachingTokenFilter buffer = null;
PositionIncrementAttribute posIncrAtt = null;
int numTokens = 0;
int positionCount = 0;
boolean severalTokensAtSamePosition = false;
try {
try (TokenStream source = this.analyzer.tokenStream(field, text)) {
buffer = new CachingTokenFilter(source);
buffer.reset();
if (buffer.hasAttribute(PositionIncrementAttribute.class)) {
posIncrAtt = buffer.getAttribute(PositionIncrementAttribute.class);
}
try {
while (buffer.incrementToken()) {
numTokens++;
int positionIncrement = (posIncrAtt != null) ? posIncrAtt.getPositionIncrement() : 1;
if (positionIncrement != 0) {
positionCount += positionIncrement;
} else {
severalTokensAtSamePosition = true;
}
}
} catch (IOException e) {
// ignore
}
// rewind the buffer stream
//will never through on subsequent reset calls
buffer.reset();
} catch (IOException e) {
throw new RuntimeException(e);
}
if (!buffer.hasAttribute(CharTermAttribute.class)) {
return new NoTokenFoundQueryNode();
}
CharTermAttribute termAtt = buffer.getAttribute(CharTermAttribute.class);
if (numTokens == 0) {
return new NoTokenFoundQueryNode();
} else if (numTokens == 1) {
String term = null;
try {
boolean hasNext;
hasNext = buffer.incrementToken();
assert hasNext == true;
term = termAtt.toString();
} catch (IOException e) {
// safe to ignore, because we know the number of tokens
}
fieldNode.setText(term);
return fieldNode;
} else if (severalTokensAtSamePosition || !(node instanceof QuotedFieldQueryNode)) {
if (positionCount == 1 || !(node instanceof QuotedFieldQueryNode)) {
if (positionCount == 1) {
// simple case: only one position, with synonyms
LinkedList<QueryNode> children = new LinkedList<>();
for (int i = 0; i < numTokens; i++) {
String term = null;
try {
boolean hasNext = buffer.incrementToken();
assert hasNext == true;
term = termAtt.toString();
} catch (IOException e) {
// safe to ignore, because we know the number of tokens
}
children.add(new FieldQueryNode(field, term, -1, -1));
}
return new GroupQueryNode(new SynonymQueryNode(children));
} else {
// multiple positions
QueryNode q = new BooleanQueryNode(Collections.<QueryNode>emptyList());
QueryNode currentQuery = null;
for (int i = 0; i < numTokens; i++) {
String term = null;
try {
boolean hasNext = buffer.incrementToken();
assert hasNext == true;
term = termAtt.toString();
} catch (IOException e) {
// safe to ignore, because we know the number of tokens
}
if (posIncrAtt != null && posIncrAtt.getPositionIncrement() == 0) {
if (!(currentQuery instanceof BooleanQueryNode)) {
QueryNode t = currentQuery;
currentQuery = new SynonymQueryNode(Collections.<QueryNode>emptyList());
((BooleanQueryNode) currentQuery).add(t);
}
((BooleanQueryNode) currentQuery).add(new FieldQueryNode(field, term, -1, -1));
} else {
if (currentQuery != null) {
if (this.defaultOperator == Operator.OR) {
q.add(currentQuery);
} else {
q.add(new ModifierQueryNode(currentQuery, Modifier.MOD_REQ));
}
}
currentQuery = new FieldQueryNode(field, term, -1, -1);
}
}
if (this.defaultOperator == Operator.OR) {
q.add(currentQuery);
} else {
q.add(new ModifierQueryNode(currentQuery, Modifier.MOD_REQ));
}
if (q instanceof BooleanQueryNode) {
q = new GroupQueryNode(q);
}
return q;
}
} else {
// phrase query:
MultiPhraseQueryNode mpq = new MultiPhraseQueryNode();
List<FieldQueryNode> multiTerms = new ArrayList<>();
int position = -1;
int i = 0;
int termGroupCount = 0;
for (; i < numTokens; i++) {
String term = null;
int positionIncrement = 1;
try {
boolean hasNext = buffer.incrementToken();
assert hasNext == true;
term = termAtt.toString();
if (posIncrAtt != null) {
positionIncrement = posIncrAtt.getPositionIncrement();
}
} catch (IOException e) {
// safe to ignore, because we know the number of tokens
}
if (positionIncrement > 0 && multiTerms.size() > 0) {
for (FieldQueryNode termNode : multiTerms) {
if (this.positionIncrementsEnabled) {
termNode.setPositionIncrement(position);
} else {
termNode.setPositionIncrement(termGroupCount);
}
mpq.add(termNode);
}
// Only increment once for each "group" of
// terms that were in the same position:
termGroupCount++;
multiTerms.clear();
}
position += positionIncrement;
multiTerms.add(new FieldQueryNode(field, term, -1, -1));
}
for (FieldQueryNode termNode : multiTerms) {
if (this.positionIncrementsEnabled) {
termNode.setPositionIncrement(position);
} else {
termNode.setPositionIncrement(termGroupCount);
}
mpq.add(termNode);
}
return mpq;
}
} else {
TokenizedPhraseQueryNode pq = new TokenizedPhraseQueryNode();
int position = -1;
for (int i = 0; i < numTokens; i++) {
String term = null;
int positionIncrement = 1;
try {
boolean hasNext = buffer.incrementToken();
assert hasNext == true;
term = termAtt.toString();
if (posIncrAtt != null) {
positionIncrement = posIncrAtt.getPositionIncrement();
}
} catch (IOException e) {
// safe to ignore, because we know the number of tokens
}
FieldQueryNode newFieldNode = new FieldQueryNode(field, term, -1, -1);
if (this.positionIncrementsEnabled) {
position += positionIncrement;
newFieldNode.setPositionIncrement(position);
} else {
newFieldNode.setPositionIncrement(i);
}
pq.add(newFieldNode);
}
return pq;
}
} finally {
if (buffer != null) {
try {
buffer.close();
} catch (IOException e) {
// safe to ignore
}
}
}
}
return node;
}
use of org.apache.lucene.queryparser.flexible.core.nodes.QuotedFieldQueryNode in project lucene-skos by behas.
the class SKOSQueryNodeProcessor method postProcessNode.
@Override
protected QueryNode postProcessNode(QueryNode node) throws QueryNodeException {
if (node instanceof TextableQueryNode && !(node instanceof WildcardQueryNode) && !(node instanceof FuzzyQueryNode) && !(node instanceof RegexpQueryNode) && !(node.getParent() instanceof RangeQueryNode)) {
FieldQueryNode fieldNode = ((FieldQueryNode) node);
String text = fieldNode.getTextAsString();
String field = fieldNode.getFieldAsString();
CachingTokenFilter buffer = null;
PositionIncrementAttribute posIncrAtt = null;
int numTokens = 0;
int positionCount = 0;
boolean severalTokensAtSamePosition = false;
try {
try (TokenStream source = this.analyzer.tokenStream(field, text)) {
buffer = new CachingTokenFilter(source);
buffer.reset();
if (buffer.hasAttribute(PositionIncrementAttribute.class)) {
posIncrAtt = buffer.getAttribute(PositionIncrementAttribute.class);
}
try {
while (buffer.incrementToken()) {
numTokens++;
int positionIncrement = (posIncrAtt != null) ? posIncrAtt.getPositionIncrement() : 1;
if (positionIncrement != 0) {
positionCount += positionIncrement;
} else {
severalTokensAtSamePosition = true;
}
}
} catch (IOException e) {
// ignore
}
// rewind the buffer stream
// will never through on subsequent reset calls
buffer.reset();
} catch (IOException e) {
throw new RuntimeException(e);
}
if (!buffer.hasAttribute(CharTermAttribute.class)) {
return new NoTokenFoundQueryNode();
}
CharTermAttribute termAtt = buffer.getAttribute(CharTermAttribute.class);
if (numTokens == 0) {
return new NoTokenFoundQueryNode();
} else if (numTokens == 1) {
String term = null;
try {
boolean hasNext;
hasNext = buffer.incrementToken();
assert hasNext == true;
term = termAtt.toString();
} catch (IOException e) {
// safe to ignore, because we know the number of tokens
}
fieldNode.setText(term);
return fieldNode;
} else if (severalTokensAtSamePosition || !(node instanceof QuotedFieldQueryNode)) {
if (positionCount == 1 || !(node instanceof QuotedFieldQueryNode)) {
if (positionCount == 1) {
// simple case: only one position, with synonyms
LinkedList<QueryNode> children = new LinkedList<>();
for (int i = 0; i < numTokens; i++) {
String term = null;
try {
boolean hasNext = buffer.incrementToken();
assert hasNext == true;
term = termAtt.toString();
} catch (IOException e) {
// safe to ignore, because we know the number of tokens
}
if (buffer.hasAttribute(SKOSTypeAttribute.class) && boosts != null) {
SKOSTypeAttribute skosAttr = buffer.getAttribute(SKOSTypeAttribute.class);
children.add(new BoostQueryNode(new FieldQueryNode(field, term, -1, -1), getBoost(skosAttr.getSkosType())));
} else {
children.add(new FieldQueryNode(field, term, -1, -1));
}
}
return new GroupQueryNode(new StandardBooleanQueryNode(children, positionCount == 1));
} else {
// multiple positions
QueryNode q = new StandardBooleanQueryNode(Collections.<QueryNode>emptyList(), false);
QueryNode currentQuery = null;
for (int i = 0; i < numTokens; i++) {
String term = null;
try {
boolean hasNext = buffer.incrementToken();
assert hasNext == true;
term = termAtt.toString();
} catch (IOException e) {
// safe to ignore, because we know the number of tokens
}
if (posIncrAtt != null && posIncrAtt.getPositionIncrement() == 0) {
if (!(currentQuery instanceof BooleanQueryNode)) {
QueryNode t = currentQuery;
currentQuery = new StandardBooleanQueryNode(Collections.<QueryNode>emptyList(), true);
((BooleanQueryNode) currentQuery).add(t);
}
((BooleanQueryNode) currentQuery).add(new FieldQueryNode(field, term, -1, -1));
} else {
if (currentQuery != null) {
if (this.defaultOperator == Operator.OR) {
q.add(currentQuery);
} else {
q.add(new ModifierQueryNode(currentQuery, Modifier.MOD_REQ));
}
}
currentQuery = new FieldQueryNode(field, term, -1, -1);
}
}
if (this.defaultOperator == Operator.OR) {
q.add(currentQuery);
} else {
q.add(new ModifierQueryNode(currentQuery, Modifier.MOD_REQ));
}
if (q instanceof BooleanQueryNode) {
q = new GroupQueryNode(q);
}
return q;
}
} else {
// phrase query:
MultiPhraseQueryNode mpq = new MultiPhraseQueryNode();
List<FieldQueryNode> multiTerms = new ArrayList<>();
int position = -1;
int i = 0;
int termGroupCount = 0;
for (; i < numTokens; i++) {
String term = null;
int positionIncrement = 1;
try {
boolean hasNext = buffer.incrementToken();
assert hasNext == true;
term = termAtt.toString();
if (posIncrAtt != null) {
positionIncrement = posIncrAtt.getPositionIncrement();
}
} catch (IOException e) {
// safe to ignore, because we know the number of tokens
}
if (positionIncrement > 0 && multiTerms.size() > 0) {
for (FieldQueryNode termNode : multiTerms) {
if (this.positionIncrementsEnabled) {
termNode.setPositionIncrement(position);
} else {
termNode.setPositionIncrement(termGroupCount);
}
mpq.add(termNode);
}
// Only increment once for each "group" of
// terms that were in the same position:
termGroupCount++;
multiTerms.clear();
}
position += positionIncrement;
multiTerms.add(new FieldQueryNode(field, term, -1, -1));
}
for (FieldQueryNode termNode : multiTerms) {
if (this.positionIncrementsEnabled) {
termNode.setPositionIncrement(position);
} else {
termNode.setPositionIncrement(termGroupCount);
}
mpq.add(termNode);
}
return mpq;
}
} else {
TokenizedPhraseQueryNode pq = new TokenizedPhraseQueryNode();
int position = -1;
for (int i = 0; i < numTokens; i++) {
String term = null;
int positionIncrement = 1;
try {
boolean hasNext = buffer.incrementToken();
assert hasNext == true;
term = termAtt.toString();
if (posIncrAtt != null) {
positionIncrement = posIncrAtt.getPositionIncrement();
}
} catch (IOException e) {
// safe to ignore, because we know the number of tokens
}
FieldQueryNode newFieldNode = new FieldQueryNode(field, term, -1, -1);
if (this.positionIncrementsEnabled) {
position += positionIncrement;
newFieldNode.setPositionIncrement(position);
} else {
newFieldNode.setPositionIncrement(i);
}
pq.add(newFieldNode);
}
return pq;
}
} finally {
if (buffer != null) {
try {
buffer.close();
} catch (IOException e) {
// safe to ignore
}
}
}
}
return node;
}
Aggregations