1 package org.apache.lucene.queryParser.standard.processors;
4 * Licensed to the Apache Software Foundation (ASF) under one or more
5 * contributor license agreements. See the NOTICE file distributed with
6 * this work for additional information regarding copyright ownership.
7 * The ASF licenses this file to You under the Apache License, Version 2.0
8 * (the "License"); you may not use this file except in compliance with
9 * the License. You may obtain a copy of the License at
11 * http://www.apache.org/licenses/LICENSE-2.0
13 * Unless required by applicable law or agreed to in writing, software
14 * distributed under the License is distributed on an "AS IS" BASIS,
15 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16 * See the License for the specific language governing permissions and
17 * limitations under the License.
20 import java.util.List;
22 import org.apache.lucene.queryParser.core.QueryNodeException;
23 import org.apache.lucene.queryParser.core.nodes.FieldQueryNode;
24 import org.apache.lucene.queryParser.core.nodes.FuzzyQueryNode;
25 import org.apache.lucene.queryParser.core.nodes.ParametricQueryNode;
26 import org.apache.lucene.queryParser.core.nodes.QueryNode;
27 import org.apache.lucene.queryParser.core.nodes.QuotedFieldQueryNode;
28 import org.apache.lucene.queryParser.core.processors.QueryNodeProcessorImpl;
29 import org.apache.lucene.queryParser.core.util.UnescapedCharSequence;
30 import org.apache.lucene.queryParser.standard.nodes.PrefixWildcardQueryNode;
31 import org.apache.lucene.queryParser.standard.nodes.WildcardQueryNode;
32 import org.apache.lucene.queryParser.standard.parser.StandardSyntaxParser;
33 import org.apache.lucene.search.PrefixQuery;
36 * The {@link StandardSyntaxParser} creates {@link PrefixWildcardQueryNode} nodes which
37 * have values containing the prefixed wildcard. However, Lucene
38 * {@link PrefixQuery} cannot contain the prefixed wildcard. So, this processor
39 * basically removed the prefixed wildcard from the
40 * {@link PrefixWildcardQueryNode} value. <br/>
43 * @see PrefixWildcardQueryNode
45 public class WildcardQueryNodeProcessor extends QueryNodeProcessorImpl {
47 public WildcardQueryNodeProcessor() {
52 protected QueryNode postProcessNode(QueryNode node) throws QueryNodeException {
54 // the old Lucene Parser ignores FuzzyQueryNode that are also PrefixWildcardQueryNode or WildcardQueryNode
55 // we do the same here, also ignore empty terms
56 if (node instanceof FieldQueryNode || node instanceof FuzzyQueryNode) {
57 FieldQueryNode fqn = (FieldQueryNode) node;
58 CharSequence text = fqn.getText();
60 // do not process wildcards for ParametricQueryNode and
61 // QuotedFieldQueryNode to reproduce the old parser behavior
62 if (fqn instanceof ParametricQueryNode
63 || fqn instanceof QuotedFieldQueryNode
64 || text.length() <= 0){
69 // Code below simulates the old lucene parser behavior for wildcards
71 if (isPrefixWildcard(text)) {
72 PrefixWildcardQueryNode prefixWildcardQN = new PrefixWildcardQueryNode(fqn);
73 return prefixWildcardQN;
75 } else if (isWildcard(text)){
76 WildcardQueryNode wildcardQN = new WildcardQueryNode(fqn);
86 private boolean isWildcard(CharSequence text) {
87 if (text ==null || text.length() <= 0) return false;
89 // If a un-escaped '*' or '?' if found return true
90 // start at the end since it's more common to put wildcards at the end
91 for(int i=text.length()-1; i>=0; i--){
92 if ((text.charAt(i) == '*' || text.charAt(i) == '?') && !UnescapedCharSequence.wasEscaped(text, i)){
100 private boolean isPrefixWildcard(CharSequence text) {
101 if (text == null || text.length() <= 0 || !isWildcard(text)) return false;
103 // Validate last character is a '*' and was not escaped
104 // If single '*' is is a wildcard not prefix to simulate old queryparser
105 if (text.charAt(text.length()-1) != '*') return false;
106 if (UnescapedCharSequence.wasEscaped(text, text.length()-1)) return false;
107 if (text.length() == 1) return false;
109 // Only make a prefix if there is only one single star at the end and no '?' or '*' characters
110 // If single wildcard return false to mimic old queryparser
111 for(int i=0; i<text.length(); i++){
112 if (text.charAt(i) == '?') return false;
113 if (text.charAt(i) == '*' && !UnescapedCharSequence.wasEscaped(text, i)){
114 if (i == text.length()-1)
125 protected QueryNode preProcessNode(QueryNode node) throws QueryNodeException {
132 protected List<QueryNode> setChildrenOrder(List<QueryNode> children)
133 throws QueryNodeException {