blob: 936b12f2f741796f3ea1f54f3d55f20ae95f8c63 [file] [log] [blame]
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package org.apache.lucene.search;
import org.apache.lucene.index.Term;
import org.apache.lucene.util.automaton.AutomatonProvider;
import org.apache.lucene.util.automaton.Operations;
import org.apache.lucene.util.automaton.RegExp;
/**
* A fast regular expression query based on the {@link org.apache.lucene.util.automaton} package.
*
* <ul>
* <li>Comparisons are <a href="http://tusker.org/regex/regex_benchmark.html">fast</a>
* <li>The term dictionary is enumerated in an intelligent way, to avoid comparisons. See {@link
* AutomatonQuery} for more details.
* </ul>
*
* <p>The supported syntax is documented in the {@link RegExp} class. Note this might be different
* than other regular expression implementations. For some alternatives with different syntax, look
* under the sandbox.
*
* <p>Note this query can be slow, as it needs to iterate over many terms. In order to prevent
* extremely slow RegexpQueries, a Regexp term should not start with the expression <code>.*</code>
*
* @see RegExp
* @lucene.experimental
*/
public class RegexpQuery extends AutomatonQuery {
/** A provider that provides no named automata */
public static final AutomatonProvider DEFAULT_PROVIDER = name -> null;
/**
* Constructs a query for terms matching <code>term</code>.
*
* <p>By default, all regular expression features are enabled.
*
* @param term regular expression.
*/
public RegexpQuery(Term term) {
this(term, RegExp.ALL);
}
/**
* Constructs a query for terms matching <code>term</code>.
*
* @param term regular expression.
* @param flags optional RegExp features from {@link RegExp}
*/
public RegexpQuery(Term term, int flags) {
this(term, flags, DEFAULT_PROVIDER, Operations.DEFAULT_DETERMINIZE_WORK_LIMIT);
}
/**
* Constructs a query for terms matching <code>term</code>.
*
* @param term regular expression.
* @param flags optional RegExp syntax features from {@link RegExp}
* @param determinizeWorkLimit maximum effort to spend while compiling the automaton from this
* regexp. Set higher to allow more complex queries and lower to prevent memory exhaustion.
* Use {@link Operations#DEFAULT_DETERMINIZE_WORK_LIMIT} as a decent default if you don't
* otherwise know what to specify.
*/
public RegexpQuery(Term term, int flags, int determinizeWorkLimit) {
this(term, flags, DEFAULT_PROVIDER, determinizeWorkLimit);
}
/**
* Constructs a query for terms matching <code>term</code>.
*
* @param term regular expression.
* @param syntax_flags optional RegExp syntax features from {@link RegExp} automaton for the
* regexp can result in. Set higher to allow more complex queries and lower to prevent memory
* exhaustion.
* @param match_flags boolean 'or' of match behavior options such as case insensitivity
* @param determinizeWorkLimit maximum effort to spend while compiling the automaton from this
* regexp. Set higher to allow more complex queries and lower to prevent memory exhaustion.
* Use {@link Operations#DEFAULT_DETERMINIZE_WORK_LIMIT} as a decent default if you don't
* otherwise know what to specify.
*/
public RegexpQuery(Term term, int syntax_flags, int match_flags, int determinizeWorkLimit) {
this(
term,
syntax_flags,
match_flags,
DEFAULT_PROVIDER,
determinizeWorkLimit,
CONSTANT_SCORE_REWRITE);
}
/**
* Constructs a query for terms matching <code>term</code>.
*
* @param term regular expression.
* @param syntax_flags optional RegExp features from {@link RegExp}
* @param provider custom AutomatonProvider for named automata
* @param determinizeWorkLimit maximum effort to spend while compiling the automaton from this
* regexp. Set higher to allow more complex queries and lower to prevent memory exhaustion.
* Use {@link Operations#DEFAULT_DETERMINIZE_WORK_LIMIT} as a decent default if you don't
* otherwise know what to specify.
*/
public RegexpQuery(
Term term, int syntax_flags, AutomatonProvider provider, int determinizeWorkLimit) {
this(term, syntax_flags, 0, provider, determinizeWorkLimit, CONSTANT_SCORE_REWRITE);
}
/**
* Constructs a query for terms matching <code>term</code>.
*
* @param term regular expression.
* @param syntax_flags optional RegExp features from {@link RegExp}
* @param match_flags boolean 'or' of match behavior options such as case insensitivity
* @param provider custom AutomatonProvider for named automata
* @param determinizeWorkLimit maximum effort to spend while compiling the automaton from this
* regexp. Set higher to allow more complex queries and lower to prevent memory exhaustion.
* Use {@link Operations#DEFAULT_DETERMINIZE_WORK_LIMIT} as a decent default if you don't
* otherwise know what to specify.
* @param rewriteMethod the rewrite method to use to build the final query
*/
public RegexpQuery(
Term term,
int syntax_flags,
int match_flags,
AutomatonProvider provider,
int determinizeWorkLimit,
RewriteMethod rewriteMethod) {
super(
term,
new RegExp(term.text(), syntax_flags, match_flags)
.toAutomaton(provider, determinizeWorkLimit),
determinizeWorkLimit,
false,
rewriteMethod);
}
/** Returns the regexp of this query wrapped in a Term. */
public Term getRegexp() {
return term;
}
/** Prints a user-readable version of this query. */
@Override
public String toString(String field) {
StringBuilder buffer = new StringBuilder();
if (!term.field().equals(field)) {
buffer.append(term.field());
buffer.append(":");
}
buffer.append('/');
buffer.append(term.text());
buffer.append('/');
return buffer.toString();
}
}