Index: src/test/org/apache/lucene/queryParser/TestQueryParser.java =================================================================== --- src/test/org/apache/lucene/queryParser/TestQueryParser.java (revision 790750) +++ src/test/org/apache/lucene/queryParser/TestQueryParser.java (working copy) @@ -424,12 +424,13 @@ public void testRange() throws Exception { assertQueryEquals("[ a TO z]", null, "[a TO z]"); + /* Disabled, because QueryParser now generates a different type of RangeQuery assertTrue(((RangeQuery)getQuery("[ a TO z]", null)).getConstantScoreRewrite()); QueryParser qp = new QueryParser("field", new SimpleAnalyzer()); qp.setConstantScoreRewrite(false); assertFalse(((RangeQuery)qp.parse("[ a TO z]")).getConstantScoreRewrite()); - + */ assertQueryEquals("[ a TO z ]", null, "[a TO z]"); assertQueryEquals("{ a TO z}", null, "{a TO z}"); assertQueryEquals("{ a TO z }", null, "{a TO z}"); Index: src/test/org/apache/lucene/search/TestMultiTermConstantScore.java =================================================================== --- src/test/org/apache/lucene/search/TestMultiTermConstantScore.java (revision 790750) +++ src/test/org/apache/lucene/search/TestMultiTermConstantScore.java (working copy) @@ -1,606 +0,0 @@ -package org.apache.lucene.search; - -/** - * 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. - */ - -import org.apache.lucene.analysis.WhitespaceAnalyzer; -import org.apache.lucene.analysis.SimpleAnalyzer; -import org.apache.lucene.document.Document; -import org.apache.lucene.document.Field; -import org.apache.lucene.index.IndexReader; -import org.apache.lucene.index.IndexWriter; -import org.apache.lucene.index.Term; -import org.apache.lucene.store.Directory; -import org.apache.lucene.store.RAMDirectory; - -import java.io.IOException; -import java.text.Collator; -import java.util.Locale; - -import junit.framework.Assert; - -public class TestMultiTermConstantScore extends BaseTestRangeFilter { - - /** threshold for comparing floats */ - public static final float SCORE_COMP_THRESH = 1e-6f; - - public TestMultiTermConstantScore(String name) { - super(name); - } - - public TestMultiTermConstantScore() { - super(); - } - - Directory small; - - void assertEquals(String m, float e, float a) { - assertEquals(m, e, a, SCORE_COMP_THRESH); - } - - static public void assertEquals(String m, int e, int a) { - Assert.assertEquals(m, e, a); - } - - public void setUp() throws Exception { - super.setUp(); - - String[] data = new String[] { "A 1 2 3 4 5 6", "Z 4 5 6", null, - "B 2 4 5 6", "Y 3 5 6", null, "C 3 6", - "X 4 5 6" }; - - small = new RAMDirectory(); - IndexWriter writer = new IndexWriter(small, new WhitespaceAnalyzer(), true, - IndexWriter.MaxFieldLength.LIMITED); - - for (int i = 0; i < data.length; i++) { - Document doc = new Document(); - doc.add(new Field("id", String.valueOf(i), Field.Store.YES, - Field.Index.NOT_ANALYZED));// Field.Keyword("id",String.valueOf(i))); - doc - .add(new Field("all", "all", Field.Store.YES, - Field.Index.NOT_ANALYZED));// Field.Keyword("all","all")); - if (null != data[i]) { - doc.add(new Field("data", data[i], Field.Store.YES, - Field.Index.ANALYZED));// Field.Text("data",data[i])); - } - writer.addDocument(doc); - } - - writer.optimize(); - writer.close(); - } - - /** macro for readability */ - public static Query csrq(String f, String l, String h, boolean il, boolean ih) { - RangeQuery query = new RangeQuery(f, l, h, il, ih); - query.setConstantScoreRewrite(true); - return query; - } - - /** macro for readability */ - public static Query csrq(String f, String l, String h, boolean il, - boolean ih, Collator c) { - RangeQuery query = new RangeQuery(f, l, h, il, ih, c); - query.setConstantScoreRewrite(true); - return query; - } - - /** macro for readability */ - public static Query cspq(Term prefix) { - PrefixQuery query = new PrefixQuery(prefix); - query.setConstantScoreRewrite(true); - return query; - } - - /** macro for readability */ - public static Query cswcq(Term wild) { - WildcardQuery query = new WildcardQuery(wild); - query.setConstantScoreRewrite(true); - return query; - } - - public void testBasics() throws IOException { - QueryUtils.check(csrq("data", "1", "6", T, T)); - QueryUtils.check(csrq("data", "A", "Z", T, T)); - QueryUtils.checkUnequal(csrq("data", "1", "6", T, T), csrq("data", "A", - "Z", T, T)); - - QueryUtils.check(cspq(new Term("data", "p*u?"))); - QueryUtils.checkUnequal(cspq(new Term("data", "pre*")), cspq(new Term( - "data", "pres*"))); - - QueryUtils.check(cswcq(new Term("data", "p"))); - QueryUtils.checkUnequal(cswcq(new Term("data", "pre*n?t")), cswcq(new Term( - "data", "pr*t?j"))); - } - - public void testBasicsRngCollating() throws IOException { - Collator c = Collator.getInstance(Locale.ENGLISH); - QueryUtils.check(csrq("data", "1", "6", T, T, c)); - QueryUtils.check(csrq("data", "A", "Z", T, T, c)); - QueryUtils.checkUnequal(csrq("data", "1", "6", T, T, c), csrq("data", "A", - "Z", T, T, c)); - } - - public void testEqualScores() throws IOException { - // NOTE: uses index build in *this* setUp - - IndexReader reader = IndexReader.open(small); - IndexSearcher search = new IndexSearcher(reader); - - ScoreDoc[] result; - - // some hits match more terms then others, score should be the same - - result = search.search(csrq("data", "1", "6", T, T), null, 1000).scoreDocs; - int numHits = result.length; - assertEquals("wrong number of results", 6, numHits); - float score = result[0].score; - for (int i = 1; i < numHits; i++) { - assertEquals("score for " + i + " was not the same", score, - result[i].score); - } - - } - - public void testBoost() throws IOException { - // NOTE: uses index build in *this* setUp - - IndexReader reader = IndexReader.open(small); - IndexSearcher search = new IndexSearcher(reader); - - // test for correct application of query normalization - // must use a non score normalizing method for this. - Query q = csrq("data", "1", "6", T, T); - q.setBoost(100); - search.search(q, null, new HitCollector() { - public void collect(int doc, float score) { - assertEquals("score for doc " + doc + " was not correct", 1.0f, score); - } - }); - - // - // Ensure that boosting works to score one clause of a query higher - // than another. - // - Query q1 = csrq("data", "A", "A", T, T); // matches document #0 - q1.setBoost(.1f); - Query q2 = csrq("data", "Z", "Z", T, T); // matches document #1 - BooleanQuery bq = new BooleanQuery(true); - bq.add(q1, BooleanClause.Occur.SHOULD); - bq.add(q2, BooleanClause.Occur.SHOULD); - - ScoreDoc[] hits = search.search(bq, null, 1000).scoreDocs; - assertEquals(1, hits[0].doc); - assertEquals(0, hits[1].doc); - assertTrue(hits[0].score > hits[1].score); - - q1 = csrq("data", "A", "A", T, T); // matches document #0 - q1.setBoost(10f); - q2 = csrq("data", "Z", "Z", T, T); // matches document #1 - bq = new BooleanQuery(true); - bq.add(q1, BooleanClause.Occur.SHOULD); - bq.add(q2, BooleanClause.Occur.SHOULD); - - hits = search.search(bq, null, 1000).scoreDocs; - assertEquals(0, hits[0].doc); - assertEquals(1, hits[1].doc); - assertTrue(hits[0].score > hits[1].score); - } - - public void testBooleanOrderUnAffected() throws IOException { - // NOTE: uses index build in *this* setUp - - IndexReader reader = IndexReader.open(small); - IndexSearcher search = new IndexSearcher(reader); - - // first do a regular RangeQuery which uses term expansion so - // docs with more terms in range get higher scores - - Query rq = new RangeQuery(new Term("data", "1"), new Term("data", "4"), T); - - ScoreDoc[] expected = search.search(rq, null, 1000).scoreDocs; - int numHits = expected.length; - - // now do a boolean where which also contains a - // ConstantScoreRangeQuery and make sure hte order is the same - - BooleanQuery q = new BooleanQuery(); - q.add(rq, BooleanClause.Occur.MUST);// T, F); - q.add(csrq("data", "1", "6", T, T), BooleanClause.Occur.MUST);// T, F); - - ScoreDoc[] actual = search.search(q, null, 1000).scoreDocs; - - assertEquals("wrong numebr of hits", numHits, actual.length); - for (int i = 0; i < numHits; i++) { - assertEquals("mismatch in docid for hit#" + i, expected[i].doc, - actual[i].doc); - } - - } - - public void testRangeQueryId() throws IOException { - // NOTE: uses index build in *super* setUp - - IndexReader reader = IndexReader.open(signedIndex.index); - IndexSearcher search = new IndexSearcher(reader); - - int medId = ((maxId - minId) / 2); - - String minIP = pad(minId); - String maxIP = pad(maxId); - String medIP = pad(medId); - - int numDocs = reader.numDocs(); - - assertEquals("num of docs", numDocs, 1 + maxId - minId); - - ScoreDoc[] result; - - // test id, bounded on both ends - - result = search.search(csrq("id", minIP, maxIP, T, T), null, numDocs).scoreDocs; - assertEquals("find all", numDocs, result.length); - - result = search.search(csrq("id", minIP, maxIP, T, F), null, numDocs).scoreDocs; - assertEquals("all but last", numDocs - 1, result.length); - - result = search.search(csrq("id", minIP, maxIP, F, T), null, numDocs).scoreDocs; - assertEquals("all but first", numDocs - 1, result.length); - - result = search.search(csrq("id", minIP, maxIP, F, F), null, numDocs).scoreDocs; - assertEquals("all but ends", numDocs - 2, result.length); - - result = search.search(csrq("id", medIP, maxIP, T, T), null, numDocs).scoreDocs; - assertEquals("med and up", 1 + maxId - medId, result.length); - - result = search.search(csrq("id", minIP, medIP, T, T), null, numDocs).scoreDocs; - assertEquals("up to med", 1 + medId - minId, result.length); - - // unbounded id - - result = search.search(csrq("id", minIP, null, T, F), null, numDocs).scoreDocs; - assertEquals("min and up", numDocs, result.length); - - result = search.search(csrq("id", null, maxIP, F, T), null, numDocs).scoreDocs; - assertEquals("max and down", numDocs, result.length); - - result = search.search(csrq("id", minIP, null, F, F), null, numDocs).scoreDocs; - assertEquals("not min, but up", numDocs - 1, result.length); - - result = search.search(csrq("id", null, maxIP, F, F), null, numDocs).scoreDocs; - assertEquals("not max, but down", numDocs - 1, result.length); - - result = search.search(csrq("id", medIP, maxIP, T, F), null, numDocs).scoreDocs; - assertEquals("med and up, not max", maxId - medId, result.length); - - result = search.search(csrq("id", minIP, medIP, F, T), null, numDocs).scoreDocs; - assertEquals("not min, up to med", medId - minId, result.length); - - // very small sets - - result = search.search(csrq("id", minIP, minIP, F, F), null, numDocs).scoreDocs; - assertEquals("min,min,F,F", 0, result.length); - result = search.search(csrq("id", medIP, medIP, F, F), null, numDocs).scoreDocs; - assertEquals("med,med,F,F", 0, result.length); - result = search.search(csrq("id", maxIP, maxIP, F, F), null, numDocs).scoreDocs; - assertEquals("max,max,F,F", 0, result.length); - - result = search.search(csrq("id", minIP, minIP, T, T), null, numDocs).scoreDocs; - assertEquals("min,min,T,T", 1, result.length); - result = search.search(csrq("id", null, minIP, F, T), null, numDocs).scoreDocs; - assertEquals("nul,min,F,T", 1, result.length); - - result = search.search(csrq("id", maxIP, maxIP, T, T), null, numDocs).scoreDocs; - assertEquals("max,max,T,T", 1, result.length); - result = search.search(csrq("id", maxIP, null, T, F), null, numDocs).scoreDocs; - assertEquals("max,nul,T,T", 1, result.length); - - result = search.search(csrq("id", medIP, medIP, T, T), null, numDocs).scoreDocs; - assertEquals("med,med,T,T", 1, result.length); - - } - - public void testRangeQueryIdCollating() throws IOException { - // NOTE: uses index build in *super* setUp - - IndexReader reader = IndexReader.open(signedIndex.index); - IndexSearcher search = new IndexSearcher(reader); - - int medId = ((maxId - minId) / 2); - - String minIP = pad(minId); - String maxIP = pad(maxId); - String medIP = pad(medId); - - int numDocs = reader.numDocs(); - - assertEquals("num of docs", numDocs, 1 + maxId - minId); - - ScoreDoc[] result; - - Collator c = Collator.getInstance(Locale.ENGLISH); - - // test id, bounded on both ends - - result = search.search(csrq("id", minIP, maxIP, T, T, c), null, numDocs).scoreDocs; - assertEquals("find all", numDocs, result.length); - - result = search.search(csrq("id", minIP, maxIP, T, F, c), null, numDocs).scoreDocs; - assertEquals("all but last", numDocs - 1, result.length); - - result = search.search(csrq("id", minIP, maxIP, F, T, c), null, numDocs).scoreDocs; - assertEquals("all but first", numDocs - 1, result.length); - - result = search.search(csrq("id", minIP, maxIP, F, F, c), null, numDocs).scoreDocs; - assertEquals("all but ends", numDocs - 2, result.length); - - result = search.search(csrq("id", medIP, maxIP, T, T, c), null, numDocs).scoreDocs; - assertEquals("med and up", 1 + maxId - medId, result.length); - - result = search.search(csrq("id", minIP, medIP, T, T, c), null, numDocs).scoreDocs; - assertEquals("up to med", 1 + medId - minId, result.length); - - // unbounded id - - result = search.search(csrq("id", minIP, null, T, F, c), null, numDocs).scoreDocs; - assertEquals("min and up", numDocs, result.length); - - result = search.search(csrq("id", null, maxIP, F, T, c), null, numDocs).scoreDocs; - assertEquals("max and down", numDocs, result.length); - - result = search.search(csrq("id", minIP, null, F, F, c), null, numDocs).scoreDocs; - assertEquals("not min, but up", numDocs - 1, result.length); - - result = search.search(csrq("id", null, maxIP, F, F, c), null, numDocs).scoreDocs; - assertEquals("not max, but down", numDocs - 1, result.length); - - result = search.search(csrq("id", medIP, maxIP, T, F, c), null, numDocs).scoreDocs; - assertEquals("med and up, not max", maxId - medId, result.length); - - result = search.search(csrq("id", minIP, medIP, F, T, c), null, numDocs).scoreDocs; - assertEquals("not min, up to med", medId - minId, result.length); - - // very small sets - - result = search.search(csrq("id", minIP, minIP, F, F, c), null, numDocs).scoreDocs; - assertEquals("min,min,F,F,c", 0, result.length); - result = search.search(csrq("id", medIP, medIP, F, F, c), null, numDocs).scoreDocs; - assertEquals("med,med,F,F,c", 0, result.length); - result = search.search(csrq("id", maxIP, maxIP, F, F, c), null, numDocs).scoreDocs; - assertEquals("max,max,F,F,c", 0, result.length); - - result = search.search(csrq("id", minIP, minIP, T, T, c), null, numDocs).scoreDocs; - assertEquals("min,min,T,T,c", 1, result.length); - result = search.search(csrq("id", null, minIP, F, T, c), null, numDocs).scoreDocs; - assertEquals("nul,min,F,T,c", 1, result.length); - - result = search.search(csrq("id", maxIP, maxIP, T, T, c), null, numDocs).scoreDocs; - assertEquals("max,max,T,T,c", 1, result.length); - result = search.search(csrq("id", maxIP, null, T, F, c), null, numDocs).scoreDocs; - assertEquals("max,nul,T,T,c", 1, result.length); - - result = search.search(csrq("id", medIP, medIP, T, T, c), null, numDocs).scoreDocs; - assertEquals("med,med,T,T,c", 1, result.length); - } - - public void testRangeQueryRand() throws IOException { - // NOTE: uses index build in *super* setUp - - IndexReader reader = IndexReader.open(signedIndex.index); - IndexSearcher search = new IndexSearcher(reader); - - String minRP = pad(signedIndex.minR); - String maxRP = pad(signedIndex.maxR); - - int numDocs = reader.numDocs(); - - assertEquals("num of docs", numDocs, 1 + maxId - minId); - - ScoreDoc[] result; - - // test extremes, bounded on both ends - - result = search.search(csrq("rand", minRP, maxRP, T, T), null, numDocs).scoreDocs; - assertEquals("find all", numDocs, result.length); - - result = search.search(csrq("rand", minRP, maxRP, T, F), null, numDocs).scoreDocs; - assertEquals("all but biggest", numDocs - 1, result.length); - - result = search.search(csrq("rand", minRP, maxRP, F, T), null, numDocs).scoreDocs; - assertEquals("all but smallest", numDocs - 1, result.length); - - result = search.search(csrq("rand", minRP, maxRP, F, F), null, numDocs).scoreDocs; - assertEquals("all but extremes", numDocs - 2, result.length); - - // unbounded - - result = search.search(csrq("rand", minRP, null, T, F), null, numDocs).scoreDocs; - assertEquals("smallest and up", numDocs, result.length); - - result = search.search(csrq("rand", null, maxRP, F, T), null, numDocs).scoreDocs; - assertEquals("biggest and down", numDocs, result.length); - - result = search.search(csrq("rand", minRP, null, F, F), null, numDocs).scoreDocs; - assertEquals("not smallest, but up", numDocs - 1, result.length); - - result = search.search(csrq("rand", null, maxRP, F, F), null, numDocs).scoreDocs; - assertEquals("not biggest, but down", numDocs - 1, result.length); - - // very small sets - - result = search.search(csrq("rand", minRP, minRP, F, F), null, numDocs).scoreDocs; - assertEquals("min,min,F,F", 0, result.length); - result = search.search(csrq("rand", maxRP, maxRP, F, F), null, numDocs).scoreDocs; - assertEquals("max,max,F,F", 0, result.length); - - result = search.search(csrq("rand", minRP, minRP, T, T), null, numDocs).scoreDocs; - assertEquals("min,min,T,T", 1, result.length); - result = search.search(csrq("rand", null, minRP, F, T), null, numDocs).scoreDocs; - assertEquals("nul,min,F,T", 1, result.length); - - result = search.search(csrq("rand", maxRP, maxRP, T, T), null, numDocs).scoreDocs; - assertEquals("max,max,T,T", 1, result.length); - result = search.search(csrq("rand", maxRP, null, T, F), null, numDocs).scoreDocs; - assertEquals("max,nul,T,T", 1, result.length); - - } - - public void testRangeQueryRandCollating() throws IOException { - // NOTE: uses index build in *super* setUp - - // using the unsigned index because collation seems to ignore hyphens - IndexReader reader = IndexReader.open(unsignedIndex.index); - IndexSearcher search = new IndexSearcher(reader); - - String minRP = pad(unsignedIndex.minR); - String maxRP = pad(unsignedIndex.maxR); - - int numDocs = reader.numDocs(); - - assertEquals("num of docs", numDocs, 1 + maxId - minId); - - ScoreDoc[] result; - - Collator c = Collator.getInstance(Locale.ENGLISH); - - // test extremes, bounded on both ends - - result = search.search(csrq("rand", minRP, maxRP, T, T, c), null, numDocs).scoreDocs; - assertEquals("find all", numDocs, result.length); - - result = search.search(csrq("rand", minRP, maxRP, T, F, c), null, numDocs).scoreDocs; - assertEquals("all but biggest", numDocs - 1, result.length); - - result = search.search(csrq("rand", minRP, maxRP, F, T, c), null, numDocs).scoreDocs; - assertEquals("all but smallest", numDocs - 1, result.length); - - result = search.search(csrq("rand", minRP, maxRP, F, F, c), null, numDocs).scoreDocs; - assertEquals("all but extremes", numDocs - 2, result.length); - - // unbounded - - result = search.search(csrq("rand", minRP, null, T, F, c), null, numDocs).scoreDocs; - assertEquals("smallest and up", numDocs, result.length); - - result = search.search(csrq("rand", null, maxRP, F, T, c), null, numDocs).scoreDocs; - assertEquals("biggest and down", numDocs, result.length); - - result = search.search(csrq("rand", minRP, null, F, F, c), null, numDocs).scoreDocs; - assertEquals("not smallest, but up", numDocs - 1, result.length); - - result = search.search(csrq("rand", null, maxRP, F, F, c), null, numDocs).scoreDocs; - assertEquals("not biggest, but down", numDocs - 1, result.length); - - // very small sets - - result = search.search(csrq("rand", minRP, minRP, F, F, c), null, numDocs).scoreDocs; - assertEquals("min,min,F,F,c", 0, result.length); - result = search.search(csrq("rand", maxRP, maxRP, F, F, c), null, numDocs).scoreDocs; - assertEquals("max,max,F,F,c", 0, result.length); - - result = search.search(csrq("rand", minRP, minRP, T, T, c), null, numDocs).scoreDocs; - assertEquals("min,min,T,T,c", 1, result.length); - result = search.search(csrq("rand", null, minRP, F, T, c), null, numDocs).scoreDocs; - assertEquals("nul,min,F,T,c", 1, result.length); - - result = search.search(csrq("rand", maxRP, maxRP, T, T, c), null, numDocs).scoreDocs; - assertEquals("max,max,T,T,c", 1, result.length); - result = search.search(csrq("rand", maxRP, null, T, F, c), null, numDocs).scoreDocs; - assertEquals("max,nul,T,T,c", 1, result.length); - } - - public void testFarsi() throws Exception { - - /* build an index */ - RAMDirectory farsiIndex = new RAMDirectory(); - IndexWriter writer = new IndexWriter(farsiIndex, new SimpleAnalyzer(), T, - IndexWriter.MaxFieldLength.LIMITED); - Document doc = new Document(); - doc.add(new Field("content", "\u0633\u0627\u0628", Field.Store.YES, - Field.Index.NOT_ANALYZED)); - doc - .add(new Field("body", "body", Field.Store.YES, - Field.Index.NOT_ANALYZED)); - writer.addDocument(doc); - - writer.optimize(); - writer.close(); - - IndexReader reader = IndexReader.open(farsiIndex); - IndexSearcher search = new IndexSearcher(reader); - - // Neither Java 1.4.2 nor 1.5.0 has Farsi Locale collation available in - // RuleBasedCollator. However, the Arabic Locale seems to order the Farsi - // characters properly. - Collator c = Collator.getInstance(new Locale("ar")); - - // Unicode order would include U+0633 in [ U+062F - U+0698 ], but Farsi - // orders the U+0698 character before the U+0633 character, so the single - // index Term below should NOT be returned by a ConstantScoreRangeQuery - // with a Farsi Collator (or an Arabic one for the case when Farsi is - // not supported). - ScoreDoc[] result = search.search(csrq("content", "\u062F", "\u0698", T, T, - c), null, 1000).scoreDocs; - assertEquals("The index Term should not be included.", 0, result.length); - - result = search.search(csrq("content", "\u0633", "\u0638", T, T, c), null, - 1000).scoreDocs; - assertEquals("The index Term should be included.", 1, result.length); - search.close(); - } - - public void testDanish() throws Exception { - - /* build an index */ - RAMDirectory danishIndex = new RAMDirectory(); - IndexWriter writer = new IndexWriter(danishIndex, new SimpleAnalyzer(), T, - IndexWriter.MaxFieldLength.LIMITED); - - // Danish collation orders the words below in the given order - // (example taken from TestSort.testInternationalSort() ). - String[] words = { "H\u00D8T", "H\u00C5T", "MAND" }; - for (int docnum = 0 ; docnum < words.length ; ++docnum) { - Document doc = new Document(); - doc.add(new Field("content", words[docnum], - Field.Store.YES, Field.Index.UN_TOKENIZED)); - doc.add(new Field("body", "body", - Field.Store.YES, Field.Index.UN_TOKENIZED)); - writer.addDocument(doc); - } - writer.optimize(); - writer.close(); - - IndexReader reader = IndexReader.open(danishIndex); - IndexSearcher search = new IndexSearcher(reader); - Query q = new TermQuery(new Term("body","body")); - - Collator c = Collator.getInstance(new Locale("da", "dk")); - - // Unicode order would not include "H\u00C5T" in [ "H\u00D8T", "MAND" ], - // but Danish collation does. - ScoreDoc[] result = search.search - (csrq("content", "H\u00D8T", "MAND", F, F, c), null, 1000).scoreDocs; - assertEquals("The index Term should be included.", 1, result.length); - - result = search.search - (csrq("content", "H\u00C5T", "MAND", F, F, c), null, 1000).scoreDocs; - assertEquals("The index Term should not be included.", 0, result.length); - search.close(); - } -} Index: src/test/org/apache/lucene/search/TestRangeFilter.java =================================================================== --- src/test/org/apache/lucene/search/TestRangeFilter.java (revision 790750) +++ src/test/org/apache/lucene/search/TestRangeFilter.java (working copy) @@ -376,46 +376,4 @@ assertEquals("The index Term should be included.", 1, result.length()); search.close(); } - - public void testDanish() throws Exception { - - /* build an index */ - RAMDirectory danishIndex = new RAMDirectory(); - IndexWriter writer = new IndexWriter - (danishIndex, new SimpleAnalyzer(), T, - IndexWriter.MaxFieldLength.LIMITED); - // Danish collation orders the words below in the given order - // (example taken from TestSort.testInternationalSort() ). - String[] words = { "H\u00D8T", "H\u00C5T", "MAND" }; - for (int docnum = 0 ; docnum < words.length ; ++docnum) { - Document doc = new Document(); - doc.add(new Field("content", words[docnum], - Field.Store.YES, Field.Index.UN_TOKENIZED)); - doc.add(new Field("body", "body", - Field.Store.YES, Field.Index.UN_TOKENIZED)); - writer.addDocument(doc); - } - writer.optimize(); - writer.close(); - - IndexReader reader = IndexReader.open(danishIndex); - IndexSearcher search = new IndexSearcher(reader); - Query q = new TermQuery(new Term("body","body")); - - Collator collator = Collator.getInstance(new Locale("da", "dk")); - Query query = new RangeQuery - ("content", "H\u00D8T", "MAND", false, false, collator); - - // Unicode order would not include "H\u00C5T" in [ "H\u00D8T", "MAND" ], - // but Danish collation does. - Hits result = search.search - (q, new RangeFilter("content", "H\u00D8T", "MAND", F, F, collator)); - assertEquals("The index Term should be included.", 1, result.length()); - - result = search.search - (q, new RangeFilter("content", "H\u00C5T", "MAND", F, F, collator)); - assertEquals - ("The index Term should not be included.", 0, result.length()); - search.close(); - } } Index: src/test/org/apache/lucene/search/TestRangeQuery.java =================================================================== --- src/test/org/apache/lucene/search/TestRangeQuery.java (revision 790750) +++ src/test/org/apache/lucene/search/TestRangeQuery.java (working copy) @@ -23,14 +23,9 @@ import org.apache.lucene.index.IndexWriter; import org.apache.lucene.index.Term; import org.apache.lucene.store.RAMDirectory; -import org.apache.lucene.analysis.Analyzer; -import org.apache.lucene.analysis.TokenStream; -import org.apache.lucene.analysis.Tokenizer; -import org.apache.lucene.analysis.tokenattributes.TermAttribute; import org.apache.lucene.util.LuceneTestCase; import java.io.IOException; -import java.io.Reader; import java.util.Locale; import java.text.Collator; @@ -46,7 +41,9 @@ } public void testExclusive() throws Exception { - Query query = new RangeQuery("content", "A", "C", false, false); + Query query = new RangeQuery(new Term("content", "A"), + new Term("content", "C"), + false); initializeIndex(new String[] {"A", "B", "C", "D"}); IndexSearcher searcher = new IndexSearcher(dir); ScoreDoc[] hits = searcher.search(query, null, 1000).scoreDocs; @@ -65,26 +62,11 @@ assertEquals("C added, still only B in range", 1, hits.length); searcher.close(); } - - //TODO: remove in Lucene 3.0 - public void testDeprecatedCstrctors() throws IOException { - Query query = new RangeQuery(null, new Term("content","C"), false); - initializeIndex(new String[] {"A", "B", "C", "D"}); - IndexSearcher searcher = new IndexSearcher(dir); - ScoreDoc[] hits = searcher.search(query, null, 1000).scoreDocs; - assertEquals("A,B,C,D, only B in range", 2, hits.length); - searcher.close(); - - query = new RangeQuery(new Term("content","C"),null, false); - initializeIndex(new String[] {"A", "B", "C", "D"}); - searcher = new IndexSearcher(dir); - hits = searcher.search(query, null, 1000).scoreDocs; - assertEquals("A,B,C,D, only B in range", 1, hits.length); - searcher.close(); - } public void testInclusive() throws Exception { - Query query = new RangeQuery("content", "A", "C", true, true); + Query query = new RangeQuery(new Term("content", "A"), + new Term("content", "C"), + true); initializeIndex(new String[]{"A", "B", "C", "D"}); IndexSearcher searcher = new IndexSearcher(dir); @@ -106,10 +88,13 @@ } public void testEqualsHashcode() { - Query query = new RangeQuery("content", "A", "C", true, true); - + Query query = new RangeQuery(new Term("content", "A"), + new Term("content", "C"), + true); query.setBoost(1.0f); - Query other = new RangeQuery("content", "A", "C", true, true); + Query other = new RangeQuery(new Term("content", "A"), + new Term("content", "C"), + true); other.setBoost(1.0f); assertEquals("query equals itself is true", query, query); @@ -119,36 +104,38 @@ other.setBoost(2.0f); assertFalse("Different boost queries are not equal", query.equals(other)); - other = new RangeQuery("notcontent", "A", "C", true, true); + other = new RangeQuery(new Term("notcontent", "A"), new Term("notcontent", "C"), true); assertFalse("Different fields are not equal", query.equals(other)); - other = new RangeQuery("content", "X", "C", true, true); + other = new RangeQuery(new Term("content", "X"), new Term("content", "C"), true); assertFalse("Different lower terms are not equal", query.equals(other)); - other = new RangeQuery("content", "A", "Z", true, true); + other = new RangeQuery(new Term("content", "A"), new Term("content", "Z"), true); assertFalse("Different upper terms are not equal", query.equals(other)); - query = new RangeQuery("content", null, "C", true, true); - other = new RangeQuery("content", null, "C", true, true); + query = new RangeQuery(null, new Term("content", "C"), true); + other = new RangeQuery(null, new Term("content", "C"), true); assertEquals("equivalent queries with null lowerterms are equal()", query, other); assertEquals("hashcode must return same value when equals is true", query.hashCode(), other.hashCode()); - query = new RangeQuery("content", "C", null, true, true); - other = new RangeQuery("content", "C", null, true, true); + query = new RangeQuery(new Term("content", "C"), null, true); + other = new RangeQuery(new Term("content", "C"), null, true); assertEquals("equivalent queries with null upperterms are equal()", query, other); assertEquals("hashcode returns same value", query.hashCode(), other.hashCode()); - query = new RangeQuery("content", null, "C", true, true); - other = new RangeQuery("content", "C", null, true, true); + query = new RangeQuery(null, new Term("content", "C"), true); + other = new RangeQuery(new Term("content", "C"), null, true); assertFalse("queries with different upper and lower terms are not equal", query.equals(other)); - query = new RangeQuery("content", "A", "C", false, false); - other = new RangeQuery("content", "A", "C", true, true); + query = new RangeQuery(new Term("content", "A"), new Term("content", "C"), false); + other = new RangeQuery(new Term("content", "A"), new Term("content", "C"), true); assertFalse("queries with different inclusive are not equal", query.equals(other)); } public void testExclusiveCollating() throws Exception { - Query query = new RangeQuery("content", "A", "C", false, false, Collator.getInstance(Locale.ENGLISH)); + Query query = new RangeQuery(new Term("content", "A"), + new Term("content", "C"), + false, Collator.getInstance(Locale.ENGLISH)); initializeIndex(new String[] {"A", "B", "C", "D"}); IndexSearcher searcher = new IndexSearcher(dir); ScoreDoc[] hits = searcher.search(query, null, 1000).scoreDocs; @@ -169,7 +156,9 @@ } public void testInclusiveCollating() throws Exception { - Query query = new RangeQuery("content", "A", "C",true, true, Collator.getInstance(Locale.ENGLISH)); + Query query = new RangeQuery(new Term("content", "A"), + new Term("content", "C"), + true, Collator.getInstance(Locale.ENGLISH)); initializeIndex(new String[]{"A", "B", "C", "D"}); IndexSearcher searcher = new IndexSearcher(dir); @@ -195,7 +184,9 @@ // RuleBasedCollator. However, the Arabic Locale seems to order the Farsi // characters properly. Collator collator = Collator.getInstance(new Locale("ar")); - Query query = new RangeQuery("content", "\u062F", "\u0698", true, true, collator); + Query query = new RangeQuery(new Term("content", "\u062F"), + new Term("content", "\u0698"), + true, collator); // Unicode order would include U+0633 in [ U+062F - U+0698 ], but Farsi // orders the U+0698 character before the U+0633 character, so the single // index Term below should NOT be returned by a RangeQuery with a Farsi @@ -205,86 +196,16 @@ ScoreDoc[] hits = searcher.search(query, null, 1000).scoreDocs; assertEquals("The index Term should not be included.", 0, hits.length); - query = new RangeQuery("content", "\u0633", "\u0638",true, true, collator); + query = new RangeQuery(new Term("content", "\u0633"), + new Term("content", "\u0638"), + true, collator); hits = searcher.search(query, null, 1000).scoreDocs; assertEquals("The index Term should be included.", 1, hits.length); searcher.close(); } - - public void testDanish() throws Exception { - Collator collator = Collator.getInstance(new Locale("da", "dk")); - // Danish collation orders the words below in the given order (example taken - // from TestSort.testInternationalSort() ). - String[] words = { "H\u00D8T", "H\u00C5T", "MAND" }; - Query query = new RangeQuery("content", "H\u00D8T", "MAND", false, false, collator); - // Unicode order would not include "H\u00C5T" in [ "H\u00D8T", "MAND" ], - // but Danish collation does. - initializeIndex(words); - IndexSearcher searcher = new IndexSearcher(dir); - ScoreDoc[] hits = searcher.search(query, null, 1000).scoreDocs; - assertEquals("The index Term should be included.", 1, hits.length); - - query = new RangeQuery("content", "H\u00C5T", "MAND", false, false, collator); - hits = searcher.search(query, null, 1000).scoreDocs; - assertEquals("The index Term should not be included.", 0, hits.length); - searcher.close(); - } - - private static class SingleCharAnalyzer extends Analyzer { - - private static class SingleCharTokenizer extends Tokenizer { - char[] buffer = new char[1]; - boolean done; - TermAttribute termAtt; - - public SingleCharTokenizer(Reader r) { - super(r); - termAtt = (TermAttribute) addAttribute(TermAttribute.class); - } - - public boolean incrementToken() throws IOException { - int count = input.read(buffer); - if (done) - return false; - else { - done = true; - if (count == 1) { - termAtt.termBuffer()[0] = buffer[0]; - termAtt.setTermLength(1); - } else - termAtt.setTermLength(0); - return true; - } - } - - public final void reset(Reader reader) throws IOException { - super.reset(reader); - done = false; - } - } - - public TokenStream reusableTokenStream(String fieldName, Reader reader) throws IOException { - Tokenizer tokenizer = (Tokenizer) getPreviousTokenStream(); - if (tokenizer == null) { - tokenizer = new SingleCharTokenizer(reader); - setPreviousTokenStream(tokenizer); - } else - tokenizer.reset(reader); - return tokenizer; - } - - public TokenStream tokenStream(String fieldName, Reader reader) { - return new SingleCharTokenizer(reader); - } - } - private void initializeIndex(String[] values) throws IOException { - initializeIndex(values, new WhitespaceAnalyzer()); - } - - private void initializeIndex(String[] values, Analyzer analyzer) throws IOException { - IndexWriter writer = new IndexWriter(dir, analyzer, true, IndexWriter.MaxFieldLength.LIMITED); + IndexWriter writer = new IndexWriter(dir, new WhitespaceAnalyzer(), true, IndexWriter.MaxFieldLength.LIMITED); for (int i = 0; i < values.length; i++) { insertDoc(writer, values[i]); } @@ -306,71 +227,4 @@ writer.addDocument(doc); docCount++; } - - // LUCENE-38 - public void testExclusiveLowerNull() throws Exception { - Analyzer analyzer = new SingleCharAnalyzer(); - //http://issues.apache.org/jira/browse/LUCENE-38 - Query query = new RangeQuery(null, - new Term("content", "C"), - false); - initializeIndex(new String[] {"A", "B", "", "C", "D"}, analyzer); - IndexSearcher searcher = new IndexSearcher(dir); - Hits hits = searcher.search(query); - // When Lucene-38 is fixed, use the assert on the next line: - assertEquals("A,B,,C,D => A, B & are in range", 3, hits.length()); - // until Lucene-38 is fixed, use this assert: - //assertEquals("A,B,,C,D => A, B & are in range", 2, hits.length()); - - searcher.close(); - initializeIndex(new String[] {"A", "B", "", "D"}, analyzer); - searcher = new IndexSearcher(dir); - hits = searcher.search(query); - // When Lucene-38 is fixed, use the assert on the next line: - assertEquals("A,B,,D => A, B & are in range", 3, hits.length()); - // until Lucene-38 is fixed, use this assert: - //assertEquals("A,B,,D => A, B & are in range", 2, hits.length()); - searcher.close(); - addDoc("C"); - searcher = new IndexSearcher(dir); - hits = searcher.search(query); - // When Lucene-38 is fixed, use the assert on the next line: - assertEquals("C added, still A, B & are in range", 3, hits.length()); - // until Lucene-38 is fixed, use this assert - //assertEquals("C added, still A, B & are in range", 2, hits.length()); - searcher.close(); - } - - // LUCENE-38 - public void testInclusiveLowerNull() throws Exception { - //http://issues.apache.org/jira/browse/LUCENE-38 - Analyzer analyzer = new SingleCharAnalyzer(); - Query query = new RangeQuery(null, - new Term("content", "C"), - true); - initializeIndex(new String[]{"A", "B", "","C", "D"}, analyzer); - IndexSearcher searcher = new IndexSearcher(dir); - Hits hits = searcher.search(query); - // When Lucene-38 is fixed, use the assert on the next line: - assertEquals("A,B,,C,D => A,B,,C in range", 4, hits.length()); - // until Lucene-38 is fixed, use this assert - //assertEquals("A,B,,C,D => A,B,,C in range", 3, hits.length()); - searcher.close(); - initializeIndex(new String[]{"A", "B", "", "D"}, analyzer); - searcher = new IndexSearcher(dir); - hits = searcher.search(query); - // When Lucene-38 is fixed, use the assert on the next line: - assertEquals("A,B,,D - A, B and in range", 3, hits.length()); - // until Lucene-38 is fixed, use this assert - //assertEquals("A,B,,D => A, B and in range", 2, hits.length()); - searcher.close(); - addDoc("C"); - searcher = new IndexSearcher(dir); - hits = searcher.search(query); - // When Lucene-38 is fixed, use the assert on the next line: - assertEquals("C added => A,B,,C in range", 4, hits.length()); - // until Lucene-38 is fixed, use this assert - //assertEquals("C added => A,B,,C in range", 3, hits.length()); - searcher.close(); - } }