[exo-jcr-commits] exo-jcr SVN: r5691 - in jcr/branches/1.15.x: exo.jcr.component.core and 9 other directories.

Author: nzamosenchuk Date: 2012-02-22 09:18:12 -0500 (Wed, 22 Feb 2012) New Revision: 5691 Added: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/AbstractFieldComparator.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/FieldComparatorBase.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SharedFieldComparatorSource.java Removed: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SharedFieldSortComparator.java Modified: jcr/branches/1.15.x/exo.jcr.component.core/developer-notes.txt jcr/branches/1.15.x/exo.jcr.component.core/pom.xml jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/AbstractExcerpt.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/CaseTermQuery.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/ChangesHolder.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/ChildAxisQuery.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/DerefQuery.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/DescendantSelfAxisQuery.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/IndexInfos.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/JcrIndexSearcher.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/JcrQueryParser.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/JcrStandartAnalyzer.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/LuceneQueryHits.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/MatchAllScorer.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/MoreLikeThis.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/MultiScorer.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/NodeIndexer.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/NotQuery.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/ParentAxisQuery.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/PersistentIndex.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/QueryHitsQuery.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/QueryImpl.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/RangeQuery.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/ReadOnlyIndexReader.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SearchIndex.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SharedFieldCache.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SingletonTokenStream.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SortedLuceneQueryHits.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/WeightedHighlighter.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/WildcardQuery.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/directory/FSDirectoryManager.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/hits/ScorerHits.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/spell/LuceneSpellChecker.java jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/synonym/WordNetSynonyms.java jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/BaseStandaloneTest.java jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/api/core/query/lucene/SlowQueryHandler.java jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/BaseQueryTest.java jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestArabicSearch.java jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestDateSearch.java jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestExcelFileSearch.java jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestIndexingConfig.java jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestMultiValueSearch.java jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestRewriteNode.java jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/lucene/TestChangesHolder.java jcr/branches/1.15.x/pom.xml Log: EXOJCR-1766 : upgrading to Lucene 3.0 Modified: jcr/branches/1.15.x/exo.jcr.component.core/developer-notes.txt =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/developer-notes.txt 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/developer-notes.txt 2012-02-22 14:18:12 UTC (rev 5691) @@ -14,3 +14,4 @@ * Removed ValueStorageCleanHelper class * Removed BackupSchedulerException class * EXOJCR-1728: Remove ReadOnly support for WorkspacePersistentDataManager +* Lucene 3.0 \ No newline at end of file Modified: jcr/branches/1.15.x/exo.jcr.component.core/pom.xml =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/pom.xml 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/pom.xml 2012-02-22 14:18:12 UTC (rev 5691) @@ -109,6 +109,10 @@ <artifactId>lucene-memory</artifactId> </dependency> <dependency> + <groupId>org.apache.lucene</groupId> + <artifactId>lucene-wordnet</artifactId> + </dependency> + <dependency> <groupId>com.sun.xml.stream</groupId> <artifactId>sjsxp</artifactId> </dependency> Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/AbstractExcerpt.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/AbstractExcerpt.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/AbstractExcerpt.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -16,8 +16,9 @@ */ package org.exoplatform.services.jcr.impl.core.query.lucene; -import org.apache.lucene.analysis.Token; import org.apache.lucene.analysis.TokenStream; +import org.apache.lucene.analysis.tokenattributes.OffsetAttribute; +import org.apache.lucene.analysis.tokenattributes.TermAttribute; import org.apache.lucene.document.Document; import org.apache.lucene.document.Fieldable; import org.apache.lucene.index.IndexReader; @@ -193,16 +194,16 @@ /** * @return the extracted terms from the query. */ - protected final Set getQueryTerms() + protected final Set<Term> getQueryTerms() { - Set extractedTerms = new HashSet(); - Set relevantTerms = new HashSet(); + Set<Term> extractedTerms = new HashSet<Term>(); + Set<Term> relevantTerms = new HashSet<Term>(); query.extractTerms(extractedTerms); // only keep terms for fulltext fields - Iterator it = extractedTerms.iterator(); + Iterator<Term> it = extractedTerms.iterator(); while (it.hasNext()) { - Term t = (Term)it.next(); + Term t = it.next(); if (t.field().equals(FieldNames.FULLTEXT)) { relevantTerms.add(t); @@ -260,16 +261,17 @@ private TermPositionVector createTermPositionVector(String text) { // term -> TermVectorOffsetInfo[] - final SortedMap termMap = new TreeMap(); + final SortedMap<String, TermVectorOffsetInfo[]> termMap = new TreeMap<String, TermVectorOffsetInfo[]>(); Reader r = new StringReader(text); TokenStream ts = index.getTextAnalyzer().tokenStream("", r); - Token t = new Token(); try { - while ((t = ts.next(t)) != null) + while (ts.incrementToken()) { - String termText = t.term(); - TermVectorOffsetInfo[] info = (TermVectorOffsetInfo[])termMap.get(termText); + OffsetAttribute offset = (OffsetAttribute)ts.getAttribute(OffsetAttribute.class); + TermAttribute term = (TermAttribute)ts.getAttribute(TermAttribute.class); + String termText = term.term(); + TermVectorOffsetInfo[] info = termMap.get(termText); if (info == null) { info = new TermVectorOffsetInfo[1]; @@ -280,22 +282,21 @@ info = new TermVectorOffsetInfo[tmp.length + 1]; System.arraycopy(tmp, 0, info, 0, tmp.length); } - info[info.length - 1] = new TermVectorOffsetInfo(t.startOffset(), t.endOffset()); + info[info.length - 1] = new TermVectorOffsetInfo(offset.startOffset(), offset.endOffset()); termMap.put(termText, info); } + ts.end(); + ts.close(); } catch (IOException e) { - if (LOG.isTraceEnabled()) - { - LOG.trace("An exception occurred: " + e.getMessage()); - } + // should never happen, we are reading from a string } return new TermPositionVector() { - private String[] terms = (String[])termMap.keySet().toArray(new String[termMap.size()]); + private String[] terms = termMap.keySet().toArray(new String[termMap.size()]); public int[] getTermPositions(int index) { @@ -307,7 +308,7 @@ TermVectorOffsetInfo[] info = TermVectorOffsetInfo.EMPTY_OFFSET_INFO; if (index >= 0 && index < terms.length) { - info = (TermVectorOffsetInfo[])termMap.get(terms[index]); + info = termMap.get(terms[index]); } return info; } @@ -332,7 +333,7 @@ int[] freqs = new int[terms.length]; for (int i = 0; i < terms.length; i++) { - freqs[i] = ((TermVectorOffsetInfo[])termMap.get(terms[i])).length; + freqs[i] = termMap.get(terms[i]).length; } return freqs; } Added: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/AbstractFieldComparator.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/AbstractFieldComparator.java (rev 0) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/AbstractFieldComparator.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -0,0 +1,156 @@ +/* + * 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.exoplatform.services.jcr.impl.core.query.lucene; + +import org.apache.lucene.index.IndexReader; + +import java.io.IOException; +import java.util.ArrayList; +import java.util.List; + +/** + * Abstract base class for <code>FieldComparator</code>s which keep their values + * (<code>Comparable</code>s) in an array. + */ +public abstract class AbstractFieldComparator extends FieldComparatorBase +{ + + /** + * The values for comparing. + */ + private final Comparable[] values; + + /** + * The index readers. + */ + + protected final List<IndexReader> readers = new ArrayList<IndexReader>(); + + /** + * The document number starts for the {@link #readers}. + */ + protected int[] starts; + + /** + * Create a new instance with the given number of values. + * + * @param numHits the number of values + */ + protected AbstractFieldComparator(int numHits) + { + values = new Comparable[numHits]; + } + + /** + * Returns the reader index for document <code>n</code>. + * + * @param n document number. + * @return the reader index. + */ + protected final int readerIndex(int n) + { + int lo = 0; + int hi = readers.size() - 1; + + while (hi >= lo) + { + int mid = (lo + hi) >> 1; + int midValue = starts[mid]; + if (n < midValue) + { + hi = mid - 1; + } + else if (n > midValue) + { + lo = mid + 1; + } + else + { + while (mid + 1 < readers.size() && starts[mid + 1] == midValue) + { + mid++; + } + return mid; + } + } + return hi; + } + + /** + * Add the given value to the values array + * + * @param slot index into values + * @param value value for adding + */ + @Override + public void setValue(int slot, Comparable value) + { + values[slot] = value; + } + + /** + * Return a value from the values array + * + * @param slot index to retrieve + * @return the retrieved value + */ + @Override + public Comparable getValue(int slot) + { + return values[slot]; + } + + @Override + public void setNextReader(IndexReader reader, int docBase) throws IOException + { + getIndexReaders(readers, reader); + + int maxDoc = 0; + starts = new int[readers.size() + 1]; + + for (int i = 0; i < readers.size(); i++) + { + IndexReader r = readers.get(i); + starts[i] = maxDoc; + maxDoc += r.maxDoc(); + } + starts[readers.size()] = maxDoc; + } + + /** + * Checks if <code>reader</code> is of type {@link MultiIndexReader} and if + * so calls itself recursively for each reader within the + * <code>MultiIndexReader</code> or otherwise adds the reader to the list. + * + * @param readers list of index readers. + * @param reader reader to decompose + */ + private static void getIndexReaders(List<IndexReader> readers, IndexReader reader) + { + if (reader instanceof MultiIndexReader) + { + for (IndexReader r : ((MultiIndexReader)reader).getIndexReaders()) + { + getIndexReaders(readers, r); + } + } + else + { + readers.add(reader); + } + } +} \ No newline at end of file Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/CaseTermQuery.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/CaseTermQuery.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/CaseTermQuery.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -21,6 +21,7 @@ import org.apache.lucene.index.TermEnum; import org.apache.lucene.search.FilteredTermEnum; import org.apache.lucene.search.MultiTermQuery; +import org.apache.lucene.util.ToStringUtils; import org.exoplatform.services.log.ExoLogger; import org.exoplatform.services.log.Log; @@ -37,180 +38,228 @@ */ abstract class CaseTermQuery extends MultiTermQuery implements TransformConstants { + private static final Log LOG = ExoLogger.getLogger("exo.jcr.component.core.CaseTermQuery"); - /** - * Indicates whether terms from the index should be lower-cased or - * upper-cased. - */ - protected final int transform; + /** + * Indicates whether terms from the index should be lower-cased or + * upper-cased. + */ + protected final int transform; - CaseTermQuery(Term term, int transform) { - super(term); - this.transform = transform; - } + private final Term term; - /** - * {@inheritDoc} - */ - protected FilteredTermEnum getEnum(IndexReader reader) throws IOException { - return new CaseTermEnum(reader); - } + CaseTermQuery(Term term, int transform) + { + this.term = term; + this.transform = transform; + } - static final class Upper extends CaseTermQuery { + /** + * {@inheritDoc} + */ + @Override + protected FilteredTermEnum getEnum(IndexReader reader) throws IOException + { + return new CaseTermEnum(reader); + } - Upper(Term term) { - super(term, TRANSFORM_UPPER_CASE); - } - } + /** Prints a user-readable version of this query. */ + @Override + public String toString(String field) + { + StringBuffer buffer = new StringBuffer(); + if (!term.field().equals(field)) + { + buffer.append(term.field()); + buffer.append(':'); + } + buffer.append(term.text()); + buffer.append(ToStringUtils.boost(getBoost())); + return buffer.toString(); + } - static final class Lower extends CaseTermQuery { + static final class Upper extends CaseTermQuery + { - Lower(Term term) { - super(term, TRANSFORM_LOWER_CASE); - } + Upper(Term term) + { + super(term, TRANSFORM_UPPER_CASE); + } + } - } + static final class Lower extends CaseTermQuery + { + Lower(Term term) + { + super(term, TRANSFORM_LOWER_CASE); + } + } + private final class CaseTermEnum extends FilteredTermEnum { - private final int nameLength; + CaseTermEnum(IndexReader reader) throws IOException + { + // gather all terms that match + // keep them in order and remember the doc frequency as value + final Map<Term, Integer> orderedTerms = new LinkedHashMap<Term, Integer>(); - private final OffsetCharSequence termText; + // there are always two range scans: one with an initial + // lower case character and another one with an initial upper case + // character + List<RangeScan> rangeScans = new ArrayList<RangeScan>(2); + int nameLength = FieldNames.getNameLength(term.text()); + String propName = term.text().substring(0, nameLength); + OffsetCharSequence termText = new OffsetCharSequence(nameLength, term.text()); + OffsetCharSequence currentTerm = new OffsetCharSequence(nameLength, term.text(), transform); - private final OffsetCharSequence currentTerm; + try + { + // start with a term using the lower case character for the first + // character of the value. + if (term.text().length() > nameLength) + { + // start with initial lower case + StringBuffer lowerLimit = new StringBuffer(propName); + String termStr = termText.toString(); + String upperTermStr = termStr.toUpperCase(); + String lowerTermStr = termStr.toLowerCase(); - CaseTermEnum(IndexReader reader) throws IOException { - // gather all terms that match - // keep them in order and remember the doc frequency as value - final Map orderedTerms = new LinkedHashMap(); + lowerLimit.append(upperTermStr); + lowerLimit.setCharAt(nameLength, Character.toLowerCase(lowerLimit.charAt(nameLength))); + StringBuffer upperLimit = new StringBuffer(propName); + upperLimit.append(lowerTermStr); + rangeScans.add(new RangeScan(reader, new Term(term.field(), lowerLimit.toString()), new Term(term + .field(), upperLimit.toString()))); - Term term = getTerm(); + // second scan with upper case start + lowerLimit = new StringBuffer(propName); + lowerLimit.append(upperTermStr); + upperLimit = new StringBuffer(propName); + upperLimit.append(lowerTermStr); + upperLimit.setCharAt(nameLength, Character.toUpperCase(upperLimit.charAt(nameLength))); + rangeScans.add(new RangeScan(reader, new Term(term.field(), lowerLimit.toString()), new Term(term + .field(), upperLimit.toString()))); - // there are always two range scanse: one with an initial - // lower case character and another one with an initial upper case - // character - List rangeScans = new ArrayList(2); - nameLength = FieldNames.getNameLength(term.text()); - String propName = term.text().substring(0, nameLength); - this.termText = new OffsetCharSequence(nameLength, term.text()); - this.currentTerm = new OffsetCharSequence(nameLength, term.text(), transform); + } + else + { + // use term as is + rangeScans.add(new RangeScan(reader, term, term)); + } - try { - // start with a term using the lower case character for the first - // character of the value. - if (term.text().length() > nameLength) { - // start with initial lower case - StringBuffer lowerLimit = new StringBuffer(propName); - String termStr = termText.toString(); - String upperTermStr = termStr.toUpperCase(); - String lowerTermStr = termStr.toLowerCase(); - - lowerLimit.append(upperTermStr); - lowerLimit.setCharAt(nameLength, Character.toLowerCase(lowerLimit.charAt(nameLength))); - StringBuffer upperLimit = new StringBuffer(propName); - upperLimit.append(lowerTermStr); - rangeScans.add(new RangeScan(reader, - new Term(term.field(), lowerLimit.toString()), - new Term(term.field(), upperLimit.toString()))); - - // second scan with upper case start - lowerLimit = new StringBuffer(propName); - lowerLimit.append(upperTermStr); - upperLimit = new StringBuffer(propName); - upperLimit.append(lowerTermStr); - upperLimit.setCharAt(nameLength, Character.toUpperCase(upperLimit.charAt(nameLength))); - rangeScans.add(new RangeScan(reader, - new Term(term.field(), lowerLimit.toString()), - new Term(term.field(), upperLimit.toString()))); - - } else { - // use term as is - rangeScans.add(new RangeScan(reader, term, term)); - } - - Iterator it = rangeScans.iterator(); - while (it.hasNext()) { - TermEnum terms = (TermEnum) it.next(); - do { - Term t = terms.term(); - if (t != null) { - currentTerm.setBase(t.text()); - int compare = currentTerm.compareTo(termText); - if (compare == 0) { - orderedTerms.put(t, new Integer(terms.docFreq())); - } else if (compare < 0) { - // try next one - } else { - // compare > 0 - } - } else { - break; - } - } while (terms.next()); - } - } finally { - Iterator it = rangeScans.iterator(); - while (it.hasNext()) { - TermEnum terms = (TermEnum) it.next(); - try { - terms.close(); - } catch (IOException e) { - if (LOG.isTraceEnabled()) - { - LOG.trace("An exception occurred: " + e.getMessage()); - } - } - } + for (TermEnum terms : rangeScans) + { + do + { + Term t = terms.term(); + if (t != null) + { + currentTerm.setBase(t.text()); + int compare = currentTerm.compareTo(termText); + if (compare == 0) + { + orderedTerms.put(t, terms.docFreq()); + } + else if (compare < 0) + { + // try next one + } + else + { + // compare > 0 + } + } + else + { + break; + } + } + while (terms.next()); } + } + finally + { + for (TermEnum terms : rangeScans) + { + try + { + terms.close(); + } + catch (IOException e) + { + if (LOG.isTraceEnabled()) + { + LOG.trace("An exception occurred: " + e.getMessage()); + } + } + } + } - final Iterator it = orderedTerms.keySet().iterator(); + final Iterator<Term> it = orderedTerms.keySet().iterator(); - setEnum(new TermEnum() { + setEnum(new TermEnum() + { - private Term current; + private Term current; - { - getNext(); - } + { + getNext(); + } - public boolean next() { - getNext(); - return current != null; - } + @Override + public boolean next() + { + getNext(); + return current != null; + } - public Term term() { - return current; - } + @Override + public Term term() + { + return current; + } - public int docFreq() { - Integer docFreq = (Integer) orderedTerms.get(current); - return docFreq != null ? docFreq.intValue() : 0; - } + @Override + public int docFreq() + { + Integer docFreq = orderedTerms.get(current); + return docFreq != null ? docFreq : 0; + } - public void close() { - // nothing to close - } + @Override + public void close() + { + // nothing to close + } - private void getNext() { - current = it.hasNext() ? (Term) it.next() : null; - } - }); - } + private void getNext() + { + current = it.hasNext() ? it.next() : null; + } + }); + } - protected boolean termCompare(Term term) { - // they all match - return true; - } + @Override + protected boolean termCompare(Term term) + { + // they all match + return true; + } - public float difference() { - return 1.0f; - } + @Override + public float difference() + { + return 1.0f; + } - protected boolean endEnum() { - // todo correct? - return false; - } - } + @Override + protected boolean endEnum() + { + // todo correct? + return false; + } + } } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/ChangesHolder.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/ChangesHolder.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/ChangesHolder.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -19,6 +19,7 @@ package org.exoplatform.services.jcr.impl.core.query.lucene; import org.apache.lucene.analysis.TokenStream; +import org.apache.lucene.document.AbstractField; import org.apache.lucene.document.Document; import org.apache.lucene.document.Field; import org.apache.lucene.document.Fieldable; @@ -43,8 +44,6 @@ private static final int STORED_FLAG = 1; - private static final int COMPRESSED_FLAG = 1 << 1; - private static final int INDEXED_FLAG = 1 << 2; private static final int TOKENIZED_FLAG = 1 << 3; @@ -120,7 +119,7 @@ } return ids; } - + /** * @return the id of the given lucene doc */ @@ -128,7 +127,7 @@ { return doc.get(FieldNames.UUID); } - + /** * {@inheritDoc} */ @@ -176,12 +175,12 @@ { // The value is a String field = - new Field(name, (String) value, getStoreParameter(flags), getIndexParameter(flags), - getTermVectorParameter(flags)); + new Field(name, (String)value, getStoreParameter(flags), getIndexParameter(flags), + getTermVectorParameter(flags)); } field.setBoost(boost); field.setOmitNorms((flags & OMIT_NORMS_FLAG) > 0); - field.setOmitTf((flags & OMIT_TF_FLAG) > 0); + field.setOmitTermFreqAndPositions((flags & OMIT_TF_FLAG) > 0); return field; } @@ -215,12 +214,8 @@ */ private static Field.Store getStoreParameter(int flags) { - if ((flags & COMPRESSED_FLAG) > 0) + if ((flags & STORED_FLAG) > 0) { - return Field.Store.COMPRESS; - } - else if ((flags & STORED_FLAG) > 0) - { return Field.Store.YES; } else @@ -237,8 +232,7 @@ */ private static Field.TermVector getTermVectorParameter(int flags) { - if (((flags & STORE_POSITION_WITH_TERM_VECTOR_FLAG) > 0) - && ((flags & STORE_OFFSET_WITH_TERM_VECTOR_FLAG) > 0)) + if (((flags & STORE_POSITION_WITH_TERM_VECTOR_FLAG) > 0) && ((flags & STORE_OFFSET_WITH_TERM_VECTOR_FLAG) > 0)) { return Field.TermVector.WITH_POSITIONS_OFFSETS; } @@ -307,7 +301,7 @@ if (field.getBoost() != 1.0f) { // Boost - out.writeFloat(field.getBoost()); + out.writeFloat(field.getBoost()); } // Value writeValue(out, field); @@ -352,10 +346,6 @@ { flags |= STORED_FLAG; } - if (field.isCompressed()) - { - flags |= COMPRESSED_FLAG; - } if (field.isIndexed()) { flags |= INDEXED_FLAG; @@ -388,7 +378,7 @@ { flags |= LAZY_FLAG; } - if (field.getOmitTf()) + if (field instanceof AbstractField && ((AbstractField)field).getOmitTermFreqAndPositions()) { flags |= OMIT_TF_FLAG; } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/ChildAxisQuery.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/ChildAxisQuery.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/ChildAxisQuery.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -418,12 +418,14 @@ this.hResolver = hResolver; } - /** - * {@inheritDoc} - */ @Override - public boolean next() throws IOException + public int nextDoc() throws IOException { + if (nextDoc == NO_MORE_DOCS) + { + return nextDoc; + } + calculateChildren(); do { @@ -431,54 +433,46 @@ } while (nextDoc > -1 && !indexIsValid(nextDoc)); - return nextDoc > -1; + if (nextDoc < 0) + { + nextDoc = NO_MORE_DOCS; + } + return nextDoc; } - /** - * {@inheritDoc} - */ @Override - public int doc() + public int docID() { return nextDoc; } - /** - * {@inheritDoc} - */ @Override public float score() throws IOException { return 1.0f; } - /** - * {@inheritDoc} - */ @Override - public boolean skipTo(int target) throws IOException + public int advance(int target) throws IOException { + if (nextDoc == NO_MORE_DOCS) + { + return nextDoc; + } + calculateChildren(); nextDoc = hits.skipTo(target); while (nextDoc > -1 && !indexIsValid(nextDoc)) { - next(); + nextDoc(); } - return nextDoc > -1; + if (nextDoc < 0) + { + nextDoc = NO_MORE_DOCS; + } + return nextDoc; } - /** - * {@inheritDoc} - * - * @throws UnsupportedOperationException this implementation always - * throws an <code>UnsupportedOperationException</code>. - */ - @Override - public Explanation explain(int doc) throws IOException - { - throw new UnsupportedOperationException(); - } - private void calculateChildren() throws IOException { if (hits == null) @@ -492,7 +486,7 @@ contextScorer.score(new AbstractHitCollector() { @Override - public void collect(int doc, float score) + protected void collect(int doc, float score) { calc[0].collectContextHit(doc); } @@ -505,22 +499,22 @@ contextScorer.score(new AbstractHitCollector() { - private List docIds = new ArrayList(); + private List<Integer> docIds = new ArrayList<Integer>(); @Override - public void collect(int doc, float score) + protected void collect(int doc, float score) { calc[0].collectContextHit(doc); if (docIds != null) { - docIds.add(new Integer(doc)); + docIds.add(doc); if (docIds.size() > CONTEXT_SIZE_THRESHOLD) { // switch calc[0] = new HierarchyResolvingChildrenCalculator(reader, hResolver); - for (Iterator it = docIds.iterator(); it.hasNext();) + for (int docId : docIds) { - calc[0].collectContextHit(((Integer)it.next()).intValue()); + calc[0].collectContextHit(docId); } // indicate that we switched docIds = null; Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/DerefQuery.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/DerefQuery.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/DerefQuery.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -241,7 +241,6 @@ } //----------------------< DerefScorer >--------------------------------- - /** * Implements a <code>Scorer</code> for this <code>DerefQuery</code>. */ @@ -261,7 +260,7 @@ /** * List of UUIDs of selected nodes */ - private List uuids = null; + private List<String> uuids = null; /** * The next document id to return @@ -281,67 +280,71 @@ this.hits = new BitSet(reader.maxDoc()); } - /** - * {@inheritDoc} - */ @Override - public boolean next() throws IOException + public int nextDoc() throws IOException { + if (nextDoc == NO_MORE_DOCS) + { + return nextDoc; + } + calculateChildren(); nextDoc = hits.nextSetBit(nextDoc + 1); - return nextDoc > -1; + if (nextDoc < 0) + { + nextDoc = NO_MORE_DOCS; + } + return nextDoc; } - /** - * {@inheritDoc} - */ @Override - public int doc() + public int docID() { return nextDoc; } - /** - * {@inheritDoc} - */ @Override public float score() throws IOException { return 1.0f; } - /** - * {@inheritDoc} - */ @Override - public boolean skipTo(int target) throws IOException + public int advance(int target) throws IOException { + if (nextDoc == NO_MORE_DOCS) + { + return nextDoc; + } + calculateChildren(); nextDoc = hits.nextSetBit(target); - return nextDoc > -1; + if (nextDoc < 0) + { + nextDoc = NO_MORE_DOCS; + } + return nextDoc; } /** - * {@inheritDoc} - * - * @throws UnsupportedOperationException this implementation always - * throws an <code>UnsupportedOperationException</code>. + * 1. do context query + * 2. go through each document from the query + * 3. find reference property UUIDs + * 4. Use UUIDs to find document number + * 5. Use the name test to filter the documents + * + * @throws IOException if an exception occurs while reading from the + * index. */ - @Override - public Explanation explain(int doc) throws IOException - { - throw new UnsupportedOperationException(); - } - private void calculateChildren() throws IOException { if (uuids == null) { - uuids = new ArrayList(); + uuids = new ArrayList<String>(); contextScorer.score(new AbstractHitCollector() { @Override - public void collect(int doc, float score) + protected void collect(int doc, float score) { hits.set(doc); } @@ -354,7 +357,7 @@ nameTestScorer.score(new AbstractHitCollector() { @Override - public void collect(int doc, float score) + protected void collect(int doc, float score) { nameTestHits.set(doc); } @@ -371,11 +374,11 @@ // no reference properties at all on this node continue; } - for (int v = 0; v < values.length; v++) + for (String value : values) { - if (values[v].startsWith(prefix)) + if (value.startsWith(prefix)) { - uuids.add(values[v].substring(prefix.length())); + uuids.add(value.substring(prefix.length())); } } } @@ -383,9 +386,9 @@ // collect the doc ids of all target nodes. we reuse the existing // bitset. hits.clear(); - for (Iterator it = uuids.iterator(); it.hasNext();) + for (String uuid : uuids) { - TermDocs node = reader.termDocs(new Term(FieldNames.UUID, (String)it.next())); + TermDocs node = reader.termDocs(new Term(FieldNames.UUID, uuid)); try { while (node.next()) Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/DescendantSelfAxisQuery.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/DescendantSelfAxisQuery.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/DescendantSelfAxisQuery.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -511,6 +511,11 @@ private final int[] singleDoc = new int[1]; /** + * The next document id to be returned + */ + private int currentDoc = -1; + + /** * Creates a new <code>DescendantSelfAxisScorer</code>. * * @param similarity the <code>Similarity</code> instance to use. @@ -529,35 +534,39 @@ * {@inheritDoc} */ @Override - public boolean next() throws IOException + public int nextDoc() throws IOException { + if (currentDoc == NO_MORE_DOCS) + { + return currentDoc; + } + collectContextHits(); - if (!subScorer.next() || contextHits.isEmpty()) + currentDoc = subScorer.nextDoc(); + if (contextHits.isEmpty()) { - return false; + currentDoc = NO_MORE_DOCS; } - int nextDoc = subScorer.doc(); - while (nextDoc > -1) + while (currentDoc != NO_MORE_DOCS) { - - if (isValid(nextDoc)) + if (isValid(currentDoc)) { - return true; + return currentDoc; } // try next - nextDoc = subScorer.next() ? subScorer.doc() : -1; + currentDoc = subScorer.nextDoc(); } - return false; + return currentDoc; } /** * {@inheritDoc} */ @Override - public int doc() + public int docID() { - return subScorer.doc(); + return currentDoc; } /** @@ -573,17 +582,22 @@ * {@inheritDoc} */ @Override - public boolean skipTo(int target) throws IOException + public int advance(int target) throws IOException { - boolean match = subScorer.skipTo(target); - if (match) + if (currentDoc == NO_MORE_DOCS) { - collectContextHits(); - return isValid(subScorer.doc()) || next(); + return currentDoc; } + + currentDoc = subScorer.nextDoc(); + if (currentDoc == NO_MORE_DOCS) + { + return NO_MORE_DOCS; + } else { - return false; + collectContextHits(); + return isValid(currentDoc) ? currentDoc : nextDoc(); } } @@ -591,40 +605,26 @@ { if (!contextHitsCalculated) { - long time = 0; - if (log.isDebugEnabled()) - { - time = System.currentTimeMillis(); - } + long time = System.currentTimeMillis(); contextScorer.score(new AbstractHitCollector() { @Override - public void collect(int doc, float score) + protected void collect(int doc, float score) { contextHits.set(doc); } }); // find all contextHitsCalculated = true; + time = System.currentTimeMillis() - time; if (log.isDebugEnabled()) { - time = System.currentTimeMillis() - time; - log.debug("Collected {} context hits in {} ms for {}", new Object[]{ - new Integer(contextHits.cardinality()), new Long(time), DescendantSelfAxisQuery.this}); + log.debug("Collected {} context hits in {} ms for {}", new Object[]{contextHits.cardinality(), time, + DescendantSelfAxisQuery.this}); } } } /** - * @throws UnsupportedOperationException this implementation always - * throws an <code>UnsupportedOperationException</code>. - */ - @Override - public Explanation explain(int doc) throws IOException - { - throw new UnsupportedOperationException(); - } - - /** * Returns <code>true</code> if <code>doc</code> is a valid match from * the sub scorer against the context hits. The caller must ensure * that the context hits are calculated before this method is called! @@ -644,7 +644,7 @@ // check if doc is a descendant of one of the context nodes pDocs = hResolver.getParents(doc, pDocs); - if (pDocs.length == 0 || pDocs[0] < 0) + if (pDocs.length == 0) { return false; } @@ -657,10 +657,9 @@ while (pDocs.length != 0) { boolean valid = false; - for (int i = 0; i < pDocs.length; i++) + for (int pDoc : pDocs) { - int pDoci = pDocs[i]; - if (pDoci >= 0 && pDoci <= contextHits.size() && ancestorCount >= minLevels && contextHits.get(pDoci)) + if (ancestorCount >= minLevels && contextHits.get(pDoc)) { valid = true; break; @@ -722,9 +721,9 @@ else { pDocs = new int[0]; - for (int i = 0; i < docs.length; i++) + for (int doc : docs) { - int[] p = hResolver.getParents(docs[i], new int[0]); + int[] p = hResolver.getParents(doc, new int[0]); int[] tmp = new int[p.length + pDocs.length]; System.arraycopy(pDocs, 0, tmp, 0, pDocs.length); System.arraycopy(p, 0, tmp, pDocs.length, p.length); Added: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/FieldComparatorBase.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/FieldComparatorBase.java (rev 0) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/FieldComparatorBase.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -0,0 +1,115 @@ +/* + * 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.exoplatform.services.jcr.impl.core.query.lucene; + +import org.apache.lucene.search.FieldComparator; + +import java.io.IOException; + +/** + * Abstract base class for <code>FieldComparator</code> implementations + * which are based on values in the form of <code>Comparables</code>. + */ +abstract public class FieldComparatorBase extends FieldComparator +{ + + /** + * The bottom value. + */ + private Comparable bottom; + + /** + * Value for a document + * + * @param doc id of the document + * @return the value for the given id + */ + protected abstract Comparable sortValue(int doc); + + /** + * Retrieves the value of a given slot + * + * @param slot index of the value to retrieve + * @return the value in the given slot + */ + protected abstract Comparable getValue(int slot); + + /** + * Puts a value into a given slot + * + * @param slot index where to put the value + * @param value the value to put into the given slot + */ + protected abstract void setValue(int slot, Comparable value); + + @Override + public int compare(int slot1, int slot2) + { + return compare(getValue(slot1), getValue(slot2)); + } + + @Override + public int compareBottom(int doc) throws IOException + { + return compare(bottom, sortValue(doc)); + } + + @Override + public void setBottom(int slot) + { + bottom = getValue(slot); + } + + /** + * Compare two values + * + * @param val1 first value + * @param val2 second value + * @return A negative integer if <code>val1</code> comes before <code>val2</code>, + * a positive integer if <code>val1</code> comes after <code>val2</code> and + * <code>0</code> if <code>val1</code> and <code>val2</code> are equal. + */ + protected int compare(Comparable val1, Comparable val2) + { + if (val1 == null) + { + if (val2 == null) + { + return 0; + } + return -1; + } + else if (val2 == null) + { + return 1; + } + return Util.compare(val1, val2); + } + + @Override + public void copy(int slot, int doc) throws IOException + { + setValue(slot, sortValue(doc)); + } + + @Override + public Comparable value(int slot) + { + return getValue(slot); + } +} Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/IndexInfos.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/IndexInfos.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/IndexInfos.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -20,6 +20,7 @@ import org.exoplatform.commons.utils.SecurityHelper; import org.exoplatform.services.jcr.impl.core.query.lucene.directory.IndexInputStream; import org.exoplatform.services.jcr.impl.core.query.lucene.directory.IndexOutputStream; +import org.exoplatform.services.jcr.impl.util.io.DirectoryHelper; import java.io.DataInputStream; import java.io.DataOutputStream; @@ -119,7 +120,7 @@ public Object run() throws Exception { // Known issue for NFS bases on ext3. Need to refresh directory to read actual data. - dir.list(); + dir.listAll(); names.clear(); indexes.clear(); @@ -186,7 +187,7 @@ { dir.deleteFile(name); } - dir.renameFile(name + ".new", name); + rename(name + ".new", name); dirty = false; return null; } @@ -194,6 +195,43 @@ } /** + * Renames file by copying. + * + * @param from + * @param to + * @throws IOException + */ + private void rename(String from, String to) throws IOException + { + IndexOutputStream out = null; + IndexInputStream in = null; + try + { + out = new IndexOutputStream(dir.createOutput(to)); + in = new IndexInputStream(dir.openInput(from)); + DirectoryHelper.transfer(in, out); + // delete old one + if (dir.fileExists(from)) + { + dir.deleteFile(from); + } + } + finally + { + if (in != null) + { + in.close(); + } + + if (out != null) + { + out.flush(); + out.close(); + } + } + } + + /** * Returns the index name at position <code>i</code>. * @param i the position. * @return the index name. Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/JcrIndexSearcher.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/JcrIndexSearcher.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/JcrIndexSearcher.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -105,7 +105,7 @@ } if (hits == null) { - if (sort == null) + if (sort == null || sort.getSort().length == 0) { hits = new LuceneQueryHits(reader, JcrIndexSearcher.this, query); } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/JcrQueryParser.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/JcrQueryParser.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/JcrQueryParser.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -16,190 +16,233 @@ */ package org.exoplatform.services.jcr.impl.core.query.lucene; -import java.util.ArrayList; -import java.util.List; - import org.apache.lucene.analysis.Analyzer; import org.apache.lucene.queryParser.ParseException; import org.apache.lucene.queryParser.QueryParser; import org.apache.lucene.search.BooleanClause; import org.apache.lucene.search.Query; +import org.apache.lucene.util.Version; +import java.util.ArrayList; +import java.util.List; + /** * <code>JackrabbitQueryParser</code> extends the standard lucene query parser * and adds JCR specific customizations. */ -public class JcrQueryParser extends QueryParser { +public class JcrQueryParser extends QueryParser +{ - /** - * The Jackrabbit synonym provider or <code>null</code> if there is none. - */ - private final SynonymProvider synonymProvider; + /** + * The Jackrabbit synonym provider or <code>null</code> if there is none. + */ + private final SynonymProvider synonymProvider; - /** - * Creates a new query parser instance. - * - * @param fieldName the field name. - * @param analyzer the analyzer. - * @param synonymProvider the synonym provider or <code>null</code> if none - * is available. - */ - public JcrQueryParser(String fieldName, - Analyzer analyzer, - SynonymProvider synonymProvider) { - super(fieldName, analyzer); - this.synonymProvider = synonymProvider; - setAllowLeadingWildcard(true); - setDefaultOperator(Operator.AND); - } + /** + * Creates a new query parser instance. + * + * @param fieldName the field name. + * @param analyzer the analyzer. + * @param synonymProvider the synonym provider or <code>null</code> if none + * is available. + */ + public JcrQueryParser(String fieldName, Analyzer analyzer, SynonymProvider synonymProvider) + { + super(Version.LUCENE_24, fieldName, analyzer); + this.synonymProvider = synonymProvider; + setAllowLeadingWildcard(true); + setDefaultOperator(Operator.AND); + } - /** - * {@inheritDoc} - */ - public Query parse(String textsearch) throws ParseException { - // replace escaped ' with just ' - StringBuffer rewritten = new StringBuffer(); - // the default lucene query parser recognizes 'AND' and 'NOT' as - // keywords. - textsearch = textsearch.replaceAll("AND", "and"); - textsearch = textsearch.replaceAll("NOT", "not"); - boolean escaped = false; - for (int i = 0; i < textsearch.length(); i++) { - if (textsearch.charAt(i) == '\\') { - if (escaped) { - rewritten.append("\\\\"); - escaped = false; - } else { - escaped = true; - } - } else if (textsearch.charAt(i) == '\'') { - if (escaped) { - escaped = false; - } - rewritten.append(textsearch.charAt(i)); - } else if (textsearch.charAt(i) == '~') { - if (i == 0 || Character.isWhitespace(textsearch.charAt(i - 1))) { - // escape tilde so we can use it for similarity query - rewritten.append("\\"); - } - rewritten.append('~'); - } else { - if (escaped) { - rewritten.append('\\'); - escaped = false; - } - rewritten.append(textsearch.charAt(i)); + /** + * {@inheritDoc} + */ + public Query parse(String textsearch) throws ParseException + { + // replace escaped ' with just ' + StringBuffer rewritten = new StringBuffer(); + // the default lucene query parser recognizes 'AND' and 'NOT' as + // keywords. + textsearch = textsearch.replaceAll("AND", "and"); + textsearch = textsearch.replaceAll("NOT", "not"); + boolean escaped = false; + for (int i = 0; i < textsearch.length(); i++) + { + if (textsearch.charAt(i) == '\\') + { + if (escaped) + { + rewritten.append("\\\\"); + escaped = false; } - } - return super.parse(rewritten.toString()); - } - - /** - * Factory method for generating a synonym query. - * Called when parser parses an input term token that has the synonym - * prefix (~term) prepended. - * - * @param field Name of the field query will use. - * @param termStr Term token to use for building term for the query - * - * @return Resulting {@link Query} built for the term - * @exception ParseException throw in overridden method to disallow - */ - protected Query getSynonymQuery(String field, String termStr) - throws ParseException { - List synonyms = new ArrayList(); - synonyms.add(new BooleanClause(getFieldQuery(field, termStr), - BooleanClause.Occur.SHOULD)); - if (synonymProvider != null) { - String[] terms = synonymProvider.getSynonyms(termStr); - for (int i = 0; i < terms.length; i++) { - synonyms.add(new BooleanClause(getFieldQuery(field, terms[i]), - BooleanClause.Occur.SHOULD)); + else + { + escaped = true; } - } - if (synonyms.size() == 1) { - return ((BooleanClause) synonyms.get(0)).getQuery(); - } else { - return getBooleanQuery(synonyms); - } - } + } + else if (textsearch.charAt(i) == '\'') + { + if (escaped) + { + escaped = false; + } + rewritten.append(textsearch.charAt(i)); + } + else if (textsearch.charAt(i) == '~') + { + if (i == 0 || Character.isWhitespace(textsearch.charAt(i - 1))) + { + // escape tilde so we can use it for similarity query + rewritten.append("\\"); + } + rewritten.append('~'); + } + else + { + if (escaped) + { + rewritten.append('\\'); + escaped = false; + } + rewritten.append(textsearch.charAt(i)); + } + } + return super.parse(rewritten.toString()); + } + /** + * Factory method for generating a synonym query. + * Called when parser parses an input term token that has the synonym + * prefix (~term) prepended. + * + * @param field Name of the field query will use. + * @param termStr Term token to use for building term for the query + * + * @return Resulting {@link Query} built for the term + * @exception ParseException throw in overridden method to disallow + */ + protected Query getSynonymQuery(String field, String termStr) throws ParseException + { + List synonyms = new ArrayList(); + synonyms.add(new BooleanClause(getFieldQuery(field, termStr), BooleanClause.Occur.SHOULD)); + if (synonymProvider != null) + { + String[] terms = synonymProvider.getSynonyms(termStr); + for (int i = 0; i < terms.length; i++) + { + synonyms.add(new BooleanClause(getFieldQuery(field, terms[i]), BooleanClause.Occur.SHOULD)); + } + } + if (synonyms.size() == 1) + { + return ((BooleanClause)synonyms.get(0)).getQuery(); + } + else + { + return getBooleanQuery(synonyms); + } + } - /** - * {@inheritDoc} - */ - protected Query getFieldQuery(String field, String queryText) - throws ParseException { - if (queryText.startsWith("~")) { - // synonym query - return getSynonymQuery(field, queryText.substring(1)); - } else { - return super.getFieldQuery(field, queryText); - } - } + /** + * {@inheritDoc} + */ + protected Query getFieldQuery(String field, String queryText) throws ParseException + { + if (queryText.startsWith("~")) + { + // synonym query + return getSynonymQuery(field, queryText.substring(1)); + } + else + { + return super.getFieldQuery(field, queryText); + } + } - /** - * {@inheritDoc} - */ - protected Query getPrefixQuery(String field, String termStr) - throws ParseException { - return getWildcardQuery(field, termStr + "*"); - } + /** + * {@inheritDoc} + */ + protected Query getPrefixQuery(String field, String termStr) throws ParseException + { + return getWildcardQuery(field, termStr + "*"); + } - /** - * {@inheritDoc} - */ - protected Query getWildcardQuery(String field, String termStr) - throws ParseException { - if (getLowercaseExpandedTerms()) { - termStr = termStr.toLowerCase(); - } - return new WildcardQuery(field, null, translateWildcards(termStr)); - } + /** + * {@inheritDoc} + */ + protected Query getWildcardQuery(String field, String termStr) throws ParseException + { + if (getLowercaseExpandedTerms()) + { + termStr = termStr.toLowerCase(); + } + return new WildcardQuery(field, null, translateWildcards(termStr)); + } - /** - * Translates unescaped wildcards '*' and '?' into '%' and '_'. - * - * @param input the input String. - * @return the translated String. - */ - private String translateWildcards(String input) { - StringBuffer translated = new StringBuffer(input.length()); - boolean escaped = false; - for (int i = 0; i < input.length(); i++) { - if (input.charAt(i) == '\\') { - if (escaped) { - translated.append("\\\\"); - escaped = false; - } else { - escaped = true; - } - } else if (input.charAt(i) == '*') { - if (escaped) { - translated.append('*'); - escaped = false; - } else { - translated.append('%'); - } - } else if (input.charAt(i) == '?') { - if (escaped) { - translated.append('?'); - escaped = false; - } else { - translated.append('_'); - } - } else if (input.charAt(i) == '%' || input.charAt(i) == '_') { - // escape every occurrence of '%' and '_' - escaped = false; - translated.append('\\').append(input.charAt(i)); - } else { - if (escaped) { - translated.append('\\'); - escaped = false; - } - translated.append(input.charAt(i)); + /** + * Translates unescaped wildcards '*' and '?' into '%' and '_'. + * + * @param input the input String. + * @return the translated String. + */ + private String translateWildcards(String input) + { + StringBuffer translated = new StringBuffer(input.length()); + boolean escaped = false; + for (int i = 0; i < input.length(); i++) + { + if (input.charAt(i) == '\\') + { + if (escaped) + { + translated.append("\\\\"); + escaped = false; } - } - return translated.toString(); - } + else + { + escaped = true; + } + } + else if (input.charAt(i) == '*') + { + if (escaped) + { + translated.append('*'); + escaped = false; + } + else + { + translated.append('%'); + } + } + else if (input.charAt(i) == '?') + { + if (escaped) + { + translated.append('?'); + escaped = false; + } + else + { + translated.append('_'); + } + } + else if (input.charAt(i) == '%' || input.charAt(i) == '_') + { + // escape every occurrence of '%' and '_' + escaped = false; + translated.append('\\').append(input.charAt(i)); + } + else + { + if (escaped) + { + translated.append('\\'); + escaped = false; + } + translated.append(input.charAt(i)); + } + } + return translated.toString(); + } } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/JcrStandartAnalyzer.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/JcrStandartAnalyzer.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/JcrStandartAnalyzer.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -19,10 +19,12 @@ import org.apache.lucene.analysis.Analyzer; import org.apache.lucene.analysis.TokenStream; import org.apache.lucene.analysis.standard.StandardAnalyzer; +import org.apache.lucene.util.Version; import org.exoplatform.commons.utils.SecurityHelper; import java.io.Reader; import java.security.PrivilegedAction; +import java.util.Collections; /** * This is the global jackrabbit lucene analyzer. By default, all @@ -34,53 +36,59 @@ * index the text of the property and to parse searchtext for this property. */ -public class JcrStandartAnalyzer extends Analyzer { +public class JcrStandartAnalyzer extends Analyzer +{ - /** - * The default Jackrabbit analyzer if none is configured in <code><SearchIndex></code> - * configuration. - */ + /** + * The default Jackrabbit analyzer if none is configured in <code><SearchIndex></code> + * configuration. + */ private Analyzer defaultAnalyzer = SecurityHelper.doPrivilegedAction(new PrivilegedAction<Analyzer>() { public Analyzer run() { - return new StandardAnalyzer(new String[]{}); + return new StandardAnalyzer(Version.LUCENE_24, Collections.EMPTY_SET); } }); - /** - * The indexing configuration. - */ - private IndexingConfiguration indexingConfig; + /** + * The indexing configuration. + */ + private IndexingConfiguration indexingConfig; - /** - * A param indexingConfig the indexing configuration. - */ - protected void setIndexingConfig(IndexingConfiguration indexingConfig) { - this.indexingConfig = indexingConfig; - } + /** + * A param indexingConfig the indexing configuration. + */ + protected void setIndexingConfig(IndexingConfiguration indexingConfig) + { + this.indexingConfig = indexingConfig; + } - /** - * @param analyzer the default jackrabbit analyzer - */ - protected void setDefaultAnalyzer(Analyzer analyzer) { - defaultAnalyzer = analyzer; - } + /** + * @param analyzer the default jackrabbit analyzer + */ + protected void setDefaultAnalyzer(Analyzer analyzer) + { + defaultAnalyzer = analyzer; + } - /** - * Creates a TokenStream which tokenizes all the text in the provided - * Reader. If the fieldName (property) is configured to have a different - * analyzer than the default, this analyzer is used for tokenization - */ - @Override - public TokenStream tokenStream(String fieldName, Reader reader) { - if (indexingConfig != null) { - Analyzer propertyAnalyzer = indexingConfig.getPropertyAnalyzer(fieldName); - if (propertyAnalyzer != null) { - return propertyAnalyzer.tokenStream(fieldName, reader); - } - } - return defaultAnalyzer.tokenStream(fieldName, reader); - } + /** + * Creates a TokenStream which tokenizes all the text in the provided + * Reader. If the fieldName (property) is configured to have a different + * analyzer than the default, this analyzer is used for tokenization + */ + @Override + public TokenStream tokenStream(String fieldName, Reader reader) + { + if (indexingConfig != null) + { + Analyzer propertyAnalyzer = indexingConfig.getPropertyAnalyzer(fieldName); + if (propertyAnalyzer != null) + { + return propertyAnalyzer.tokenStream(fieldName, reader); + } + } + return defaultAnalyzer.tokenStream(fieldName, reader); + } } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/LuceneQueryHits.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/LuceneQueryHits.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/LuceneQueryHits.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -17,6 +17,7 @@ package org.exoplatform.services.jcr.impl.core.query.lucene; import org.apache.lucene.index.IndexReader; +import org.apache.lucene.search.DocIdSetIterator; import org.apache.lucene.search.IndexSearcher; import org.apache.lucene.search.Query; import org.apache.lucene.search.Scorer; @@ -73,11 +74,15 @@ */ public ScoreNode nextScoreNode() throws IOException { - if (!scorer.next()) + if (scorer == null) { return null; } - int doc = scorer.doc(); + int doc = scorer.nextDoc(); + if (doc == DocIdSetIterator.NO_MORE_DOCS) + { + return null; + } String uuid = reader.document(doc).get(FieldNames.UUID); return new ScoreNode(uuid, scorer.score(), doc); } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/MatchAllScorer.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/MatchAllScorer.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/MatchAllScorer.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -34,147 +34,174 @@ * The MatchAllScorer implements a Scorer that scores / collects all * documents in the index that match a field. */ -class MatchAllScorer extends Scorer { +class MatchAllScorer extends Scorer +{ - /** - * next doc number - */ - private int nextDoc = -1; + /** + * next doc number + */ + private int nextDoc = -1; - /** - * IndexReader giving access to index - */ - private IndexReader reader; + /** + * IndexReader giving access to index + */ + private IndexReader reader; - /** - * The field to match - */ - private String field; + /** + * The field to match + */ + private String field; - /** - * BitSet filtering documents without content is specified field - */ - private BitSet docFilter; + /** + * BitSet filtering documents without content is specified field + */ + private BitSet docFilter; - /** - * Explanation object. the same for all docs - */ - private final Explanation matchExpl; + /** + * Explanation object. the same for all docs + */ + private final Explanation matchExpl; - /** - * Creates a new MatchAllScorer. - * - * @param reader the IndexReader - * @param field the field name to match. - * @throws IOException if an error occurs while collecting hits. - * e.g. while reading from the search index. - */ - MatchAllScorer(IndexReader reader, String field) - throws IOException { - super(Similarity.getDefault()); - this.reader = reader; - this.field = field; - matchExpl - = new Explanation(Similarity.getDefault().idf(reader.maxDoc(), - reader.maxDoc()), - "matchAll"); - calculateDocFilter(); - } + /** + * Creates a new MatchAllScorer. + * + * @param reader the IndexReader + * @param field the field name to match. + * @throws IOException if an error occurs while collecting hits. + * e.g. while reading from the search index. + */ + MatchAllScorer(IndexReader reader, String field) throws IOException + { + super(Similarity.getDefault()); + this.reader = reader; + this.field = field; + matchExpl = new Explanation(Similarity.getDefault().idf(reader.maxDoc(), reader.maxDoc()), "matchAll"); + calculateDocFilter(); + } - /** - * {@inheritDoc} - */ - public void score(Collector hc) throws IOException { - while (next()) { - hc.collect(doc()); - } - } + /** + * {@inheritDoc} + */ + @Override + public void score(Collector collector) throws IOException + { + collector.setScorer(this); // TODO: WTF? missing in exo but present in jcrbt + while (nextDoc() != NO_MORE_DOCS) + { + collector.collect(docID()); + } + } - /** - * {@inheritDoc} - */ - public boolean next() throws IOException { - nextDoc = docFilter.nextSetBit(nextDoc + 1); - return nextDoc > -1; - } + /** + * {@inheritDoc} + */ + @Override + public int nextDoc() throws IOException + { + if (nextDoc == NO_MORE_DOCS) + { + return nextDoc; + } - /** - * {@inheritDoc} - */ - public int doc() { - return nextDoc; - } + nextDoc = docFilter.nextSetBit(nextDoc + 1); + if (nextDoc < 0) + { + nextDoc = NO_MORE_DOCS; + } + return nextDoc; + } - /** - * {@inheritDoc} - */ - public float score() throws IOException { - return 1.0f; - } + /** + * {@inheritDoc} + */ + @Override + public int docID() + { + return nextDoc; + } - /** - * {@inheritDoc} - */ - public boolean skipTo(int target) throws IOException { - nextDoc = target - 1; - return next(); - } + /** + * {@inheritDoc} + */ + @Override + public float score() throws IOException + { + return 1.0f; + } - /** - * {@inheritDoc} - */ - public Explanation explain(int doc) { - return matchExpl; - } + /** + * {@inheritDoc} + */ + @Override + public int advance(int target) throws IOException + { + if (nextDoc == NO_MORE_DOCS) + { + return nextDoc; + } - /** - * Calculates a BitSet filter that includes all the nodes - * that have content in properties according to the field name - * passed in the constructor of this MatchAllScorer. - * - * @throws IOException if an error occurs while reading from - * the search index. - */ - private void calculateDocFilter() throws IOException { - PerQueryCache cache = PerQueryCache.getInstance(); - Map readerCache = (Map) cache.get(MatchAllScorer.class, reader); - if (readerCache == null) { - readerCache = new HashMap(); - cache.put(MatchAllScorer.class, reader, readerCache); - } - // get BitSet for field - docFilter = (BitSet) readerCache.get(field); + nextDoc = target - 1; + return nextDoc(); + } - if (docFilter != null) { - // use cached BitSet; - return; - } + /** + * Calculates a BitSet filter that includes all the nodes + * that have content in properties according to the field name + * passed in the constructor of this MatchAllScorer. + * + * @throws IOException if an error occurs while reading from + * the search index. + */ + private void calculateDocFilter() throws IOException + { + PerQueryCache cache = PerQueryCache.getInstance(); + Map readerCache = (Map)cache.get(MatchAllScorer.class, reader); + if (readerCache == null) + { + readerCache = new HashMap(); + cache.put(MatchAllScorer.class, reader, readerCache); + } + // get BitSet for field + docFilter = (BitSet)readerCache.get(field); - // otherwise calculate new - docFilter = new BitSet(reader.maxDoc()); - // we match all terms - String namedValue = FieldNames.createNamedValue(field, ""); - TermEnum terms = reader.terms(new Term(FieldNames.PROPERTIES, namedValue)); - try { - TermDocs docs = reader.termDocs(); - try { - while (terms.term() != null - && terms.term().field() == FieldNames.PROPERTIES - && terms.term().text().startsWith(namedValue)) { - docs.seek(terms); - while (docs.next()) { - docFilter.set(docs.doc()); - } - terms.next(); - } - } finally { - docs.close(); + if (docFilter != null) + { + // use cached BitSet; + return; + } + + // otherwise calculate new + docFilter = new BitSet(reader.maxDoc()); + // we match all terms + String namedValue = FieldNames.createNamedValue(field, ""); + TermEnum terms = reader.terms(new Term(FieldNames.PROPERTIES, namedValue)); + try + { + TermDocs docs = reader.termDocs(); + try + { + while (terms.term() != null && terms.term().field() == FieldNames.PROPERTIES + && terms.term().text().startsWith(namedValue)) + { + docs.seek(terms); + while (docs.next()) + { + docFilter.set(docs.doc()); + } + terms.next(); } - } finally { - terms.close(); - } + } + finally + { + docs.close(); + } + } + finally + { + terms.close(); + } - // put BitSet into cache - readerCache.put(field, docFilter); - } + // put BitSet into cache + readerCache.put(field, docFilter); + } } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/MoreLikeThis.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/MoreLikeThis.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/MoreLikeThis.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -16,24 +16,10 @@ */ package org.exoplatform.services.jcr.impl.core.query.lucene; -import java.io.File; -import java.io.FileReader; -import java.io.IOException; -import java.io.InputStreamReader; -import java.io.Reader; -import java.io.StringReader; -import java.net.URL; -import java.util.ArrayList; -import java.util.Collection; -import java.util.HashMap; -import java.util.Iterator; -import java.util.Map; -import java.util.Set; - import org.apache.lucene.analysis.Analyzer; -import org.apache.lucene.analysis.Token; import org.apache.lucene.analysis.TokenStream; import org.apache.lucene.analysis.standard.StandardAnalyzer; +import org.apache.lucene.analysis.tokenattributes.TermAttribute; import org.apache.lucene.document.Document; import org.apache.lucene.index.IndexReader; import org.apache.lucene.index.Term; @@ -45,7 +31,21 @@ import org.apache.lucene.search.Similarity; import org.apache.lucene.search.TermQuery; import org.apache.lucene.util.PriorityQueue; +import org.apache.lucene.util.Version; +import java.io.File; +import java.io.FileReader; +import java.io.IOException; +import java.io.InputStreamReader; +import java.io.Reader; +import java.io.StringReader; +import java.net.URL; +import java.util.ArrayList; +import java.util.Collection; +import java.util.HashMap; +import java.util.Iterator; +import java.util.Map; +import java.util.Set; /** * Generate "more like this" similarity queries. @@ -140,740 +140,816 @@ * </pre> * */ -public final class MoreLikeThis { +public final class MoreLikeThis +{ - /** - * Default maximum number of tokens to parse in each example doc field that is not stored with TermVector support. - * @see #getMaxNumTokensParsed - */ - public static final int DEFAULT_MAX_NUM_TOKENS_PARSED = 5000; + /** + * Default maximum number of tokens to parse in each example doc field that is not stored with TermVector support. + * @see #getMaxNumTokensParsed + */ + public static final int DEFAULT_MAX_NUM_TOKENS_PARSED = 5000; - /** - * Default analyzer to parse source doc with. - * @see #getAnalyzer - */ - public static final Analyzer DEFAULT_ANALYZER = new StandardAnalyzer(); + /** + * Default analyzer to parse source doc with. + * @see #getAnalyzer + */ + public static final Analyzer DEFAULT_ANALYZER = new StandardAnalyzer(Version.LUCENE_24); - /** - * Ignore terms with less than this frequency in the source doc. - * @see #getMinTermFreq - * @see #setMinTermFreq - */ - public static final int DEFAULT_MIN_TERM_FREQ = 2; + /** + * Ignore terms with less than this frequency in the source doc. + * @see #getMinTermFreq + * @see #setMinTermFreq + */ + public static final int DEFAULT_MIN_TERM_FREQ = 2; - /** - * Ignore words which do not occur in at least this many docs. - * @see #getMinDocFreq - * @see #setMinDocFreq - */ - public static final int DEFAULT_MIN_DOC_FREQ = 5; + /** + * Ignore words which do not occur in at least this many docs. + * @see #getMinDocFreq + * @see #setMinDocFreq + */ + public static final int DEFAULT_MIN_DOC_FREQ = 5; - /** - * Boost terms in query based on score. - * @see #isBoost - * @see #setBoost - */ - public static final boolean DEFAULT_BOOST = false; + /** + * Boost terms in query based on score. + * @see #isBoost + * @see #setBoost + */ + public static final boolean DEFAULT_BOOST = false; - /** - * Default field names. Null is used to specify that the field names should be looked - * up at runtime from the provided reader. - */ - public static final String[] DEFAULT_FIELD_NAMES = new String[] { "contents"}; + /** + * Default field names. Null is used to specify that the field names should be looked + * up at runtime from the provided reader. + */ + public static final String[] DEFAULT_FIELD_NAMES = new String[]{"contents"}; - /** - * Ignore words less than this length or if 0 then this has no effect. - * @see #getMinWordLen - * @see #setMinWordLen - */ - public static final int DEFAULT_MIN_WORD_LENGTH = 0; + /** + * Ignore words less than this length or if 0 then this has no effect. + * @see #getMinWordLen + * @see #setMinWordLen + */ + public static final int DEFAULT_MIN_WORD_LENGTH = 0; - /** - * Ignore words greater than this length or if 0 then this has no effect. - * @see #getMaxWordLen - * @see #setMaxWordLen - */ - public static final int DEFAULT_MAX_WORD_LENGTH = 0; + /** + * Ignore words greater than this length or if 0 then this has no effect. + * @see #getMaxWordLen + * @see #setMaxWordLen + */ + public static final int DEFAULT_MAX_WORD_LENGTH = 0; - /** - * Default set of stopwords. - * If null means to allow stop words. - * - * @see #setStopWords - * @see #getStopWords - */ - public static final Set DEFAULT_STOP_WORDS = null; + /** + * Default set of stopwords. + * If null means to allow stop words. + * + * @see #setStopWords + * @see #getStopWords + */ + public static final Set DEFAULT_STOP_WORDS = null; - /** - * Current set of stop words. - */ - private Set stopWords = DEFAULT_STOP_WORDS; + /** + * Current set of stop words. + */ + private Set stopWords = DEFAULT_STOP_WORDS; - /** - * Return a Query with no more than this many terms. - * - * @see BooleanQuery#getMaxClauseCount - * @see #getMaxQueryTerms - * @see #setMaxQueryTerms - */ - public static final int DEFAULT_MAX_QUERY_TERMS = 25; + /** + * Return a Query with no more than this many terms. + * + * @see BooleanQuery#getMaxClauseCount + * @see #getMaxQueryTerms + * @see #setMaxQueryTerms + */ + public static final int DEFAULT_MAX_QUERY_TERMS = 25; - /** - * Analyzer that will be used to parse the doc. - */ - private Analyzer analyzer = DEFAULT_ANALYZER; + /** + * Analyzer that will be used to parse the doc. + */ + private Analyzer analyzer = DEFAULT_ANALYZER; - /** - * Ignore words less freqent that this. - */ - private int minTermFreq = DEFAULT_MIN_TERM_FREQ; + /** + * Ignore words less freqent that this. + */ + private int minTermFreq = DEFAULT_MIN_TERM_FREQ; - /** - * Ignore words which do not occur in at least this many docs. - */ - private int minDocFreq = DEFAULT_MIN_DOC_FREQ; + /** + * Ignore words which do not occur in at least this many docs. + */ + private int minDocFreq = DEFAULT_MIN_DOC_FREQ; - /** - * Should we apply a boost to the Query based on the scores? - */ - private boolean boost = DEFAULT_BOOST; + /** + * Should we apply a boost to the Query based on the scores? + */ + private boolean boost = DEFAULT_BOOST; - /** - * Field name we'll analyze. - */ - private String[] fieldNames = DEFAULT_FIELD_NAMES; + /** + * Field name we'll analyze. + */ + private String[] fieldNames = DEFAULT_FIELD_NAMES; - /** - * The maximum number of tokens to parse in each example doc field that is not stored with TermVector support - */ - private int maxNumTokensParsed = DEFAULT_MAX_NUM_TOKENS_PARSED; + /** + * The maximum number of tokens to parse in each example doc field that is not stored with TermVector support + */ + private int maxNumTokensParsed = DEFAULT_MAX_NUM_TOKENS_PARSED; - /** - * Ignore words if less than this len. - */ - private int minWordLen = DEFAULT_MIN_WORD_LENGTH; + /** + * Ignore words if less than this len. + */ + private int minWordLen = DEFAULT_MIN_WORD_LENGTH; - /** - * Ignore words if greater than this len. - */ - private int maxWordLen = DEFAULT_MAX_WORD_LENGTH; + /** + * Ignore words if greater than this len. + */ + private int maxWordLen = DEFAULT_MAX_WORD_LENGTH; - /** - * Don't return a query longer than this. - */ - private int maxQueryTerms = DEFAULT_MAX_QUERY_TERMS; + /** + * Don't return a query longer than this. + */ + private int maxQueryTerms = DEFAULT_MAX_QUERY_TERMS; - /** - * For idf() calculations. - */ - private Similarity similarity;// = new DefaultSimilarity(); + /** + * For idf() calculations. + */ + private Similarity similarity;// = new DefaultSimilarity(); - /** - * IndexReader to use - */ - private final IndexReader ir; + /** + * IndexReader to use + */ + private final IndexReader ir; - /** - * Constructor requiring an IndexReader. - */ - public MoreLikeThis(IndexReader ir) { - this(ir, new DefaultSimilarity()); - } + /** + * Constructor requiring an IndexReader. + */ + public MoreLikeThis(IndexReader ir) + { + this(ir, new DefaultSimilarity()); + } - public MoreLikeThis(IndexReader ir, Similarity sim){ + public MoreLikeThis(IndexReader ir, Similarity sim) + { this.ir = ir; this.similarity = sim; - } + } + public Similarity getSimilarity() + { + return similarity; + } - public Similarity getSimilarity() { - return similarity; - } + public void setSimilarity(Similarity similarity) + { + this.similarity = similarity; + } - public void setSimilarity(Similarity similarity) { - this.similarity = similarity; - } + /** + * Returns an analyzer that will be used to parse source doc with. The default analyzer + * is the {@link #DEFAULT_ANALYZER}. + * + * @return the analyzer that will be used to parse source doc with. + * @see #DEFAULT_ANALYZER + */ + public Analyzer getAnalyzer() + { + return analyzer; + } - /** - * Returns an analyzer that will be used to parse source doc with. The default analyzer - * is the {@link #DEFAULT_ANALYZER}. - * - * @return the analyzer that will be used to parse source doc with. - * @see #DEFAULT_ANALYZER - */ - public Analyzer getAnalyzer() { - return analyzer; - } + /** + * Sets the analyzer to use. An analyzer is not required for generating a query with the + * {@link #like(int)} method, all other 'like' methods require an analyzer. + * + * @param analyzer the analyzer to use to tokenize text. + */ + public void setAnalyzer(Analyzer analyzer) + { + this.analyzer = analyzer; + } - /** - * Sets the analyzer to use. An analyzer is not required for generating a query with the - * {@link #like(int)} method, all other 'like' methods require an analyzer. - * - * @param analyzer the analyzer to use to tokenize text. - */ - public void setAnalyzer(Analyzer analyzer) { - this.analyzer = analyzer; - } + /** + * Returns the frequency below which terms will be ignored in the source doc. The default + * frequency is the {@link #DEFAULT_MIN_TERM_FREQ}. + * + * @return the frequency below which terms will be ignored in the source doc. + */ + public int getMinTermFreq() + { + return minTermFreq; + } - /** - * Returns the frequency below which terms will be ignored in the source doc. The default - * frequency is the {@link #DEFAULT_MIN_TERM_FREQ}. - * - * @return the frequency below which terms will be ignored in the source doc. - */ - public int getMinTermFreq() { - return minTermFreq; - } + /** + * Sets the frequency below which terms will be ignored in the source doc. + * + * @param minTermFreq the frequency below which terms will be ignored in the source doc. + */ + public void setMinTermFreq(int minTermFreq) + { + this.minTermFreq = minTermFreq; + } - /** - * Sets the frequency below which terms will be ignored in the source doc. - * - * @param minTermFreq the frequency below which terms will be ignored in the source doc. - */ - public void setMinTermFreq(int minTermFreq) { - this.minTermFreq = minTermFreq; - } + /** + * Returns the frequency at which words will be ignored which do not occur in at least this + * many docs. The default frequency is {@link #DEFAULT_MIN_DOC_FREQ}. + * + * @return the frequency at which words will be ignored which do not occur in at least this + * many docs. + */ + public int getMinDocFreq() + { + return minDocFreq; + } - /** - * Returns the frequency at which words will be ignored which do not occur in at least this - * many docs. The default frequency is {@link #DEFAULT_MIN_DOC_FREQ}. - * - * @return the frequency at which words will be ignored which do not occur in at least this - * many docs. - */ - public int getMinDocFreq() { - return minDocFreq; - } + /** + * Sets the frequency at which words will be ignored which do not occur in at least this + * many docs. + * + * @param minDocFreq the frequency at which words will be ignored which do not occur in at + * least this many docs. + */ + public void setMinDocFreq(int minDocFreq) + { + this.minDocFreq = minDocFreq; + } - /** - * Sets the frequency at which words will be ignored which do not occur in at least this - * many docs. - * - * @param minDocFreq the frequency at which words will be ignored which do not occur in at - * least this many docs. - */ - public void setMinDocFreq(int minDocFreq) { - this.minDocFreq = minDocFreq; - } + /** + * Returns whether to boost terms in query based on "score" or not. The default is + * {@link #DEFAULT_BOOST}. + * + * @return whether to boost terms in query based on "score" or not. + * @see #setBoost + */ + public boolean isBoost() + { + return boost; + } - /** - * Returns whether to boost terms in query based on "score" or not. The default is - * {@link #DEFAULT_BOOST}. - * - * @return whether to boost terms in query based on "score" or not. - * @see #setBoost - */ - public boolean isBoost() { - return boost; - } + /** + * Sets whether to boost terms in query based on "score" or not. + * + * @param boost true to boost terms in query based on "score", false otherwise. + * @see #isBoost + */ + public void setBoost(boolean boost) + { + this.boost = boost; + } - /** - * Sets whether to boost terms in query based on "score" or not. - * - * @param boost true to boost terms in query based on "score", false otherwise. - * @see #isBoost - */ - public void setBoost(boolean boost) { - this.boost = boost; - } + /** + * Returns the field names that will be used when generating the 'More Like This' query. + * The default field names that will be used is {@link #DEFAULT_FIELD_NAMES}. + * + * @return the field names that will be used when generating the 'More Like This' query. + */ + public String[] getFieldNames() + { + return fieldNames; + } - /** - * Returns the field names that will be used when generating the 'More Like This' query. - * The default field names that will be used is {@link #DEFAULT_FIELD_NAMES}. - * - * @return the field names that will be used when generating the 'More Like This' query. - */ - public String[] getFieldNames() { - return fieldNames; - } + /** + * Sets the field names that will be used when generating the 'More Like This' query. + * Set this to null for the field names to be determined at runtime from the IndexReader + * provided in the constructor. + * + * @param fieldNames the field names that will be used when generating the 'More Like This' + * query. + */ + public void setFieldNames(String[] fieldNames) + { + this.fieldNames = fieldNames; + } - /** - * Sets the field names that will be used when generating the 'More Like This' query. - * Set this to null for the field names to be determined at runtime from the IndexReader - * provided in the constructor. - * - * @param fieldNames the field names that will be used when generating the 'More Like This' - * query. - */ - public void setFieldNames(String[] fieldNames) { - this.fieldNames = fieldNames; - } + /** + * Returns the minimum word length below which words will be ignored. Set this to 0 for no + * minimum word length. The default is {@link #DEFAULT_MIN_WORD_LENGTH}. + * + * @return the minimum word length below which words will be ignored. + */ + public int getMinWordLen() + { + return minWordLen; + } - /** - * Returns the minimum word length below which words will be ignored. Set this to 0 for no - * minimum word length. The default is {@link #DEFAULT_MIN_WORD_LENGTH}. - * - * @return the minimum word length below which words will be ignored. - */ - public int getMinWordLen() { - return minWordLen; - } + /** + * Sets the minimum word length below which words will be ignored. + * + * @param minWordLen the minimum word length below which words will be ignored. + */ + public void setMinWordLen(int minWordLen) + { + this.minWordLen = minWordLen; + } - /** - * Sets the minimum word length below which words will be ignored. - * - * @param minWordLen the minimum word length below which words will be ignored. - */ - public void setMinWordLen(int minWordLen) { - this.minWordLen = minWordLen; - } + /** + * Returns the maximum word length above which words will be ignored. Set this to 0 for no + * maximum word length. The default is {@link #DEFAULT_MAX_WORD_LENGTH}. + * + * @return the maximum word length above which words will be ignored. + */ + public int getMaxWordLen() + { + return maxWordLen; + } - /** - * Returns the maximum word length above which words will be ignored. Set this to 0 for no - * maximum word length. The default is {@link #DEFAULT_MAX_WORD_LENGTH}. - * - * @return the maximum word length above which words will be ignored. - */ - public int getMaxWordLen() { - return maxWordLen; - } + /** + * Sets the maximum word length above which words will be ignored. + * + * @param maxWordLen the maximum word length above which words will be ignored. + */ + public void setMaxWordLen(int maxWordLen) + { + this.maxWordLen = maxWordLen; + } - /** - * Sets the maximum word length above which words will be ignored. - * - * @param maxWordLen the maximum word length above which words will be ignored. - */ - public void setMaxWordLen(int maxWordLen) { - this.maxWordLen = maxWordLen; - } + /** + * Set the set of stopwords. + * Any word in this set is considered "uninteresting" and ignored. + * Even if your Analyzer allows stopwords, you might want to tell the MoreLikeThis code to ignore them, as + * for the purposes of document similarity it seems reasonable to assume that "a stop word is never interesting". + * + * @param stopWords set of stopwords, if null it means to allow stop words + * + * @see org.apache.lucene.analysis.StopFilter#makeStopSet StopFilter.makeStopSet() + * @see #getStopWords + */ + public void setStopWords(Set stopWords) + { + this.stopWords = stopWords; + } - /** - * Set the set of stopwords. - * Any word in this set is considered "uninteresting" and ignored. - * Even if your Analyzer allows stopwords, you might want to tell the MoreLikeThis code to ignore them, as - * for the purposes of document similarity it seems reasonable to assume that "a stop word is never interesting". - * - * @param stopWords set of stopwords, if null it means to allow stop words - * - * @see org.apache.lucene.analysis.StopFilter#makeStopSet StopFilter.makeStopSet() - * @see #getStopWords - */ - public void setStopWords(Set stopWords) { - this.stopWords = stopWords; - } + /** + * Get the current stop words being used. + * @see #setStopWords + */ + public Set getStopWords() + { + return stopWords; + } - /** - * Get the current stop words being used. - * @see #setStopWords - */ - public Set getStopWords() { - return stopWords; - } + /** + * Returns the maximum number of query terms that will be included in any generated query. + * The default is {@link #DEFAULT_MAX_QUERY_TERMS}. + * + * @return the maximum number of query terms that will be included in any generated query. + */ + public int getMaxQueryTerms() + { + return maxQueryTerms; + } - /** - * Returns the maximum number of query terms that will be included in any generated query. - * The default is {@link #DEFAULT_MAX_QUERY_TERMS}. - * - * @return the maximum number of query terms that will be included in any generated query. - */ - public int getMaxQueryTerms() { - return maxQueryTerms; - } + /** + * Sets the maximum number of query terms that will be included in any generated query. + * + * @param maxQueryTerms the maximum number of query terms that will be included in any + * generated query. + */ + public void setMaxQueryTerms(int maxQueryTerms) + { + this.maxQueryTerms = maxQueryTerms; + } - /** - * Sets the maximum number of query terms that will be included in any generated query. - * - * @param maxQueryTerms the maximum number of query terms that will be included in any - * generated query. - */ - public void setMaxQueryTerms(int maxQueryTerms) { - this.maxQueryTerms = maxQueryTerms; - } + /** + * @return The maximum number of tokens to parse in each example doc field that is not stored with TermVector support + * @see #DEFAULT_MAX_NUM_TOKENS_PARSED + */ + public int getMaxNumTokensParsed() + { + return maxNumTokensParsed; + } - /** - * @return The maximum number of tokens to parse in each example doc field that is not stored with TermVector support - * @see #DEFAULT_MAX_NUM_TOKENS_PARSED - */ - public int getMaxNumTokensParsed() { - return maxNumTokensParsed; - } + /** + * @param i The maximum number of tokens to parse in each example doc field that is not stored with TermVector support + */ + public void setMaxNumTokensParsed(int i) + { + maxNumTokensParsed = i; + } - /** - * @param i The maximum number of tokens to parse in each example doc field that is not stored with TermVector support - */ - public void setMaxNumTokensParsed(int i) { - maxNumTokensParsed = i; - } + /** + * Return a query that will return docs like the passed lucene document ID. + * + * @param docNum the documentID of the lucene doc to generate the 'More Like This" query for. + * @return a query that will return docs like the passed lucene document ID. + */ + public Query like(int docNum) throws IOException + { + if (fieldNames == null) + { + // gather list of valid fields from lucene + Collection fields = ir.getFieldNames(IndexReader.FieldOption.INDEXED); + fieldNames = (String[])fields.toArray(new String[fields.size()]); + } - /** - * Return a query that will return docs like the passed lucene document ID. - * - * @param docNum the documentID of the lucene doc to generate the 'More Like This" query for. - * @return a query that will return docs like the passed lucene document ID. - */ - public Query like(int docNum) throws IOException { - if (fieldNames == null) { - // gather list of valid fields from lucene - Collection fields = ir.getFieldNames( IndexReader.FieldOption.INDEXED); - fieldNames = (String[]) fields.toArray(new String[fields.size()]); - } + return createQuery(retrieveTerms(docNum)); + } - return createQuery(retrieveTerms(docNum)); - } + /** + * Return a query that will return docs like the passed file. + * + * @return a query that will return docs like the passed file. + */ + public Query like(File f) throws IOException + { + if (fieldNames == null) + { + // gather list of valid fields from lucene + Collection fields = ir.getFieldNames(IndexReader.FieldOption.INDEXED); + fieldNames = (String[])fields.toArray(new String[fields.size()]); + } - /** - * Return a query that will return docs like the passed file. - * - * @return a query that will return docs like the passed file. - */ - public Query like(File f) throws IOException { - if (fieldNames == null) { - // gather list of valid fields from lucene - Collection fields = ir.getFieldNames( IndexReader.FieldOption.INDEXED); - fieldNames = (String[]) fields.toArray(new String[fields.size()]); - } + return like(new FileReader(f)); + } - return like(new FileReader(f)); - } + /** + * Return a query that will return docs like the passed URL. + * + * @return a query that will return docs like the passed URL. + */ + public Query like(URL u) throws IOException + { + return like(new InputStreamReader(u.openConnection().getInputStream())); + } - /** - * Return a query that will return docs like the passed URL. - * - * @return a query that will return docs like the passed URL. - */ - public Query like(URL u) throws IOException { - return like(new InputStreamReader(u.openConnection().getInputStream())); - } + /** + * Return a query that will return docs like the passed stream. + * + * @return a query that will return docs like the passed stream. + */ + public Query like(java.io.InputStream is) throws IOException + { + return like(new InputStreamReader(is)); + } - /** - * Return a query that will return docs like the passed stream. - * - * @return a query that will return docs like the passed stream. - */ - public Query like(java.io.InputStream is) throws IOException { - return like(new InputStreamReader(is)); - } + /** + * Return a query that will return docs like the passed Reader. + * + * @return a query that will return docs like the passed Reader. + */ + public Query like(Reader r) throws IOException + { + return createQuery(retrieveTerms(r)); + } - /** - * Return a query that will return docs like the passed Reader. - * - * @return a query that will return docs like the passed Reader. - */ - public Query like(Reader r) throws IOException { - return createQuery(retrieveTerms(r)); - } + /** + * Create the More like query from a PriorityQueue + */ + private Query createQuery(PriorityQueue q) + { + BooleanQuery query = new BooleanQuery(); + Object cur; + int qterms = 0; + float bestScore = 0; - /** - * Create the More like query from a PriorityQueue - */ - private Query createQuery(PriorityQueue q) { - BooleanQuery query = new BooleanQuery(); - Object cur; - int qterms = 0; - float bestScore = 0; + while (((cur = q.pop()) != null)) + { + Object[] ar = (Object[])cur; + TermQuery tq = new JcrTermQuery(new Term((String)ar[1], (String)ar[0])); - while (((cur = q.pop()) != null)) { - Object[] ar = (Object[]) cur; - TermQuery tq = new JcrTermQuery(new Term((String) ar[1], (String) ar[0])); + if (boost) + { + if (qterms == 0) + { + bestScore = ((Float)ar[2]).floatValue(); + } + float myScore = ((Float)ar[2]).floatValue(); - if (boost) { - if (qterms == 0) { - bestScore = ((Float) ar[2]).floatValue(); - } - float myScore = ((Float) ar[2]).floatValue(); + tq.setBoost(myScore / bestScore); + } - tq.setBoost(myScore / bestScore); - } + try + { + query.add(tq, BooleanClause.Occur.SHOULD); + } + catch (BooleanQuery.TooManyClauses ignore) + { + break; + } - try { - query.add(tq, BooleanClause.Occur.SHOULD); - } - catch (BooleanQuery.TooManyClauses ignore) { - break; - } + qterms++; + if (maxQueryTerms > 0 && qterms >= maxQueryTerms) + { + break; + } + } - qterms++; - if (maxQueryTerms > 0 && qterms >= maxQueryTerms) { - break; - } - } + return query; + } - return query; - } + /** + * Create a PriorityQueue from a word->tf map. + * + * @param words a map of words keyed on the word(String) with Int objects as the values. + */ + private PriorityQueue createQueue(Map words) throws IOException + { + // have collected all words in doc and their freqs + int numDocs = ir.numDocs(); + FreqQ res = new FreqQ(words.size()); // will order words by score - /** - * Create a PriorityQueue from a word->tf map. - * - * @param words a map of words keyed on the word(String) with Int objects as the values. - */ - private PriorityQueue createQueue(Map words) throws IOException { - // have collected all words in doc and their freqs - int numDocs = ir.numDocs(); - FreqQ res = new FreqQ(words.size()); // will order words by score + Iterator it = words.keySet().iterator(); + while (it.hasNext()) + { // for every word + String word = (String)it.next(); - Iterator it = words.keySet().iterator(); - while (it.hasNext()) { // for every word - String word = (String) it.next(); + int tf = ((Int)words.get(word)).x; // term freq in the source doc + if (minTermFreq > 0 && tf < minTermFreq) + { + continue; // filter out words that don't occur enough times in the source + } - int tf = ((Int) words.get(word)).x; // term freq in the source doc - if (minTermFreq > 0 && tf < minTermFreq) { - continue; // filter out words that don't occur enough times in the source - } + // go through all the fields and find the largest document frequency + String topField = fieldNames[0]; + int docFreq = 0; + for (int i = 0; i < fieldNames.length; i++) + { + int freq = ir.docFreq(new Term(fieldNames[i], word)); + topField = (freq > docFreq) ? fieldNames[i] : topField; //NOSONAR + docFreq = (freq > docFreq) ? freq : docFreq; + } - // go through all the fields and find the largest document frequency - String topField = fieldNames[0]; - int docFreq = 0; - for (int i = 0; i < fieldNames.length; i++) - { - int freq = ir.docFreq(new Term(fieldNames[i], word)); - topField = (freq > docFreq) ? fieldNames[i] : topField; //NOSONAR - docFreq = (freq > docFreq) ? freq : docFreq; - } + if (minDocFreq > 0 && docFreq < minDocFreq) + { + continue; // filter out words that don't occur in enough docs + } - if (minDocFreq > 0 && docFreq < minDocFreq) { - continue; // filter out words that don't occur in enough docs - } + if (docFreq == 0) + { + continue; // index update problem? + } - if (docFreq == 0) { - continue; // index update problem? - } + float idf = similarity.idf(docFreq, numDocs); + float score = tf * idf; - float idf = similarity.idf(docFreq, numDocs); - float score = tf * idf; + // only really need 1st 3 entries, other ones are for troubleshooting + res.insertWithOverflow(new Object[]{word, // the word + topField, // the top field + new Float(score), // overall score + new Float(idf), // idf + new Integer(docFreq), // freq in all docs + new Integer(tf)}); + } + return res; + } - // only really need 1st 3 entries, other ones are for troubleshooting - res.insert(new Object[]{word, // the word - topField, // the top field - new Float(score), // overall score - new Float(idf), // idf - new Integer(docFreq), // freq in all docs - new Integer(tf) - }); - } - return res; - } + /** + * Describe the parameters that control how the "more like this" query is formed. + */ + public String describeParams() + { + StringBuffer sb = new StringBuffer(); + sb.append("\t" + "maxQueryTerms : " + maxQueryTerms + "\n"); + sb.append("\t" + "minWordLen : " + minWordLen + "\n"); + sb.append("\t" + "maxWordLen : " + maxWordLen + "\n"); + sb.append("\t" + "fieldNames : "); + String delim = ""; + for (int i = 0; i < fieldNames.length; i++) + { + String fieldName = fieldNames[i]; + sb.append(delim).append(fieldName); + delim = ", "; + } + sb.append("\n"); + sb.append("\t" + "boost : " + boost + "\n"); + sb.append("\t" + "minTermFreq : " + minTermFreq + "\n"); + sb.append("\t" + "minDocFreq : " + minDocFreq + "\n"); + return sb.toString(); + } - /** - * Describe the parameters that control how the "more like this" query is formed. - */ - public String describeParams() { - StringBuffer sb = new StringBuffer(); - sb.append("\t" + "maxQueryTerms : " + maxQueryTerms + "\n"); - sb.append("\t" + "minWordLen : " + minWordLen + "\n"); - sb.append("\t" + "maxWordLen : " + maxWordLen + "\n"); - sb.append("\t" + "fieldNames : "); - String delim = ""; - for (int i = 0; i < fieldNames.length; i++) { - String fieldName = fieldNames[i]; - sb.append(delim).append(fieldName); - delim = ", "; - } - sb.append("\n"); - sb.append("\t" + "boost : " + boost + "\n"); - sb.append("\t" + "minTermFreq : " + minTermFreq + "\n"); - sb.append("\t" + "minDocFreq : " + minDocFreq + "\n"); - return sb.toString(); - } + /** + * Find words for a more-like-this query former. + * + * @param docNum the id of the lucene document from which to find terms + */ + public PriorityQueue retrieveTerms(int docNum) throws IOException + { + Map termFreqMap = new HashMap(); + for (int i = 0; i < fieldNames.length; i++) + { + String fieldName = fieldNames[i]; + TermFreqVector vector = ir.getTermFreqVector(docNum, fieldName); - /** - * Find words for a more-like-this query former. - * - * @param docNum the id of the lucene document from which to find terms - */ - public PriorityQueue retrieveTerms(int docNum) throws IOException { - Map termFreqMap = new HashMap(); - for (int i = 0; i < fieldNames.length; i++) { - String fieldName = fieldNames[i]; - TermFreqVector vector = ir.getTermFreqVector(docNum, fieldName); - - // field does not store term vector info - if (vector == null) { - Document d = ir.document(docNum); - String[] text = d.getValues(fieldName); - if (text != null) { - for (int j = 0; j < text.length; j++) { - addTermFrequencies(new StringReader(text[j]), termFreqMap, fieldName); - } - } + // field does not store term vector info + if (vector == null) + { + Document d = ir.document(docNum); + String[] text = d.getValues(fieldName); + if (text != null) + { + for (int j = 0; j < text.length; j++) + { + addTermFrequencies(new StringReader(text[j]), termFreqMap, fieldName); + } } - else { - addTermFrequencies(termFreqMap, vector); - } + } + else + { + addTermFrequencies(termFreqMap, vector); + } - } + } - return createQueue(termFreqMap); - } + return createQueue(termFreqMap); + } - /** - * Adds terms and frequencies found in vector into the Map termFreqMap - * @param termFreqMap a Map of terms and their frequencies - * @param vector List of terms and their frequencies for a doc/field - */ - private void addTermFrequencies(Map termFreqMap, TermFreqVector vector) { - String[] terms = vector.getTerms(); - int[] freqs = vector.getTermFrequencies(); - for (int j = 0; j < terms.length; j++) { - String term = terms[j]; + /** + * Adds terms and frequencies found in vector into the Map termFreqMap + * @param termFreqMap a Map of terms and their frequencies + * @param vector List of terms and their frequencies for a doc/field + */ + private void addTermFrequencies(Map termFreqMap, TermFreqVector vector) + { + String[] terms = vector.getTerms(); + int[] freqs = vector.getTermFrequencies(); + for (int j = 0; j < terms.length; j++) + { + String term = terms[j]; - if (isNoiseWord(term)) { - continue; - } - // increment frequency - Int cnt = (Int) termFreqMap.get(term); - if (cnt == null) { - cnt = new Int(); - termFreqMap.put(term, cnt); - cnt.x = freqs[j]; - } - else { - cnt.x += freqs[j]; - } - } - } + if (isNoiseWord(term)) + { + continue; + } + // increment frequency + Int cnt = (Int)termFreqMap.get(term); + if (cnt == null) + { + cnt = new Int(); + termFreqMap.put(term, cnt); + cnt.x = freqs[j]; + } + else + { + cnt.x += freqs[j]; + } + } + } - /** - * Adds term frequencies found by tokenizing text from reader into the Map words - * @param r a source of text to be tokenized - * @param termFreqMap a Map of terms and their frequencies - * @param fieldName Used by analyzer for any special per-field analysis - */ - private void addTermFrequencies(Reader r, Map termFreqMap, String fieldName) - throws IOException { - TokenStream ts = analyzer.tokenStream(fieldName, r); - int tokenCount = 0; - // for every token - final Token reusableToken = new Token(); - for (Token nextToken = ts.next(reusableToken); nextToken != null; nextToken = ts.next(reusableToken)) { - String word = nextToken.term(); - tokenCount++; - if (tokenCount > maxNumTokensParsed) { - break; - } - if (isNoiseWord(word)) { - continue; - } + /** + * Adds term frequencies found by tokenizing text from reader into the Map words + * @param r a source of text to be tokenized + * @param termFreqMap a Map of terms and their frequencies + * @param fieldName Used by analyzer for any special per-field analysis + */ + private void addTermFrequencies(Reader r, Map<String, Int> termFreqMap, String fieldName) throws IOException + { + TokenStream ts = analyzer.tokenStream(fieldName, r); + int tokenCount = 0; + // for every token + while (ts.incrementToken()) + { + TermAttribute term = (TermAttribute)ts.getAttribute(TermAttribute.class); + String word = term.term(); + tokenCount++; + if (tokenCount > maxNumTokensParsed) + { + break; + } + if (isNoiseWord(word)) + { + continue; + } - // increment frequency - Int cnt = (Int) termFreqMap.get(word); - if (cnt == null) { - termFreqMap.put(word, new Int()); - } else { - cnt.x++; - } - } - } + // increment frequency + Int cnt = termFreqMap.get(word); + if (cnt == null) + { + termFreqMap.put(word, new Int()); + } + else + { + cnt.x++; + } + } + ts.end(); + ts.close(); + } - /** determines if the passed term is likely to be of interest in "more like" comparisons - * - * @param term The word being considered - * @return true if should be ignored, false if should be used in further analysis - */ - private boolean isNoiseWord(String term) { - int len = term.length(); - if (minWordLen > 0 && len < minWordLen) { - return true; - } - if (maxWordLen > 0 && len > maxWordLen) { - return true; - } - if (stopWords != null && stopWords.contains( term)) { - return true; - } - return false; - } + /** determines if the passed term is likely to be of interest in "more like" comparisons + * + * @param term The word being considered + * @return true if should be ignored, false if should be used in further analysis + */ + private boolean isNoiseWord(String term) + { + int len = term.length(); + if (minWordLen > 0 && len < minWordLen) + { + return true; + } + if (maxWordLen > 0 && len > maxWordLen) + { + return true; + } + if (stopWords != null && stopWords.contains(term)) + { + return true; + } + return false; + } + /** + * Find words for a more-like-this query former. + * The result is a priority queue of arrays with one entry for <b>every word</b> in the document. + * Each array has 6 elements. + * The elements are: + * <ol> + * <li> The word (String) + * <li> The top field that this word comes from (String) + * <li> The score for this word (Float) + * <li> The IDF value (Float) + * <li> The frequency of this word in the index (Integer) + * <li> The frequency of this word in the source document (Integer) + * </ol> + * This is a somewhat "advanced" routine, and in general only the 1st entry in the array is of interest. + * This method is exposed so that you can identify the "interesting words" in a document. + * For an easier method to call see {@link #retrieveInterestingTerms retrieveInterestingTerms()}. + * + * @param r the reader that has the content of the document + * @return the most interesting words in the document ordered by score, with the highest scoring, or best entry, first + * + * @see #retrieveInterestingTerms + */ + public PriorityQueue retrieveTerms(Reader r) throws IOException + { + Map words = new HashMap(); + for (int i = 0; i < fieldNames.length; i++) + { + String fieldName = fieldNames[i]; + addTermFrequencies(r, words, fieldName); + } + return createQueue(words); + } - /** - * Find words for a more-like-this query former. - * The result is a priority queue of arrays with one entry for <b>every word</b> in the document. - * Each array has 6 elements. - * The elements are: - * <ol> - * <li> The word (String) - * <li> The top field that this word comes from (String) - * <li> The score for this word (Float) - * <li> The IDF value (Float) - * <li> The frequency of this word in the index (Integer) - * <li> The frequency of this word in the source document (Integer) - * </ol> - * This is a somewhat "advanced" routine, and in general only the 1st entry in the array is of interest. - * This method is exposed so that you can identify the "interesting words" in a document. - * For an easier method to call see {@link #retrieveInterestingTerms retrieveInterestingTerms()}. - * - * @param r the reader that has the content of the document - * @return the most interesting words in the document ordered by score, with the highest scoring, or best entry, first - * - * @see #retrieveInterestingTerms - */ - public PriorityQueue retrieveTerms(Reader r) throws IOException { - Map words = new HashMap(); - for (int i = 0; i < fieldNames.length; i++) { - String fieldName = fieldNames[i]; - addTermFrequencies(r, words, fieldName); - } - return createQueue(words); - } + /** + * @see #retrieveInterestingTerms(java.io.Reader) + */ + public String[] retrieveInterestingTerms(int docNum) throws IOException + { + ArrayList al = new ArrayList(maxQueryTerms); + PriorityQueue pq = retrieveTerms(docNum); + Object cur; + // have to be careful, retrieveTerms returns all words but that's probably not useful to our caller... + int lim = maxQueryTerms; + // we just want to return the top words + while (((cur = pq.pop()) != null) && lim-- > 0) + { + Object[] ar = (Object[])cur; + al.add(ar[0]); // the 1st entry is the interesting word + } + String[] res = new String[al.size()]; + return (String[])al.toArray(res); + } - /** - * @see #retrieveInterestingTerms(java.io.Reader) - */ - public String[] retrieveInterestingTerms(int docNum) throws IOException { - ArrayList al = new ArrayList(maxQueryTerms); - PriorityQueue pq = retrieveTerms(docNum); - Object cur; - // have to be careful, retrieveTerms returns all words but that's probably not useful to our caller... - int lim = maxQueryTerms; - // we just want to return the top words - while (((cur = pq.pop()) != null) && lim-- > 0) { - Object[] ar = (Object[]) cur; - al.add(ar[0]); // the 1st entry is the interesting word - } - String[] res = new String[al.size()]; - return (String[]) al.toArray(res); - } + /** + * Convenience routine to make it easy to return the most interesting words in a document. + * More advanced users will call {@link #retrieveTerms(java.io.Reader) retrieveTerms()} directly. + * @param r the source document + * @return the most interesting words in the document + * + * @see #retrieveTerms(java.io.Reader) + * @see #setMaxQueryTerms + */ + public String[] retrieveInterestingTerms(Reader r) throws IOException + { + ArrayList al = new ArrayList(maxQueryTerms); + PriorityQueue pq = retrieveTerms(r); + Object cur; + // have to be careful, retrieveTerms returns all words but that's probably not useful to our caller... + int lim = maxQueryTerms; + // we just want to return the top words + while (((cur = pq.pop()) != null) && lim-- > 0) + { + Object[] ar = (Object[])cur; + al.add(ar[0]); // the 1st entry is the interesting word + } + String[] res = new String[al.size()]; + return (String[])al.toArray(res); + } - /** - * Convenience routine to make it easy to return the most interesting words in a document. - * More advanced users will call {@link #retrieveTerms(java.io.Reader) retrieveTerms()} directly. - * @param r the source document - * @return the most interesting words in the document - * - * @see #retrieveTerms(java.io.Reader) - * @see #setMaxQueryTerms - */ - public String[] retrieveInterestingTerms(Reader r) throws IOException { - ArrayList al = new ArrayList(maxQueryTerms); - PriorityQueue pq = retrieveTerms(r); - Object cur; - // have to be careful, retrieveTerms returns all words but that's probably not useful to our caller... - int lim = maxQueryTerms; - // we just want to return the top words - while (((cur = pq.pop()) != null) && lim-- > 0) { - Object[] ar = (Object[]) cur; - al.add(ar[0]); // the 1st entry is the interesting word - } - String[] res = new String[al.size()]; - return (String[]) al.toArray(res); - } + /** + * PriorityQueue that orders words by score. + */ + private static class FreqQ extends PriorityQueue + { + FreqQ(int s) + { + initialize(s); + } - /** - * PriorityQueue that orders words by score. - */ - private static class FreqQ extends PriorityQueue { - FreqQ (int s) { - initialize(s); - } + protected boolean lessThan(Object a, Object b) + { + Object[] aa = (Object[])a; + Object[] bb = (Object[])b; + Float fa = (Float)aa[2]; + Float fb = (Float)bb[2]; + return fa.floatValue() > fb.floatValue(); + } + } - protected boolean lessThan(Object a, Object b) { - Object[] aa = (Object[]) a; - Object[] bb = (Object[]) b; - Float fa = (Float) aa[2]; - Float fb = (Float) bb[2]; - return fa.floatValue() > fb.floatValue(); - } - } + /** + * Use for frequencies and to avoid renewing Integers. + */ + private static class Int + { + int x; - /** - * Use for frequencies and to avoid renewing Integers. - */ - private static class Int { - int x; + Int() + { + x = 1; + } + } - Int() { - x = 1; - } - } - - } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/MultiScorer.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/MultiScorer.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/MultiScorer.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -16,138 +16,156 @@ */ package org.exoplatform.services.jcr.impl.core.query.lucene; -import java.io.IOException; - -import org.apache.lucene.search.Explanation; import org.apache.lucene.search.Scorer; import org.apache.lucene.search.Similarity; +import java.io.IOException; + /** * <code>MultiScorer</code> spans multiple Scorers and returns document numbers * and score values in the order as supplied to the constructor of this * <code>MultiScorer</code>. */ -class MultiScorer extends Scorer { +class MultiScorer extends Scorer +{ - /** - * The sub scorers. - */ - private final Scorer[] scorers; + /** + * The sub scorers. + */ + private final Scorer[] scorers; - /** - * The document start numbers of the sub scorers. - */ - private final int[] starts; + /** + * The document start numbers of the sub scorers. + */ + private final int[] starts; - /** - * Index of the current scorer. - */ - private int current = 0; + /** + * Index of the current scorer. + */ + private int currentScorer; - /** - * Indicates if there are more documents. - */ - private boolean hasNext = true; + /** + * The next document id to be returned + */ + private int currentDoc = -1; - /** - * Creates a new <code>MultiScorer</code> that spans multiple - * <code>scorers</code>. - * - * @param similarity the similarity implementation that should be use. - * @param scorers the sub scorers. - * @param starts the document number start for each sub scorer. - */ - MultiScorer(Similarity similarity, Scorer[] scorers, int[] starts) { - super(similarity); - this.scorers = scorers; - this.starts = starts; - } + /** + * Creates a new <code>MultiScorer</code> that spans multiple + * <code>scorers</code>. + * + * @param similarity the similarity implementation that should be use. + * @param scorers the sub scorers. + * @param starts the document number start for each sub scorer. + */ + MultiScorer(Similarity similarity, Scorer[] scorers, int[] starts) + { + super(similarity); + this.scorers = scorers; + this.starts = starts; + } - /** - * {@inheritDoc} - */ - public boolean next() throws IOException { - while (hasNext) { - if (scorers[current].next()) { - return true; - } else if (++current < scorers.length) { - // advance to next scorer - } else { - // no more scorers - hasNext = false; - } - } - return hasNext; - } + @Override + public int nextDoc() throws IOException + { + while (currentDoc != NO_MORE_DOCS) + { + if (scorers[currentScorer].nextDoc() != NO_MORE_DOCS) + { + currentDoc = scorers[currentScorer].docID() + starts[currentScorer]; + return currentDoc; + } + else if (++currentScorer < scorers.length) + { + // advance to next scorer + } + else + { + // no more scorers + currentDoc = NO_MORE_DOCS; + } + } - /** - * {@inheritDoc} - */ - public int doc() { - return scorers[current].doc() + starts[current]; - } + return currentDoc; + } - /** - * {@inheritDoc} - */ - public float score() throws IOException { - return scorers[current].score(); - } + @Override + public int docID() + { + return currentDoc; + } - /** - * {@inheritDoc} - */ - public boolean skipTo(int target) throws IOException { - current = scorerIndex(target); - if (scorers[current].skipTo(target - starts[current])) { - return true; - } else { - if (++current < scorers.length) { - // simply move to the next if there is any - return next(); - } else { - // no more document - hasNext = false; - return hasNext; - } - } - } + @Override + public float score() throws IOException + { + return scorers[currentScorer].score(); + } - /** - * {@inheritDoc} - */ - public Explanation explain(int doc) throws IOException { - int scorerIndex = scorerIndex(doc); - return scorers[scorerIndex].explain(doc - starts[scorerIndex]); - } + @Override + public int advance(int target) throws IOException + { + if (currentDoc == NO_MORE_DOCS) + { + return currentDoc; + } - //--------------------------< internal >------------------------------------ + currentScorer = scorerIndex(target); + if (scorers[currentScorer].advance(target - starts[currentScorer]) != NO_MORE_DOCS) + { + currentDoc = scorers[currentScorer].docID() + starts[currentScorer]; + return currentDoc; + } + else + { + if (++currentScorer < scorers.length) + { + // simply move to the next if there is any + currentDoc = nextDoc(); + return currentDoc; + } + else + { + // no more document + currentDoc = NO_MORE_DOCS; + return currentDoc; + } + } + } - /** - * Returns the scorer index for document <code>n</code>. - * Implementation copied from lucene MultiReader class. - * - * @param n document number. - * @return the scorer index. - */ - private int scorerIndex(int n) { - int lo = 0; // search starts array - int hi = scorers.length - 1; // for first element less + //--------------------------< internal >------------------------------------ - while (hi >= lo) { - int mid = (lo + hi) >> 1; - int midValue = starts[mid]; - if (n < midValue) { - hi = mid - 1; - } else if (n > midValue) { - lo = mid + 1; - } else { // found a match - while (mid + 1 < scorers.length && starts[mid + 1] == midValue) { - mid++; // scan to last match - } - return mid; + /** + * Returns the scorer index for document <code>n</code>. + * Implementation copied from lucene MultiReader class. + * + * @param n document number. + * @return the scorer index. + */ + private int scorerIndex(int n) + { + int lo = 0; // search starts array + int hi = scorers.length - 1; // for first element less + + while (hi >= lo) + { + int mid = (lo + hi) >> 1; + int midValue = starts[mid]; + if (n < midValue) + { + hi = mid - 1; + } + else if (n > midValue) + { + lo = mid + 1; + } + else + { // found a match + while (mid + 1 < scorers.length && starts[mid + 1] == midValue) + { + mid++; // scan to last match } - } - return hi; - } -} + return mid; + } + } + return hi; + } +} \ No newline at end of file Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/NodeIndexer.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/NodeIndexer.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/NodeIndexer.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -264,7 +264,7 @@ * * @param doc the lucene document. * @param name the name of the multi-value property. - * @throws RepositoryException + * @throws RepositoryException */ private void addMVPName(Document doc, InternalQName name) throws RepositoryException { @@ -712,7 +712,6 @@ doc.add(createFieldWithoutNorms(fieldName, pathString.toString(), PropertyType.PATH)); } - /** * Adds the string value to the document both as the named field and * optionally for full text indexing if <code>tokenized</code> is @@ -809,33 +808,8 @@ */ protected Field createFulltextField(String value, boolean store, boolean withOffsets) { - Field.TermVector tv; - if (withOffsets) - { - tv = Field.TermVector.WITH_OFFSETS; - } - else - { - tv = Field.TermVector.NO; - } - if (store) - { - // store field compressed if greater than 16k - Field.Store stored; - if (value.length() > 0x4000) - { - stored = Field.Store.COMPRESS; - } - else - { - stored = Field.Store.YES; - } - return new Field(FieldNames.FULLTEXT, value, stored, Field.Index.ANALYZED, tv); - } - else - { - return new Field(FieldNames.FULLTEXT, value, Field.Store.NO, Field.Index.ANALYZED, tv); - } + return new Field(FieldNames.FULLTEXT, value, store ? Field.Store.YES : Field.Store.NO, Field.Index.ANALYZED, + withOffsets ? Field.TermVector.WITH_OFFSETS : Field.TermVector.NO); } /** Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/NotQuery.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/NotQuery.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/NotQuery.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -210,14 +210,20 @@ * {@inheritDoc} */ @Override - public boolean next() throws IOException + public int nextDoc() throws IOException { + if (docNo == NO_MORE_DOCS) + { + return docNo; + } + if (docNo == -1) { // get first doc of context scorer - if (contextScorer.next()) + int docId = contextScorer.nextDoc(); + if (docId != NO_MORE_DOCS) { - contextNo = contextScorer.doc(); + contextNo = docId; } } // move to next candidate @@ -231,23 +237,21 @@ while (contextNo != -1 && contextNo == docNo) { docNo++; - if (contextScorer.next()) - { - contextNo = contextScorer.doc(); - } - else - { - contextNo = -1; - } + int docId = contextScorer.nextDoc(); + contextNo = docId == NO_MORE_DOCS ? -1 : docId; } - return docNo < reader.maxDoc(); + if (docNo >= reader.maxDoc()) + { + docNo = NO_MORE_DOCS; + } + return docNo; } /** * {@inheritDoc} */ @Override - public int doc() + public int docID() { return docNo; } @@ -265,30 +269,20 @@ * {@inheritDoc} */ @Override - public boolean skipTo(int target) throws IOException + public int advance(int target) throws IOException { + if (docNo == NO_MORE_DOCS) + { + return docNo; + } + if (contextNo != -1 && contextNo < target) { - if (contextScorer.skipTo(target)) - { - contextNo = contextScorer.doc(); - } - else - { - contextNo = -1; - } + int docId = contextScorer.advance(target); + contextNo = docId == NO_MORE_DOCS ? -1 : docId; } docNo = target - 1; - return next(); + return nextDoc(); } - - /** - * @throws UnsupportedOperationException always - */ - @Override - public Explanation explain(int doc) throws IOException - { - throw new UnsupportedOperationException(); - } } } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/ParentAxisQuery.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/ParentAxisQuery.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/ParentAxisQuery.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -256,13 +256,13 @@ /** * Map that contains the scores from matching documents from the context - * query. To save memory only scores that are not equal to 1.0f are put - * to this map. + * query. To save memory only scores that are not equal to the score + * value of the first match are put to this map. * <p/> * key=[Integer] id of selected document from context query<br> * value=[Float] score for that document */ - private final Map scores = new HashMap(); + private final Map<Integer, Float> scores = new HashMap<Integer, Float>(); /** * The next document id to return @@ -270,6 +270,11 @@ private int nextDoc = -1; /** + * The score of the first match. + */ + private Float firstScore; + + /** * Creates a new <code>ParentAxisScorer</code>. * * @param similarity the <code>Similarity</code> instance to use. @@ -286,63 +291,57 @@ this.hResolver = resolver; } - /** - * {@inheritDoc} - */ @Override - public boolean next() throws IOException + public int nextDoc() throws IOException { + if (nextDoc == NO_MORE_DOCS) + { + return nextDoc; + } + calculateParent(); nextDoc = hits.nextSetBit(nextDoc + 1); - return nextDoc > -1; + if (nextDoc < 0) + { + nextDoc = NO_MORE_DOCS; + } + return nextDoc; } - /** - * {@inheritDoc} - */ @Override - public int doc() + public int docID() { return nextDoc; } - /** - * {@inheritDoc} - */ @Override public float score() throws IOException { - Float score = (Float)scores.get(new Integer(nextDoc)); + Float score = scores.get(nextDoc); if (score == null) { - score = DEFAULT_SCORE; + score = firstScore; } - return score.floatValue(); + return score; } - /** - * {@inheritDoc} - */ @Override - public boolean skipTo(int target) throws IOException + public int advance(int target) throws IOException { + if (nextDoc == NO_MORE_DOCS) + { + return nextDoc; + } + calculateParent(); nextDoc = hits.nextSetBit(target); - return nextDoc > -1; + if (nextDoc < 0) + { + nextDoc = NO_MORE_DOCS; + } + return nextDoc; } - /** - * {@inheritDoc} - * - * @throws UnsupportedOperationException this implementation always - * throws an <code>UnsupportedOperationException</code>. - */ - @Override - public Explanation explain(int doc) throws IOException - { - throw new UnsupportedOperationException(); - } - private void calculateParent() throws IOException { if (hits == null) @@ -350,44 +349,54 @@ hits = new BitSet(reader.maxDoc()); final IOException[] ex = new IOException[1]; - contextScorer.score(new AbstractHitCollector() + if (contextScorer != null) { + contextScorer.score(new AbstractHitCollector() + { + private int[] docs = new int[1]; - private int[] docs = new int[1]; - - @Override - public void collect(int doc, float score) - { - try + @Override + protected void collect(int doc, float score) { - docs = hResolver.getParents(doc, docs); - if (docs.length == 1) + try { - // optimize single value - hits.set(docs[0]); - if (score != DEFAULT_SCORE.floatValue()) + docs = hResolver.getParents(doc, docs); + if (docs.length == 1) { - scores.put(new Integer(docs[0]), new Float(score)); + // optimize single value + hits.set(docs[0]); + if (firstScore == null) + { + firstScore = score; + } + else if (firstScore != score) + { + scores.put(doc, score); + } } - } - else - { - for (int i = 0; i < docs.length; i++) + else { - hits.set(docs[i]); - if (score != DEFAULT_SCORE.floatValue()) + for (int docNum : docs) { - scores.put(new Integer(docs[i]), new Float(score)); + hits.set(docNum); + if (firstScore == null) + { + firstScore = score; + } + else if (firstScore != score) + { + scores.put(doc, score); + } } } } + catch (IOException e) + { + ex[0] = e; + } } - catch (IOException e) - { - ex[0] = e; - } - } - }); + }); + } if (ex[0] != null) { Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/PersistentIndex.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/PersistentIndex.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/PersistentIndex.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -110,7 +110,7 @@ byte[] buffer = new byte[1024]; Directory dir = index.getDirectory(); Directory dest = getDirectory(); - String[] files = dir.list(); + String[] files = dir.listAll(); for (int i = 0; i < files.length; i++) { IndexInput in = dir.openInput(files[i]); Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/QueryHitsQuery.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/QueryHitsQuery.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/QueryHitsQuery.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -39,235 +39,264 @@ * <code>QueryHitsQuery</code> exposes a {@link QueryHits} implementation again * as a Lucene Query. */ -public class QueryHitsQuery extends Query implements JcrQuery{ +public class QueryHitsQuery extends Query implements JcrQuery +{ - /** - * The underlying query hits. - */ - private final QueryHits hits; + /** + * The underlying query hits. + */ + private final QueryHits hits; - /** - * Creates a new query based on {@link QueryHits}. - * - * @param hits the query hits. - */ - public QueryHitsQuery(QueryHits hits) { - this.hits = hits; - } + /** + * Creates a new query based on {@link QueryHits}. + * + * @param hits the query hits. + */ + public QueryHitsQuery(QueryHits hits) + { + this.hits = hits; + } - /** - * {@inheritDoc} - */ - @Override - public Weight createWeight(Searcher searcher) throws IOException { - return new QueryHitsQueryWeight(searcher.getSimilarity()); - } + /** + * {@inheritDoc} + */ + @Override + public Weight createWeight(Searcher searcher) throws IOException + { + return new QueryHitsQueryWeight(searcher.getSimilarity()); + } - /** - * {@inheritDoc} - */ - @Override - public String toString(String field) { - return "QueryHitsQuery"; - } + /** + * {@inheritDoc} + */ + @Override + public String toString(String field) + { + return "QueryHitsQuery"; + } - /** - * {@inheritDoc} - */ - @Override - public void extractTerms(Set terms) { - // no terms - } + /** + * {@inheritDoc} + */ + @Override + public void extractTerms(Set terms) + { + // no terms + } - //-----------------------< JackrabbitQuery >-------------------------------- + //-----------------------< JackrabbitQuery >-------------------------------- - /** - * {@inheritDoc} - */ - public QueryHits execute(JcrIndexSearcher searcher, - SessionImpl session, - Sort sort) throws IOException { - if (sort.getSort().length == 0) { - return hits; - } else { - return null; - } - } + /** + * {@inheritDoc} + */ + public QueryHits execute(JcrIndexSearcher searcher, SessionImpl session, Sort sort) throws IOException + { + if (sort.getSort().length == 0) + { + return hits; + } + else + { + return null; + } + } - //------------------------< QueryHitsQueryWeight >-------------------------- + //------------------------< QueryHitsQueryWeight >-------------------------- - /** - * The Weight implementation for this query. - */ - public class QueryHitsQueryWeight extends Weight { + /** + * The Weight implementation for this query. + */ + public class QueryHitsQueryWeight extends Weight + { - /** - * The similarity. - */ - private final Similarity similarity; + /** + * The similarity. + */ + private final Similarity similarity; - /** - * Creates a new weight with the given <code>similarity</code>. - * - * @param similarity the similarity. - */ - public QueryHitsQueryWeight(Similarity similarity) { - this.similarity = similarity; - } + /** + * Creates a new weight with the given <code>similarity</code>. + * + * @param similarity the similarity. + */ + public QueryHitsQueryWeight(Similarity similarity) + { + this.similarity = similarity; + } - /** - * {@inheritDoc} - */ - @Override - public Query getQuery() { - return QueryHitsQuery.this; - } + /** + * {@inheritDoc} + */ + @Override + public Query getQuery() + { + return QueryHitsQuery.this; + } - /** - * {@inheritDoc} - */ - @Override - public float getValue() { - return 1.0f; - } + /** + * {@inheritDoc} + */ + @Override + public float getValue() + { + return 1.0f; + } - /** - * {@inheritDoc} - */ - @Override - public float sumOfSquaredWeights() throws IOException { - return 1.0f; - } + /** + * {@inheritDoc} + */ + @Override + public float sumOfSquaredWeights() throws IOException + { + return 1.0f; + } - /** - * {@inheritDoc} - */ - @Override - public void normalize(float norm) { - } + /** + * {@inheritDoc} + */ + @Override + public void normalize(float norm) + { + } - /** - * {@inheritDoc} - */ - @Override - public Scorer scorer(IndexReader reader, boolean scoreDocsInOrder, boolean topScorer) throws IOException { - return new QueryHitsQueryScorer(reader, similarity); - } + /** + * {@inheritDoc} + */ + @Override + public Scorer scorer(IndexReader reader, boolean scoreDocsInOrder, boolean topScorer) throws IOException + { + return new QueryHitsQueryScorer(reader, similarity); + } - /** - * {@inheritDoc} - */ - @Override - public Explanation explain(IndexReader reader, int doc) throws IOException { - return new Explanation(); - } - } + /** + * {@inheritDoc} + */ + @Override + public Explanation explain(IndexReader reader, int doc) throws IOException + { + return new Explanation(); + } + } - //-------------------< QueryHitsQueryScorer >------------------------------- + //-------------------< QueryHitsQueryScorer >------------------------------- - /** - * the scorer implementation for this query. - */ - public class QueryHitsQueryScorer extends Scorer { + /** + * the scorer implementation for this query. + */ + public class QueryHitsQueryScorer extends Scorer + { - /** - * Iterator over <code>Integer</code> instances identifying the - * lucene documents. Document numbers are iterated in ascending order. - */ - private final Iterator docs; + /** + * Iterator over <code>Integer</code> instances identifying the + * lucene documents. Document numbers are iterated in ascending order. + */ + private final Iterator<Integer> docs; - /** - * Maps <code>Integer</code> document numbers to <code>Float</code> - * scores. - */ - private final Map scores = new HashMap(); + /** + * Maps <code>Integer</code> document numbers to <code>Float</code> + * scores. + */ + private final Map<Integer, Float> scores = new HashMap<Integer, Float>(); - /** - * The current document number. - */ - private Integer currentDoc = null; + /** + * The current document number. + */ + private Integer currentDoc = null; - /** - * Creates a new scorer. - * - * @param reader the index reader. - * @param similarity the similarity implementation. - * @throws IOException if an error occurs while reading from the index. - */ - protected QueryHitsQueryScorer(IndexReader reader, - Similarity similarity) - throws IOException { - super(similarity); - ScoreNode node; - Set sortedDocs = new TreeSet(); - try { - while ((node = hits.nextScoreNode()) != null) { - String uuid = node.getNodeId(); - Term id = new Term(FieldNames.UUID, uuid); - TermDocs tDocs = reader.termDocs(id); - try { - if (tDocs.next()) { - Integer doc = new Integer(tDocs.doc()); - sortedDocs.add(doc); - scores.put(doc, new Float(node.getScore())); - } - } finally { - tDocs.close(); - } - } - } finally { - hits.close(); + /** + * Creates a new scorer. + * + * @param reader the index reader. + * @param similarity the similarity implementation. + * @throws IOException if an error occurs while reading from the index. + */ + protected QueryHitsQueryScorer(IndexReader reader, Similarity similarity) throws IOException + { + super(similarity); + ScoreNode node; + Set<Integer> sortedDocs = new TreeSet(); + try + { + while ((node = hits.nextScoreNode()) != null) + { + String uuid = node.getNodeId(); + Term id = new Term(FieldNames.UUID, uuid); + TermDocs tDocs = reader.termDocs(id); + try + { + if (tDocs.next()) + { + Integer doc = new Integer(tDocs.doc()); + sortedDocs.add(doc); + scores.put(doc, new Float(node.getScore())); + } + } + finally + { + tDocs.close(); + } } - docs = sortedDocs.iterator(); - } + } + finally + { + hits.close(); + } + docs = sortedDocs.iterator(); + } - /** - * {@inheritDoc} - */ - @Override - public boolean next() throws IOException { - if (docs.hasNext()) { - currentDoc = (Integer) docs.next(); - return true; - } - return false; - } + /** + * {@inheritDoc} + */ + @Override + public int nextDoc() throws IOException + { + if (currentDoc == NO_MORE_DOCS) + { + return currentDoc; + } - /** - * {@inheritDoc} - */ - @Override - public int doc() { - return currentDoc.intValue(); - } + currentDoc = docs.hasNext() ? docs.next() : NO_MORE_DOCS; + return currentDoc; + } - /** - * {@inheritDoc} - */ - @Override - public float score() throws IOException { - return ((Float) scores.get(currentDoc)).floatValue(); - } + /** + * {@inheritDoc} + */ + @Override + public int docID() + { + return currentDoc == null ? -1 : currentDoc; + } - /** - * {@inheritDoc} - */ - @Override - public boolean skipTo(int target) throws IOException { - do { - if (!next()) { - return false; - } - } while (target > doc()); - return true; - } + /** + * {@inheritDoc} + */ + @Override + public float score() throws IOException + { + return scores.get(currentDoc).floatValue(); + } - /** - * {@inheritDoc} - */ - @Override - public Explanation explain(int doc) throws IOException { - return new Explanation(); - } - } + /** + * {@inheritDoc} + */ + @Override + public int advance(int target) throws IOException + { + if (currentDoc == NO_MORE_DOCS) + { + return currentDoc; + } + + do + { + if (nextDoc() == NO_MORE_DOCS) + { + return NO_MORE_DOCS; + } + } + while (target > docID()); + return docID(); + } + } } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/QueryImpl.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/QueryImpl.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/QueryImpl.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -117,9 +117,9 @@ // build lucene query Query query = - LuceneQueryBuilder.createQuery(root, session, index.getContext().getItemStateManager(), index - .getNamespaceMappings(), index.getTextAnalyzer(), propReg, index.getSynonymProvider(), index - .getIndexFormatVersion(), index.getContext().getVirtualTableResolver()); + LuceneQueryBuilder.createQuery(root, session, index.getContext().getItemStateManager(), + index.getNamespaceMappings(), index.getTextAnalyzer(), propReg, index.getSynonymProvider(), + index.getIndexFormatVersion(), index.getContext().getVirtualTableResolver()); OrderQueryNode orderNode = root.getOrderNode(); @@ -142,7 +142,7 @@ return new SingleColumnQueryResult(index, itemMgr, session, session.getAccessManager(), this, query, new SpellSuggestion(index.getSpellChecker(), root), getSelectProperties(), orderProperties, ascSpecs, - getRespectDocumentOrder(), offset, limit); + orderProperties.length == 0 && getRespectDocumentOrder(), offset, limit); } /** Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/RangeQuery.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/RangeQuery.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/RangeQuery.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -20,19 +20,19 @@ import org.apache.lucene.index.Term; import org.apache.lucene.index.TermDocs; import org.apache.lucene.index.TermEnum; -import org.apache.lucene.search.ConstantScoreRangeQuery; +import org.apache.lucene.search.BooleanQuery; import org.apache.lucene.search.Explanation; import org.apache.lucene.search.Query; import org.apache.lucene.search.Scorer; import org.apache.lucene.search.Searcher; import org.apache.lucene.search.Similarity; +import org.apache.lucene.search.TermRangeQuery; import org.apache.lucene.search.Weight; import java.io.IOException; import java.util.ArrayList; import java.util.BitSet; import java.util.HashMap; -import java.util.Iterator; import java.util.List; import java.util.Map; import java.util.Set; @@ -40,464 +40,541 @@ /** * Implements a lucene range query. */ -public class RangeQuery extends Query implements Transformable { +public class RangeQuery extends Query implements Transformable +{ - /** - * The lower term. May be <code>null</code> if <code>upperTerm</code> is not - * <code>null</code>. - */ - private Term lowerTerm; + /** + * The lower term. May be <code>null</code> if <code>upperTerm</code> is not + * <code>null</code>. + */ + private Term lowerTerm; - /** - * The upper term. May be <code>null</code> if <code>lowerTerm</code> is not - * <code>null</code>. - */ - private Term upperTerm; + /** + * The upper term. May be <code>null</code> if <code>lowerTerm</code> is not + * <code>null</code>. + */ + private Term upperTerm; - /** - * If <code>true</code> the range interval is inclusive. - */ - private boolean inclusive; + /** + * If <code>true</code> the range interval is inclusive. + */ + private boolean inclusive; - /** - * How the term enum is transformed before it is compared to lower and upper - * term. - */ - private int transform = TRANSFORM_NONE; + /** + * How the term enum is transformed before it is compared to lower and upper + * term. + */ + private int transform = TRANSFORM_NONE; - /** - * Creates a new RangeQuery. The lower or the upper term may be - * <code>null</code>, but not both! - * - * @param lowerTerm the lower term of the interval, or <code>null</code> - * @param upperTerm the upper term of the interval, or <code>null</code>. - * @param inclusive if <code>true</code> the interval is inclusive. - */ - public RangeQuery(Term lowerTerm, Term upperTerm, boolean inclusive) { - this(lowerTerm, upperTerm, inclusive, TRANSFORM_NONE); - } + /** + * The rewritten range query or <code>null</code> if the range spans more + * than {@link org.apache.lucene.search.BooleanQuery#maxClauseCount} terms. + */ + private Query stdRangeQuery; - /** - * Creates a new RangeQuery. The lower or the upper term may be - * <code>null</code>, but not both! - * - * @param lowerTerm the lower term of the interval, or <code>null</code> - * @param upperTerm the upper term of the interval, or <code>null</code>. - * @param inclusive if <code>true</code> the interval is inclusive. - * @param transform how term enums are transformed when read from the index. - */ - public RangeQuery(Term lowerTerm, Term upperTerm, boolean inclusive, int transform) { - if (lowerTerm == null && upperTerm == null) { - throw new IllegalArgumentException("At least one term must be non-null"); - } - if (lowerTerm != null && upperTerm != null && lowerTerm.field() != upperTerm.field()) { - throw new IllegalArgumentException("Both terms must be for the same field"); - } + /** + * Creates a new RangeQuery. The lower or the upper term may be + * <code>null</code>, but not both! + * + * @param lowerTerm the lower term of the interval, or <code>null</code> + * @param upperTerm the upper term of the interval, or <code>null</code>. + * @param inclusive if <code>true</code> the interval is inclusive. + */ + public RangeQuery(Term lowerTerm, Term upperTerm, boolean inclusive) + { + this(lowerTerm, upperTerm, inclusive, TRANSFORM_NONE); + } - // if we have a lowerTerm, start there. otherwise, start at beginning - if (lowerTerm != null) { - this.lowerTerm = lowerTerm; - } else { - this.lowerTerm = new Term(upperTerm.field(), ""); - } + /** + * Creates a new RangeQuery. The lower or the upper term may be + * <code>null</code>, but not both! + * + * @param lowerTerm the lower term of the interval, or <code>null</code> + * @param upperTerm the upper term of the interval, or <code>null</code>. + * @param inclusive if <code>true</code> the interval is inclusive. + * @param transform how term enums are transformed when read from the index. + */ + public RangeQuery(Term lowerTerm, Term upperTerm, boolean inclusive, int transform) + { + if (lowerTerm == null && upperTerm == null) + { + throw new IllegalArgumentException("At least one term must be non-null"); + } + if (lowerTerm != null && upperTerm != null && lowerTerm.field() != upperTerm.field()) + { + throw new IllegalArgumentException("Both terms must be for the same field"); + } - this.upperTerm = upperTerm; - this.inclusive = inclusive; - this.transform = transform; - } + // if we have a lowerTerm, start there. otherwise, start at beginning + if (lowerTerm != null) + { + this.lowerTerm = lowerTerm; + } + else + { + this.lowerTerm = new Term(upperTerm.field(), ""); + } - /** - * {@inheritDoc} - */ - public void setTransformation(int transformation) { - this.transform = transformation; - } + this.upperTerm = upperTerm; + this.inclusive = inclusive; + this.transform = transform; + } - /** - * Rewrites this query into a {@link ConstantScoreRangeQuery} if - * {@link #transform} is {@link #TRANSFORM_NONE}. - * - * @param reader the index reader. - * @return the rewritten query or this query if rewriting is not possible. - * @throws IOException if an error occurs. - */ - @Override - public Query rewrite(IndexReader reader) throws IOException { - if (transform == TRANSFORM_NONE) { - return new ConstantScoreRangeQuery(lowerTerm.field(), - lowerTerm.text(), upperTerm.text(), inclusive, - inclusive).rewrite(reader); - } else { - // always use our implementation when we need to transform the - // term enum + /** + * {@inheritDoc} + */ + public void setTransformation(int transformation) + { + this.transform = transformation; + } + + /** + * Tries to rewrite this query into a standard lucene RangeQuery. + * This rewrite might fail with a TooManyClauses exception. If that + * happens, we use our own implementation. + * + * @param reader the index reader. + * @return the rewritten query or this query if rewriting is not possible. + * @throws IOException if an error occurs. + */ + public Query rewrite(IndexReader reader) throws IOException + { + if (transform == TRANSFORM_NONE) + { + Query stdRangeQueryImpl = + new TermRangeQuery(lowerTerm.field(), lowerTerm.text(), upperTerm.text(), inclusive, inclusive); + try + { + stdRangeQuery = stdRangeQueryImpl.rewrite(reader); + return stdRangeQuery; + } + catch (BooleanQuery.TooManyClauses e) + { + // failed, use own implementation return this; - } - } + } + } + else + { + // always use our implementation when we need to transform the + // term enum + return this; + } + } - /** - * Creates the <code>Weight</code> for this query. - * - * @param searcher the searcher to use for the <code>Weight</code>. - * @return the <code>Weigth</code> for this query. - */ - @Override - public Weight createWeight(Searcher searcher) { - return new RangeQueryWeight(searcher); - } + /** + * Creates the <code>Weight</code> for this query. + * + * @param searcher the searcher to use for the <code>Weight</code>. + * @return the <code>Weigth</code> for this query. + */ + public Weight createWeight(Searcher searcher) + { + return new RangeQueryWeight(searcher); + } - /** - * Returns a string representation of this query. - * @param field the field name for which to create a string representation. - * @return a string representation of this query. - */ - @Override - public String toString(String field) { - StringBuffer buffer = new StringBuffer(); - if (!getField().equals(field)) { - buffer.append(getField()); - buffer.append(":"); - } - buffer.append(inclusive ? "[" : "{"); - buffer.append(lowerTerm != null ? lowerTerm.text() : "null"); - buffer.append(" TO "); - buffer.append(upperTerm != null ? upperTerm.text() : "null"); - buffer.append(inclusive ? "]" : "}"); - if (getBoost() != 1.0f) { - buffer.append("^"); - buffer.append(Float.toString(getBoost())); - } - return buffer.toString(); - } + /** + * Returns a string representation of this query. + * @param field the field name for which to create a string representation. + * @return a string representation of this query. + */ + public String toString(String field) + { + StringBuffer buffer = new StringBuffer(); + if (!getField().equals(field)) + { + buffer.append(getField()); + buffer.append(":"); + } + buffer.append(inclusive ? "[" : "{"); + buffer.append(lowerTerm != null ? lowerTerm.text() : "null"); + buffer.append(" TO "); + buffer.append(upperTerm != null ? upperTerm.text() : "null"); + buffer.append(inclusive ? "]" : "}"); + if (getBoost() != 1.0f) + { + buffer.append("^"); + buffer.append(Float.toString(getBoost())); + } + return buffer.toString(); + } - /** - * {@inheritDoc} - */ - @Override - public void extractTerms(Set terms) { - // cannot extract terms - } + /** + * {@inheritDoc} + */ + public void extractTerms(Set terms) + { + if (stdRangeQuery != null) + { + stdRangeQuery.extractTerms(terms); + } + } - /** - * Returns the field name for this query. - */ - private String getField() { - return (lowerTerm != null ? lowerTerm.field() : upperTerm.field()); - } + /** + * Returns the field name for this query. + */ + private String getField() + { + return (lowerTerm != null ? lowerTerm.field() : upperTerm.field()); + } - //--------------------------< RangeQueryWeight >---------------------------- + //--------------------------< RangeQueryWeight >---------------------------- - /** - * The <code>Weight</code> implementation for this <code>RangeQuery</code>. - */ - private class RangeQueryWeight extends AbstractWeight { + /** + * The <code>Weight</code> implementation for this <code>RangeQuery</code>. + */ + private class RangeQueryWeight extends AbstractWeight + { - /** - * Creates a new <code>RangeQueryWeight</code> instance using - * <code>searcher</code>. - * - * @param searcher a <code>Searcher</code> instance. - */ - RangeQueryWeight(Searcher searcher) { - super(searcher); - } + /** + * Creates a new <code>RangeQueryWeight</code> instance using + * <code>searcher</code>. + * + * @param searcher a <code>Searcher</code> instance. + */ + RangeQueryWeight(Searcher searcher) + { + super(searcher); + } - /** - * Creates a {@link RangeQueryScorer} instance. - * - * @param reader index reader - * @return a {@link RangeQueryScorer} instance - */ - @Override - protected Scorer createScorer(IndexReader reader, boolean scoreDocsInOrder, boolean topScorer) { - return new RangeQueryScorer(searcher.getSimilarity(), reader); - }; + /** + * Creates a {@link RangeQueryScorer} instance. + * + * @param reader index reader + * @return a {@link RangeQueryScorer} instance + */ + @Override + protected Scorer createScorer(IndexReader reader, boolean scoreDocsInOrder, boolean topScorer) + { + return new RangeQueryScorer(searcher.getSimilarity(), reader); + }; - /** - * Returns this <code>RangeQuery</code>. - * - * @return this <code>RangeQuery</code>. - */ - @Override - public Query getQuery() { - return RangeQuery.this; - } + /** + * Returns this <code>RangeQuery</code>. + * + * @return this <code>RangeQuery</code>. + */ + @Override + public Query getQuery() + { + return RangeQuery.this; + } - /** - * {@inheritDoc} - */ - @Override - public float getValue() { - return 1.0f; - } + /** + * {@inheritDoc} + */ + @Override + public float getValue() + { + return 1.0f; + } - /** - * {@inheritDoc} - */ - @Override - public float sumOfSquaredWeights() throws IOException { - return 1.0f; - } + /** + * {@inheritDoc} + */ + @Override + public float sumOfSquaredWeights() throws IOException + { + return 1.0f; + } - /** - * {@inheritDoc} - */ - @Override - public void normalize(float norm) { - } + /** + * {@inheritDoc} + */ + @Override + public void normalize(float norm) + { + } - /** - * {@inheritDoc} - */ - @Override - public Explanation explain(IndexReader reader, int doc) throws IOException { - return new Explanation(); - } - } + /** + * {@inheritDoc} + */ + @Override + public Explanation explain(IndexReader reader, int doc) throws IOException + { + return new Explanation(); + } + } - //------------------------< RangeQueryScorer >------------------------------ + //------------------------< RangeQueryScorer >------------------------------ + /** + * Implements a <code>Scorer</code> for this <code>RangeQuery</code>. + */ + private final class RangeQueryScorer extends Scorer + { - /** - * Implements a <code>Scorer</code> for this <code>RangeQuery</code>. - */ - private final class RangeQueryScorer extends Scorer { + /** + * The index reader to use for calculating the matching documents. + */ + private final IndexReader reader; - /** - * The index reader to use for calculating the matching documents. - */ - private final IndexReader reader; + /** + * The documents ids that match this range query. + */ + private final BitSet hits; - /** - * The documents ids that match this range query. - */ - private final BitSet hits; + /** + * Set to <code>true</code> when the hits have been calculated. + */ + private boolean hitsCalculated = false; - /** - * Set to <code>true</code> when the hits have been calculated. - */ - private boolean hitsCalculated = false; + /** + * The next document id to return + */ + private int nextDoc = -1; - /** - * The next document id to return - */ - private int nextDoc = -1; + /** + * The cache key to use to store the results. + */ + private final String cacheKey; - /** - * The cache key to use to store the results. - */ - private final String cacheKey; + /** + * The map to store the results. + */ + private final Map<String, BitSet> resultMap; - /** - * The map to store the results. - */ - private final Map resultMap; + /** + * Creates a new RangeQueryScorer. + * @param similarity the similarity implementation. + * @param reader the index reader to use. + */ + @SuppressWarnings({"unchecked"}) + RangeQueryScorer(Similarity similarity, IndexReader reader) + { + super(similarity); + this.reader = reader; + StringBuffer key = new StringBuffer(); + key.append(lowerTerm != null ? lowerTerm.field() : upperTerm.field()); + key.append('\uFFFF'); + key.append(lowerTerm != null ? lowerTerm.text() : ""); + key.append('\uFFFF'); + key.append(upperTerm != null ? upperTerm.text() : ""); + key.append('\uFFFF'); + key.append(inclusive); + key.append('\uFFFF'); + key.append(transform); + this.cacheKey = key.toString(); + // check cache + PerQueryCache cache = PerQueryCache.getInstance(); + Map<String, BitSet> m = (Map<String, BitSet>)cache.get(RangeQueryScorer.class, reader); + if (m == null) + { + m = new HashMap<String, BitSet>(); + cache.put(RangeQueryScorer.class, reader, m); + } + resultMap = m; - /** - * Creates a new RangeQueryScorer. - * @param similarity the similarity implementation. - * @param reader the index reader to use. - */ - RangeQueryScorer(Similarity similarity, IndexReader reader) { - super(similarity); - this.reader = reader; - StringBuffer key = new StringBuffer(); - key.append(lowerTerm != null ? lowerTerm.field() : upperTerm.field()); - key.append('\uFFFF'); - key.append(lowerTerm != null ? lowerTerm.text() : ""); - key.append('\uFFFF'); - key.append(upperTerm != null ? upperTerm.text() : ""); - key.append('\uFFFF'); - key.append(inclusive); - key.append('\uFFFF'); - key.append(transform); - this.cacheKey = key.toString(); - // check cache - PerQueryCache cache = PerQueryCache.getInstance(); - Map m = (Map) cache.get(RangeQueryScorer.class, reader); - if (m == null) { - m = new HashMap(); - cache.put(RangeQueryScorer.class, reader, m); - } - resultMap = m; + BitSet result = resultMap.get(cacheKey); + if (result == null) + { + result = new BitSet(reader.maxDoc()); + } + else + { + hitsCalculated = true; + } + hits = result; + } - BitSet result = (BitSet) resultMap.get(cacheKey); - if (result == null) { - result = new BitSet(reader.maxDoc()); - } else { - hitsCalculated = true; - } - hits = result; - } + @Override + public int nextDoc() throws IOException + { + if (nextDoc == NO_MORE_DOCS) + { + return nextDoc; + } - /** - * {@inheritDoc} - */ - @Override - public boolean next() throws IOException { - calculateHits(); - nextDoc = hits.nextSetBit(nextDoc + 1); - return nextDoc > -1; - } + calculateHits(); + nextDoc = hits.nextSetBit(nextDoc + 1); + if (nextDoc < 0) + { + nextDoc = NO_MORE_DOCS; + } + return nextDoc; + } - /** - * {@inheritDoc} - */ - @Override - public int doc() { - return nextDoc; - } + @Override + public int docID() + { + return nextDoc; + } - /** - * {@inheritDoc} - */ - @Override - public float score() { - return 1.0f; - } + @Override + public float score() + { + return 1.0f; + } - /** - * {@inheritDoc} - */ - @Override - public boolean skipTo(int target) throws IOException { - calculateHits(); - nextDoc = hits.nextSetBit(target); - return nextDoc > -1; - } + @Override + public int advance(int target) throws IOException + { + if (nextDoc == NO_MORE_DOCS) + { + return nextDoc; + } - /** - * Returns an empty Explanation object. - * @return an empty Explanation object. - */ - @Override - public Explanation explain(int doc) { - return new Explanation(); - } + calculateHits(); + nextDoc = hits.nextSetBit(target); + if (nextDoc < 0) + { + nextDoc = NO_MORE_DOCS; + } + return nextDoc; + } - /** - * Calculates the ids of the documents matching this range query. - * @throws IOException if an error occurs while reading from the index. - */ - private void calculateHits() throws IOException { - if (hitsCalculated) { - return; - } + /** + * Calculates the ids of the documents matching this range query. + * @throws IOException if an error occurs while reading from the index. + */ + private void calculateHits() throws IOException + { + if (hitsCalculated) + { + return; + } - String testField = getField(); + String testField = getField(); - boolean checkLower = false; - if (!inclusive || transform != TRANSFORM_NONE) { - // make adjustments to set to exclusive - checkLower = true; - } + boolean checkLower = false; + if (!inclusive || transform != TRANSFORM_NONE) + { + // make adjustments to set to exclusive + checkLower = true; + } - int propNameLength = FieldNames.getNameLength(lowerTerm.text()); - String namePrefix = ""; - if (propNameLength > 0) { - namePrefix = lowerTerm.text().substring(0, propNameLength); - } - List startTerms = new ArrayList(2); + int propNameLength = FieldNames.getNameLength(lowerTerm.text()); + String namePrefix = ""; + if (propNameLength > 0) + { + namePrefix = lowerTerm.text().substring(0, propNameLength); + } + List<Term> startTerms = new ArrayList<Term>(2); - if (transform == TRANSFORM_NONE || lowerTerm.text().length() <= propNameLength) { - // use lowerTerm as is - startTerms.add(lowerTerm); - } else { - // first enumerate terms using lower case start character - StringBuffer termText = new StringBuffer(propNameLength + 1); - termText.append(lowerTerm.text().subSequence(0, propNameLength)); - char startCharacter = lowerTerm.text().charAt(propNameLength); - termText.append(Character.toLowerCase(startCharacter)); - startTerms.add(new Term(lowerTerm.field(), termText.toString())); - // second enumerate terms using upper case start character - termText.setCharAt(termText.length() - 1, Character.toUpperCase(startCharacter)); - startTerms.add(new Term(lowerTerm.field(), termText.toString())); - } + if (transform == TRANSFORM_NONE || lowerTerm.text().length() <= propNameLength) + { + // use lowerTerm as is + startTerms.add(lowerTerm); + } + else + { + // first enumerate terms using lower case start character + StringBuffer termText = new StringBuffer(propNameLength + 1); + termText.append(lowerTerm.text().subSequence(0, propNameLength)); + char startCharacter = lowerTerm.text().charAt(propNameLength); + termText.append(Character.toLowerCase(startCharacter)); + startTerms.add(new Term(lowerTerm.field(), termText.toString())); + // second enumerate terms using upper case start character + termText.setCharAt(termText.length() - 1, Character.toUpperCase(startCharacter)); + startTerms.add(new Term(lowerTerm.field(), termText.toString())); + } - Iterator it = startTerms.iterator(); - while (it.hasNext()) { - Term startTerm = (Term) it.next(); + for (Term startTerm : startTerms) + { + TermEnum terms = reader.terms(startTerm); + try + { + TermDocs docs = reader.termDocs(); + try + { + do + { + Term term = terms.term(); + if (term != null && term.field() == testField && term.text().startsWith(namePrefix)) + { + if (checkLower) + { + int compare = termCompare(term.text(), lowerTerm.text(), propNameLength); + if (compare > 0 || compare == 0 && inclusive) + { + // do not check lower term anymore if no + // transformation is done on the term enum + checkLower = transform != TRANSFORM_NONE; + } + else + { + // continue with next term + continue; + } + } + if (upperTerm != null) + { + int compare = termCompare(term.text(), upperTerm.text(), propNameLength); + // if beyond the upper term, or is exclusive and + // this is equal to the upper term + if ((compare > 0) || (!inclusive && compare == 0)) + { + // only break out if no transformation + // was done on the term from the enum + if (transform == TRANSFORM_NONE) + { + break; + } + else + { + // because of the transformation + // it is possible that the next + // term will be included again if + // we still enumerate on the same + // property name + if (term.text().startsWith(namePrefix)) + { + continue; + } + else + { + break; + } + } + } + } - TermEnum terms = reader.terms(startTerm); - try { - TermDocs docs = reader.termDocs(); - try { - do { - Term term = terms.term(); - if (term != null - && term.field() == testField - && term.text().startsWith(namePrefix)) { - if (checkLower) { - int compare = termCompare(term.text(), lowerTerm.text(), propNameLength); - if (compare > 0 || compare == 0 && inclusive) { - // do not check lower term anymore if no - // transformation is done on the term enum - checkLower = transform == TRANSFORM_NONE ? false : true; - } else { - // continue with next term - continue; - } - } - if (upperTerm != null) { - int compare = termCompare(term.text(), upperTerm.text(), propNameLength); - // if beyond the upper term, or is exclusive and - // this is equal to the upper term - if ((compare > 0) || (!inclusive && compare == 0)) { - // only break out if no transformation - // was done on the term from the enum - if (transform == TRANSFORM_NONE) { - break; - } else { - // because of the transformation - // it is possible that the next - // term will be included again if - // we still enumerate on the same - // property name - if (term.text().startsWith(namePrefix)) { - continue; - } else { - break; - } - } - } - } - - docs.seek(terms); - while (docs.next()) { - hits.set(docs.doc()); - } - } else { - break; - } - } while(terms.next()); - } finally { - docs.close(); - } - } finally { - terms.close(); - } + docs.seek(terms); + while (docs.next()) + { + hits.set(docs.doc()); + } + } + else + { + break; + } + } + while (terms.next()); + } + finally + { + docs.close(); + } } + finally + { + terms.close(); + } + } - hitsCalculated = true; - // put to cache - resultMap.put(cacheKey, hits); - } + hitsCalculated = true; + // put to cache + resultMap.put(cacheKey, hits); + } - /** - * Compares the <code>text</code> with the <code>other</code> String. This - * implementation behaves like {@link String#compareTo(Object)} but also - * respects the {@link RangeQuery#transform} property. - * - * @param text the text to compare to <code>other</code>. The - * transformation function is applied to this parameter before - * it is compared to <code>other</code>. - * @param other the other String. - * @param offset start comparing the two strings at <code>offset</code>. - * @return see {@link String#compareTo(Object)}. But also respects {@link - * RangeQuery#transform}. - */ - private int termCompare(String text, String other, int offset) { - OffsetCharSequence seq1 = new OffsetCharSequence(offset, text, transform); - OffsetCharSequence seq2 = new OffsetCharSequence(offset, other); - return seq1.compareTo(seq2); - } - } + /** + * Compares the <code>text</code> with the <code>other</code> String. This + * implementation behaves like {@link String#compareTo(Object)} but also + * respects the {@link RangeQuery#transform} property. + * + * @param text the text to compare to <code>other</code>. The + * transformation function is applied to this parameter before + * it is compared to <code>other</code>. + * @param other the other String. + * @param offset start comparing the two strings at <code>offset</code>. + * @return see {@link String#compareTo(Object)}. But also respects {@link + * RangeQuery#transform}. + */ + private int termCompare(String text, String other, int offset) + { + OffsetCharSequence seq1 = new OffsetCharSequence(offset, text, transform); + OffsetCharSequence seq2 = new OffsetCharSequence(offset, other); + return seq1.compareTo(seq2); + } + } } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/ReadOnlyIndexReader.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/ReadOnlyIndexReader.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/ReadOnlyIndexReader.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -16,306 +16,337 @@ */ package org.exoplatform.services.jcr.impl.core.query.lucene; -import java.io.IOException; -import java.util.BitSet; - import org.apache.lucene.index.Term; import org.apache.lucene.index.TermDocs; import org.apache.lucene.index.TermPositions; +import java.io.IOException; +import java.util.BitSet; +import java.util.Map; + /** * Overwrites the methods that would modify the index and throws an * {@link UnsupportedOperationException} in each of those methods. A * <code>ReadOnlyIndexReader</code> will always show all documents that have * not been deleted at the time when the index reader is created. */ -class ReadOnlyIndexReader extends RefCountingIndexReader { +class ReadOnlyIndexReader extends RefCountingIndexReader +{ - /** - * The underlying shared reader. - */ - private final SharedIndexReader reader; + /** + * The underlying shared reader. + */ + private final SharedIndexReader reader; - /** - * The deleted documents as initially read from the IndexReader passed - * in the constructor of this class. - */ - private final BitSet deleted; + /** + * The deleted documents as initially read from the IndexReader passed + * in the constructor of this class. + */ + private final BitSet deleted; - /** - * The version of the index reader from where the deleted BitSet was - * obtained from. - */ - private long deletedDocsVersion; + /** + * The version of the index reader from where the deleted BitSet was + * obtained from. + */ + private long deletedDocsVersion; - /** - * Creates a new index reader based on <code>reader</code> at - * <code>modificationTick</code>. - * - * @param reader the underlying <code>IndexReader</code>. - * @param deleted the documents that are deleted in - * <code>reader</code>. - * @param deletedDocsVersion the version of the index reader from where the - * deleted BitSet was obtained from. - */ - public ReadOnlyIndexReader(SharedIndexReader reader, - BitSet deleted, - long deletedDocsVersion) { - super(reader); - this.reader = reader; - this.deleted = deleted; - this.deletedDocsVersion = deletedDocsVersion; - // acquire underlying reader - reader.acquire(); - } + /** + * Creates a new index reader based on <code>reader</code> at + * <code>modificationTick</code>. + * + * @param reader the underlying <code>IndexReader</code>. + * @param deleted the documents that are deleted in + * <code>reader</code>. + * @param deletedDocsVersion the version of the index reader from where the + * deleted BitSet was obtained from. + */ + public ReadOnlyIndexReader(SharedIndexReader reader, BitSet deleted, long deletedDocsVersion) + { + super(reader); + this.reader = reader; + this.deleted = deleted; + this.deletedDocsVersion = deletedDocsVersion; + // acquire underlying reader + reader.acquire(); + } - /** - * @return version of the deleted docs. - */ - long getDeletedDocsVersion() { - return deletedDocsVersion; - } + /** + * @return version of the deleted docs. + */ + long getDeletedDocsVersion() + { + return deletedDocsVersion; + } - /** - * Returns the tick value when the underlying {@link CachingIndexReader} was - * created. - * - * @return the creation tick for the underlying reader. - */ - long getCreationTick() { - return reader.getCreationTick(); - } + /** + * Returns the tick value when the underlying {@link CachingIndexReader} was + * created. + * + * @return the creation tick for the underlying reader. + */ + long getCreationTick() + { + return reader.getCreationTick(); + } - /** - * Updates the deleted documents in this index reader. When this method - * returns this index reader will have the same documents marked as deleted - * as the passed <code>reader</code>. - * <p/> - * This method is not thread-safe! Make sure no other thread is concurrently - * using this reader at the same time. - * - * @param reader the reader from where to obtain the deleted documents - * info. - */ - void updateDeletedDocs(CommittableIndexReader reader) { - int maxDoc = reader.maxDoc(); - for (int i = 0; i < maxDoc; i++) { - if (reader.isDeleted(i)) { - deleted.set(i); - } - } - deletedDocsVersion = reader.getModificationCount(); - } + /** + * Updates the deleted documents in this index reader. When this method + * returns this index reader will have the same documents marked as deleted + * as the passed <code>reader</code>. + * <p/> + * This method is not thread-safe! Make sure no other thread is concurrently + * using this reader at the same time. + * + * @param reader the reader from where to obtain the deleted documents + * info. + */ + void updateDeletedDocs(CommittableIndexReader reader) + { + int maxDoc = reader.maxDoc(); + for (int i = 0; i < maxDoc; i++) + { + if (reader.isDeleted(i)) + { + deleted.set(i); + } + } + deletedDocsVersion = reader.getModificationCount(); + } - /** - * Returns the <code>DocId</code> of the parent of <code>n</code> or - * {@link DocId#NULL} if <code>n</code> does not have a parent - * (<code>n</code> is the root node). - * - * @param n the document number. - * @return the <code>DocId</code> of <code>n</code>'s parent. - * @throws IOException if an error occurs while reading from the index. - */ - public DocId getParent(int n) throws IOException { - return getBase().getParent(n, deleted); - } + /** + * Returns the <code>DocId</code> of the parent of <code>n</code> or + * {@link DocId#NULL} if <code>n</code> does not have a parent + * (<code>n</code> is the root node). + * + * @param n the document number. + * @return the <code>DocId</code> of <code>n</code>'s parent. + * @throws IOException if an error occurs while reading from the index. + */ + public DocId getParent(int n) throws IOException + { + return getBase().getParent(n, deleted); + } - /** - * Returns the {@link SharedIndexReader} this reader is based on. - * - * @return the {@link SharedIndexReader} this reader is based on. - */ - public SharedIndexReader getBase() { - return (SharedIndexReader) in; - } + /** + * Returns the {@link SharedIndexReader} this reader is based on. + * + * @return the {@link SharedIndexReader} this reader is based on. + */ + public SharedIndexReader getBase() + { + return (SharedIndexReader)in; + } - //---------------------< IndexReader overwrites >--------------------------- + //---------------------< IndexReader overwrites >--------------------------- - /** - * Returns true if document <code>n</code> has been deleted - * @param n the document number - * @return true if document <code>n</code> has been deleted - */ - public boolean isDeleted(int n) { - return deleted.get(n); - } + /** + * Returns true if document <code>n</code> has been deleted + * @param n the document number + * @return true if document <code>n</code> has been deleted + */ + public boolean isDeleted(int n) + { + return deleted.get(n); + } - /** - * Returns <code>true</code> if any documents have been deleted. - * - * @return <code>true</code> if any documents have been deleted. - */ - public boolean hasDeletions() { - return !deleted.isEmpty(); - } + /** + * Returns <code>true</code> if any documents have been deleted. + * + * @return <code>true</code> if any documents have been deleted. + */ + public boolean hasDeletions() + { + return !deleted.isEmpty(); + } - /** - * Returns the number of documents in this index reader. - * - * @return the number of documents in this index reader. - */ - public int numDocs() { - return maxDoc() - deleted.cardinality(); - } + /** + * Returns the number of documents in this index reader. + * + * @return the number of documents in this index reader. + */ + public int numDocs() + { + return maxDoc() - deleted.cardinality(); + } - /** - * @exception UnsupportedOperationException always - */ - protected final void doDelete(int docNum) { - throw new UnsupportedOperationException("IndexReader is read-only"); - } + /** + * @exception UnsupportedOperationException always + */ + protected final void doDelete(int docNum) + { + throw new UnsupportedOperationException("IndexReader is read-only"); + } - /** - * @exception UnsupportedOperationException always - */ - protected final void doUndeleteAll() { - throw new UnsupportedOperationException("IndexReader is read-only"); - } + /** + * @exception UnsupportedOperationException always + */ + protected final void doUndeleteAll() + { + throw new UnsupportedOperationException("IndexReader is read-only"); + } - /** - * @exception UnsupportedOperationException always - */ - protected final void doCommit() { - throw new UnsupportedOperationException("IndexReader is read-only"); - } + /** + * @exception UnsupportedOperationException always + */ + protected final void doCommit(Map commitUserData) + { + throw new UnsupportedOperationException("IndexReader is read-only"); + } - /** - * Wraps the underlying <code>TermDocs</code> and filters out documents - * marked as deleted.<br/> - * If <code>term</code> is for a {@link FieldNames#UUID} field and this - * <code>ReadOnlyIndexReader</code> does not have such a document, - * {@link EmptyTermDocs#INSTANCE} is returned. - * - * @param term the term to enumerate the docs for. - * @return TermDocs for <code>term</code>. - * @throws IOException if an error occurs while reading from the index. - */ - public TermDocs termDocs(Term term) throws IOException { - // do not wrap for empty TermDocs - TermDocs td = reader.termDocs(term); - if (td != EmptyTermDocs.INSTANCE) { - td = new FilteredTermDocs(td); - } - return td; - } + /** + * Wraps the underlying <code>TermDocs</code> and filters out documents + * marked as deleted.<br/> + * If <code>term</code> is for a {@link FieldNames#UUID} field and this + * <code>ReadOnlyIndexReader</code> does not have such a document, + * {@link EmptyTermDocs#INSTANCE} is returned. + * + * @param term the term to enumerate the docs for. + * @return TermDocs for <code>term</code>. + * @throws IOException if an error occurs while reading from the index. + */ + public TermDocs termDocs(Term term) throws IOException + { + // do not wrap for empty TermDocs + TermDocs td = reader.termDocs(term); + if (td != EmptyTermDocs.INSTANCE) + { + td = new FilteredTermDocs(td); + } + return td; + } - /** - * Wraps the underlying <code>TermDocs</code> and filters out documents - * marked as deleted. - * - * @return TermDocs over the whole index. - * @throws IOException if an error occurs while reading from the index. - */ - public TermDocs termDocs() throws IOException { - return new FilteredTermDocs(super.termDocs()); - } + /** + * Wraps the underlying <code>TermDocs</code> and filters out documents + * marked as deleted. + * + * @return TermDocs over the whole index. + * @throws IOException if an error occurs while reading from the index. + */ + public TermDocs termDocs() throws IOException + { + return new FilteredTermDocs(super.termDocs()); + } - /** - * Wraps the underlying <code>TermPositions</code> and filters out documents - * marked as deleted. - * - * @return TermPositions over the whole index. - * @throws IOException if an error occurs while reading from the index. - */ - public TermPositions termPositions() throws IOException { - return new FilteredTermPositions(super.termPositions()); - } + /** + * Wraps the underlying <code>TermPositions</code> and filters out documents + * marked as deleted. + * + * @return TermPositions over the whole index. + * @throws IOException if an error occurs while reading from the index. + */ + public TermPositions termPositions() throws IOException + { + return new FilteredTermPositions(super.termPositions()); + } - //----------------------< FilteredTermDocs >-------------------------------- + //----------------------< FilteredTermDocs >-------------------------------- - /** - * Filters a wrapped TermDocs by omitting documents marked as deleted. - */ - private class FilteredTermDocs extends FilterTermDocs { + /** + * Filters a wrapped TermDocs by omitting documents marked as deleted. + */ + private class FilteredTermDocs extends FilterTermDocs + { - /** - * Creates a new filtered TermDocs based on <code>in</code>. - * - * @param in the TermDocs to filter. - */ - public FilteredTermDocs(TermDocs in) { - super(in); - } + /** + * Creates a new filtered TermDocs based on <code>in</code>. + * + * @param in the TermDocs to filter. + */ + public FilteredTermDocs(TermDocs in) + { + super(in); + } - /** - * @inheritDoc - */ - public boolean next() throws IOException { - boolean hasNext = super.next(); - while (hasNext && deleted.get(super.doc())) { - hasNext = super.next(); - } - return hasNext; - } + /** + * @inheritDoc + */ + public boolean next() throws IOException + { + boolean hasNext = super.next(); + while (hasNext && deleted.get(super.doc())) + { + hasNext = super.next(); + } + return hasNext; + } - /** - * @inheritDoc - */ - public int read(int[] docs, int[] freqs) throws IOException { - int count; - for (count = 0; count < docs.length && next(); count++) { - docs[count] = doc(); - freqs[count] = freq(); - } - return count; - } + /** + * @inheritDoc + */ + public int read(int[] docs, int[] freqs) throws IOException + { + int count; + for (count = 0; count < docs.length && next(); count++) + { + docs[count] = doc(); + freqs[count] = freq(); + } + return count; + } - /** - * @inheritDoc - */ - public boolean skipTo(int i) throws IOException { - boolean exists = super.skipTo(i); - while (exists && deleted.get(doc())) { - exists = next(); - } - return exists; - } - } + /** + * @inheritDoc + */ + public boolean skipTo(int i) throws IOException + { + boolean exists = super.skipTo(i); + while (exists && deleted.get(doc())) + { + exists = next(); + } + return exists; + } + } - //---------------------< FilteredTermPositions >---------------------------- + //---------------------< FilteredTermPositions >---------------------------- - /** - * Filters a wrapped TermPositions by omitting documents marked as deleted. - */ - private final class FilteredTermPositions extends FilteredTermDocs - implements TermPositions { + /** + * Filters a wrapped TermPositions by omitting documents marked as deleted. + */ + private final class FilteredTermPositions extends FilteredTermDocs implements TermPositions + { - /** - * Creates a new filtered TermPositions based on <code>in</code>. - * - * @param in the TermPositions to filter. - */ - public FilteredTermPositions(TermPositions in) { - super(in); - } + /** + * Creates a new filtered TermPositions based on <code>in</code>. + * + * @param in the TermPositions to filter. + */ + public FilteredTermPositions(TermPositions in) + { + super(in); + } - /** - * @inheritDoc - */ - public int nextPosition() throws IOException { - return ((TermPositions) this.in).nextPosition(); - } + /** + * @inheritDoc + */ + public int nextPosition() throws IOException + { + return ((TermPositions)this.in).nextPosition(); + } - /** - * @inheritDoc - */ - public int getPayloadLength() { - return ((TermPositions) in).getPayloadLength(); - } + /** + * @inheritDoc + */ + public int getPayloadLength() + { + return ((TermPositions)in).getPayloadLength(); + } - /** - * @inheritDoc - */ - public byte[] getPayload(byte[] data, int offset) throws IOException { - return ((TermPositions) in).getPayload(data, offset); - } + /** + * @inheritDoc + */ + public byte[] getPayload(byte[] data, int offset) throws IOException + { + return ((TermPositions)in).getPayload(data, offset); + } - /** - * @inheritDoc - */ - public boolean isPayloadAvailable() { - return ((TermPositions) in).isPayloadAvailable(); - } + /** + * @inheritDoc + */ + public boolean isPayloadAvailable() + { + return ((TermPositions)in).isPayloadAvailable(); + } - } + } } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SearchIndex.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SearchIndex.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SearchIndex.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -22,6 +22,8 @@ import org.apache.commons.collections.iterators.TransformIterator; import org.apache.lucene.analysis.Analyzer; import org.apache.lucene.analysis.Token; +import org.apache.lucene.analysis.tokenattributes.PayloadAttribute; +import org.apache.lucene.analysis.tokenattributes.TermAttribute; import org.apache.lucene.document.Document; import org.apache.lucene.document.Field; import org.apache.lucene.document.Fieldable; @@ -29,11 +31,11 @@ import org.apache.lucene.index.MultiReader; import org.apache.lucene.index.Term; import org.apache.lucene.index.TermDocs; +import org.apache.lucene.search.FieldComparatorSource; import org.apache.lucene.search.IndexSearcher; import org.apache.lucene.search.Query; import org.apache.lucene.search.Similarity; import org.apache.lucene.search.Sort; -import org.apache.lucene.search.SortComparatorSource; import org.apache.lucene.search.SortField; import org.exoplatform.commons.utils.ClassLoading; import org.exoplatform.commons.utils.PrivilegedFileHelper; @@ -82,16 +84,7 @@ import java.lang.reflect.Constructor; import java.lang.reflect.InvocationTargetException; import java.security.PrivilegedAction; -import java.util.ArrayList; -import java.util.Arrays; -import java.util.Collection; -import java.util.Collections; -import java.util.HashMap; -import java.util.HashSet; -import java.util.Iterator; -import java.util.List; -import java.util.Map; -import java.util.Set; +import java.util.*; import java.util.concurrent.CountDownLatch; import javax.jcr.RepositoryException; @@ -452,7 +445,7 @@ /** * The sort comparator source for indexed properties. */ - private SortComparatorSource scs; + private FieldComparatorSource scs; /** * Flag that indicates whether the hierarchy cache should be initialized @@ -641,7 +634,7 @@ } } - scs = new SharedFieldSortComparator(FieldNames.PROPERTIES, context.getItemStateManager(), nsMappings); + scs = new SharedFieldComparatorSource(FieldNames.PROPERTIES, context.getItemStateManager(), nsMappings); npResolver = new LocationFactory(nsMappings); indexingConfig = createIndexingConfiguration(nsMappings); @@ -672,10 +665,9 @@ } catch (IOException e) { - log - .error( - "Error while reindexing the workspace. Please fix the problem, delete index and restart server.", - e); + log.error( + "Error while reindexing the workspace. Please fix the problem, delete index and restart server.", + e); } } }, "Reindexing-" + context.getRepositoryName() + "-" + context.getContainer().getWorkspaceName()).start(); @@ -1109,8 +1101,7 @@ } catch (RepositoryException e) { - log - .warn("Exception while creating document for node: " + state.getIdentifier() + ": " + e.toString(), e); + log.warn("Exception while creating document for node: " + state.getIdentifier() + ": " + e.toString(), e); } return doc; } @@ -1137,8 +1128,7 @@ } catch (RepositoryException e) { - log - .warn("Exception while creating document for node: " + state.getIdentifier() + ": " + e.toString()); + log.warn("Exception while creating document for node: " + state.getIdentifier() + ": " + e.toString()); } return null; } @@ -1287,9 +1277,7 @@ waitForResuming(); checkOpen(); - Sort sort = new Sort(createSortFields(orderProps, orderSpecs)); - final IndexReader reader = getIndexReader(queryImpl.needsSystemTree()); JcrIndexSearcher searcher = new JcrIndexSearcher(session, reader, getContext().getItemStateManager()); searcher.setSimilarity(getSimilarity()); @@ -1651,28 +1639,11 @@ /** * @return the sort comparator source for this index. */ - protected SortComparatorSource getSortComparatorSource() + protected FieldComparatorSource getSortComparatorSource() { return scs; } - // /** - // * Factory method to create the <code>TextExtractor</code> instance. - // * - // * @return the <code>TextExtractor</code> instance this index should use. - // */ - // protected TextExtractor createTextExtractor() - // { - // TextExtractor txtExtr = new JackrabbitTextExtractor(textFilterClasses); - // if (extractorPoolSize > 0) - // { - // // wrap with pool - // txtExtr = new PooledTextExtractor(txtExtr, extractorPoolSize, - // extractorBackLog, extractorTimeout); - // } - // return txtExtr; - // } - /** * @param namespaceMappings * The namespace mappings @@ -1981,7 +1952,13 @@ Fieldable field = fields[k]; // assume properties fields use // SingleTokenStream - t = field.tokenStreamValue().next(t); + //t = field.tokenStreamValue().next(t); + field.tokenStreamValue().incrementToken(); + TermAttribute term = + (TermAttribute)field.tokenStreamValue().getAttribute(TermAttribute.class); + PayloadAttribute payload = + (PayloadAttribute)field.tokenStreamValue().getAttribute(PayloadAttribute.class); + String value = new String(t.termBuffer(), 0, t.termLength()); if (value.startsWith(namePrefix)) { @@ -1992,7 +1969,8 @@ String path = getNamespaceMappings().translatePath(p); value = FieldNames.createNamedValue(path, value); t.setTermBuffer(value); - doc.add(new Field(field.name(), new SingletonTokenStream(t))); + doc.add(new Field(field.name(), new SingletonTokenStream(term.term(), payload + .getPayload()))); doc.add(new Field(FieldNames.AGGREGATED_NODE_UUID, parent.getIdentifier(), Field.Store.NO, Field.Index.NOT_ANALYZED_NO_NORMS)); } @@ -2015,8 +1993,7 @@ catch (Exception e) { // do not fail if aggregate cannot be created - log - .warn("Exception while building indexing aggregate for" + " node with UUID: " + state.getIdentifier(), e); + log.warn("Exception while building indexing aggregate for" + " node with UUID: " + state.getIdentifier(), e); } } } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SharedFieldCache.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SharedFieldCache.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SharedFieldCache.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -16,6 +16,13 @@ */ package org.exoplatform.services.jcr.impl.core.query.lucene; +import org.apache.lucene.index.IndexReader; +import org.apache.lucene.index.Term; +import org.apache.lucene.index.TermDocs; +import org.apache.lucene.index.TermEnum; +import org.apache.lucene.index.TermPositions; +import org.apache.lucene.search.FieldComparator; + import java.io.IOException; import java.util.HashMap; import java.util.Map; @@ -23,298 +30,335 @@ import javax.jcr.PropertyType; -import org.apache.lucene.index.IndexReader; -import org.apache.lucene.index.Term; -import org.apache.lucene.index.TermDocs; -import org.apache.lucene.index.TermEnum; -import org.apache.lucene.index.TermPositions; -import org.apache.lucene.search.SortComparator; - /** * Implements a variant of the lucene class <code>org.apache.lucene.search.FieldCacheImpl</code>. * The lucene FieldCache class has some sort of support for custom comparators * but it only works on the basis of a field name. There is no further control * over the terms to iterate, that's why we use our own implementation. */ -public class SharedFieldCache { +public class SharedFieldCache +{ - /** - * Expert: Stores term text values and document ordering data. - */ - public static class ValueIndex { + /** + * Expert: Stores term text values and document ordering data. + */ + public static class ValueIndex + { - /** - * Some heuristic factor that determines whether the array is sparse. Note that if less then - * 1% is set, we already count the array as sparse. This is because it will become memory consuming - * quickly by keeping the (sparse) arrays - */ - private static final int SPARSE_FACTOR = 100; + /** + * Some heuristic factor that determines whether the array is sparse. Note that if less then + * 1% is set, we already count the array as sparse. This is because it will become memory consuming + * quickly by keeping the (sparse) arrays + */ + private static final int SPARSE_FACTOR = 100; - /** - * Values indexed by document id. - */ - private final Comparable[] values; + /** + * Values indexed by document id. + */ + private final Comparable[] values; - /** - * Values (Comparable) map indexed by document id. - */ - public final Map valuesMap; + /** + * Values (Comparable) map indexed by document id. + */ + public final Map<Integer, Comparable> valuesMap; - /** - * Boolean indicating whether the {@link #valuesMap} impl has to be used - */ - public final boolean sparse; + /** + * Boolean indicating whether the {@link #valuesMap} impl has to be used + */ + public final boolean sparse; - /** - * Creates one of these objects - */ - public ValueIndex(Comparable[] values, int setValues) { - if (isSparse(values, setValues)) { - this.sparse = true; - this.values = null; - if (setValues == 0) { - this.valuesMap = null; - } else { - this.valuesMap = getValuesMap(values, setValues); - } - } else { - this.sparse = false; - this.values = values; - this.valuesMap = null; + /** + * Creates one of these objects + */ + public ValueIndex(Comparable[] values, int setValues) + { + if (isSparse(values, setValues)) + { + this.sparse = true; + this.values = null; + if (setValues == 0) + { + this.valuesMap = null; } - } - - public Comparable getValue(int i) { - if (sparse) { - return valuesMap == null ? null : (Comparable) valuesMap.get(new Integer(i)); - } else { - return values[i]; + else + { + this.valuesMap = getValuesMap(values, setValues); } - } + } + else + { + this.sparse = false; + this.values = values; + this.valuesMap = null; + } + } - private Map getValuesMap(Comparable[] values, int setValues) { - Map map = new HashMap(setValues); - for (int i = 0; i < values.length && setValues > 0; i++) { - if (values[i] != null) { - map.put(new Integer(i), values[i]); - setValues--; - } - } - return map; - } + public Comparable getValue(int i) + { + if (sparse) + { + return valuesMap == null ? null : valuesMap.get(i); + } + else + { + return values[i]; + } + } - private boolean isSparse(Comparable[] values, int setValues) { - // some really simple test to test whether the array is sparse. - // Currently, when less then 1% is set, the array is already sparse - // for this typical cache to avoid memory issues - if (setValues * SPARSE_FACTOR < values.length) { - return true; + private Map<Integer, Comparable> getValuesMap(Comparable[] values, int setValues) + { + Map<Integer, Comparable> map = new HashMap<Integer, Comparable>(setValues); + for (int i = 0; i < values.length && setValues > 0; i++) + { + if (values[i] != null) + { + map.put(i, values[i]); + setValues--; } - return false; - } - } + } + return map; + } - /** - * Reference to the single instance of <code>SharedFieldCache</code>. - */ - public static final SharedFieldCache INSTANCE = new SharedFieldCache(); + private boolean isSparse(Comparable[] values, int setValues) + { + // some really simple test to test whether the array is sparse. Currently, when less then 1% is set, the array is already sparse + // for this typical cache to avoid memory issues + if (setValues * SPARSE_FACTOR < values.length) + { + return true; + } + return false; + } + } - /** - * The internal cache. Maps Entry to array of interpreted term values. - */ - private final Map cache = new WeakHashMap(); + /** + * Reference to the single instance of <code>SharedFieldCache</code>. + */ + public static final SharedFieldCache INSTANCE = new SharedFieldCache(); - /** - * Private constructor. - */ - private SharedFieldCache() { - } + /** + * The internal cache. Maps Entry to array of interpreted term values. + */ + private final Map<IndexReader, Map<Key, ValueIndex>> cache = new WeakHashMap<IndexReader, Map<Key, ValueIndex>>(); - /** - * Creates a <code>ValueIndex</code> for a <code>field</code> and a term - * <code>prefix</code>. The term prefix acts as the property name for the - * shared <code>field</code>. - * <p/> - * This method is an adapted version of: <code>FieldCacheImpl.getStringIndex()</code> - * - * @param reader the <code>IndexReader</code>. - * @param field name of the shared field. - * @param prefix the property name, will be used as term prefix. - * @param comparator the sort comparator instance. - * @return a ValueIndex that contains the field values and order - * information. - * @throws IOException if an error occurs while reading from the index. - */ - public ValueIndex getValueIndex(IndexReader reader, - String field, - String prefix, - SortComparator comparator) - throws IOException { + /** + * Private constructor. + */ + private SharedFieldCache() + { + } - if (reader instanceof ReadOnlyIndexReader) { - reader = ((ReadOnlyIndexReader) reader).getBase(); - } + /** + * Creates a <code>ValueIndex</code> for a <code>field</code> and a term + * <code>prefix</code>. The term prefix acts as the property name for the + * shared <code>field</code>. + * <p/> + * This method is an adapted version of: <code>FieldCacheImpl.getStringIndex()</code> + * + * @param reader the <code>IndexReader</code>. + * @param field name of the shared field. + * @param prefix the property name, will be used as term prefix. + * @param comparator the field comparator instance. + * @return a ValueIndex that contains the field values and order + * information. + * @throws IOException if an error occurs while reading from the index. + */ + public ValueIndex getValueIndex(IndexReader reader, String field, String prefix, FieldComparator comparator) + throws IOException + { - field = field.intern(); - ValueIndex ret = lookup(reader, field, prefix, comparator); - if (ret == null) { - Comparable[] retArray = new Comparable[reader.maxDoc()]; - int setValues = 0; - if (retArray.length > 0) { - IndexFormatVersion version = IndexFormatVersion.getVersion(reader); - boolean hasPayloads = version.isAtLeast(IndexFormatVersion.V3); - TermDocs termDocs; - byte[] payload = null; - int type; - if (hasPayloads) { - termDocs = reader.termPositions(); - payload = new byte[1]; - } else { - termDocs = reader.termDocs(); - } - TermEnum termEnum = reader.terms(new Term(field, prefix)); + if (reader instanceof ReadOnlyIndexReader) + { + reader = ((ReadOnlyIndexReader)reader).getBase(); + } - char[] tmp = new char[16]; - try { - if (termEnum.term() == null) { - throw new RuntimeException("no terms in field " + field); - } - do { - Term term = termEnum.term(); - if (term.field() != field || !term.text().startsWith(prefix)) { - break; - } + field = field.intern(); + ValueIndex ret = lookup(reader, field, prefix, comparator); + if (ret == null) + { + Comparable[] retArray = new Comparable[reader.maxDoc()]; + int setValues = 0; + if (retArray.length > 0) + { + IndexFormatVersion version = IndexFormatVersion.getVersion(reader); + boolean hasPayloads = version.isAtLeast(IndexFormatVersion.V3); + TermDocs termDocs; + byte[] payload = null; + int type; + if (hasPayloads) + { + termDocs = reader.termPositions(); + payload = new byte[1]; + } + else + { + termDocs = reader.termDocs(); + } + TermEnum termEnum = reader.terms(new Term(field, prefix)); - // make sure term is compacted - String text = term.text(); - int len = text.length() - prefix.length(); - if (tmp.length < len) { - // grow tmp - tmp = new char[len]; - } - text.getChars(prefix.length(), text.length(), tmp, 0); - String value = new String(tmp, 0, len); + char[] tmp = new char[16]; + try + { + if (termEnum.term() == null) + { + throw new RuntimeException("no terms in field " + field); + } + do + { + Term term = termEnum.term(); + if (term.field() != field || !term.text().startsWith(prefix)) + { + break; + } - termDocs.seek(termEnum); - while (termDocs.next()) { - type = PropertyType.UNDEFINED; - if (hasPayloads) { - TermPositions termPos = (TermPositions) termDocs; - termPos.nextPosition(); - if (termPos.isPayloadAvailable()) { - payload = termPos.getPayload(payload, 0); - type = PropertyMetaData.fromByteArray(payload).getPropertyType(); - } - } - setValues++; - retArray[termDocs.doc()] = getValue(value, type); + // make sure term is compacted + String text = term.text(); + int len = text.length() - prefix.length(); + if (tmp.length < len) + { + // grow tmp + tmp = new char[len]; + } + text.getChars(prefix.length(), text.length(), tmp, 0); + String value = new String(tmp, 0, len); + + termDocs.seek(termEnum); + while (termDocs.next()) + { + type = PropertyType.UNDEFINED; + if (hasPayloads) + { + TermPositions termPos = (TermPositions)termDocs; + termPos.nextPosition(); + if (termPos.isPayloadAvailable()) + { + payload = termPos.getPayload(payload, 0); + type = PropertyMetaData.fromByteArray(payload).getPropertyType(); } - } while (termEnum.next()); - } finally { - termDocs.close(); - termEnum.close(); - } + } + setValues++; + retArray[termDocs.doc()] = getValue(value, type); + } + } + while (termEnum.next()); } - ValueIndex value = new ValueIndex(retArray, setValues); - store(reader, field, prefix, comparator, value); + finally + { + termDocs.close(); + termEnum.close(); + } + } + ValueIndex value = new ValueIndex(retArray, setValues); + store(reader, field, prefix, comparator, value); + return value; + } + return ret; + } + + /** + * See if a <code>ValueIndex</code> object is in the cache. + */ + ValueIndex lookup(IndexReader reader, String field, String prefix, FieldComparator comparer) + { + Key key = new Key(field, prefix, comparer); + synchronized (this) + { + Map<Key, ValueIndex> readerCache = cache.get(reader); + if (readerCache == null) + { + return null; + } + return readerCache.get(key); + } + } + + /** + * Put a <code>ValueIndex</code> <code>value</code> to cache. + */ + ValueIndex store(IndexReader reader, String field, String prefix, FieldComparator comparer, ValueIndex value) + { + Key key = new Key(field, prefix, comparer); + synchronized (this) + { + Map<Key, ValueIndex> readerCache = cache.get(reader); + if (readerCache == null) + { + readerCache = new HashMap<Key, ValueIndex>(); + cache.put(reader, readerCache); + } + return readerCache.put(key, value); + } + } + + /** + * Returns a comparable for the given <code>value</code> that is read from + * the index. + * + * @param value the value as read from the index. + * @param type the property type. + * @return a comparable for the <code>value</code>. + */ + private Comparable getValue(String value, int type) + { + switch (type) + { + case PropertyType.BOOLEAN : + return ComparableBoolean.valueOf(Boolean.valueOf(value).booleanValue()); + case PropertyType.DATE : + return new Long(DateField.stringToTime(value)); + case PropertyType.LONG : + return new Long(LongField.stringToLong(value)); + case PropertyType.DOUBLE : + return new Double(DoubleField.stringToDouble(value)); + default : return value; - } - return ret; - } + } + } - /** - * See if a <code>ValueIndex</code> object is in the cache. - */ - ValueIndex lookup(IndexReader reader, String field, - String prefix, SortComparator comparer) { - Key key = new Key(field, prefix, comparer); - synchronized (this) { - HashMap readerCache = (HashMap) cache.get(reader); - if (readerCache == null) { - return null; - } - return (ValueIndex) readerCache.get(key); - } - } + /** + * A compound <code>Key</code> that consist of <code>field</code> + * <code>prefix</code> and <code>comparator</code>. + */ + static class Key + { - /** - * Put a <code>ValueIndex</code> <code>value</code> to cache. - */ - Object store(IndexReader reader, String field, String prefix, - SortComparator comparer, ValueIndex value) { - Key key = new Key(field, prefix, comparer); - synchronized (this) { - HashMap readerCache = (HashMap) cache.get(reader); - if (readerCache == null) { - readerCache = new HashMap(); - cache.put(reader, readerCache); - } - return readerCache.put(key, value); - } - } + private final String field; - /** - * Returns a comparable for the given <code>value</code> that is read from - * the index. - * - * @param value the value as read from the index. - * @param type the property type. - * @return a comparable for the <code>value</code>. - */ - private Comparable getValue(String value, int type) { - switch (type) { - case PropertyType.BOOLEAN: - return ComparableBoolean.valueOf(Boolean.valueOf(value).booleanValue()); - case PropertyType.DATE: - return new Long(DateField.stringToTime(value)); - case PropertyType.LONG: - return new Long(LongField.stringToLong(value)); - case PropertyType.DOUBLE: - return new Double(DoubleField.stringToDouble(value)); - default: - return value; - } - } + private final String prefix; - /** - * A compound <code>Key</code> that consist of <code>field</code> - * <code>prefix</code> and <code>comparator</code>. - */ - static class Key { + private final Object comparator; - private final String field; - private final String prefix; - private final SortComparator comparator; + /** + * Creates <code>Key</code> for ValueIndex lookup. + */ + Key(String field, String prefix, FieldComparator comparator) + { + this.field = field.intern(); + this.prefix = prefix.intern(); + this.comparator = comparator; + } - /** - * Creates <code>Key</code> for ValueIndex lookup. - */ - Key(String field, String prefix, SortComparator comparator) { - this.field = field.intern(); - this.prefix = prefix.intern(); - this.comparator = comparator; - } + /** + * Returns <code>true</code> if <code>o</code> is a <code>Key</code> + * instance and refers to the same field, prefix and comparator object. + */ + public boolean equals(Object o) + { + if (o instanceof Key) + { + Key other = (Key)o; + return other.field == field && other.prefix == prefix && other.comparator.equals(comparator); + } + return false; + } - /** - * Returns <code>true</code> if <code>o</code> is a <code>Key</code> - * instance and refers to the same field, prefix and comparator object. - */ - public boolean equals(Object o) { - if (o instanceof Key) { - Key other = (Key) o; - return other.field == field - && other.prefix == prefix - && other.comparator.equals(comparator); - } - return false; - } + /** + * Composes a hashcode based on the field, prefix and comparator. + */ + public int hashCode() + { + return field.hashCode() ^ prefix.hashCode() ^ comparator.hashCode(); + } + } - /** - * Composes a hashcode based on the field, prefix and comparator. - */ - public int hashCode() { - return field.hashCode() ^ prefix.hashCode() ^ comparator.hashCode(); - } - } - } Added: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SharedFieldComparatorSource.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SharedFieldComparatorSource.java (rev 0) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SharedFieldComparatorSource.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -0,0 +1,403 @@ +/* + * 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.exoplatform.services.jcr.impl.core.query.lucene; + +import org.apache.lucene.document.Document; +import org.apache.lucene.index.IndexReader; +import org.apache.lucene.search.FieldComparator; +import org.apache.lucene.search.FieldComparatorSource; +import org.exoplatform.services.jcr.dataflow.ItemDataConsumer; +import org.exoplatform.services.jcr.datamodel.IllegalNameException; +import org.exoplatform.services.jcr.datamodel.IllegalPathException; +import org.exoplatform.services.jcr.datamodel.ItemData; +import org.exoplatform.services.jcr.datamodel.ItemType; +import org.exoplatform.services.jcr.datamodel.NodeData; +import org.exoplatform.services.jcr.datamodel.PropertyData; +import org.exoplatform.services.jcr.datamodel.QPath; +import org.exoplatform.services.jcr.datamodel.QPathEntry; +import org.exoplatform.services.jcr.datamodel.ValueData; +import org.exoplatform.services.jcr.impl.Constants; +import org.exoplatform.services.jcr.impl.core.JCRPath; +import org.exoplatform.services.jcr.impl.core.LocationFactory; +import org.exoplatform.services.log.ExoLogger; +import org.exoplatform.services.log.Log; + +import java.io.IOException; +import java.util.List; + +import javax.jcr.InvalidItemStateException; +import javax.jcr.RepositoryException; + +/** + * Created by The eXo Platform SAS + * Author : eXoPlatform + * exo(a)exoplatform.com + * Feb 18, 2012 + */ +public class SharedFieldComparatorSource extends FieldComparatorSource +{ + + /** + * The logger + */ + private static Log LOG = ExoLogger.getLogger("exo.jcr.component.core.SharedFieldSortComparator"); + + /** + * The name of the shared field in the lucene index. + */ + private final String field; + + /** + * The item state manager. + */ + private final ItemDataConsumer ism; + + /** + * LocationFactory. + */ + private final LocationFactory locationFactory; + + /** + * The index internal namespace mappings. + */ + private final NamespaceMappings nsMappings; + + /** + * Creates a new <code>SharedFieldSortComparator</code> for a given shared + * field. + * + * @param fieldname the shared field. + * @param ism the item state manager of this workspace. + * @param hmgr the hierarchy manager of this workspace. + * @param nsMappings the index internal namespace mappings. + */ + public SharedFieldComparatorSource(String fieldname, ItemDataConsumer ism, NamespaceMappings nsMappings) + { + this.field = fieldname; + this.ism = ism; + this.locationFactory = new LocationFactory(nsMappings); + this.nsMappings = nsMappings; + } + + /** + * Create a new <code>FieldComparator</code> for an embedded <code>propertyName</code> + * and a <code>reader</code>. + * + * @param propertyName the relative path to the property to sort on as returned + * by {@link org.apache.jackrabbit.spi.Path#getString()}. + * @return a <code>FieldComparator</code> + * @throws java.io.IOException if an error occurs + */ + @Override + public FieldComparator newComparator(String propertyName, int numHits, int sortPos, boolean reversed) + throws IOException + { + + try + { + QPath path = locationFactory.parseJCRPath(propertyName).getInternalPath(); + SimpleFieldComparator simple = new SimpleFieldComparator(nsMappings.translatePath(path), field, numHits); + if (path.getEntries().length == 1) + { + return simple; + } + else + { + return new CompoundScoreFieldComparator(new FieldComparator[]{simple, + new RelPathFieldComparator(path, numHits)}, numHits); + } + } + catch (IllegalNameException e) + { + throw Util.createIOException(e); + } + catch (RepositoryException e) + { + throw Util.createIOException(e); + } + // + // PathFactory factory = PathFactoryImpl.getInstance(); + // Path path = factory.create(propertyName); + // + // try { + // SimpleFieldComparator simple = new SimpleFieldComparator(nsMappings.translatePath(path), field, numHits); + // + // return path.getLength() == 1 + // ? simple + // : new CompoundScoreFieldComparator( + // new FieldComparator[] { simple, new RelPathFieldComparator(path, numHits) }, numHits); + // + // } + // catch (IllegalNameException e) { + // throw Util.createIOException(e); + // } + } + + // /** + // * Creates a new <code>ScoreDocComparator</code> for an embedded + // * <code>propertyName</code> and a <code>reader</code>. + // * + // */ + // public ScoreDocComparator newComparator(IndexReader reader, String relPath) throws IOException + // { + // + // try + // { + // QPath p = locationFactory.parseJCRPath(relPath).getInternalPath(); + // ScoreDocComparator simple = new SimpleScoreDocComparator(reader, nsMappings.translatePath(p)); + // if (p.getEntries().length == 1) + // { + // return simple; + // } + // else + // { + // return new CompoundScoreDocComparator(reader, new ScoreDocComparator[]{simple, + // new RelPathScoreDocComparator(reader, p)}); + // } + // } + // catch (IllegalNameException e) + // { + // throw Util.createIOException(e); + // } + // catch (RepositoryException e) + // { + // throw Util.createIOException(e); + // } + // } + + private ItemData getItemData(NodeData parent, QPathEntry name, ItemType itemType) throws RepositoryException + { + if (name.getName().equals(JCRPath.PARENT_RELPATH) && name.getNamespace().equals(Constants.NS_DEFAULT_URI)) + { + if (parent.getIdentifier().equals(Constants.ROOT_UUID)) + { + return null; + } + else + { + return ism.getItemData(parent.getParentIdentifier()); + } + } + + return ism.getItemData(parent, name, itemType); + + } + + private ItemData getItemData(NodeData parent, QPath relPath, ItemType itemType) throws RepositoryException + { + + QPathEntry[] relPathEntries = relPath.getEntries(); //relPath.getRelPath(relPath.getDepth()); + + ItemData item = parent; + for (int i = 0; i < relPathEntries.length; i++) + { + if (i == relPathEntries.length - 1) + { + item = getItemData(parent, relPathEntries[i], itemType); + } + else + { + item = getItemData(parent, relPathEntries[i], ItemType.UNKNOWN); + } + + if (item == null) + { + break; + } + + if (item.isNode()) + { + parent = (NodeData)item; + } + else if (i < relPathEntries.length - 1) + { + throw new IllegalPathException("Path can not contains a property as the intermediate element"); + } + } + return item; + + } + + static final class SimpleFieldComparator extends AbstractFieldComparator + { + + /** + * The term look ups of the index segments. + */ + protected SharedFieldCache.ValueIndex[] indexes; + + /** + * The name of the property + */ + private final String propertyName; + + /** + * The name of the field in the index + */ + private final String fieldName; + + /** + * Create a new instance of the <code>FieldComparator</code>. + * + * @param propertyName the name of the property + * @param fieldName the name of the field in the index + * @param numHits the number of values + */ + public SimpleFieldComparator(String propertyName, String fieldName, int numHits) + { + super(numHits); + this.propertyName = propertyName; + this.fieldName = fieldName; + } + + @Override + public void setNextReader(IndexReader reader, int docBase) throws IOException + { + super.setNextReader(reader, docBase); + + indexes = new SharedFieldCache.ValueIndex[readers.size()]; + + String namedValue = FieldNames.createNamedValue(propertyName, ""); + for (int i = 0; i < readers.size(); i++) + { + IndexReader r = readers.get(i); + indexes[i] = SharedFieldCache.INSTANCE.getValueIndex(r, fieldName, namedValue, this); + } + } + + @Override + protected Comparable sortValue(int doc) + { + int idx = readerIndex(doc); + return indexes[idx].getValue(doc - starts[idx]); + } + } + + /** + * Implements a compound <code>FieldComparator</code> which delegates to several + * other comparators. The comparators are asked for a sort value in the + * sequence they are passed to the constructor. + */ + private static final class CompoundScoreFieldComparator extends AbstractFieldComparator + { + private final FieldComparator[] fieldComparators; + + /** + * Create a new instance of the <code>FieldComparator</code>. + * + * @param fieldComparators delegatees + * @param numHits the number of values + */ + public CompoundScoreFieldComparator(FieldComparator[] fieldComparators, int numHits) + { + super(numHits); + this.fieldComparators = fieldComparators; + } + + @Override + public Comparable sortValue(int doc) + { + for (FieldComparator fieldComparator : fieldComparators) + { + if (fieldComparator instanceof FieldComparatorBase) + { + Comparable c = ((FieldComparatorBase)fieldComparator).sortValue(doc); + + if (c != null) + { + return c; + } + } + } + return null; + } + + @Override + public void setNextReader(IndexReader reader, int docBase) throws IOException + { + for (FieldComparator fieldComparator : fieldComparators) + { + fieldComparator.setNextReader(reader, docBase); + } + } + } + + /** + * A <code>FieldComparator</code> which works with order by clauses that use a + * relative path to a property to sort on. + */ + private final class RelPathFieldComparator extends AbstractFieldComparator + { + + /** + * Relative path to the property + */ + private final QPath relPath; + + /** + * Create a new instance of the <code>FieldComparator</code>. + * + * @param propertyName relative path of the property + * @param numHits the number of values + */ + public RelPathFieldComparator(QPath relPath, int numHits) + { + super(numHits); + this.relPath = relPath; + } + + @Override + protected Comparable sortValue(int doc) + { + try + { + int idx = readerIndex(doc); + IndexReader reader = readers.get(idx); + Document document = reader.document(doc - starts[idx], FieldSelectors.UUID); + String uuid = document.get(FieldNames.UUID); + ItemData parent = ism.getItemData(uuid); + if (!parent.isNode()) + { + throw new InvalidItemStateException(); + } + ItemData property = getItemData((NodeData)parent, relPath, ItemType.PROPERTY); + if (property != null) + { + if (property.isNode()) + { + throw new InvalidItemStateException(); + } + PropertyData propertyData = (PropertyData)property; + List<ValueData> values = propertyData.getValues(); + if (values.size() > 0) + { + return Util.getComparable(values.get(0), propertyData.getType()); + } + } + return null; + } + catch (Exception ignore) + { + LOG.error(ignore.getLocalizedMessage(), ignore); + } + + return null; + } + + } + +} Deleted: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SharedFieldSortComparator.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SharedFieldSortComparator.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SharedFieldSortComparator.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -1,428 +0,0 @@ -/* - * 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.exoplatform.services.jcr.impl.core.query.lucene; - -import org.apache.lucene.document.Document; -import org.apache.lucene.index.IndexReader; -import org.apache.lucene.search.ScoreDoc; -import org.apache.lucene.search.ScoreDocComparator; -import org.apache.lucene.search.SortComparator; -import org.apache.lucene.search.SortField; -import org.exoplatform.services.jcr.dataflow.ItemDataConsumer; -import org.exoplatform.services.jcr.datamodel.IllegalNameException; -import org.exoplatform.services.jcr.datamodel.IllegalPathException; -import org.exoplatform.services.jcr.datamodel.ItemData; -import org.exoplatform.services.jcr.datamodel.ItemType; -import org.exoplatform.services.jcr.datamodel.NodeData; -import org.exoplatform.services.jcr.datamodel.PropertyData; -import org.exoplatform.services.jcr.datamodel.QPath; -import org.exoplatform.services.jcr.datamodel.QPathEntry; -import org.exoplatform.services.jcr.datamodel.ValueData; -import org.exoplatform.services.jcr.impl.Constants; -import org.exoplatform.services.jcr.impl.core.JCRPath; -import org.exoplatform.services.jcr.impl.core.LocationFactory; -import org.exoplatform.services.log.ExoLogger; -import org.exoplatform.services.log.Log; - -import java.io.IOException; -import java.util.ArrayList; -import java.util.List; - -import javax.jcr.InvalidItemStateException; -import javax.jcr.RepositoryException; - -/** - * Implements a <code>SortComparator</code> which knows how to sort on a lucene - * field that contains values for multiple properties. - */ -public class SharedFieldSortComparator extends SortComparator -{ - - /** - * The logger - */ - private static Log LOG = ExoLogger.getLogger("exo.jcr.component.core.SharedFieldSortComparator"); - - /** - * The name of the shared field in the lucene index. - */ - private final String field; - - /** - * The item state manager. - */ - private final ItemDataConsumer ism; - - /** - * LocationFactory. - */ - private final LocationFactory locationFactory; - - /** - * The index internal namespace mappings. - */ - private final NamespaceMappings nsMappings; - - /** - * Creates a new <code>SharedFieldSortComparator</code> for a given shared - * field. - * - * @param fieldname the shared field. - * @param ism the item state manager of this workspace. - * @param hmgr the hierarchy manager of this workspace. - * @param nsMappings the index internal namespace mappings. - */ - public SharedFieldSortComparator(String fieldname, ItemDataConsumer ism, NamespaceMappings nsMappings) - { - this.field = fieldname; - this.ism = ism; - this.locationFactory = new LocationFactory(nsMappings); - this.nsMappings = nsMappings; - } - - /** - * Creates a new <code>ScoreDocComparator</code> for an embedded - * <code>propertyName</code> and a <code>reader</code>. - * - * @param reader the index reader. - * @param relPath the relative path to the property to sort on as returned - * by {@link Path#getString()}. - * @return a <code>ScoreDocComparator</code> for the - * @throws IOException if an error occurs while reading from the index. - */ - @Override - public ScoreDocComparator newComparator(IndexReader reader, String relPath) throws IOException - { - - try - { - QPath p = locationFactory.parseJCRPath(relPath).getInternalPath(); - ScoreDocComparator simple = new SimpleScoreDocComparator(reader, nsMappings.translatePath(p)); - if (p.getEntries().length == 1) - { - return simple; - } - else - { - return new CompoundScoreDocComparator(reader, new ScoreDocComparator[]{simple, - new RelPathScoreDocComparator(reader, p)}); - } - } - catch (IllegalNameException e) - { - throw Util.createIOException(e); - } - catch (RepositoryException e) - { - throw Util.createIOException(e); - } - } - - /** - * @throws UnsupportedOperationException always. - */ - @Override - protected Comparable getComparable(String termtext) - { - throw new UnsupportedOperationException(); - } - - /** - * Checks if <code>reader</code> is of type {@link MultiIndexReader} and if - * that's the case calls this method recursively for each reader within the - * multi index reader; otherwise the reader is simply added to the list. - * - * @param readers the list of index readers. - * @param reader the reader to check. - */ - private static void getIndexReaders(List readers, IndexReader reader) - { - if (reader instanceof MultiIndexReader) - { - IndexReader[] r = ((MultiIndexReader)reader).getIndexReaders(); - for (int i = 0; i < r.length; i++) - { - getIndexReaders(readers, r[i]); - } - } - else - { - readers.add(reader); - } - } - - /** - * Abstract base class of {@link ScoreDocComparator} implementations. - */ - abstract class AbstractScoreDocComparator implements ScoreDocComparator - { - - /** - * The index readers. - */ - protected final List readers = new ArrayList(); - - /** - * The document number starts for the {@link #readers}. - */ - protected final int[] starts; - - public AbstractScoreDocComparator(IndexReader reader) throws IOException - { - getIndexReaders(readers, reader); - - int maxDoc = 0; - this.starts = new int[readers.size() + 1]; - - for (int i = 0; i < readers.size(); i++) - { - IndexReader r = (IndexReader)readers.get(i); - starts[i] = maxDoc; - maxDoc += r.maxDoc(); - } - starts[readers.size()] = maxDoc; - } - - /** - * Compares sort values of <code>i</code> and <code>j</code>. If the - * sort values have differing types, then the sort order is defined on - * the type itself by calling <code>compareTo()</code> on the respective - * type class names. - * - * @param i first score doc. - * @param j second score doc. - * @return a negative integer if <code>i</code> should come before - * <code>j</code><br> a positive integer if <code>i</code> - * should come after <code>j</code><br> <code>0</code> if they - * are equal - */ - public int compare(ScoreDoc i, ScoreDoc j) - { - return Util.compare(sortValue(i), sortValue(j)); - } - - public int sortType() - { - return SortField.CUSTOM; - } - - /** - * Returns the reader index for document <code>n</code>. - * - * @param n document number. - * @return the reader index. - */ - protected int readerIndex(int n) - { - int lo = 0; - int hi = readers.size() - 1; - - while (hi >= lo) - { - int mid = (lo + hi) >> 1; - int midValue = starts[mid]; - if (n < midValue) - { - hi = mid - 1; - } - else if (n > midValue) - { - lo = mid + 1; - } - else - { - while (mid + 1 < readers.size() && starts[mid + 1] == midValue) - { - mid++; - } - return mid; - } - } - return hi; - } - } - - /** - * A score doc comparator that works for order by clauses with properties - * directly on the result nodes. - */ - private final class SimpleScoreDocComparator extends AbstractScoreDocComparator - { - - /** - * The term look ups of the index segments. - */ - protected final SharedFieldCache.ValueIndex[] indexes; - - public SimpleScoreDocComparator(IndexReader reader, String propertyName) throws IOException - { - super(reader); - this.indexes = new SharedFieldCache.ValueIndex[readers.size()]; - - String namedValue = FieldNames.createNamedValue(propertyName, ""); - for (int i = 0; i < readers.size(); i++) - { - IndexReader r = (IndexReader)readers.get(i); - indexes[i] = SharedFieldCache.INSTANCE.getValueIndex(r, field, namedValue, SharedFieldSortComparator.this); - } - } - - /** - * Returns the index term for the score doc <code>i</code>. - * - * @param i the score doc. - * @return the sort value if available. - */ - public Comparable sortValue(ScoreDoc i) - { - int idx = readerIndex(i.doc); - return indexes[idx].getValue(i.doc - starts[idx]); - } - } - - /** - * A score doc comparator that works with order by clauses that use a - * relative path to a property to sort on. - */ - private final class RelPathScoreDocComparator extends AbstractScoreDocComparator - { - - private final QPath relPath; - - public RelPathScoreDocComparator(IndexReader reader, QPath relPath) throws IOException - { - super(reader); - this.relPath = relPath; - } - - /** - * Returns the sort value for the given {@link ScoreDoc}. The value is - * retrieved from the item state manager. - * - * @param i the score doc. - * @return the sort value for the score doc. - */ - public Comparable sortValue(ScoreDoc i) - { - try - { - int idx = readerIndex(i.doc); - IndexReader reader = (IndexReader)readers.get(idx); - Document doc = reader.document(i.doc - starts[idx], FieldSelectors.UUID); - String uuid = doc.get(FieldNames.UUID); - ItemData parent = ism.getItemData(uuid); - if (!parent.isNode()) - throw new InvalidItemStateException(); - ItemData property = getItemData((NodeData)parent, relPath, ItemType.PROPERTY); - if (property != null) - { - if (property.isNode()) - throw new InvalidItemStateException(); - PropertyData propertyData = (PropertyData)property; - List<ValueData> values = propertyData.getValues(); - if (values.size() > 0) - { - return Util.getComparable(values.get(0), propertyData.getType()); - } - } - return null; - } - catch (Exception e) - { - LOG.error(e.getLocalizedMessage(), e); - return null; - } - } - } - - private ItemData getItemData(NodeData parent, QPathEntry name, ItemType itemType) throws RepositoryException - { - if (name.getName().equals(JCRPath.PARENT_RELPATH) && name.getNamespace().equals(Constants.NS_DEFAULT_URI)) - { - if (parent.getIdentifier().equals(Constants.ROOT_UUID)) - return null; - else - return ism.getItemData(parent.getParentIdentifier()); - } - - return ism.getItemData(parent, name, itemType); - - } - - private ItemData getItemData(NodeData parent, QPath relPath, ItemType itemType) throws RepositoryException - { - - QPathEntry[] relPathEntries = relPath.getEntries(); //relPath.getRelPath(relPath.getDepth()); - - ItemData item = parent; - for (int i = 0; i < relPathEntries.length; i++) - { - if (i == relPathEntries.length - 1) - { - item = getItemData(parent, relPathEntries[i], itemType); - } - else - { - item = getItemData(parent, relPathEntries[i], ItemType.UNKNOWN); - } - - if (item == null) - break; - - if (item.isNode()) - parent = (NodeData)item; - else if (i < relPathEntries.length - 1) - throw new IllegalPathException("Path can not contains a property as the intermediate element"); - } - return item; - - } - - /** - * Implements a compound score doc comparator that delegates to several - * other comparators. The comparators are asked for a sort value in the - * sequence they are passed to the constructor. The first non-null value - * will be returned by {@link #sortValue(ScoreDoc)}. - */ - private final class CompoundScoreDocComparator extends AbstractScoreDocComparator - { - - private final ScoreDocComparator[] comparators; - - public CompoundScoreDocComparator(IndexReader reader, ScoreDocComparator[] comparators) throws IOException - { - super(reader); - this.comparators = comparators; - } - - /** - * {@inheritDoc} - */ - public Comparable sortValue(ScoreDoc i) - { - for (int j = 0; j < comparators.length; j++) - { - Comparable c = comparators[j].sortValue(i); - if (c != null) - { - return c; - } - } - return null; - } - } -} Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SingletonTokenStream.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SingletonTokenStream.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SingletonTokenStream.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -18,6 +18,8 @@ import org.apache.lucene.analysis.Token; import org.apache.lucene.analysis.TokenStream; +import org.apache.lucene.analysis.tokenattributes.PayloadAttribute; +import org.apache.lucene.analysis.tokenattributes.TermAttribute; import org.apache.lucene.index.Payload; import org.exoplatform.services.jcr.impl.Constants; @@ -33,7 +35,6 @@ */ public final class SingletonTokenStream extends TokenStream implements Externalizable { - /** * The string value of the token. */ @@ -44,26 +45,54 @@ */ private Payload payload; - private boolean hasNext = true; + /** + * The term attribute of the current token + */ + private TermAttribute termAttribute; /** - * for serialization + * The payload attribute of the current token */ + private PayloadAttribute payloadAttribute; + + private boolean consumed = false; + + /** + * Default constructor for serialization + */ public SingletonTokenStream() { + } /** + * Creates a new SingleTokenStream with the given value and payload. + * + * @param value + * the string value that will be returned with the token. + * @param payload + * the payload that will be attached to this token + */ + public SingletonTokenStream(String value, Payload payload) + { + this.value = value; + this.payload = payload; + termAttribute = (TermAttribute)addAttribute(TermAttribute.class); + payloadAttribute = (PayloadAttribute)addAttribute(PayloadAttribute.class); + } + + /** * Creates a new SingleTokenStream with the given value and a property * <code>type</code>. - * - * @param value the string value that will be returned with the token. - * @param type the JCR property type. + * + * @param value + * the string value that will be returned with the token. + * @param type + * the JCR property type. */ public SingletonTokenStream(String value, int type) { - this.value = value; - this.payload = new Payload(new PropertyMetaData(type).toByteArray()); + this(value, new Payload(new PropertyMetaData(type).toByteArray())); } /** @@ -71,33 +100,51 @@ * * @param t the token. */ + @Deprecated public SingletonTokenStream(Token t) { - this.value = t.term(); - this.payload = t.getPayload(); + this(t.term(), t.getPayload()); } + @Override + public boolean incrementToken() throws IOException + { + if (consumed) + { + return false; + } + clearAttributes(); + termAttribute.setTermBuffer(value); + payloadAttribute.setPayload(payload); + consumed = true; + return true; + } + /** * {@inheritDoc} */ - public Token next(Token reusableToken) throws IOException + @Override + public void reset() throws IOException { - if (hasNext) - { - reusableToken.clear(); - reusableToken.setTermBuffer(value); - reusableToken.setPayload(payload); - reusableToken.setStartOffset(0); - reusableToken.setEndOffset(value.length()); - hasNext = false; - return reusableToken; - } - return null; + consumed = false; } /** * {@inheritDoc} */ + @Override + public void close() throws IOException + { + consumed = true; + value = null; + payload = null; + payloadAttribute = null; + termAttribute = null; + } + + /** + * {@inheritDoc} + */ public void readExternal(ObjectInput in) throws IOException, ClassNotFoundException { payload = (Payload)in.readObject(); Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SortedLuceneQueryHits.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SortedLuceneQueryHits.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/SortedLuceneQueryHits.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -20,7 +20,7 @@ import org.apache.lucene.search.Query; import org.apache.lucene.search.ScoreDoc; import org.apache.lucene.search.Sort; -import org.apache.lucene.search.TopFieldDocCollector; +import org.apache.lucene.search.TopFieldCollector; import org.slf4j.Logger; import org.slf4j.LoggerFactory; @@ -154,8 +154,9 @@ private void getHits() throws IOException { - TopFieldDocCollector collector = new TopFieldDocCollector(reader, sort, numHits); + TopFieldCollector collector = TopFieldCollector.create(sort, numHits, false, true, false, false); searcher.search(query, collector); + this.size = collector.getTotalHits(); ScoreDoc[] docs = collector.topDocs().scoreDocs; for (int i = scoreNodes.size(); i < docs.length; i++) @@ -167,5 +168,6 @@ log.debug("getHits() {}/{}", new Integer(scoreNodes.size()), new Integer(numHits)); // double hits for next round numHits *= 2; + } } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/WeightedHighlighter.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/WeightedHighlighter.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/WeightedHighlighter.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -138,7 +138,7 @@ break; } } - bestFragments.insert(fi); + bestFragments.insertWithOverflow(fi); } } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/WildcardQuery.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/WildcardQuery.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/WildcardQuery.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -17,7 +17,6 @@ package org.exoplatform.services.jcr.impl.core.query.lucene; import org.apache.lucene.index.IndexReader; -import org.apache.lucene.index.Term; import org.apache.lucene.index.TermDocs; import org.apache.lucene.index.TermEnum; import org.apache.lucene.search.BooleanQuery; @@ -29,6 +28,7 @@ import org.apache.lucene.search.Searcher; import org.apache.lucene.search.Similarity; import org.apache.lucene.search.Weight; +import org.apache.lucene.util.ToStringUtils; import org.slf4j.Logger; import org.slf4j.LoggerFactory; @@ -132,13 +132,24 @@ @Override public Query rewrite(IndexReader reader) throws IOException { - Query stdWildcardQuery = new MultiTermQuery(new Term(field, pattern)) + Query stdWildcardQuery = new MultiTermQuery() { @Override protected FilteredTermEnum getEnum(IndexReader reader) throws IOException { return new WildcardTermEnum(reader, field, propName, pattern, transform); } + + /** Prints a user-readable version of this query. */ + @Override + public String toString(String field) + { + StringBuffer buffer = new StringBuffer(); + buffer.append(field); + buffer.append(':'); + buffer.append(ToStringUtils.boost(getBoost())); + return buffer.toString(); + } }; try { @@ -338,18 +349,27 @@ * {@inheritDoc} */ @Override - public boolean next() throws IOException + public int nextDoc() throws IOException { + if (nextDoc == NO_MORE_DOCS) + { + return nextDoc; + } + calculateHits(); nextDoc = hits.nextSetBit(nextDoc + 1); - return nextDoc > -1; + if (nextDoc < 0) + { + nextDoc = NO_MORE_DOCS; + } + return nextDoc; } /** * {@inheritDoc} */ @Override - public int doc() + public int docID() { return nextDoc; } @@ -367,24 +387,23 @@ * {@inheritDoc} */ @Override - public boolean skipTo(int target) throws IOException + public int advance(int target) throws IOException { + if (nextDoc == NO_MORE_DOCS) + { + return nextDoc; + } + calculateHits(); nextDoc = hits.nextSetBit(target); - return nextDoc > -1; + if (nextDoc < 0) + { + nextDoc = NO_MORE_DOCS; + } + return nextDoc; } /** - * Returns an empty Explanation object. - * @return an empty Explanation object. - */ - @Override - public Explanation explain(int doc) - { - return new Explanation(); - } - - /** * Calculates the ids of the documents matching this wildcard query. * @throws IOException if an error occurs while reading from the index. */ Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/directory/FSDirectoryManager.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/directory/FSDirectoryManager.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/directory/FSDirectoryManager.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -18,6 +18,8 @@ import org.apache.lucene.store.Directory; import org.apache.lucene.store.FSDirectory; +import org.apache.lucene.store.FSLockFactory; +import org.apache.lucene.store.LockFactory; import org.apache.lucene.store.NativeFSLockFactory; import org.exoplatform.commons.utils.PropertyManager; import org.exoplatform.commons.utils.SecurityHelper; @@ -27,6 +29,7 @@ import java.io.File; import java.io.FileFilter; import java.io.IOException; +import java.lang.reflect.Constructor; import java.security.PrivilegedAction; import java.security.PrivilegedExceptionAction; @@ -37,38 +40,48 @@ public class FSDirectoryManager implements DirectoryManager { - /** - * The full qualified name of the lock factory to use by default, if not - * specified org.apache.lucene.store.NativeFSLockFactory will be used - */ - public static final String LOCK_FACTORY_CLASS; + private static Class<? extends FSDirectory> FS_DIRECTORY_CLASS; - /** - * The full qualified name of the lock factory to use by default, if not - * specified org.apache.lucene.store.NativeFSLockFactory will be used - */ - public static final String FS_DIRECTORY_CLASS; + private static Class<? extends LockFactory> LOCK_FACTORY_CLASS; - /** - * Static block, used to initialize (map) org.exoplatform.jcr.lucene* - * properties to org.apache.lucene.* and make it only once at a system - * start - * - * Required to set custom Index Directory and Lock Factory implementations for Lucene 2.x. - */ static { // get eXo system properties - LOCK_FACTORY_CLASS = PropertyManager.getProperty("org.exoplatform.jcr.lucene.store.FSDirectoryLockFactoryClass"); - FS_DIRECTORY_CLASS = PropertyManager.getProperty("org.exoplatform.jcr.lucene.FSDirectory.class"); + String lockFactoryClassName = + PropertyManager.getProperty("org.exoplatform.jcr.lucene.store.FSDirectoryLockFactoryClass"); + String fsDirectoryClassName = PropertyManager.getProperty("org.exoplatform.jcr.lucene.FSDirectory.class"); // map to Lucene ones. Works only with Lucene 2.x. - if (LOCK_FACTORY_CLASS != null) + if (lockFactoryClassName != null) { - PropertyManager.setProperty("org.apache.lucene.store.FSDirectoryLockFactoryClass", LOCK_FACTORY_CLASS); + try + { + // avoid case when abstract base class used + if (!FSLockFactory.class.getName().equals(lockFactoryClassName)) + { + LOCK_FACTORY_CLASS = (Class<? extends LockFactory>)Class.forName(lockFactoryClassName); + } + } + catch (ClassNotFoundException e) + { + throw new RuntimeException("cannot load LockFactory class: " + e.toString(), e); + } } - if (FS_DIRECTORY_CLASS != null) + + if (fsDirectoryClassName != null) { - PropertyManager.setProperty("org.apache.lucene.FSDirectory.class", FS_DIRECTORY_CLASS); + try + { + // avoid case when abstract base class used + if (!FSDirectory.class.getName().equals(fsDirectoryClassName)) + { + FS_DIRECTORY_CLASS = (Class<? extends FSDirectory>)Class.forName(fsDirectoryClassName); + } + // else rely on Lucene FSDirectory instantiation + } + catch (ClassNotFoundException e) + { + throw new RuntimeException("cannot load FSDirectory class: " + e.toString(), e); + } } } @@ -133,15 +146,18 @@ throw new IOException("Cannot create directory: " + dir); } } - // if both not defined, using FSDirectory.open - if (FS_DIRECTORY_CLASS == null && LOCK_FACTORY_CLASS == null) + LockFactory lockFactory = + (LOCK_FACTORY_CLASS == null) ? new NativeFSLockFactory() : (LockFactory)LOCK_FACTORY_CLASS.newInstance(); + + if (FS_DIRECTORY_CLASS == null) { - return FSDirectory.open(dir, new NativeFSLockFactory(dir)); + return FSDirectory.open(dir, lockFactory); } - // LOCK FACTORY only defined, using deprecated getDirectory method else { - return FSDirectory.getDirectory(dir, LOCK_FACTORY_CLASS != null ? null : new NativeFSLockFactory(dir)); + Constructor<? extends FSDirectory> constructor = + FS_DIRECTORY_CLASS.getConstructor(File.class, LockFactory.class); + return constructor.newInstance(dir, lockFactory); } } }); Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/hits/ScorerHits.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/hits/ScorerHits.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/hits/ScorerHits.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -16,47 +16,61 @@ */ package org.exoplatform.services.jcr.impl.core.query.lucene.hits; +import org.apache.lucene.search.DocIdSetIterator; +import org.apache.lucene.search.Scorer; + import java.io.IOException; -import org.apache.lucene.search.Scorer; - /** * Wraps a {@link org.apache.lucene.search.Scorer} in a {@link Hits} instance. */ -public class ScorerHits implements Hits { +public class ScorerHits implements Hits +{ - private final Scorer scorer; + private final Scorer scorer; - public ScorerHits(Scorer scorer) { - this.scorer = scorer; - } + public ScorerHits(Scorer scorer) + { + this.scorer = scorer; + } - /** - * {@inheritDoc} - */ - public void set(int doc) { - throw new UnsupportedOperationException(); - } + /** + * {@inheritDoc} + */ + public void set(int doc) + { + throw new UnsupportedOperationException(); + } - /** - * {@inheritDoc} - */ - public int next() throws IOException { - if (scorer.next()) { - return scorer.doc(); - } else { - return -1; - } - } + /** + * {@inheritDoc} + */ + public int next() throws IOException + { + int docId = scorer.nextDoc(); + if (docId != DocIdSetIterator.NO_MORE_DOCS) + { + return docId; + } + else + { + return -1; + } + } - /** - * {@inheritDoc} - */ - public int skipTo(int target) throws IOException { - if (scorer.skipTo(target)) { - return scorer.doc(); - } else { - return -1; - } - } + /** + * {@inheritDoc} + */ + public int skipTo(int target) throws IOException + { + int docId = scorer.advance(target); + if (docId != DocIdSetIterator.NO_MORE_DOCS) + { + return docId; + } + else + { + return -1; + } + } } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/spell/LuceneSpellChecker.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/spell/LuceneSpellChecker.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/spell/LuceneSpellChecker.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -16,8 +16,10 @@ */ package org.exoplatform.services.jcr.impl.core.query.lucene.spell; -import org.apache.lucene.analysis.Token; import org.apache.lucene.analysis.TokenStream; +import org.apache.lucene.analysis.tokenattributes.OffsetAttribute; +import org.apache.lucene.analysis.tokenattributes.PositionIncrementAttribute; +import org.apache.lucene.analysis.tokenattributes.TermAttribute; import org.apache.lucene.index.IndexReader; import org.apache.lucene.search.spell.Dictionary; import org.apache.lucene.search.spell.LuceneDictionary; @@ -54,7 +56,7 @@ * Logger instance for this class. */ private static final Log LOG = ExoLogger.getLogger("exo.jcr.component.core.LuceneSpellChecker"); - + public static final class FiveSecondsRefreshInterval extends LuceneSpellChecker { public FiveSecondsRefreshInterval() @@ -256,7 +258,7 @@ public Object run() throws Exception { spellIndexDirectory = handler.getDirectoryManager().getDirectory("spellchecker"); - + if (IndexReader.indexExists(spellIndexDirectory)) { lastRefresh = System.currentTimeMillis(); @@ -283,7 +285,7 @@ { // tokenize the statement (field name doesn't matter actually...) List<String> words = new ArrayList<String>(); - List<Token> tokens = new ArrayList<Token>(); + List<TokenData> tokens = new ArrayList<TokenData>(); tokenize(statement, words, tokens); String[] suggestions = check(words.toArray(new String[words.size()])); @@ -294,11 +296,11 @@ StringBuffer sb = new StringBuffer(statement); for (int i = suggestions.length - 1; i >= 0; i--) { - Token t = tokens.get(i); + TokenData t = tokens.get(i); // only replace if word acutally changed - if (!t.termText().equalsIgnoreCase(suggestions[i])) + if (!t.word.equalsIgnoreCase(suggestions[i])) { - sb.replace(t.startOffset(), t.endOffset(), suggestions[i]); + sb.replace(t.startOffset, t.endOffset, suggestions[i]); } } // if suggestion is same as a statement return null @@ -357,41 +359,74 @@ * @throws IOException * if an error occurs while parsing the statement. */ - private void tokenize(String statement, List<String> words, List<Token> tokens) throws IOException + private void tokenize(String statement, List<String> words, List<TokenData> tokens) throws IOException { TokenStream ts = handler.getTextAnalyzer().tokenStream(FieldNames.FULLTEXT, new StringReader(statement)); + TermAttribute term = (TermAttribute)ts.getAttribute(TermAttribute.class); + PositionIncrementAttribute positionIncrement = + (PositionIncrementAttribute)ts.getAttribute(PositionIncrementAttribute.class); + OffsetAttribute offset = (OffsetAttribute)ts.getAttribute(OffsetAttribute.class); try { - Token t; - while ((t = ts.next()) != null) + while (ts.incrementToken()) { - String origWord = statement.substring(t.startOffset(), t.endOffset()); - if (t.getPositionIncrement() > 0) + + String word = term.term(); + // while ((t = ts.next()) != null) + // { + String origWord = statement.substring(offset.startOffset(), offset.endOffset()); + if (positionIncrement.getPositionIncrement() > 0) { - words.add(t.termText()); - tokens.add(t); + words.add(word); + tokens.add(new TokenData(offset.startOffset(), offset.endOffset(), term.term())); } else { // very simple implementation: use termText with length // closer to original word - Token current = tokens.get(tokens.size() - 1); - if (Math.abs(origWord.length() - current.termText().length()) > Math.abs(origWord.length() - - t.termText().length())) + TokenData current = tokens.get(tokens.size() - 1); + if (Math.abs(origWord.length() - current.termLength()) > Math.abs(origWord.length() - word.length())) { // replace current token and word - words.set(words.size() - 1, t.termText()); - tokens.set(tokens.size() - 1, t); + words.set(words.size() - 1, word); + tokens + .set(tokens.size() - 1, new TokenData(offset.startOffset(), offset.endOffset(), term.term())); } } } } finally { + ts.end(); ts.close(); } } + class TokenData + { + int startOffset; + + int endOffset; + + String word; + + public TokenData(int startOffset, int endOffset, String word) + { + this.startOffset = startOffset; + this.endOffset = endOffset; + this.word = word; + } + + /** + * @return + */ + public int termLength() + { + return word.length(); + } + + } + /** * Checks the spelling of the passed <code>words</code> and returns a * suggestion. Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/synonym/WordNetSynonyms.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/synonym/WordNetSynonyms.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/main/java/org/exoplatform/services/jcr/impl/core/query/lucene/synonym/WordNetSynonyms.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -16,7 +16,7 @@ */ package org.exoplatform.services.jcr.impl.core.query.lucene.synonym; -import org.apache.lucene.index.memory.SynonymMap; +import org.apache.lucene.wordnet.SynonymMap; import org.exoplatform.services.jcr.impl.core.query.lucene.SynonymProvider; import org.exoplatform.services.log.ExoLogger; import org.exoplatform.services.log.Log; Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/BaseStandaloneTest.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/BaseStandaloneTest.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/BaseStandaloneTest.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -366,7 +366,9 @@ } if (dread < eread) + { dbuff = new byte[eread - dread]; + } } } @@ -446,7 +448,9 @@ for (String nodeMixin : nodeMixins) { if (mixin.equals(nodeMixin)) + { continue nextMixin; + } } fail("Mixin '" + mixin + "' isn't accessible"); @@ -530,61 +534,52 @@ */ public class TestWorkspaceDataContainer implements WorkspaceDataContainer { - @Override + public String getInfo() { return null; } - @Override public String getName() { return null; } - @Override public String getUniqueName() { return null; } - @Override public String getStorageVersion() { return null; } - @Override public Calendar getCurrentTime() { return null; } - @Override public boolean isSame(WorkspaceDataContainer another) { return false; } - @Override public WorkspaceStorageConnection openConnection() throws RepositoryException { return null; } - @Override public WorkspaceStorageConnection openConnection(boolean readOnly) throws RepositoryException { return null; } - @Override public WorkspaceStorageConnection reuseConnection(WorkspaceStorageConnection original) throws RepositoryException { return null; } - @Override public boolean isCheckSNSNewConnection() { return false; @@ -601,145 +596,123 @@ public class TestWorkspaceStorageConnection implements WorkspaceStorageConnection { - @Override public ItemData getItemData(NodeData parentData, QPathEntry name, ItemType itemType) throws RepositoryException, IllegalStateException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public ItemData getItemData(String identifier) throws RepositoryException, IllegalStateException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public List<NodeData> getChildNodesData(NodeData parent) throws RepositoryException, IllegalStateException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public List<NodeData> getChildNodesData(NodeData parent, List<QPathEntryFilter> pattern) throws RepositoryException, IllegalStateException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public int getChildNodesCount(NodeData parent) throws RepositoryException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public int getLastOrderNumber(NodeData parent) throws RepositoryException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public List<PropertyData> getChildPropertiesData(NodeData parent) throws RepositoryException, IllegalStateException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public List<PropertyData> getChildPropertiesData(NodeData parent, List<QPathEntryFilter> pattern) throws RepositoryException, IllegalStateException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public List<PropertyData> listChildPropertiesData(NodeData parent) throws RepositoryException, IllegalStateException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public List<PropertyData> getReferencesData(String nodeIdentifier) throws RepositoryException, IllegalStateException, UnsupportedOperationException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public void add(NodeData data) throws RepositoryException, UnsupportedOperationException, InvalidItemStateException, IllegalStateException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public void add(PropertyData data) throws RepositoryException, UnsupportedOperationException, InvalidItemStateException, IllegalStateException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public void update(NodeData data) throws RepositoryException, UnsupportedOperationException, InvalidItemStateException, IllegalStateException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public void update(PropertyData data) throws RepositoryException, UnsupportedOperationException, InvalidItemStateException, IllegalStateException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public void rename(NodeData data) throws RepositoryException, UnsupportedOperationException, InvalidItemStateException, IllegalStateException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public void delete(NodeData data) throws RepositoryException, UnsupportedOperationException, InvalidItemStateException, IllegalStateException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public void delete(PropertyData data) throws RepositoryException, UnsupportedOperationException, InvalidItemStateException, IllegalStateException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public void commit() throws IllegalStateException, RepositoryException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public void rollback() throws IllegalStateException, RepositoryException { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public void close() throws IllegalStateException, RepositoryException { } - @Override public boolean isOpened() { throw new UnsupportedOperationException("TestWorkspaceStorageConnection: operation is unsupported."); } - @Override public List<ACLHolder> getACLHolders() throws RepositoryException, IllegalStateException, UnsupportedOperationException { @@ -749,7 +722,7 @@ /** * @see org.exoplatform.services.jcr.storage.WorkspaceStorageConnection#getValue(java.lang.String, int, int) */ - @Override + public ValueData getValue(String propertyId, int orderNumb, int persistedVersion) throws IllegalStateException, RepositoryException { @@ -759,7 +732,7 @@ /** * @see org.exoplatform.services.jcr.storage.WorkspaceStorageConnection#getChildNodesDataByPage(org.exoplatform.services.jcr.datamodel.NodeData, int, int, java.util.List) */ - @Override + public boolean getChildNodesDataByPage(NodeData parent, int fromOrderNum, int toOrderNum, List<NodeData> childs) throws RepositoryException { Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/api/core/query/lucene/SlowQueryHandler.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/api/core/query/lucene/SlowQueryHandler.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/api/core/query/lucene/SlowQueryHandler.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -100,7 +100,7 @@ public void apply(ChangesHolder changes) throws RepositoryException, IOException { // TODO Auto-generated method stub - + } public ChangesHolder getChanges(Iterator<String> remove, Iterator<NodeData> add) @@ -123,16 +123,14 @@ public void setOnline(boolean isOnline, boolean allowQuery, boolean dropStaleIndexes) throws IOException { // TODO Auto-generated method stub - + } /** * @see org.exoplatform.services.jcr.impl.core.query.QueryHandler#checkIndex(org.exoplatform.services.jcr.dataflow.ItemDataConsumer, boolean, InspectionLog) */ - @Override public void checkIndex(ItemDataConsumer itemStateManager, boolean isSystem, InspectionReport inspectionLog) - throws RepositoryException, - IOException + throws RepositoryException, IOException { // do nothing } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/BaseQueryTest.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/BaseQueryTest.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/BaseQueryTest.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -16,12 +16,12 @@ */ package org.exoplatform.services.jcr.impl.core.query; -import org.apache.lucene.document.Document; import org.apache.lucene.index.IndexReader; import org.apache.lucene.index.Term; -import org.apache.lucene.search.Hits; import org.apache.lucene.search.IndexSearcher; +import org.apache.lucene.search.ScoreDoc; import org.apache.lucene.search.TermQuery; +import org.apache.lucene.search.TopDocs; import org.exoplatform.services.jcr.JcrImplBaseTest; import org.exoplatform.services.jcr.impl.core.query.lucene.FieldNames; import org.exoplatform.services.jcr.impl.core.query.lucene.SearchIndex; @@ -52,21 +52,21 @@ protected SearchIndex defaultSearchIndex; - protected Document getDocument(String nodeIdentifer, boolean includeSystemIndex) throws IOException, + protected ScoreDoc getDocument(String nodeIdentifer, boolean includeSystemIndex) throws IOException, RepositoryException { IndexReader reader = defaultSearchIndex.getIndexReader(); IndexSearcher is = new IndexSearcher(reader); TermQuery query = new TermQuery(new Term(FieldNames.UUID, nodeIdentifer)); - Hits result = is.search(query); + TopDocs result = is.search(query, null, Integer.MAX_VALUE); try { - if (result.length() == 1) + if (result.totalHits == 1) { - return result.doc(0); + return result.scoreDocs[0]; } - else if (result.length() > 1) + else if (result.totalHits > 1) { throw new RepositoryException("Results more then one"); } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestArabicSearch.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestArabicSearch.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestArabicSearch.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -17,12 +17,12 @@ package org.exoplatform.services.jcr.impl.core.query; -import org.apache.lucene.document.Document; import org.apache.lucene.index.IndexReader; import org.apache.lucene.index.Term; -import org.apache.lucene.search.Hits; import org.apache.lucene.search.IndexSearcher; +import org.apache.lucene.search.ScoreDoc; import org.apache.lucene.search.TermQuery; +import org.apache.lucene.search.TopDocs; import org.exoplatform.services.jcr.impl.core.NodeImpl; import org.exoplatform.services.jcr.impl.core.query.lucene.FieldNames; import org.exoplatform.services.jcr.impl.core.query.lucene.Util; @@ -67,21 +67,21 @@ String word = "\u0627\u0644\u0644\u0627\u062a\u064a\u0646\u064a\u0629"; // Check is node indexed - Document doc = getDocument(cont.getInternalIdentifier(), false); + ScoreDoc doc = getDocument(cont.getInternalIdentifier(), false); assertNotNull("Node is not indexed", doc); IndexReader reader = defaultSearchIndex.getIndexReader(); IndexSearcher is = new IndexSearcher(reader); TermQuery query = new TermQuery(new Term(FieldNames.FULLTEXT, word)); - Hits result = is.search(query); - assertEquals(1, result.length()); + TopDocs search = is.search(query, null, Integer.MAX_VALUE); + assertEquals(1, search.totalHits); QueryManager qman = this.workspace.getQueryManager(); Query q = qman.createQuery("SELECT * FROM nt:resource " + " WHERE CONTAINS(., '" + word + "')", Query.SQL); QueryResult res = q.execute(); assertEquals(1, res.getNodes().getSize()); - + is.close(); Util.closeOrRelease(reader); } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestDateSearch.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestDateSearch.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestDateSearch.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -17,12 +17,12 @@ package org.exoplatform.services.jcr.impl.core.query; -import org.apache.lucene.document.Document; import org.apache.lucene.index.IndexReader; import org.apache.lucene.index.Term; -import org.apache.lucene.search.Hits; import org.apache.lucene.search.IndexSearcher; +import org.apache.lucene.search.ScoreDoc; import org.apache.lucene.search.TermQuery; +import org.apache.lucene.search.TopDocs; import org.exoplatform.services.jcr.impl.core.NodeImpl; import org.exoplatform.services.jcr.impl.core.query.lucene.FieldNames; import org.exoplatform.services.jcr.impl.core.query.lucene.Util; @@ -64,22 +64,22 @@ String word = "2005-10-02".toLowerCase();// "ronaldo";//-10-06T00:00:00.000+0300 // Check is node indexed - Document doc = getDocument(cont.getInternalIdentifier(), false); + ScoreDoc doc = getDocument(cont.getInternalIdentifier(), false); assertNotNull("Node is not indexed", doc); System.out.println("its doc " + doc); IndexReader reader = defaultSearchIndex.getIndexReader(); IndexSearcher is = new IndexSearcher(reader); TermQuery query = new TermQuery(new Term(FieldNames.FULLTEXT, word)); - Hits result = is.search(query); - assertEquals(1, result.length()); + TopDocs topDocs = is.search(query, Integer.MAX_VALUE); + assertEquals(1, topDocs.totalHits); QueryManager qman = this.workspace.getQueryManager(); Query q = qman.createQuery("SELECT * FROM nt:resource " + " WHERE CONTAINS(., '" + word + "')", Query.SQL); QueryResult res = q.execute(); assertEquals(1, res.getNodes().getSize()); - + is.close(); Util.closeOrRelease(reader); } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestExcelFileSearch.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestExcelFileSearch.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestExcelFileSearch.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -17,12 +17,12 @@ package org.exoplatform.services.jcr.impl.core.query; -import org.apache.lucene.document.Document; import org.apache.lucene.index.IndexReader; import org.apache.lucene.index.Term; -import org.apache.lucene.search.Hits; import org.apache.lucene.search.IndexSearcher; +import org.apache.lucene.search.ScoreDoc; import org.apache.lucene.search.TermQuery; +import org.apache.lucene.search.TopDocs; import org.exoplatform.services.document.DocumentReader; import org.exoplatform.services.document.DocumentReaderService; import org.exoplatform.services.document.impl.MSExcelDocumentReader; @@ -93,14 +93,14 @@ String word = "eric"; // Check is node indexed - Document doc = getDocument(cont.getInternalIdentifier(), false); + ScoreDoc doc = getDocument(cont.getInternalIdentifier(), false); assertNotNull("Node is not indexed", doc); IndexReader reader = defaultSearchIndex.getIndexReader(); IndexSearcher is = new IndexSearcher(reader); TermQuery query = new TermQuery(new Term(FieldNames.FULLTEXT, word)); - Hits result = is.search(query); - assertEquals(1, result.length()); + TopDocs topDocs = is.search(query, null, Integer.MAX_VALUE); + assertEquals(1, topDocs.totalHits); is.close(); Util.closeOrRelease(reader); } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestIndexingConfig.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestIndexingConfig.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestIndexingConfig.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -22,11 +22,13 @@ import org.apache.lucene.document.Document; import org.apache.lucene.index.IndexReader; import org.apache.lucene.index.Term; +import org.apache.lucene.search.BooleanClause.Occur; import org.apache.lucene.search.BooleanQuery; -import org.apache.lucene.search.Hits; import org.apache.lucene.search.IndexSearcher; +import org.apache.lucene.search.ScoreDoc; import org.apache.lucene.search.TermQuery; -import org.apache.lucene.search.BooleanClause.Occur; +import org.apache.lucene.search.TopDocs; +import org.apache.lucene.util.Version; import org.exoplatform.services.jcr.impl.core.NodeImpl; import org.exoplatform.services.jcr.impl.core.RepositoryImpl; import org.exoplatform.services.jcr.impl.core.query.lucene.FieldNames; @@ -101,7 +103,7 @@ indexingConfigurationImpl.addPropertyAnalyzer("FULL:" + simple, new SimpleAnalyzer()); indexingConfigurationImpl.addPropertyAnalyzer("FULL:" + whitespace, new WhitespaceAnalyzer()); - indexingConfigurationImpl.addPropertyAnalyzer("FULL:" + stop, new StopAnalyzer()); + indexingConfigurationImpl.addPropertyAnalyzer("FULL:" + stop, new StopAnalyzer(Version.LUCENE_24)); testRoot = testSession.getRootNode().addNode("testrootAnalyzers"); root.save(); } @@ -130,7 +132,7 @@ // There must be [the] [quick] [brown] [fox] [jumped] [over] [the] [lazy] [dogs] // in Node1 - Document doc = this.getDocument(testNode1.getInternalIdentifier(), false); + ScoreDoc doc = this.getDocument(testNode1.getInternalIdentifier(), false); assertNotNull(doc); TermQuery the = new TermQuery(new Term("FULL:" + simple, "the")); @@ -155,8 +157,8 @@ IndexReader ir = searchIndex.getIndexReader(); IndexSearcher is = new IndexSearcher(ir); - Hits hits = is.search(compl); - assertEquals(1, hits.length()); + TopDocs search = is.search(compl, null, Integer.MAX_VALUE); + assertEquals(1, search.totalHits); // Test is there are all terms // There must be [xy] [z] [corporation] [xyz] [example] [com] @@ -176,8 +178,8 @@ compl.add(example, Occur.MUST); compl.add(com, Occur.MUST); - hits = is.search(compl); - assertEquals(1, hits.length()); + search = is.search(compl, null, Integer.MAX_VALUE); + assertEquals(1, search.totalHits); is.close(); Util.closeOrRelease(ir); @@ -230,8 +232,8 @@ IndexReader ir = searchIndex.getIndexReader(); IndexSearcher is = new IndexSearcher(ir); - Hits hits = is.search(compl); - assertEquals(1, hits.length()); + TopDocs search = is.search(compl, null, Integer.MAX_VALUE); + assertEquals(1, search.totalHits); // Test is there are all terms // There must be [XY&Z] [Corporation] [-] [xyz(a)example.com] @@ -247,8 +249,8 @@ compl.add(defiz, Occur.MUST); compl.add(example, Occur.MUST); - hits = is.search(compl); - assertEquals(1, hits.length()); + search = is.search(compl, null, Integer.MAX_VALUE); + assertEquals(1, search.totalHits); is.close(); Util.closeOrRelease(ir); @@ -296,8 +298,8 @@ IndexReader ir = searchIndex.getIndexReader(); IndexSearcher is = new IndexSearcher(ir); - Hits hits = is.search(compl); - assertEquals(1, hits.length()); + TopDocs search = is.search(compl, null, Integer.MAX_VALUE); + assertEquals(1, search.totalHits); // Test is there are all terms // There must be [xy] [z] [corporation] [xyz] [example] [com] @@ -317,8 +319,8 @@ compl.add(example, Occur.MUST); compl.add(com, Occur.MUST); - hits = is.search(compl); - assertEquals(1, hits.length()); + search = is.search(compl, null, Integer.MAX_VALUE); + assertEquals(1, search.totalHits); is.close(); Util.closeOrRelease(ir); @@ -370,8 +372,8 @@ IndexReader ir = searchIndex.getIndexReader(); IndexSearcher is = new IndexSearcher(ir); - Hits hits = is.search(compl); - assertEquals(1, hits.length()); + TopDocs search = is.search(compl, null, Integer.MAX_VALUE); + assertEquals(1, search.totalHits); // Test is there are all terms // Terms [xy&z] [corporation] [xyz@example] [com] - it's a default @@ -388,8 +390,8 @@ compl.add(corporation, Occur.MUST); compl.add(com, Occur.MUST); - hits = is.search(compl); - assertEquals(1, hits.length()); + search = is.search(compl, null, Integer.MAX_VALUE); + assertEquals(1, search.totalHits); is.close(); Util.closeOrRelease(ir); @@ -401,22 +403,22 @@ } } - protected Document getDocument(String nodeIdentifer, boolean includeSystemIndex) throws IOException, + protected ScoreDoc getDocument(String nodeIdentifer, boolean includeSystemIndex) throws IOException, RepositoryException { IndexReader reader = ((SearchIndex)searchManager.getHandler()).getIndexReader(); IndexSearcher is = new IndexSearcher(reader); TermQuery query = new TermQuery(new Term(FieldNames.UUID, nodeIdentifer)); - Hits result = is.search(query); + TopDocs topDocs = is.search(query, null, Integer.MAX_VALUE); try { - if (result.length() == 1) + if (topDocs.totalHits == 1) { - return result.doc(0); + return topDocs.scoreDocs[0]; } - else if (result.length() > 1) + else if (topDocs.totalHits > 1) { throw new RepositoryException("Results more then one"); } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestMultiValueSearch.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestMultiValueSearch.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestMultiValueSearch.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -17,7 +17,7 @@ package org.exoplatform.services.jcr.impl.core.query; -import org.apache.lucene.document.Document; +import org.apache.lucene.search.ScoreDoc; import org.exoplatform.services.jcr.impl.core.NodeImpl; import javax.jcr.PropertyType; @@ -39,7 +39,7 @@ root.save(); // Check is node indexed - Document doc = getDocument(node.getInternalIdentifier(), false); + ScoreDoc doc = getDocument(node.getInternalIdentifier(), false); assertNotNull("Node is not indexed", doc); System.out.println("its doc " + doc); @@ -65,7 +65,7 @@ root.save(); // Check is node indexed - Document doc = getDocument(node.getInternalIdentifier(), false); + ScoreDoc doc = getDocument(node.getInternalIdentifier(), false); assertNotNull("Node is not indexed", doc); System.out.println("its doc " + doc); Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestRewriteNode.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestRewriteNode.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/TestRewriteNode.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -19,9 +19,9 @@ import org.apache.lucene.index.IndexReader; import org.apache.lucene.index.Term; -import org.apache.lucene.search.Hits; import org.apache.lucene.search.IndexSearcher; import org.apache.lucene.search.TermQuery; +import org.apache.lucene.search.TopDocs; import org.exoplatform.services.jcr.impl.core.NodeImpl; import org.exoplatform.services.jcr.impl.core.query.lucene.FieldNames; import org.exoplatform.services.jcr.impl.core.query.lucene.Util; @@ -54,8 +54,8 @@ IndexReader reader = defaultSearchIndex.getIndexReader(); IndexSearcher is = new IndexSearcher(reader); TermQuery query = new TermQuery(new Term(FieldNames.FULLTEXT, "fox")); - Hits result = is.search(query); - assertEquals(1, result.length()); + TopDocs topDocs = is.search(query, null, Integer.MAX_VALUE); + assertEquals(1, topDocs.totalHits); cont.setProperty("jcr:data", "Bahama mama"); root.save(); @@ -63,18 +63,18 @@ reader = defaultSearchIndex.getIndexReader(); is = new IndexSearcher(reader); query = new TermQuery(new Term(FieldNames.FULLTEXT, "mama")); - result = is.search(query); - assertEquals(1, result.length()); + topDocs = is.search(query, null, Integer.MAX_VALUE); + assertEquals(1, topDocs.totalHits); reader = defaultSearchIndex.getIndexReader(); is = new IndexSearcher(reader); query = new TermQuery(new Term(FieldNames.FULLTEXT, "fox")); - result = is.search(query); - assertEquals(0, result.length()); + topDocs = is.search(query, null, Integer.MAX_VALUE); + assertEquals(0, topDocs.totalHits); is.close(); Util.closeOrRelease(reader); - + } } Modified: jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/lucene/TestChangesHolder.java =================================================================== --- jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/lucene/TestChangesHolder.java 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/exo.jcr.component.core/src/test/java/org/exoplatform/services/jcr/impl/core/query/lucene/TestChangesHolder.java 2012-02-22 14:18:12 UTC (rev 5691) @@ -20,11 +20,13 @@ import junit.framework.TestCase; +import org.apache.lucene.document.AbstractField; import org.apache.lucene.document.Document; import org.apache.lucene.document.Field; import org.apache.lucene.document.Field.Index; import org.apache.lucene.document.Field.Store; import org.apache.lucene.document.Field.TermVector; +import org.apache.lucene.document.Fieldable; import java.io.ByteArrayInputStream; import java.io.ByteArrayOutputStream; @@ -46,13 +48,14 @@ public void testSerNDeserializeDocs() throws Exception { - System.out.println("### testSerNDeserializeDocs ###"); + //System.out.println("### testSerNDeserializeDocs ###"); Collection<Document> add = new ArrayList<Document>(3); Document doc = new Document(); doc.setBoost(2.0f); - Field fieldFull = new Field("full", "full-value", Store.COMPRESS, Index.ANALYZED_NO_NORMS, TermVector.WITH_POSITIONS_OFFSETS); + Field fieldFull = + new Field("full", "full-value", Store.YES, Index.ANALYZED_NO_NORMS, TermVector.WITH_POSITIONS_OFFSETS); fieldFull.setBoost(2.0f); - fieldFull.setOmitTf(true); + fieldFull.setOmitTermFreqAndPositions(true); doc.add(fieldFull); Field fieldEmpty = new Field("empty", "empty-value", Store.NO, Index.NOT_ANALYZED, TermVector.NO); doc.add(fieldEmpty); @@ -63,9 +66,9 @@ doc = new Document(); doc.add(fieldEmpty); add.add(doc); - + ByteArrayOutputStream baos = null; - + int total = 100000; long start; Collection<String> remove = Collections.emptyList(); @@ -78,7 +81,7 @@ oos.writeObject(new ChangesHolder(remove, add)); oos.close(); } - System.out.println("Custom serialization: total time = " + (System.currentTimeMillis() - start) + ", size = " + baos.size()); + //System.out.println("Custom serialization: total time = " + (System.currentTimeMillis() - start) + ", size = " + baos.size()); start = System.currentTimeMillis(); for (int i = 0; i < total; i++) @@ -87,7 +90,7 @@ addResult = ((ChangesHolder)ois.readObject()).getAdd(); ois.close(); } - System.out.println("Custom deserialization: total time = " + (System.currentTimeMillis() - start)); + //System.out.println("Custom deserialization: total time = " + (System.currentTimeMillis() - start)); checkDocs(addResult); start = System.currentTimeMillis(); for (int i = 0; i < total; i++) @@ -97,7 +100,7 @@ oos.writeObject(add); oos.close(); } - System.out.println("Native serialization: total time = " + (System.currentTimeMillis() - start) + ", size = " + baos.size()); + //System.out.println("Native serialization: total time = " + (System.currentTimeMillis() - start) + ", size = " + baos.size()); start = System.currentTimeMillis(); for (int i = 0; i < total; i++) { @@ -105,7 +108,7 @@ addResult = (Collection<Document>)ois.readObject(); ois.close(); } - System.out.println("Native deserialization: total time = " + (System.currentTimeMillis() - start)); + //System.out.println("Native deserialization: total time = " + (System.currentTimeMillis() - start)); checkDocs(addResult); } @@ -116,7 +119,7 @@ Iterator<Document> it = addResult.iterator(); Document doc = it.next(); assertEquals(2.0f, doc.getBoost()); - List<Field> fields = doc.getFields(); + List<Fieldable> fields = doc.getFields(); assertNotNull(fields); assertEquals(2, fields.size()); checkFieldFull(fields.get(0)); @@ -135,19 +138,21 @@ checkFieldEmpty(fields.get(0)); } - private void checkFieldFull(Field field) + private void checkFieldFull(Fieldable field) { assertEquals("full", field.name()); assertEquals("full-value", field.stringValue()); assertTrue(field.isStored()); - assertTrue(field.isCompressed()); assertTrue(field.isIndexed()); assertTrue(field.isTokenized()); assertTrue(field.getOmitNorms()); assertTrue(field.isTermVectorStored()); assertTrue(field.isStoreOffsetWithTermVector()); assertTrue(field.isStorePositionWithTermVector()); - assertTrue(field.getOmitTf()); + if (field instanceof AbstractField) + { + assertTrue(((AbstractField)field).getOmitTermFreqAndPositions()); + } assertFalse(field.isBinary()); assertFalse(field.isLazy()); assertEquals(2.0f, field.getBoost()); @@ -155,35 +160,37 @@ assertEquals(0, field.getBinaryOffset()); } - private void checkFieldEmpty(Field field) + private void checkFieldEmpty(Fieldable field) { assertEquals("empty", field.name()); assertEquals("empty-value", field.stringValue()); assertFalse(field.isStored()); - assertFalse(field.isCompressed()); assertTrue(field.isIndexed()); assertFalse(field.isTokenized()); assertFalse(field.getOmitNorms()); assertFalse(field.isTermVectorStored()); assertFalse(field.isStoreOffsetWithTermVector()); assertFalse(field.isStorePositionWithTermVector()); - assertFalse(field.getOmitTf()); + if (field instanceof AbstractField) + { + assertFalse(((AbstractField)field).getOmitTermFreqAndPositions()); + } assertFalse(field.isBinary()); assertFalse(field.isLazy()); assertEquals(1.0f, field.getBoost()); assertEquals(0, field.getBinaryLength()); assertEquals(0, field.getBinaryOffset()); } - + public void testSerNDeserializeIds() throws Exception { - System.out.println("### testSerNDeserializeIds ###"); + //System.out.println("### testSerNDeserializeIds ###"); Collection<String> remove = new ArrayList<String>(3); remove.add(UUID.randomUUID().toString()); remove.add(UUID.randomUUID().toString()); remove.add(UUID.randomUUID().toString()); ByteArrayOutputStream baos = null; - + int total = 100000; long start; Collection<Document> add = Collections.emptyList(); @@ -196,7 +203,7 @@ oos.writeObject(new ChangesHolder(remove, add)); oos.close(); } - System.out.println("Custom serialization: total time = " + (System.currentTimeMillis() - start) + ", size = " + baos.size()); + //System.out.println("Custom serialization: total time = " + (System.currentTimeMillis() - start) + ", size = " + baos.size()); start = System.currentTimeMillis(); for (int i = 0; i < total; i++) @@ -205,7 +212,7 @@ addResult = ((ChangesHolder)ois.readObject()).getRemove(); ois.close(); } - System.out.println("Custom deserialization: total time = " + (System.currentTimeMillis() - start)); + //System.out.println("Custom deserialization: total time = " + (System.currentTimeMillis() - start)); checkIds(remove, addResult); start = System.currentTimeMillis(); for (int i = 0; i < total; i++) @@ -215,7 +222,7 @@ oos.writeObject(remove); oos.close(); } - System.out.println("Native serialization: total time = " + (System.currentTimeMillis() - start) + ", size = " + baos.size()); + //System.out.println("Native serialization: total time = " + (System.currentTimeMillis() - start) + ", size = " + baos.size()); start = System.currentTimeMillis(); for (int i = 0; i < total; i++) { @@ -223,8 +230,8 @@ addResult = (Collection<String>)ois.readObject(); ois.close(); } - System.out.println("Native deserialization: total time = " + (System.currentTimeMillis() - start)); - checkIds(remove, addResult); + //System.out.println("Native deserialization: total time = " + (System.currentTimeMillis() - start)); + checkIds(remove, addResult); } private void checkIds(Collection<String> remove, Collection<String> addResult) Modified: jcr/branches/1.15.x/pom.xml =================================================================== --- jcr/branches/1.15.x/pom.xml 2012-02-22 13:02:08 UTC (rev 5690) +++ jcr/branches/1.15.x/pom.xml 2012-02-22 14:18:12 UTC (rev 5691) @@ -274,19 +274,24 @@ <dependency> <groupId>org.apache.lucene</groupId> <artifactId>lucene-core</artifactId> - <version>2.9.4</version> + <version>3.0.3</version> </dependency> <dependency> <groupId>org.apache.lucene</groupId> <artifactId>lucene-spellchecker</artifactId> - <version>2.9.4</version> + <version>3.0.3</version> </dependency> <dependency> <groupId>org.apache.lucene</groupId> <artifactId>lucene-memory</artifactId> - <version>2.9.4</version> + <version>3.0.3</version> </dependency> <dependency> + <groupId>org.apache.lucene</groupId> + <artifactId>lucene-wordnet</artifactId> + <version>3.0.3</version> + </dependency> + <dependency> <groupId>com.sun.xml.stream</groupId> <artifactId>sjsxp</artifactId> <version>1.0.1</version>