This package provides fast numeric range queries/filters on long, double or Date
+fields based on trie structures.
See the publication about panFMP, where this algorithm was described: + +
Schindler, U, Diepenbroek, M, 2008. Generic XML-based Framework for Metadata Portals. +Computers & Geosciences 34 (12), 1947-1955. +doi:10.1016/j.cageo.2008.02.023+ +
A quote from this paper: Because Apache Lucene is a full-text search engine and not a conventional database,
+it cannot handle numerical ranges (e.g., field value is inside user defined bounds, even dates are numerical values).
+We have developed an extension to Apache Lucene that stores
+the numerical values in a special string-encoded format with variable precision
+(all numerical values like doubles, longs, and timestamps are converted to lexicographic sortable string representations of
+long long words and stored with precisions from one byte to the full 8 bytes).
+For a more detailed description, how the values are stored, see {@link org.apache.lucene.search.trie.TrieUtils}.
+A range is then divided recursively into multiple intervals for searching:
+The center of the range is searched only with the lowest possible precision in the trie, the boundaries are matched
+more exactly. This reduces the number of terms dramatically (in the lowest precision of 1-byte the index only
+contains a maximum of 256 distinct values). Overall, a range could consist of a theoretical maximum of
+7*256*2 + 255 = 3825 distinct terms (when there is a term for every distinct value of an
+8-byte-number in the index and the range covers all of them; a maximum of 255 distinct values is used
+because it would always be possible to reduce the full 256 values to one term with degraded precision).
+In practise, we have seen up to 300 terms in most cases (index with 500,000 metadata records
+and a homogeneous dispersion of values).
This dramatically improves the performance of Apache Lucene with range queries, which +is no longer dependent on the index size and number of distinct values because there is +an upper limit not related to any of these properties.
+ +To use the new query types the numerical values, which may be long, double or Date,
+during indexing the values must be stored in a special format in index (using {@link org.apache.lucene.search.trie.TrieUtils}).
+This can be done like this:
+ Document doc=new Document();
+ // add some standard fields:
+ String svalue="anything to index";
+ doc.add(new Field("exampleString",
+ svalue, Field.Store.YES, Field.Index.ANALYZED) ;
+ // add some numerical fields:
+ double fvalue=1.057E17;
+ TrieUtils.addDoubleTrieCodedDocumentField(doc, "exampleDouble",
+ fvalue, true /* index the field */, Field.Store.YES);
+ double lvalue=121345;
+ TrieUtils.addLongTrieCodedDocumentField(doc, "exampleLong",
+ lvalue, true /* index the field */, Field.Store.YES);
+ Date dvalue=new Date(); // actual time
+ TrieUtils.addDateTrieCodedDocumentField(doc, "exampleDate",
+ dvalue, true /* index the field */, Field.Store.YES);
+ // add document to IndexWriter
+
+
+The numeric index fields you prepared in this way can be searched by {@link org.apache.lucene.search.trie.TrieRangeQuery}:
+ +
+ Query q=new TrieRangeQuery("exampleDouble", Double.valueOf(1.0E17), Double.valueOf(2.0E17));
+ TopDocs docs=searcher.search(q, 10);
+ for (int i=0; i<docs.scoreDocs.length; i++) {
+ Document doc=searcher.doc(docs.scoreDocs[i].doc);
+ System.out.println(doc.get("exampleString"));
+ // decode the stored numerical value (important!!!):
+ System.out.println( TrieUtils.trieCodedToDouble(doc.get("exampleDouble")) );
+ }
+
+
+Comparisions of the different types of RangeQueries on an index with about 500,000 docs showed, +that the old {@link org.apache.lucene.search.RangeQuery} (with raised +{@link org.apache.lucene.search.BooleanQuery} clause count) took about 30-40 secs to complete, +{@link org.apache.lucene.search.ConstantScoreRangeQuery} took 5 secs and +{@link org.apache.lucene.search.trie.TrieRangeQuery} took <100ms to +complete (on an Opteron64 machine, Java 1.5). +This query type was developed for a geographic portal, where the performance for +e.g. bounding boxes or exact date/time stamps is important.
+ + + \ No newline at end of file Index: contrib/queries/src/java/org/apache/lucene/search/trie/TrieRangeFilter.java =================================================================== --- contrib/queries/src/java/org/apache/lucene/search/trie/TrieRangeFilter.java (revision 0) +++ contrib/queries/src/java/org/apache/lucene/search/trie/TrieRangeFilter.java (revision 0) @@ -0,0 +1,203 @@ +package org.apache.lucene.search.trie; + +/** + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +import java.io.IOException; +import java.util.Date; + +import org.apache.lucene.search.Filter; +import org.apache.lucene.search.DocIdSet; +import org.apache.lucene.index.IndexReader; +import org.apache.lucene.index.TermDocs; +import org.apache.lucene.index.TermEnum; +import org.apache.lucene.index.Term; +import org.apache.lucene.util.OpenBitSet; + +/** + * Implementation of a Lucene {@link Filter} that implements trie-based range filtering. + * This filter depends on a specific structure of terms in the index that can only be created + * by {@link TrieUtils} methods. + * For more information, how the algorithm works, see the package description {@link org.apache.lucene.search.trie}. + * @author Uwe Schindler (panFMP developer) + */ +public final class TrieRangeFilter extends Filter { + + /** Generic constructor (internal use only): Uses already trie-converted min/max values */ + public TrieRangeFilter(final String field, final String min, final String max) { + if (min==null && max==null) throw new IllegalArgumentException("The min and max values cannot be both null."); + this.minUnconverted=min; + this.maxUnconverted=max; + this.min=(min==null) ? TrieUtils.TRIE_CODED_NUMERIC_MIN : min; + this.max=(max==null) ? TrieUtils.TRIE_CODED_NUMERIC_MAX : max; + this.field=field.intern(); + } + + /** + * Generates a trie query using the supplied field with range bounds in numeric form (double). + * You can setmin or max (but not both) to null to leave one bound open.
+ */
+ public TrieRangeFilter(final String field, final Double min, final Double max) {
+ if (min==null && max==null) throw new IllegalArgumentException("The min and max double values cannot be both null.");
+ this.minUnconverted=min;
+ this.maxUnconverted=max;
+ this.min=(min==null) ? TrieUtils.TRIE_CODED_NUMERIC_MIN : TrieUtils.doubleToTrieCoded(min.doubleValue());
+ this.max=(max==null) ? TrieUtils.TRIE_CODED_NUMERIC_MAX : TrieUtils.doubleToTrieCoded(max.doubleValue());
+ this.field=field.intern();
+ }
+
+ /**
+ * Generates a trie query using the supplied field with range bounds in date/time form.
+ * You can set min or max (but not both) to null to leave one bound open.
+ */
+ public TrieRangeFilter(final String field, final Date min, final Date max) {
+ if (min==null && max==null) throw new IllegalArgumentException("The min and max date values cannot be both null.");
+ this.minUnconverted=min;
+ this.maxUnconverted=max;
+ this.min=(min==null) ? TrieUtils.TRIE_CODED_NUMERIC_MIN : TrieUtils.dateToTrieCoded(min);
+ this.max=(max==null) ? TrieUtils.TRIE_CODED_NUMERIC_MAX : TrieUtils.dateToTrieCoded(max);
+ this.field=field.intern();
+ }
+
+ /**
+ * Generates a trie query using the supplied field with range bounds in integer form (long).
+ * You can set min or max (but not both) to null to leave one bound open.
+ */
+ public TrieRangeFilter(final String field, final Long min, final Long max) {
+ if (min==null && max==null) throw new IllegalArgumentException("The min and max long values cannot be both null.");
+ this.minUnconverted=min;
+ this.maxUnconverted=max;
+ this.min=(min==null) ? TrieUtils.TRIE_CODED_NUMERIC_MIN : TrieUtils.longToTrieCoded(min.longValue());
+ this.max=(max==null) ? TrieUtils.TRIE_CODED_NUMERIC_MAX : TrieUtils.longToTrieCoded(max.longValue());
+ this.field=field.intern();
+ }
+
+ //@Override
+ public String toString() {
+ return toString(null);
+ }
+
+ public String toString(final String field) {
+ final StringBuffer sb=new StringBuffer();
+ if (!this.field.equals(field)) sb.append(this.field).append(':');
+ return sb.append('[').append(minUnconverted).append(" TO ").append(maxUnconverted).append(']').toString();
+ }
+
+ //@Override
+ public final boolean equals(final Object o) {
+ if (o instanceof TrieRangeFilter) {
+ TrieRangeFilter q=(TrieRangeFilter)o;
+ return (field==q.field && min.equals(q.min) && max.equals(q.max));
+ } else return false;
+ }
+
+ //@Override
+ public final int hashCode() {
+ return field.hashCode()+(min.hashCode()^0x14fa55fb)+(max.hashCode()^0x733fa5fe);
+ }
+
+ /** Marks documents in a specific range. Code borrowed from original RangeFilter and simplified (and returns number of terms) */
+ private int setBits(final IndexReader reader, final TermDocs termDocs, final OpenBitSet bits, String lowerTerm, String upperTerm) throws IOException {
+ //System.out.println(lowerTerm+" TO "+upperTerm);
+ int count=0,len=lowerTerm.length();
+ final String field;
+ if (len<8) {
+ // lower precision value is in helper field
+ field=(this.field + TrieUtils.LOWER_PRECISION_FIELD_NAME_SUFFIX).intern();
+ // add padding before lower precision values to group them
+ lowerTerm=new StringBuffer(len).append((char)(TrieUtils.TRIE_CODED_PADDING_START+len)).append(lowerTerm).toString();
+ upperTerm=new StringBuffer(len).append((char)(TrieUtils.TRIE_CODED_PADDING_START+len)).append(upperTerm).toString();
+ // length is longer by 1 char because of padding, the field name must be suffixed!
+ len++;
+ } else {
+ // full precision value is in original field
+ field=this.field;
+ }
+ final TermEnum enumerator = reader.terms(new Term(field, lowerTerm));
+ try {
+ do {
+ final Term term = enumerator.term();
+ if (term!=null && term.field()==field) {
+ // break out when upperTerm reached or length of term is different
+ String t=term.text();
+ if (len!=t.length() || t.compareTo(upperTerm)>0) break;
+ // we have a good term, find the docs
+ count++;
+ termDocs.seek(enumerator);
+ while (termDocs.next()) bits.set(termDocs.doc());
+ } else break;
+ } while (enumerator.next());
+ } finally {
+ enumerator.close();
+ }
+ return count;
+ }
+
+ /** Splits range recursively (and returns number of terms) */
+ private int splitRange(
+ final IndexReader reader, final TermDocs termDocs, final OpenBitSet bits,
+ final String min, final boolean lowerBoundOpen, final String max, final boolean upperBoundOpen
+ ) throws IOException {
+ int count=0;
+ final int length=min.length();
+ final String minShort=lowerBoundOpen ? min.substring(0,length-1) : TrieUtils.incrementTrieCoded(min.substring(0,length-1));
+ final String maxShort=upperBoundOpen ? max.substring(0,length-1) : TrieUtils.decrementTrieCoded(max.substring(0,length-1));
+
+ if (length==1 || minShort.compareTo(maxShort)>=0) {
+ count+=setBits(reader, termDocs, bits, min, max);
+ } else {
+ if (!lowerBoundOpen) {
+ count+=setBits(reader, termDocs, bits, min, TrieUtils.decrementTrieCoded(minShort+TrieUtils.TRIE_CODED_SYMBOL_MIN));
+ }
+ count+=splitRange(
+ reader,termDocs,bits,
+ minShort,lowerBoundOpen,
+ maxShort,upperBoundOpen
+ );
+ if (!upperBoundOpen) {
+ count+=setBits(reader, termDocs, bits, TrieUtils.incrementTrieCoded(maxShort+TrieUtils.TRIE_CODED_SYMBOL_MAX), max);
+ }
+ }
+ return count;
+ }
+
+ /**
+ * Returns a DocIdSet that provides the documents which should be permitted or prohibited in search results.
+ */
+ //@Override
+ public DocIdSet getDocIdSet(IndexReader reader) throws IOException {
+ final OpenBitSet bits = new OpenBitSet(reader.maxDoc());
+ final TermDocs termDocs=reader.termDocs();
+ try {
+ final int count=splitRange(
+ reader,termDocs,bits,
+ min,TrieUtils.TRIE_CODED_NUMERIC_MIN.equals(min),
+ max,TrieUtils.TRIE_CODED_NUMERIC_MAX.equals(max)
+ );
+ // count is not used yet, we can make statistics on it or debug it like so:
+ //System.out.println("Found "+count+" distinct terms in filtered range for field '"+field+"'.");
+ } finally {
+ termDocs.close();
+ }
+ return bits;
+ }
+
+ // members
+ private String field,min,max;
+ private Object minUnconverted,maxUnconverted;
+
+}
\ No newline at end of file
Index: contrib/queries/src/java/org/apache/lucene/search/trie/TrieRangeQuery.java
===================================================================
--- contrib/queries/src/java/org/apache/lucene/search/trie/TrieRangeQuery.java (revision 0)
+++ contrib/queries/src/java/org/apache/lucene/search/trie/TrieRangeQuery.java (revision 0)
@@ -0,0 +1,100 @@
+package org.apache.lucene.search.trie;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import java.util.Date;
+import java.io.IOException;
+
+import org.apache.lucene.search.Query;
+import org.apache.lucene.search.Filter;
+import org.apache.lucene.search.ConstantScoreQuery;
+import org.apache.lucene.util.ToStringUtils;
+import org.apache.lucene.index.IndexReader;
+
+/**
+ * Implementation of a Lucene {@link Query} that implements a trie-based range query.
+ * This query depends on a specific structure of terms in the index that can only be created
+ * by {@link TrieUtils} methods.
+ * This class wraps a {@link TrieRangeFilter} using a {@link ConstantScoreQuery}.
+ * @see TrieRangeFilter
+ * @author Uwe Schindler (panFMP developer)
+ */
+public final class TrieRangeQuery extends Query {
+
+ /** Generic constructor (internal use only): Uses already trie-converted min/max values */
+ public TrieRangeQuery(final String field, final String min, final String max) {
+ filter=new TrieRangeFilter(field,min,max);
+ }
+
+ /**
+ * Generates a trie query using the supplied field with range bounds in numeric form (double).
+ * You can set min or max (but not both) to null to leave one bound open.
+ */
+ public TrieRangeQuery(final String field, final Double min, final Double max) {
+ filter=new TrieRangeFilter(field,min,max);
+ }
+
+ /**
+ * Generates a trie query using the supplied field with range bounds in date/time form.
+ * You can set min or max (but not both) to null to leave one bound open.
+ */
+ public TrieRangeQuery(final String field, final Date min, final Date max) {
+ filter=new TrieRangeFilter(field,min,max);
+ }
+
+ /**
+ * Generates a trie query using the supplied field with range bounds in integer form (long).
+ * You can set min or max (but not both) to null to leave one bound open.
+ */
+ public TrieRangeQuery(final String field, final Long min, final Long max) {
+ filter=new TrieRangeFilter(field,min,max);
+ }
+
+ //@Override
+ public String toString(final String field) {
+ return filter.toString(field)+ToStringUtils.boost(getBoost());
+ }
+
+ //@Override
+ public final boolean equals(final Object o) {
+ if (o instanceof TrieRangeQuery) {
+ TrieRangeQuery q=(TrieRangeQuery)o;
+ return (filter.equals(q.filter) && getBoost()==q.getBoost());
+ } else return false;
+ }
+
+ //@Override
+ public final int hashCode() {
+ return filter.hashCode()^0x1756fa55+Float.floatToIntBits(getBoost());
+ }
+
+ /**
+ * Rewrites the query to native Lucene {@link Query}'s. This implementation uses a {@link ConstantScoreQuery} with
+ * a {@link TrieRangeFilter} as implementation of the trie algorithm.
+ */
+ //@Override
+ public Query rewrite(final IndexReader reader) throws IOException {
+ final ConstantScoreQuery q = new ConstantScoreQuery(filter);
+ q.setBoost(getBoost());
+ return q.rewrite(reader);
+ }
+
+ // members
+ private final TrieRangeFilter filter;
+
+}
\ No newline at end of file
Index: contrib/queries/src/java/org/apache/lucene/search/trie/TrieUtils.java
===================================================================
--- contrib/queries/src/java/org/apache/lucene/search/trie/TrieUtils.java (revision 0)
+++ contrib/queries/src/java/org/apache/lucene/search/trie/TrieUtils.java (revision 0)
@@ -0,0 +1,262 @@
+package org.apache.lucene.search.trie;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import java.util.Date;
+
+import org.apache.lucene.document.Document;
+import org.apache.lucene.document.Field;
+
+/**
+ *This is a helper class to construct the trie-based index entries for numerical values.
+ *
For more information, how the algorithm works, see the package description {@link org.apache.lucene.search.trie}. The format of how the + * numerical values are stored in index is documented here: + *
All numerical values are first converted to special unsigned longs by applying some bit-wise transformations. This means:
signed long is transformed to the unsigned form like so:signed longs are shifted, so that {@link Long#MIN_VALUE} is mapped to 0x0000000000000000,
+ * {@link Long#MAX_VALUE} is mapped to 0xffffffffffffffff.doubles are converted by getting their IEEE 754 floating-point "double format" bit layout and then some bits
+ * are swapped, to be able to compare the result as unsigned longs.Each byte of this unsigned long (starting with the most-significant half-byte) is converted to ASCII chars
+ * between 0x30 ({@link #TRIE_CODED_SYMBOL_MIN}) and 0x12f==0x30+0x0ff ({@link #TRIE_CODED_SYMBOL_MAX}).
+ * The resulting {@link String} is comparable like the corresponding unsigned long.
+ *
To store the different precisions of the long values (from one byte [only the most significant one] to the full eight bytes),
+ * each lower precision is prefixed by the length ({@link #TRIE_CODED_PADDING_START} To store the different precisions of the long values (from one byte [only the most significant one] to the full eight bytes),
+ * each lower precision is prefixed by the length ({@link #TRIE_CODED_PADDING_START} To store the different precisions of the long values (from one byte [only the most significant one] to the full eight bytes),
+ * each lower precision is prefixed by the length ({@link #TRIE_CODED_PADDING_START} To store the different precisions of the long values (from one byte [only the most significant one] to the full eight bytes),
+ * each lower precision is prefixed by the length ({@link #TRIE_CODED_PADDING_START}+precision == 0x20+precision),
+ * in an extra "helper" field with a suffixed field name (i.e. fieldname "numeric" => lower precision's name "numeric#trie").
+ * The full long is not prefixed at all and indexed and stored according to the given flags in the original field name.
+ * By this it is possible to get the correct enumeration of terms in correct precision
+ * of the term list by just jumping to the correct fieldname and/or prefix. The full precision value may also be
+ * stored in the document. Having the full precision value as term in a separate field with the original name,
+ * sorting of query results agains such fields is possible using the original field name.
+ * @author Uwe Schindler (panFMP developer)
+ */
+public final class TrieUtils {
+
+ /** Marker (PADDING) before lower-precision trie entries to signal the precision value. See class description! */
+ public static final char TRIE_CODED_PADDING_START=(char)0x20;
+
+ /** The "helper" field containing the lower precision terms is the original fieldname with this appended. */
+ public static final String LOWER_PRECISION_FIELD_NAME_SUFFIX="#trie";
+
+ /** Character used as lower end [0x30 to 0x12f] */
+ public static final char TRIE_CODED_SYMBOL_MIN=(char)0x30;
+ /** Character used as upper end [0x30 to 0x12f] */
+ public static final char TRIE_CODED_SYMBOL_MAX=(char)(TRIE_CODED_SYMBOL_MIN+0x0ff);
+
+ /** minimum encoded value of a numerical index entry: {@link Long#MIN_VALUE} */
+ public static String TRIE_CODED_NUMERIC_MIN=longToTrieCoded(Long.MIN_VALUE);
+ /** maximum encoded value of a numerical index entry: {@link Long#MAX_VALUE} */
+ public static String TRIE_CODED_NUMERIC_MAX=longToTrieCoded(Long.MAX_VALUE);
+
+ private TrieUtils() {} // no instance
+
+ // internal conversion to/from strings
+
+ private static final String internalLongToTrieCoded(final long l) {
+ final StringBuffer sb=new StringBuffer(8);
+ synchronized(sb) {
+ for (int j=56; j>=0; j-=8) {
+ sb.append((char)(TRIE_CODED_SYMBOL_MIN+((l >>> j) & 0x0ff)));
+ }
+ return sb.toString();
+ }
+ }
+
+ private static final long internalTrieCodedToLong(final String s) {
+ if (s==null) throw new NullPointerException("Trie encoded string may not be NULL");
+ final int len=s.length();
+ if (len!=8) throw new NumberFormatException("Invalid trie encoded numerical value representation: '"+s+"' (incompatible length, must be 8)");
+ long l=0L;
+ for (int i=0; ilong value encoded to a String (with 8 bytes). */
+ public static String longToTrieCoded(final long l) {
+ return internalLongToTrieCoded(l ^ 0x8000000000000000L);
+ }
+
+ /** Converts a encoded String (8 bytes) value back to a long. */
+ public static long trieCodedToLong(final String s) {
+ return internalTrieCodedToLong(s) ^ 0x8000000000000000L;
+ }
+
+ // Double's
+
+ /** Converts a double value encoded to a String (with 8 bytes). */
+ public static String doubleToTrieCoded(final double d) {
+ long l=Double.doubleToLongBits(d);
+ if ((l & 0x8000000000000000L) == 0L) {
+ // >0
+ l |= 0x8000000000000000L;
+ } else {
+ // <0
+ l = ~l;
+ }
+ return internalLongToTrieCoded(l);
+ }
+
+ /** Converts a encoded String (8 bytes) value back to a double. */
+ public static double trieCodedToDouble(final String s) {
+ long l=internalTrieCodedToLong(s);
+ if ((l & 0x8000000000000000L) != 0L) {
+ // >0
+ l &= 0x7fffffffffffffffL;
+ } else {
+ // <0
+ l = ~l;
+ }
+ return Double.longBitsToDouble(l);
+ }
+
+ // Date's
+
+ /** Converts a Date value encoded to a String (with 8 bytes). */
+ public static String dateToTrieCoded(final Date d) {
+ return longToTrieCoded(d.getTime());
+ }
+
+ /** Converts a encoded String (8 bytes) value back to a Date. */
+ public static Date trieCodedToDate(final String s) {
+ return new Date(trieCodedToLong(s));
+ }
+
+ // increment / decrement
+
+ /** Increments an encoded String value by 1. Needed by {@link TrieRangeFilter}. */
+ protected static String incrementTrieCoded(final String v) {
+ final int l=v.length();
+ final StringBuffer sb=new StringBuffer(l);
+ synchronized(sb) {
+ boolean inc=true;
+ for (int i=l-1; i>=0; i--) {
+ int b=v.charAt(i)-TRIE_CODED_SYMBOL_MIN;
+ if (inc) b++;
+ if (inc=(b>0x0ff)) b=0;
+ sb.insert(0,(char)(TRIE_CODED_SYMBOL_MIN+b));
+ }
+ return sb.toString();
+ }
+ }
+
+ /** Decrements an encoded String value by 1. Needed by {@link TrieRangeFilter}. */
+ protected static String decrementTrieCoded(final String v) {
+ final int l=v.length();
+ final StringBuffer sb=new StringBuffer(l);
+ synchronized(sb) {
+ boolean dec=true;
+ for (int i=l-1; i>=0; i--) {
+ int b=v.charAt(i)-TRIE_CODED_SYMBOL_MIN;
+ if (dec) b--;
+ if (dec=(b<0)) b=0x0ff;
+ sb.insert(0,(char)(TRIE_CODED_SYMBOL_MIN+b));
+ }
+ return sb.toString();
+ }
+ }
+
+ private static void addConvertedTrieCodedDocumentField(
+ final Document ldoc, final String fieldname, final String val,
+ final boolean index, final Field.Store store
+ ) {
+ Field f=new Field(fieldname, val, store, index?Field.Index.NOT_ANALYZED:Field.Index.NO);
+ if (index) {
+ f.setOmitTf(true);
+ ldoc.add(f);
+ // add the lower precision values in the helper field with prefix
+ final StringBuffer sb=new StringBuffer(8);
+ synchronized(sb) {
+ for (int i=7; i>0; i--) {
+ sb.setLength(0);
+ f=new Field(
+ fieldname + LOWER_PRECISION_FIELD_NAME_SUFFIX,
+ sb.append( (char)(TRIE_CODED_PADDING_START+i) ).append( val.substring(0,i) ).toString(),
+ Field.Store.NO, Field.Index.NOT_ANALYZED
+ );
+ f.setOmitTf(true);
+ ldoc.add(f);
+ }
+ }
+ } else {
+ ldoc.add(f);
+ }
+ }
+
+ /**
+ * Stores a double value in trie-form in document for indexing.
+ * +precision),
+ * in an extra "helper" field with a name of fieldname+LOWER_PRECISION_FIELD_NAME_SUFFIX
+ * (i.e. fieldname "numeric" => lower precision's name "numeric#trie").
+ * The full long is not prefixed at all and indexed and stored according to the given flags in the original field name.
+ * If the field should not be searchable, set index to false. It is then only stored (for convenience).
+ * Fields added to a document using this method can be queried by {@link TrieRangeQuery}.
+ */
+ public static void addDoubleTrieCodedDocumentField(
+ final Document ldoc, final String fieldname, final double val,
+ final boolean index, final Field.Store store
+ ) {
+ addConvertedTrieCodedDocumentField(ldoc, fieldname, doubleToTrieCoded(val), index, store);
+ }
+
+ /**
+ * Stores a Date value in trie-form in document for indexing.
+ * +precision),
+ * in an extra "helper" field with a name of fieldname+LOWER_PRECISION_FIELD_NAME_SUFFIX
+ * (i.e. fieldname "numeric" => lower precision's name "numeric#trie").
+ * The full long is not prefixed at all and indexed and stored according to the given flags in the original field name.
+ * If the field should not be searchable, set index to false. It is then only stored (for convenience).
+ * Fields added to a document using this method can be queried by {@link TrieRangeQuery}.
+ */
+ public static void addDateTrieCodedDocumentField(
+ final Document ldoc, final String fieldname,
+ final Date val, final boolean index, final Field.Store store
+ ) {
+ addConvertedTrieCodedDocumentField(ldoc, fieldname, dateToTrieCoded(val), index, store);
+ }
+
+ /**
+ * Stores a long value in trie-form in document for indexing.
+ * +precision),
+ * in an extra "helper" field with a name of fieldname+LOWER_PRECISION_FIELD_NAME_SUFFIX
+ * (i.e. fieldname "numeric" => lower precision's name "numeric#trie").
+ * The full long is not prefixed at all and indexed and stored according to the given flags in the original field name.
+ * If the field should not be searchable, set index to false. It is then only stored (for convenience).
+ * Fields added to a document using this method can be queried by {@link TrieRangeQuery}.
+ */
+ public static void addLongTrieCodedDocumentField(
+ final Document ldoc, final String fieldname,
+ final long val, final boolean index, final Field.Store store
+ ) {
+ addConvertedTrieCodedDocumentField(ldoc, fieldname, longToTrieCoded(val), index, store);
+ }
+
+}
+
Index: contrib/queries/src/test/org/apache/lucene/search/trie/TestTrieRangeQuery.java
===================================================================
--- contrib/queries/src/test/org/apache/lucene/search/trie/TestTrieRangeQuery.java (revision 0)
+++ contrib/queries/src/test/org/apache/lucene/search/trie/TestTrieRangeQuery.java (revision 0)
@@ -0,0 +1,93 @@
+package org.apache.lucene.search.trie;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import java.io.IOException;
+
+import junit.framework.TestCase;
+
+import org.apache.lucene.analysis.WhitespaceAnalyzer;
+import org.apache.lucene.document.Document;
+import org.apache.lucene.document.Field;
+import org.apache.lucene.index.IndexWriter;
+import org.apache.lucene.index.IndexWriter.MaxFieldLength;
+import org.apache.lucene.store.RAMDirectory;
+import org.apache.lucene.search.IndexSearcher;
+import org.apache.lucene.search.ScoreDoc;
+import org.apache.lucene.search.TopDocs;
+import org.apache.lucene.search.Sort;
+import org.apache.lucene.search.SortField;
+
+public class TestTrieRangeQuery extends TestCase
+{
+ private RAMDirectory directory;
+ private IndexSearcher searcher;
+
+ private static final long distance=66666;
+ private static final String field="long";
+
+ // @Override
+ protected void setUp() throws Exception
+ {
+ directory = new RAMDirectory();
+ IndexWriter writer = new IndexWriter(directory, new WhitespaceAnalyzer(),
+ true, MaxFieldLength.UNLIMITED);
+
+ // Add a series of 10000 docs with increasing long values
+ for (long l=0L; l