Apache
/
lucene
mirror of https://github.com/apache/lucene.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

#26396 - HighFreqTerms fixup from Jean-Fran��ois Halleux 

git-svn-id: https://svn.apache.org/repos/asf/lucene/java/trunk@150941 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Erik Hatcher 2004-01-25 13:16:00 +00:00
parent 8fa85d5cdb
commit 507626829b
1 changed files with 47 additions and 65 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

112
sandbox/contributions/miscellaneous/src/java/org/apache/lucene/misc/HighFreqTerms.java
View File

@ -3,7 +3,7 @@ package org.apache.lucene.misc;
/* ==================================================================== /* ====================================================================
* The Apache Software License, Version 1.1 * The Apache Software License, Version 1.1
* *
* Copyright (c) 2001 The Apache Software Foundation. All rights * Copyright (c) 2001,2004 The Apache Software Foundation. All rights
* reserved. * reserved.
* *
* Redistribution and use in source and binary forms, with or without * Redistribution and use in source and binary forms, with or without
@ -54,10 +54,10 @@ package org.apache.lucene.misc;
* <http://www.apache.org/>. * <http://www.apache.org/>.
*/ */
import org.apache.lucene.util.PriorityQueue;
import org.apache.lucene.index.IndexReader; import org.apache.lucene.index.IndexReader;
import org.apache.lucene.index.Term; import org.apache.lucene.index.Term;
import org.apache.lucene.index.TermEnum; import org.apache.lucene.index.TermEnum;
import org.apache.lucene.util.PriorityQueue;
/** /**
* <code>HighFreqTerms</code> class extracts terms and their frequencies out * <code>HighFreqTerms</code> class extracts terms and their frequencies out
@ -65,77 +65,59 @@ import org.apache.lucene.index.TermEnum;
* *
* @version $Id$ * @version $Id$
*/ */
public class HighFreqTerms public class HighFreqTerms {
{
public static int numTerms = 100; // The top numTerms will be displayed
public static final int numTerms = 100;
public static void main(String[] args) throws Exception public static void main(String[] args) throws Exception {
{ IndexReader reader = null;
IndexReader reader = null; if (args.length == 1) {
if (args.length == 1) reader = IndexReader.open(args[0]);
{ } else {
reader = IndexReader.open(args[0]); usage();
} System.exit(1);
else }
{
usage();
System.exit(1);
}
TermInfoQueue tiq = new TermInfoQueue(numTerms); TermInfoQueue tiq = new TermInfoQueue(numTerms);
TermEnum terms = reader.terms(); TermEnum terms = reader.terms();
int minFreq = 0; while (terms.next()) {
while (terms.next()) tiq.insert(new TermInfo(terms.term(), terms.docFreq()));
{ }
if (terms.docFreq() > minFreq)
{
tiq.put(new TermInfo(terms.term(), terms.docFreq()));
if (tiq.size() > numTerms) // if tiq overfull
{
tiq.pop(); // remove lowest in tiq
minFreq = ((TermInfo)tiq.top()).docFreq; // reset minFreq
}
}
}
while (tiq.size() != 0) while (tiq.size() != 0) {
{ TermInfo termInfo = (TermInfo) tiq.pop();
TermInfo termInfo = (TermInfo)tiq.pop(); System.out.println(termInfo.term + " " + termInfo.docFreq);
System.out.println(termInfo.term + " " + termInfo.docFreq); }
}
reader.close(); reader.close();
} }
private static void usage() private static void usage() {
{ System.out.println(
System.out.println("\n\n" + "\n\n"
"java org.apache.lucene.misc.HighFreqTerms <index dir>\n\n"); + "java org.apache.lucene.misc.HighFreqTerms <index dir>\n\n");
} }
} }
final class TermInfo final class TermInfo {
{ TermInfo(Term t, int df) {
TermInfo(Term t, int df) term = t;
{ docFreq = df;
term = t; }
docFreq = df; int docFreq;
} Term term;
int docFreq;
Term term;
} }
final class TermInfoQueue extends PriorityQueue final class TermInfoQueue extends PriorityQueue {
{ TermInfoQueue(int size) {
TermInfoQueue(int size) initialize(size);
{ }
initialize(size);
} protected final boolean lessThan(Object a, Object b) {
protected final boolean lessThan(Object a, Object b) TermInfo termInfoA = (TermInfo) a;
{ TermInfo termInfoB = (TermInfo) b;
TermInfo termInfoA = (TermInfo)a; return termInfoA.docFreq < termInfoB.docFreq;
TermInfo termInfoB = (TermInfo)b; }
return termInfoA.docFreq < termInfoB.docFreq;
}
} }