cutting 2004/10/13 10:12:04
Modified: src/java/org/apache/lucene/search BooleanScorer.java
Log:
Reverse a mistaken optimization. With a larger collection and set of
queries I find that a table size of 2k works best.
Revision Changes Path
1.13 +1 -1
jakarta-lucene/src/java/org/apache/lucene/search/BooleanScorer.java
Index: BooleanScorer.java
===================================================================
RCS file:
/home/cvs/jakarta-lucene/src/java/org/apache/lucene/search/BooleanScorer.java,v
retrieving revision 1.12
retrieving revision 1.13
diff -u -r1.12 -r1.13
--- BooleanScorer.java 22 Sep 2004 17:56:00 -0000 1.12
+++ BooleanScorer.java 13 Oct 2004 17:12:04 -0000 1.13
@@ -176,7 +176,7 @@
/** A simple hash table of document scores within a range. */
static final class BucketTable {
- public static final int SIZE = 1 << 8;
+ public static final int SIZE = 1 << 11;
public static final int MASK = SIZE - 1;
final Bucket[] buckets = new Bucket[SIZE];
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
