(lucene) branch branch_10x updated: Reduce long[] array allocation for bitset in readBitSetIterator (#13828)

[zh...@apache.org]

This is an automated email from the ASF dual-hosted git repository.

zhangchao pushed a commit to branch branch_10x
in repository https://gitbox.apache.org/repos/asf/lucene.git


The following commit(s) were added to refs/heads/branch_10x by this push:
     new e10e9d136d7 Reduce long[] array allocation for bitset in readBitSetIterator (#13828)
e10e9d136d7 is described below

commit e10e9d136d7ee6959267044e1bea055b9907aee2
Author: Zhang Chao <80...@qq.com>
AuthorDate: Tue Oct 8 10:19:28 2024 +0800

    Reduce long[] array allocation for bitset in readBitSetIterator (#13828)
---
 lucene/CHANGES.txt                                 |  2 +-
 .../org/apache/lucene/util/bkd/DocIdsWriter.java   | 22 +++++++++++++++-------
 2 files changed, 16 insertions(+), 8 deletions(-)

diff --git a/lucene/CHANGES.txt b/lucene/CHANGES.txt
index 98854bf8758..694143586d3 100644
--- a/lucene/CHANGES.txt
+++ b/lucene/CHANGES.txt
@@ -19,7 +19,7 @@ Improvements
 
 Optimizations
 ---------------------
-(No changes)
+* GITHUB#13828: Reduce long[] array allocation for bitset in readBitSetIterator. (Zhang Chao)
 
 Bug Fixes
 ---------------------
diff --git a/lucene/core/src/java/org/apache/lucene/util/bkd/DocIdsWriter.java b/lucene/core/src/java/org/apache/lucene/util/bkd/DocIdsWriter.java
index 9f6a10b9ddc..b9ea0d9aa08 100644
--- a/lucene/core/src/java/org/apache/lucene/util/bkd/DocIdsWriter.java
+++ b/lucene/core/src/java/org/apache/lucene/util/bkd/DocIdsWriter.java
@@ -17,13 +17,16 @@
 package org.apache.lucene.util.bkd;
 
 import java.io.IOException;
+import java.util.Arrays;
 import org.apache.lucene.index.PointValues.IntersectVisitor;
 import org.apache.lucene.search.DocIdSetIterator;
 import org.apache.lucene.store.DataOutput;
 import org.apache.lucene.store.IndexInput;
+import org.apache.lucene.util.ArrayUtil;
 import org.apache.lucene.util.DocBaseBitSetIterator;
 import org.apache.lucene.util.FixedBitSet;
 import org.apache.lucene.util.IntsRef;
+import org.apache.lucene.util.LongsRef;
 
 final class DocIdsWriter {
 
@@ -36,6 +39,7 @@ final class DocIdsWriter {
   private static final byte LEGACY_DELTA_VINT = (byte) 0;
 
   private final int[] scratch;
+  private final LongsRef scratchLongs = new LongsRef();
 
   /**
    * IntsRef to be used to iterate over the scratch buffer. A single instance is reused to avoid
@@ -205,12 +209,17 @@ final class DocIdsWriter {
     }
   }
 
-  private static DocIdSetIterator readBitSetIterator(IndexInput in, int count) throws IOException {
+  private DocIdSetIterator readBitSetIterator(IndexInput in, int count) throws IOException {
     int offsetWords = in.readVInt();
     int longLen = in.readVInt();
-    long[] bits = new long[longLen];
-    in.readLongs(bits, 0, longLen);
-    FixedBitSet bitSet = new FixedBitSet(bits, longLen << 6);
+    scratchLongs.longs = ArrayUtil.growNoCopy(scratchLongs.longs, longLen);
+    in.readLongs(scratchLongs.longs, 0, longLen);
+    // make ghost bits clear for FixedBitSet.
+    if (longLen < scratchLongs.length) {
+      Arrays.fill(scratchLongs.longs, longLen, scratchLongs.longs.length, 0);
+    }
+    scratchLongs.length = longLen;
+    FixedBitSet bitSet = new FixedBitSet(scratchLongs.longs, longLen << 6);
     return new DocBaseBitSetIterator(bitSet, count, offsetWords << 6);
   }
 
@@ -230,7 +239,7 @@ final class DocIdsWriter {
     }
   }
 
-  private static void readBitSet(IndexInput in, int count, int[] docIDs) throws IOException {
+  private void readBitSet(IndexInput in, int count, int[] docIDs) throws IOException {
     DocIdSetIterator iterator = readBitSetIterator(in, count);
     int docId, pos = 0;
     while ((docId = iterator.nextDoc()) != DocIdSetIterator.NO_MORE_DOCS) {
@@ -307,8 +316,7 @@ final class DocIdsWriter {
     }
   }
 
-  private static void readBitSet(IndexInput in, int count, IntersectVisitor visitor)
-      throws IOException {
+  private void readBitSet(IndexInput in, int count, IntersectVisitor visitor) throws IOException {
     DocIdSetIterator bitSetIterator = readBitSetIterator(in, count);
     visitor.visit(bitSetIterator);
   }