use lucene for xor distance

benwtrent · Jun 17, 2024 · d639369 · d639369
1 parent a928a99
commit d639369
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 22 deletions.
diff --git a/server/src/main/java/org/elasticsearch/script/field/vectors/ByteBinaryDenseVector.java b/server/src/main/java/org/elasticsearch/script/field/vectors/ByteBinaryDenseVector.java
@@ -8,7 +8,6 @@
 
 package org.elasticsearch.script.field.vectors;
 
-import org.apache.lucene.util.BitUtil;
 import org.apache.lucene.util.BytesRef;
 import org.apache.lucene.util.VectorUtil;
 import org.elasticsearch.core.SuppressForbidden;
@@ -103,17 +102,7 @@ public double l1Norm(List<Number> queryVector) {
 
     @Override
     public int hamming(byte[] queryVector) {
-        int distance = 0, i = 0;
-        for (final int upperBound = queryVector.length & ~(Long.BYTES - 1); i < upperBound; i += Long.BYTES) {
-            distance += Long.bitCount(
-                (long) BitUtil.VH_NATIVE_LONG.get(queryVector, i) ^ (long) BitUtil.VH_NATIVE_LONG.get(vectorValue, i)
-            );
-        }
-        // tail:
-        for (; i < queryVector.length; i++) {
-            distance += Integer.bitCount((queryVector[i] ^ vectorValue[i]) & 0xFF);
-        }
-        return distance;
+        return VectorUtil.xorBitCount(queryVector, vectorValue);
     }
 
     @Override

diff --git a/server/src/main/java/org/elasticsearch/script/field/vectors/ByteKnnDenseVector.java b/server/src/main/java/org/elasticsearch/script/field/vectors/ByteKnnDenseVector.java
@@ -8,7 +8,6 @@
 
 package org.elasticsearch.script.field.vectors;
 
-import org.apache.lucene.util.BitUtil;
 import org.apache.lucene.util.VectorUtil;
 import org.elasticsearch.core.SuppressForbidden;
 
@@ -104,15 +103,7 @@ public double l1Norm(List<Number> queryVector) {
 
     @Override
     public int hamming(byte[] queryVector) {
-        int distance = 0, i = 0;
-        for (final int upperBound = queryVector.length & ~(Long.BYTES - 1); i < upperBound; i += Long.BYTES) {
-            distance += Long.bitCount((long) BitUtil.VH_NATIVE_LONG.get(queryVector, i) ^ (long) BitUtil.VH_NATIVE_LONG.get(docVector, i));
-        }
-        // tail:
-        for (; i < queryVector.length; i++) {
-            distance += Integer.bitCount((queryVector[i] ^ docVector[i]) & 0xFF);
-        }
-        return distance;
+        return VectorUtil.xorBitCount(queryVector, docVector);
     }
 
     @Override