Combining filter rewrite and skip list to optimize sub aggregation

CHANGELOG.md

@@ -16,6 +16,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 - Add pointer based lag metric in pull-based ingestion ([#19635](https://github.com/opensearch-project/OpenSearch/pull/19635))
 
 ### Changed
+- Combining filter rewrite and skip list to optimize sub aggregation([#19573](https://github.com/opensearch-project/OpenSearch/pull/19573))
 - Faster `terms` query creation for `keyword` field with index and docValues enabled ([#19350](https://github.com/opensearch-project/OpenSearch/pull/19350))
 - Refactor to move prepareIndex and prepareDelete methods to Engine class ([#19551](https://github.com/opensearch-project/OpenSearch/pull/19551))
 - Omit maxScoreCollector in SimpleTopDocsCollectorContext when concurrent segment search enabled ([#19584](https://github.com/opensearch-project/OpenSearch/pull/19584))

gradle/libs.versions.toml

@@ -1,6 +1,6 @@
 [versions]
 opensearch        = "3.4.0"
-lucene            = "10.3.1"
+lucene            = "10.4.0-snapshot-a8c1455"
 
 bundled_jdk_vendor = "adoptium"
 bundled_jdk = "24.0.2+12"

libs/core/licenses/lucene-core-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+96781fdefe4a7e2ac3482ccb41801b2cd1607501

libs/core/src/main/java/org/opensearch/Version.java

@@ -123,7 +123,7 @@ public class Version implements Comparable<Version>, ToXContentFragment {
     public static final Version V_3_2_0 = new Version(3020099, org.apache.lucene.util.Version.LUCENE_10_2_2);
     public static final Version V_3_3_0 = new Version(3030099, org.apache.lucene.util.Version.LUCENE_10_3_1);
     public static final Version V_3_3_1 = new Version(3030199, org.apache.lucene.util.Version.LUCENE_10_3_1);
-    public static final Version V_3_4_0 = new Version(3040099, org.apache.lucene.util.Version.LUCENE_10_3_1);
+    public static final Version V_3_4_0 = new Version(3040099, org.apache.lucene.util.Version.LUCENE_10_4_0);
     public static final Version CURRENT = V_3_4_0;
 
     public static Version fromId(int id) {

modules/lang-expression/licenses/lucene-expressions-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+886e976447a505c02e636ba67b4d381b257a3f44

plugins/analysis-icu/licenses/lucene-analysis-icu-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+2ad93f9924ecc4491b26dc540bef1fabe2afd2ea

plugins/analysis-kuromoji/licenses/lucene-analysis-kuromoji-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+6efc194346481fb1ca2a174182ad2429294021bb

plugins/analysis-nori/licenses/lucene-analysis-nori-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+11a5ecc992d8725bfb760925f07fe13a40ce4b06

plugins/analysis-phonetic/licenses/lucene-analysis-phonetic-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+7821a0b53f5c3e25418659163968f60327484da1

plugins/analysis-smartcn/licenses/lucene-analysis-smartcn-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+6bff6a2e336dee93a4dff882138296d80ab62fd6

plugins/analysis-stempel/licenses/lucene-analysis-stempel-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+5d76ae18c4e5119845cd278ed276929505c88bfe

plugins/analysis-ukrainian/licenses/lucene-analysis-morfologik-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+9513d1dcd7dd057a9026fa3c0764d778540d3b8d

server/licenses/lucene-analysis-common-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+7142df4d1076f734aabd863fc56516b58dcae605

server/licenses/lucene-backward-codecs-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+3d9f4cf13f20e1b384021a808d8f57a73427287e

server/licenses/lucene-core-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+96781fdefe4a7e2ac3482ccb41801b2cd1607501

server/licenses/lucene-grouping-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+beba73c7a31f7f72d21a1485b200d864706638b3

server/licenses/lucene-highlighter-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+077fca639cabd72a1c8521e6e3e8aacc36d0b7eb

server/licenses/lucene-join-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+79ab48828901f08ca79b74fd0d1de855a7ed8cda

server/licenses/lucene-memory-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+de41ae03a22b20fa52140f178ae62cad7724cf4e

server/licenses/lucene-misc-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+7ae8474b4a17d92394126f9c6cc8e76e479df388

server/licenses/lucene-queries-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+24e09399d8c70f5a6a0d72546cec522696bf4bac

server/licenses/lucene-queryparser-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+86cf4d847c39a0081d23535be84bf36dbb6a262a

server/licenses/lucene-sandbox-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+c1078394a5db5510b3b8d508440d53be642fc1eb

server/licenses/lucene-spatial-extras-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+8b7925295d45c23b9fc087e9762afc74da578ae5

server/licenses/lucene-spatial3d-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+055860e2000dfc62982ccb1e2ba4ea8bf75acd8e

server/licenses/lucene-suggest-10.4.0-snapshot-a8c1455.jar.sha1

@@ -0,0 +1 @@
+197e5f4d7f5dab0442c1865fce9dd941d7620022

server/src/main/java/org/opensearch/common/lucene/Lucene.java

@@ -111,7 +111,7 @@
  * @opensearch.internal
  */
 public class Lucene {
-    public static final String LATEST_CODEC = "Lucene103";
+    public static final String LATEST_CODEC = "Lucene104";
 
     public static final String SOFT_DELETES_FIELD = "__soft_deletes";
     public static final String PARENT_FIELD = "__nested_parent";

server/src/main/java/org/opensearch/index/codec/CodecService.java

@@ -34,7 +34,7 @@
 
 import org.apache.logging.log4j.Logger;
 import org.apache.lucene.codecs.Codec;
-import org.apache.lucene.codecs.lucene103.Lucene103Codec;
+import org.apache.lucene.codecs.lucene104.Lucene104Codec;
 import org.opensearch.common.Nullable;
 import org.opensearch.common.annotation.PublicApi;
 import org.opensearch.common.collect.MapBuilder;
@@ -71,23 +71,23 @@ public CodecService(@Nullable MapperService mapperService, IndexSettings indexSe
         final MapBuilder<String, Codec> codecs = MapBuilder.<String, Codec>newMapBuilder();
         assert null != indexSettings;
         if (mapperService == null) {
-            codecs.put(DEFAULT_CODEC, new Lucene103Codec());
-            codecs.put(LZ4, new Lucene103Codec());
-            codecs.put(BEST_COMPRESSION_CODEC, new Lucene103Codec(Lucene103Codec.Mode.BEST_COMPRESSION));
-            codecs.put(ZLIB, new Lucene103Codec(Lucene103Codec.Mode.BEST_COMPRESSION));
+            codecs.put(DEFAULT_CODEC, new Lucene104Codec());
+            codecs.put(LZ4, new Lucene104Codec());
+            codecs.put(BEST_COMPRESSION_CODEC, new Lucene104Codec(Lucene104Codec.Mode.BEST_COMPRESSION));
+            codecs.put(ZLIB, new Lucene104Codec(Lucene104Codec.Mode.BEST_COMPRESSION));
         } else {
             // CompositeCodec still delegates to PerFieldMappingPostingFormatCodec
             // We can still support all the compression codecs when composite index is present
             if (mapperService.isCompositeIndexPresent()) {
                 codecs.putAll(compositeCodecFactory.getCompositeIndexCodecs(mapperService, logger));
             } else {
-                codecs.put(DEFAULT_CODEC, new PerFieldMappingPostingFormatCodec(Lucene103Codec.Mode.BEST_SPEED, mapperService, logger));
-                codecs.put(LZ4, new PerFieldMappingPostingFormatCodec(Lucene103Codec.Mode.BEST_SPEED, mapperService, logger));
+                codecs.put(DEFAULT_CODEC, new PerFieldMappingPostingFormatCodec(Lucene104Codec.Mode.BEST_SPEED, mapperService, logger));
+                codecs.put(LZ4, new PerFieldMappingPostingFormatCodec(Lucene104Codec.Mode.BEST_SPEED, mapperService, logger));
                 codecs.put(
                     BEST_COMPRESSION_CODEC,
-                    new PerFieldMappingPostingFormatCodec(Lucene103Codec.Mode.BEST_COMPRESSION, mapperService, logger)
+                    new PerFieldMappingPostingFormatCodec(Lucene104Codec.Mode.BEST_COMPRESSION, mapperService, logger)
                 );
-                codecs.put(ZLIB, new PerFieldMappingPostingFormatCodec(Lucene103Codec.Mode.BEST_COMPRESSION, mapperService, logger));
+                codecs.put(ZLIB, new PerFieldMappingPostingFormatCodec(Lucene104Codec.Mode.BEST_COMPRESSION, mapperService, logger));
             }
         }
         codecs.put(LUCENE_DEFAULT_CODEC, Codec.getDefault());

server/src/main/java/org/opensearch/index/codec/PerFieldMappingPostingFormatCodec.java

@@ -36,7 +36,7 @@
 import org.apache.lucene.codecs.Codec;
 import org.apache.lucene.codecs.DocValuesFormat;
 import org.apache.lucene.codecs.PostingsFormat;
-import org.apache.lucene.codecs.lucene103.Lucene103Codec;
+import org.apache.lucene.codecs.lucene104.Lucene104Codec;
 import org.apache.lucene.codecs.lucene90.Lucene90DocValuesFormat;
 import org.opensearch.common.lucene.Lucene;
 import org.opensearch.index.codec.fuzzy.FuzzyFilterPostingsFormat;
@@ -59,7 +59,7 @@
  *
  * @opensearch.internal
  */
-public class PerFieldMappingPostingFormatCodec extends Lucene103Codec {
+public class PerFieldMappingPostingFormatCodec extends Lucene104Codec {
     private final Logger logger;
     private final MapperService mapperService;
     private final DocValuesFormat dvFormat = new Lucene90DocValuesFormat();

server/src/main/java/org/opensearch/index/codec/composite/CompositeCodecFactory.java

@@ -10,10 +10,9 @@
 
 import org.apache.logging.log4j.Logger;
 import org.apache.lucene.codecs.Codec;
-import org.apache.lucene.codecs.lucene103.Lucene103Codec;
+import org.apache.lucene.codecs.lucene104.Lucene104Codec;
 import org.opensearch.common.annotation.ExperimentalApi;
-import org.opensearch.index.codec.composite.backward_codecs.composite101.Composite101Codec;
-import org.opensearch.index.codec.composite.composite103.Composite103Codec;
+import org.opensearch.index.codec.composite.composite104.Composite104Codec;
 import org.opensearch.index.mapper.MapperService;
 
 import java.util.HashMap;
@@ -33,16 +32,16 @@
 public class CompositeCodecFactory {
 
     // we can use this to track the latest composite codec
-    public static final String COMPOSITE_CODEC = Composite101Codec.COMPOSITE_INDEX_CODEC_NAME;
+    public static final String COMPOSITE_CODEC = Composite104Codec.COMPOSITE_INDEX_CODEC_NAME;
 
     public CompositeCodecFactory() {}
 
     public Map<String, Codec> getCompositeIndexCodecs(MapperService mapperService, Logger logger) {
         Map<String, Codec> codecs = new HashMap<>();
-        codecs.put(DEFAULT_CODEC, new Composite103Codec(Lucene103Codec.Mode.BEST_SPEED, mapperService, logger));
-        codecs.put(LZ4, new Composite103Codec(Lucene103Codec.Mode.BEST_SPEED, mapperService, logger));
-        codecs.put(BEST_COMPRESSION_CODEC, new Composite103Codec(Lucene103Codec.Mode.BEST_COMPRESSION, mapperService, logger));
-        codecs.put(ZLIB, new Composite103Codec(Lucene103Codec.Mode.BEST_COMPRESSION, mapperService, logger));
+        codecs.put(DEFAULT_CODEC, new Composite104Codec(Lucene104Codec.Mode.BEST_SPEED, mapperService, logger));
+        codecs.put(LZ4, new Composite104Codec(Lucene104Codec.Mode.BEST_SPEED, mapperService, logger));
+        codecs.put(BEST_COMPRESSION_CODEC, new Composite104Codec(Lucene104Codec.Mode.BEST_COMPRESSION, mapperService, logger));
+        codecs.put(ZLIB, new Composite104Codec(Lucene104Codec.Mode.BEST_COMPRESSION, mapperService, logger));
         return codecs;
     }
 }

server/src/main/java/org/opensearch/index/codec/composite/backward_codecs/composite101/package-info.java

@@ -7,6 +7,6 @@
  */
 
 /**
- *  Responsible for handling all composite index codecs and operations associated with Composite100 codec
+ *  Responsible for handling all composite index codecs and operations associated with Composite101 codec
  */
 package org.opensearch.index.codec.composite.backward_codecs.composite101;

server/src/main/java/org/opensearch/index/codec/composite/composite103/Composite103Codec.java → server/src/main/java/org/opensearch/index/codec/composite/backward_codecs/composite103/Composite103Codec.java

@@ -6,15 +6,13 @@
  * compatible open source license.
  */
 
-package org.opensearch.index.codec.composite.composite103;
+package org.opensearch.index.codec.composite.backward_codecs.composite103;
 
-import org.apache.logging.log4j.Logger;
+import org.apache.lucene.backward_codecs.lucene103.Lucene103Codec;
 import org.apache.lucene.codecs.Codec;
 import org.apache.lucene.codecs.DocValuesFormat;
 import org.apache.lucene.codecs.FilterCodec;
-import org.apache.lucene.codecs.lucene103.Lucene103Codec;
 import org.opensearch.common.annotation.ExperimentalApi;
-import org.opensearch.index.codec.PerFieldMappingPostingFormatCodec;
 import org.opensearch.index.codec.composite.composite912.Composite912DocValuesFormat;
 import org.opensearch.index.mapper.MapperService;
 
@@ -34,10 +32,6 @@ public Composite103Codec() {
         this(COMPOSITE_INDEX_CODEC_NAME, new Lucene103Codec(), null);
     }
 
-    public Composite103Codec(Lucene103Codec.Mode compressionMode, MapperService mapperService, Logger logger) {
-        this(COMPOSITE_INDEX_CODEC_NAME, new PerFieldMappingPostingFormatCodec(compressionMode, mapperService, logger), mapperService);
-    }
-
     /**
      * Sole constructor. When subclassing this codec, create a no-arg ctor and pass the delegate codec and a unique name to
      * this ctor.

server/src/main/java/org/opensearch/index/codec/composite/backward_codecs/composite103/package-info.java

@@ -0,0 +1,12 @@
+/*
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * The OpenSearch Contributors require contributions made to
+ * this file be licensed under the Apache-2.0 license or a
+ * compatible open source license.
+ */
+
+/**
+ *  Responsible for handling all composite index codecs and operations associated with Composite103 codec
+ */
+package org.opensearch.index.codec.composite.backward_codecs.composite103;

server/src/main/java/org/opensearch/index/codec/composite/composite104/Composite104Codec.java

@@ -0,0 +1,58 @@
+/*
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * The OpenSearch Contributors require contributions made to
+ * this file be licensed under the Apache-2.0 license or a
+ * compatible open source license.
+ */
+
+package org.opensearch.index.codec.composite.composite104;
+
+import org.apache.logging.log4j.Logger;
+import org.apache.lucene.codecs.Codec;
+import org.apache.lucene.codecs.DocValuesFormat;
+import org.apache.lucene.codecs.FilterCodec;
+import org.apache.lucene.codecs.lucene104.Lucene104Codec;
+import org.opensearch.common.annotation.ExperimentalApi;
+import org.opensearch.index.codec.PerFieldMappingPostingFormatCodec;
+import org.opensearch.index.codec.composite.composite912.Composite912DocValuesFormat;
+import org.opensearch.index.mapper.MapperService;
+
+/**
+ *  Extends the Codec to support new file formats for composite indices eg: star tree index
+ *  based on the mappings.
+ *
+ * @opensearch.experimental
+ */
+@ExperimentalApi
+public class Composite104Codec extends FilterCodec {
+    public static final String COMPOSITE_INDEX_CODEC_NAME = "Composite104Codec";
+    private final MapperService mapperService;
+
+    // needed for SPI - this is used in reader path
+    public Composite104Codec() {
+        this(COMPOSITE_INDEX_CODEC_NAME, new Lucene104Codec(), null);
+    }
+
+    public Composite104Codec(Lucene104Codec.Mode compressionMode, MapperService mapperService, Logger logger) {
+        this(COMPOSITE_INDEX_CODEC_NAME, new PerFieldMappingPostingFormatCodec(compressionMode, mapperService, logger), mapperService);
+    }
+
+    /**
+     * Sole constructor. When subclassing this codec, create a no-arg ctor and pass the delegate codec and a unique name to
+     * this ctor.
+     *
+     * @param name name of the codec
+     * @param delegate codec delegate
+     * @param mapperService mapper service instance
+     */
+    protected Composite104Codec(String name, Codec delegate, MapperService mapperService) {
+        super(name, delegate);
+        this.mapperService = mapperService;
+    }
+
+    @Override
+    public DocValuesFormat docValuesFormat() {
+        return new Composite912DocValuesFormat(mapperService);
+    }
+}

server/src/main/java/org/opensearch/index/codec/composite/composite103/package-info.java → server/src/main/java/org/opensearch/index/codec/composite/composite104/package-info.java

@@ -7,6 +7,6 @@
  */
 
 /**
- *  Responsible for handling all composite index codecs and operations associated with Composite100 codec
+ *  Responsible for handling all composite index codecs and operations associated with Composite104 codec
  */
-package org.opensearch.index.codec.composite.composite103;
+package org.opensearch.index.codec.composite.composite104;