Store offsets in sorted doc values field instead of binary doc values field

Author: martijnvg

server/src/main/java/org/elasticsearch/index/mapper/FieldArrayContext.java

@@ -9,7 +9,7 @@
 
 package org.elasticsearch.index.mapper;
 
-import org.apache.lucene.document.BinaryDocValuesField;
+import org.apache.lucene.document.SortedDocValuesField;
 import org.apache.lucene.util.BitUtil;
 import org.elasticsearch.common.io.stream.BytesStreamOutput;
 
@@ -65,7 +65,7 @@ void addToLuceneDocument(DocumentParserContext context) throws IOException {
                 for (int ord : offsetToOrd) {
                     streamOutput.writeVInt(BitUtil.zigZagEncode(ord));
                 }
-                context.doc().add(new BinaryDocValuesField(fieldName, streamOutput.bytes().toBytesRef()));
+                context.doc().add(new SortedDocValuesField(fieldName, streamOutput.bytes().toBytesRef()));
             }
         }
     }

server/src/main/java/org/elasticsearch/index/mapper/SortedSetWithOffsetsDocValuesSyntheticFieldLoaderLayer.java

@@ -12,6 +12,7 @@
 import org.apache.lucene.index.BinaryDocValues;
 import org.apache.lucene.index.DocValues;
 import org.apache.lucene.index.LeafReader;
+import org.apache.lucene.index.SortedDocValues;
 import org.apache.lucene.index.SortedSetDocValues;
 import org.apache.lucene.util.BitUtil;
 import org.apache.lucene.util.BytesRef;
@@ -46,7 +47,7 @@ public String fieldName() {
     @Override
     public DocValuesLoader docValuesLoader(LeafReader leafReader, int[] docIdsInLeaf) throws IOException {
         SortedSetDocValues valueDocValues = DocValues.getSortedSet(leafReader, name);
-        BinaryDocValues offsetDocValues = DocValues.getBinary(leafReader, offsetsFieldName);
+        SortedDocValues offsetDocValues = DocValues.getSorted(leafReader, offsetsFieldName);
 
         return docValues = new DocValuesWithOffsetsLoader(valueDocValues, offsetDocValues);
     }
@@ -77,15 +78,15 @@ public void write(XContentBuilder b) throws IOException {
     }
 
     static final class DocValuesWithOffsetsLoader implements DocValuesLoader {
-        private final BinaryDocValues offsetDocValues;
+        private final SortedDocValues offsetDocValues;
         private final SortedSetDocValues valueDocValues;
         private final ByteArrayStreamInput scratch = new ByteArrayStreamInput();
 
         private boolean hasValue;
         private boolean hasOffset;
         private int[] offsetToOrd;
 
-        DocValuesWithOffsetsLoader(SortedSetDocValues valueDocValues, BinaryDocValues offsetDocValues) {
+        DocValuesWithOffsetsLoader(SortedSetDocValues valueDocValues, SortedDocValues offsetDocValues) {
             this.valueDocValues = valueDocValues;
             this.offsetDocValues = offsetDocValues;
         }
@@ -96,7 +97,8 @@ public boolean advanceToDoc(int docId) throws IOException {
             hasOffset = offsetDocValues.advanceExact(docId);
             if (hasValue || hasOffset) {
                 if (hasOffset) {
-                    var encodedValue = offsetDocValues.binaryValue();
+                    int offsetOrd = offsetDocValues.ordValue();
+                    var encodedValue = offsetDocValues.lookupOrd(offsetOrd);
                     scratch.reset(encodedValue.bytes, encodedValue.offset, encodedValue.length);
                     offsetToOrd = parseOffsetArray(scratch);
                 } else {

server/src/test/java/org/elasticsearch/index/mapper/SyntheticSourceNativeArrayIntegrationTests.java

@@ -186,7 +186,7 @@ private void verifySyntheticArray(Object[][] arrays, XContentBuilder mapping, In
                 assertThat(storedFieldNames, contains(expectedStoredFields));
             }
             var fieldInfo = FieldInfos.getMergedFieldInfos(reader).fieldInfo("field.offsets");
-            assertThat(fieldInfo.getDocValuesType(), equalTo(DocValuesType.BINARY));
+            assertThat(fieldInfo.getDocValuesType(), equalTo(DocValuesType.SORTED));
         }
     }
 
@@ -324,7 +324,7 @@ private void verifySyntheticArrayInObject(List<Object[]> documents) throws IOExc
                 assertThat(storedFieldNames, contains("_id"));
             }
             var fieldInfo = FieldInfos.getMergedFieldInfos(reader).fieldInfo("object.field.offsets");
-            assertThat(fieldInfo.getDocValuesType(), equalTo(DocValuesType.BINARY));
+            assertThat(fieldInfo.getDocValuesType(), equalTo(DocValuesType.SORTED));
         }
     }