Use optimizedText in match_only_text fields

Author: jordan-powers

modules/mapper-extras/src/main/java/org/elasticsearch/index/mapper/extras/MatchOnlyTextFieldMapper.java

@@ -31,6 +31,7 @@
 import org.apache.lucene.util.IOFunction;
 import org.elasticsearch.common.CheckedIntFunction;
 import org.elasticsearch.common.lucene.Lucene;
+import org.elasticsearch.common.text.UTF8DecodingReader;
 import org.elasticsearch.common.unit.Fuzziness;
 import org.elasticsearch.index.IndexVersion;
 import org.elasticsearch.index.IndexVersions;
@@ -384,7 +385,7 @@ public IndexFieldData.Builder fielddataBuilder(FieldDataContext fieldDataContext
                 ) {
                     @Override
                     protected BytesRef storedToBytesRef(Object stored) {
-                        return new BytesRef((String) stored);
+                        return (BytesRef) stored;
                     }
                 };
             }
@@ -442,18 +443,20 @@ public FieldMapper.Builder getMergeBuilder() {
 
     @Override
     protected void parseCreateField(DocumentParserContext context) throws IOException {
-        final String value = context.parser().textOrNull();
+        final var value = context.parser().optimizedTextOrNull();
 
         if (value == null) {
             return;
         }
 
-        Field field = new Field(fieldType().name(), value, fieldType);
+        final var utfBytes = value.bytes();
+        Field field = new Field(fieldType().name(), new UTF8DecodingReader(utfBytes), fieldType);
         context.doc().add(field);
         context.addToFieldNames(fieldType().name());
 
         if (storeSource) {
-            context.doc().add(new StoredField(fieldType().storedFieldNameForSyntheticSource(), value));
+            final var bytesRef = new BytesRef(utfBytes.bytes(), utfBytes.offset(), utfBytes.length());
+            context.doc().add(new StoredField(fieldType().storedFieldNameForSyntheticSource(), bytesRef));
         }
     }
 
@@ -473,7 +476,7 @@ protected SyntheticSourceSupport syntheticSourceSupport() {
             () -> new StringStoredFieldFieldLoader(fieldType().storedFieldNameForSyntheticSource(), fieldType().name(), leafName()) {
                 @Override
                 protected void write(XContentBuilder b, Object value) throws IOException {
-                    b.value((String) value);
+                    b.value(((BytesRef) value).utf8ToString());
                 }
             }
         );

modules/mapper-extras/src/main/java/org/elasticsearch/index/mapper/extras/SourceConfirmedTextQuery.java

@@ -41,6 +41,7 @@
 import org.apache.lucene.search.Weight;
 import org.apache.lucene.search.similarities.Similarity;
 import org.apache.lucene.search.similarities.Similarity.SimScorer;
+import org.apache.lucene.util.BytesRef;
 import org.apache.lucene.util.IOFunction;
 import org.elasticsearch.common.CheckedIntFunction;
 import org.elasticsearch.common.lucene.search.MultiPhrasePrefixQuery;
@@ -438,7 +439,13 @@ private MemoryIndex getOrCreateMemoryIndex() throws IOException {
                     if (value == null) {
                         continue;
                     }
-                    cacheEntry.memoryIndex.addField(field, value.toString(), indexAnalyzer);
+                    String valueStr;
+                    if (value instanceof BytesRef valueRef) {
+                        valueStr = valueRef.utf8ToString();
+                    } else {
+                        valueStr = value.toString();
+                    }
+                    cacheEntry.memoryIndex.addField(field, valueStr, indexAnalyzer);
                 }
             }
             return cacheEntry.memoryIndex;

modules/mapper-extras/src/test/java/org/elasticsearch/index/mapper/extras/MatchOnlyTextFieldMapperTests.java

@@ -123,7 +123,12 @@ public void testDefaults() throws IOException {
         ParsedDocument doc = mapper.parse(source(b -> b.field("field", "1234")));
         List<IndexableField> fields = doc.rootDoc().getFields("field");
         assertEquals(1, fields.size());
-        assertEquals("1234", fields.get(0).stringValue());
+
+        var reader = fields.get(0).readerValue();
+        char[] buff = new char[20];
+        assertEquals(4, reader.read(buff));
+        assertEquals("1234", new String(buff, 0, 4));
+
         IndexableFieldType fieldType = fields.get(0).fieldType();
         assertThat(fieldType.omitNorms(), equalTo(true));
         assertTrue(fieldType.tokenized());

server/src/main/java/org/elasticsearch/common/text/UTF8DecodingReader.java

@@ -0,0 +1,53 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the "Elastic License
+ * 2.0", the "GNU Affero General Public License v3.0 only", and the "Server Side
+ * Public License v 1"; you may not use this file except in compliance with, at
+ * your election, the "Elastic License 2.0", the "GNU Affero General Public
+ * License v3.0 only", or the "Server Side Public License, v 1".
+ */
+
+package org.elasticsearch.common.text;
+
+import org.elasticsearch.xcontent.XContentString;
+
+import java.io.Reader;
+import java.nio.ByteBuffer;
+import java.nio.CharBuffer;
+import java.nio.charset.CharsetDecoder;
+import java.nio.charset.StandardCharsets;
+
+/**
+ * Reader that decodes UTF-8 formatted bytes into chars.
+ */
+public class UTF8DecodingReader extends Reader {
+    private CharsetDecoder decoder = StandardCharsets.UTF_8.newDecoder();
+    private ByteBuffer bytes;
+
+    public UTF8DecodingReader(ByteBuffer bytes) {
+        this.bytes = bytes;
+    }
+
+    public UTF8DecodingReader(XContentString.UTF8Bytes utf8bytes) {
+        this.bytes = ByteBuffer.wrap(utf8bytes.bytes(), utf8bytes.offset(), utf8bytes.length());
+    }
+
+    @Override
+    public int read(char[] cbuf, int off, int len) {
+        return read(CharBuffer.wrap(cbuf, off, len));
+    }
+
+    @Override
+    public int read(CharBuffer cbuf) {
+        if (bytes.hasRemaining() == false) {
+            return -1;
+        }
+
+        int startPos = cbuf.position();
+        decoder.decode(bytes, cbuf, true);
+        return cbuf.position() - startPos;
+    }
+
+    @Override
+    public void close() {}
+}