Fix decoding of non-ascii field names in ignored source (#132018) (#132034)

When encoding an ignored source entry, we write the string length of the
field name, not the encoded byte count; however, the decode logic treats
this encoded value as the byte length. This patch updates the decode logic
to instead properly treat the value as the string length.

(cherry picked from commit 178c0c9)

Author: jordan-powers

docs/changelog/132018.yaml

@@ -0,0 +1,5 @@
+pr: 132018
+summary: Fix decoding of non-ascii field names in ignored source
+area: Mapping
+type: bug
+issues: []

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

@@ -181,8 +181,12 @@ static NameValue decode(Object field) {
         int encodedSize = ByteUtils.readIntLE(bytes, 0);
         int nameSize = encodedSize % PARENT_OFFSET_IN_NAME_OFFSET;
         int parentOffset = encodedSize / PARENT_OFFSET_IN_NAME_OFFSET;
-        String name = new String(bytes, 4, nameSize, StandardCharsets.UTF_8);
-        BytesRef value = new BytesRef(bytes, 4 + nameSize, bytes.length - nameSize - 4);
+
+        String decoded = new String(bytes, 4, bytes.length - 4, StandardCharsets.UTF_8);
+        String name = decoded.substring(0, nameSize);
+        int nameByteCount = name.getBytes(StandardCharsets.UTF_8).length;
+
+        BytesRef value = new BytesRef(bytes, 4 + nameByteCount, bytes.length - nameByteCount - 4);
         return new NameValue(name, parentOffset, value, null);
     }
 

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

@@ -10,12 +10,14 @@
 package org.elasticsearch.index.mapper;
 
 import org.apache.lucene.index.DirectoryReader;
+import org.elasticsearch.common.Strings;
 import org.elasticsearch.common.settings.Settings;
 import org.elasticsearch.core.CheckedConsumer;
 import org.elasticsearch.core.Nullable;
 import org.elasticsearch.search.lookup.SourceFilter;
 import org.elasticsearch.test.FieldMaskingReader;
 import org.elasticsearch.xcontent.XContentBuilder;
+import org.elasticsearch.xcontent.json.JsonXContent;
 import org.hamcrest.Matchers;
 
 import java.io.IOException;
@@ -122,6 +124,15 @@ public void testIgnoredString() throws IOException {
         );
     }
 
+    public void testIgnoredStringFullUnicode() throws IOException {
+        String value = randomUnicodeOfCodepointLengthBetween(5, 20);
+        String fieldName = randomUnicodeOfCodepointLength(5);
+
+        String expected = Strings.toString(JsonXContent.contentBuilder().startObject().field(fieldName, value).endObject());
+
+        assertEquals(expected, getSyntheticSourceWithFieldLimit(b -> b.field(fieldName, value)));
+    }
+
     public void testIgnoredInt() throws IOException {
         int value = randomInt();
         assertEquals("{\"my_value\":" + value + "}", getSyntheticSourceWithFieldLimit(b -> b.field("my_value", value)));