Check UTF16 string size before converting to string to avoid OOME

Signed-off-by: Rishab Nahata <[email protected]>

server/src/main/java/org/opensearch/common/bytes/AbstractBytesReference.java

@@ -33,6 +33,7 @@
 
 import org.apache.lucene.util.BytesRef;
 import org.apache.lucene.util.BytesRefIterator;
+import org.apache.lucene.util.UnicodeUtil;
 import org.opensearch.common.io.stream.StreamInput;
 import org.opensearch.core.xcontent.XContentBuilder;
 
@@ -49,6 +50,7 @@
 public abstract class AbstractBytesReference implements BytesReference {
 
     private Integer hash = null; // we cache the hash of this reference since it can be quite costly to re-calculated it
+    static final int MAX_LENGTH = Integer.MAX_VALUE >> 1;
 
     @Override
     public int getInt(int index) {
@@ -82,6 +84,13 @@ public void writeTo(OutputStream os) throws IOException {
 
     @Override
     public String utf8ToString() {
+        BytesRef bytesRef = toBytesRef();
+        char[] ref = new char[bytesRef.length];
+        int len = UnicodeUtil.UTF8toUTF16(bytesRef.bytes, bytesRef.offset, bytesRef.length, ref);
+        if (len > MAX_LENGTH) {
+            throw new IllegalStateException("UTF16 String size is " + len +
+                ", should be less than " + MAX_LENGTH);
+        }
         return toBytesRef().utf8ToString();
     }