Only for protobuf:

According to the wire format of the Confluent serializers message-indices are injected as additional bytes after magic bytes and schema id:

magic-byte, schema-id, message-indexes, protobuf-payload

We need to strip these message-indexes before attempting to parse the data using the plain Google protobuf library if they are present.

See: https://docs.confluent.io/platform/current/schema-registry/fundamentals/serdes-develop/index.html#wire-format