Chapter 4. Data conversion

Data Grid uses transcoders to convert data between various encodings that are identified by media types.

4.1. Hot Rod DataFormat API

Read and write operations on remote caches via the Hot Rod endpoint use the client marshaller by default. Hot Rod provides a DataFormat API for Java clients that you can use to perform cache operations with different media type encodings and/or marshallers.

Different marshallers for key and values

You can override marshallers for keys and values at run time.

For example, to bypass all serialization in the Hot Rod client and read the byte[] array stored in the remote cache:

// Existing RemoteCache instance
RemoteCache<String, Pojo> remoteCache = ...

// IdentityMarshaller is a no-op marshaller
DataFormat rawKeyAndValues =
DataFormat.builder()
          .keyMarshaller(IdentityMarshaller.INSTANCE)
          .valueMarshaller(IdentityMarshaller.INSTANCE)
          .build();

// Creates a new instance of RemoteCache with the supplied DataFormat
RemoteCache<byte[], byte[]> rawResultsCache =
remoteCache.withDataFormat(rawKeyAndValues);
Important

Using different marshallers and data formats for keys with keyMarshaller() and keyType() methods can interfere with client intelligence routing mechanisms, causing extra network hops within the Data Grid cluster. If performance is critical, you should use the same encoding for keys on the client and on the server.

Reading data in different encodings

Request and send data in different encodings specified by a org.infinispan.commons.dataconversion.MediaType as follows:

// Existing remote cache using ProtostreamMarshaller
RemoteCache<String, Pojo> protobufCache = ...

// Request values returned as JSON
// Use the UTF8StringMarshaller to convert UTF-8 to String
DataFormat jsonString =
DataFormat.builder()
          .valueType(MediaType.APPLICATION_JSON)
          .valueMarshaller(new UTF8StringMarshaller())
          .build();
RemoteCache<byte[], byte[]> rawResultsCache =
protobufCache.withDataFormat(jsonString);

Using custom value marshallers

You can use custom marshallers for values, as in the following example that returns values as org.codehaus.jackson.JsonNode objects.

In this example, Data Grid Server handles the data conversion and throws an exception if it does not support the specified media type.

DataFormat jsonNode =
DataFormat.builder()
          .valueType(MediaType.APPLICATION_JSON)
          .valueMarshaller(new CustomJacksonMarshaller()
          .build();

RemoteCache<String, JsonNode> jsonNodeCache =
remoteCache.withDataFormat(jsonNode);

Returning values as XML

The following code snippet returns values as XML:

Object xmlValue = remoteCache
      .withDataFormat(DataFormat.builder()
      .valueType(MediaType.APPLICATION_XML)
      .valueMarshaller(new UTF8StringMarshaller())
      .build())
      .get(key);

For example, the preceding get(key) call returns values such as:

<?xml version="1.0" ?><string>Hello!</string>

4.2. Converting data on demand with embedded caches

Embedded caches have a default request encoding of application/x-java-object and a storage encoding that corresponds to the media type that you configure for the cache. This means that Data Grid marshalls POJOs from the application to the storage encoding for the cache and then returns POJOs back to the application. In some complex scenarios you can use the AdvancedCache API to change the default conversion to and from POJOs to other encodings.

The following example uses the withMediaType() method to return values as application/json on demand.

Advanced cache with MediaType

DefaultCacheManager cacheManager = new DefaultCacheManager();

// Encode keys and values as Protobuf
ConfigurationBuilder cfg = new ConfigurationBuilder();
cfg.encoding().key().mediaType("application/x-protostream");
cfg.encoding().value().mediaType("application/x-protostream");

cacheManager.defineConfiguration("mycache", cfg.build());

Cache<Integer, Person> cache = cacheManager.getCache("mycache");

cache.put(1, new Person("John","Doe"));

// Use Protobuf for keys and JSON for values
Cache<Integer, byte[]> jsonValuesCache = (Cache<Integer, byte[]>) cache.getAdvancedCache().withMediaType("application/x-protostream", "application/json");

byte[] json = jsonValuesCache.get(1);

Value returned in JSON format

{
   "_type":"org.infinispan.sample.Person",
   "name":"John",
   "surname":"Doe"
}

Additional resources