JDO Objects

JDO requires that implementations support the persistence of java.lang.Object as first class objects (FCO's). DataNucleus provides this capability and also provides that java.lang.Object can be stored as serialised. It follows the same general process as for Interfaces since both interfaces and java.lang.Object are basically references to some persistable object.

JDO doesn't define how an object FCO is persisted in the datastore. Obviously there can be many "implementations" and so no obvious solution. DataNucleus allows the following

  • per-implementation : a FK is created for each "implementation" so that the datastore can provide referential integrity. The other advantage is that since there are FKs then querying can be performed. The disadvantage is that if there are many implementations then the table can become large with many columns not used
  • identity : a single column is added and this stores the class name of the "implementation" stored, as well as the identity of the object. The disadvantages are that no querying can be performed, and that there is no referential integrity.
  • xcalia : a slight variation on "identity" whereby there is a single column yet the contents of that column are consistent with what Xcalia XIC JDO implementation stored there.

The user controls which one of these is to be used by specifying the extension mapping-strategy on the field containing the interface. The default is "per-implementation"



FCO

Let's suppose you have a field in a class and you have a selection of possible persistable class that could be stored there, so you decide to make the field a java.lang.Object . So let's take an example. We have the following class

public class ParkingSpace
{
    String location;
    Object occupier;
}

So we have a space in a car park, and in that space we have an occupier of the space. We have some legacy data and so can't make the type of this "occupier" an interface type, so we just use java.lang.Object . Now we know that we can only have particular types of objects stored there (since there are only a few types of vehicle that can enter the car park). So we define our MetaData like this

<package name="mydomain.samples.object">
    <class name="ParkingSpace">
        <field name="location"/>
        <field name="occupier" persistence-modifier="persistent"
               field-type="mydomain.samples.vehicles.Car,
                       mydomain.samples.vehicles.Motorbike"/>
        </field>
</class>

or using annotations

@Persistent(types={mydomain.samples.vehicles.Car.class, mydomain.samples.vehicles.Motorbike.class})
Object occupier;

This will result in the following database schema.





So DataNucleus adds foreign keys from the ParkingSpace table to all of the possible implementation tables for the occupier field.

In conclusion, when using "per-implementation" mapping for any java.lang.Object field in a class to be persisted (as non-serialised), you must define the possible "implementation" classes that can be stored there.



If we use mapping-strategy of "identity" then we get a different datastore schema.

    <class name="ParkingSpace">
        <field name="location"/>
        <field name="occupier" persistence-modifier="persistent">
            <extension vendor-name="datanucleus" key="mapping-strategy" value="identity"/>
        </field>
    </class>

and the datastore schema becomes

and the column "OCCUPIER" will contain strings such as com.mydomain.samples.object.Car:1 allowing retrieval of the related implementation object.

Collections of Objects

You can have a Collection/Map containing elements of java.lang.Object. You specify this in the same way as you would any Collection/Map. DataNucleus supports having a Collection of references with multiple implementation types as long as you use a join table relation.

Serialised Objects

By default a field of type java.lang.Object is stored as an instance of the underlying PersistenceCapable in the table of that object. If either your Object field represents non-PersistenceCapable objects or you simply wish to serialise the Object into the same table as the owning object, you need to specify the "serialized" attribute, like this

<class name="MyClass">
    <field name="myObject" serialized="true"/>
</class>

Similarly, where you have a collection of Objects using a join table, the objects are, by default, stored in the table of the PersistenceCapable instance. If instead you want them to occupy a single BLOB column of the join table, you should specify the "embedded-element" attribute of <collection> like this

<class name="MyClass">
    <field name="myCollection">
        <collection element-type="java.lang.Object" serialized-element="true"/>
        <join/>
    </field>
</class> 

Please refer to the serialised fields guide for more details of storing objects in this way.