OpenJPA extends standard metadata to allow you to access advanced OpenJPA functionality. This section covers persistence metadata extensions; we discuss mapping metadata extensions in Section 9, “ Mapping Extensions ”. All metadata extensions are optional; OpenJPA will rely on its defaults when no explicit data is provided.
OpenJPA recognizes the following class extensions:
The
org.apache.openjpa.persistence.FetchGroups
and
org.apache.openjpa.persistence.FetchGroup
annotations allow you to define fetch groups in your JPA entities.
Section 7, “
Fetch Groups
” discusses OpenJPA's support for fetch
groups in general; see Section 7.1, “
Custom Fetch Groups
” for how to
use these annotations in particular.
Section 1, “ Data Cache ” examines caching in OpenJPA. Metadata extensions allow individual classes to override system caching defaults.
OpenJPA defines the
org.apache.openjpa.persistence.DataCache
annotation for caching information. This annotation has the following
properties:
boolean enabled
: Whether to cache data for instances of the
class. Defaults to true
for base classes, or the superclass
value for subclasses. If you set this property to false
, all
other properties are ignored.
int timeout
: The number of milliseconds data for the class
remains valid. Use -1 for no timeout. Defaults to the
openjpa.DataCacheTimeout
property value.
The OpenJPA enhancer may add a synthetic field to detachable classes to hold detached state (see Section 1.3, “ Defining the Detached Object Graph ” for details). You can instead declare your own detached state field or suppress the creation of a detached state field altogether. In the latter case, your class must not use datastore identity, and should declare a version field to detect optimistic concurrency errors during detached modifications.
OpenJPA defines the
org.apache.openjpa.persistence.DetachedState
annotation for controlling detached state. When used to annotate a class,
DetachedState
recognizes the following properties:
boolean enabled
: Set to false to suppress the use of
detached state.
String fieldName
: Use this property to declare your own
detached state field. The field must be of type Object
.
Typically this property is only used if the field is inherited from a
non-persisted superclass. If the field is declared in your entity class, you
will typically annotate the field directly, as described below.
If you declare your own detached state field, you can annotate that field with
DetachedState
directly, rather than placing the
annotation at the class level and using the fieldName
property. When placed on a field, DetachedState
acts as a
marker annotation; it does not recognize any properties. Your annotated field
must be of type Object
.
OpenJPA recognizes the following field extensions:
In a dependent relation, the referenced object is deleted
whenever the owning object is deleted, or whenever the relation is severed by
nulling or resetting the owning field. For example, if the
Magazine.coverArticle
field is marked dependent, then setting
Magazine.coverArticle
to a new Article
instance will automatically delete the old Article
stored
in the field. Similarly, deleting a Magazine
object will
automatically delete its current cover Article
. (This
latter processing is analogous to using JPA's CascadeType.REMOVE functionality
as described in Section 2.9.1, “
Cascade Type
”.) You can
prevent an orphaned dependent object from being automatically deleted by
assigning it to another relation in the same transaction.
OpenJPA offers a family of marker annotations to denote dependent relations in JPA entities:
org.apache.openjpa.persistence.Dependent
: Marks
a direct relation as dependent.
org.apache.openjpa.persistence.ElementDependent
: Marks the entity elements of a collection, array, or map field as dependent.
org.apache.openjpa.persistence.KeyDependent
:
Marks the key entities in a map field as dependent.
The
org.apache.openjpa.persistence.LoadFetchGroup
annotation specifies a field's load fetch group.
Section 7, “
Fetch Groups
” discusses OpenJPA's support for fetch groups
in general; see Section 7.1, “
Custom Fetch Groups
” for how to use this
annotation in particular.
This boolean extension, denoted by the OpenJPA
org.apache.openjpa.persistence.LRS
annotation,
indicates that a field should use OpenJPA's special large result set collection
or map proxies. A complete description of large result set proxies is available
in Section 6.4.2, “
Large Result Set Proxies
”.
This extension names the inverse field in a logical bidirectional relation.
To create a logical bidirectional relation in OpenJPA, use the
org.apache.openjpa.persistence.InverseLogical
annotation. We discuss logical bidirectional relations and this
extension in detail in Section 5, “
Managed Inverses
”.
The read-only extension makes a field unwritable. The extension only applies to existing persistent objects; new object fields are always writeable.
To mark a field read-only in JPA metadata, set the
org.apache.openjpa.persistence.ReadOnly
annotation to an
org.apache.openjpa.persistence.UpdateAction
enum
value. The UpdateAction
enum includes:
UpdateAction.IGNORE
: Updates to the field are completely
ignored. The field is not considered dirty. The new value will not even get
stored in the OpenJPA data cache.
UpdateAction.RESTRICT
: Any attempt to change the field will
result in an immediate exception.
OpenJPA has three levels of support for relations:
Relations that hold a reference to an object of a concrete persistent class are supported by storing the primary key values of the related instance in the database.
Relations that hold a reference to an object of an unknown persistent class are supported by storing the stringified identity value of the related instance. This level of support does not allow queries across the relation.
Relations that hold an unknown object or interface. The only way to support these relations is to serialize their value to the database. This does not allow you to query the field, and is not very efficient.
Clearly, when you declare a field's type to be another persistence-capable
class, OpenJPA uses level 1 support. By default, OpenJPA assumes that any
interface-typed fields you declare will be implemented only by other persistent
classes, and assigns interfaces level 2 support. The exception to this rule is
the java.io.Serializable
interface. If you declare a
field to be of type Serializable
, OpenJPA lumps it
together with java.lang.Object
fields and other
non-interface, unrecognized field types, which are all assigned level 3 support.
With OpenJPA's type family of metadata extensions, you can control the level of
support given to your unknown/interface-typed fields. Setting the value of this
extension to Entity
indicates that the
field value will always be some persistent object, and gives level 2 support.
Setting the value of this extension to the class of a concrete persistent type
is even better; it gives you level 1 support (just as if you had declared your
field to be of that type in the first place). Setting this extension to
Object
uses level 3 support. This is useful when you have
an interface relation that may not hold other
persistent objects (recall that OpenJPA assumes interface fields will always
hold persistent instances by default).
This extension is also used with OpenJPA's externalization feature, described in Section 6.5, “ Externalization ”.
OpenJPA defines the following type annotations for field values, collection, array, and map elements, and map keys, respectively:
The OpenJPA
org.apache.openjpa.persistence.Externalizer
annotation names a method to transform a field value into a value of
another type. See Section 6.5, “
Externalization
” for details.
The OpenJPA
org.apache.openjpa.persistence.Factory
annotation
names a method to re-create a field value from its externalized form. See
Section 6.5, “
Externalization
” for details.
The OpenJPA
org.apache.openjpa.persistence.ExternalValues
annotation declares values for transformation of simple fields to
different constant values in the datastore. See
Section 6.5.1, “
External Values
” for details.
The following example shows you how to specify extensions in metadata.
Example 6.4. OpenJPA Metadata Extensions
import org.apache.openjpa.persistence.*; @Entity @DataCache(enabled=false) public class Magazine { @ManyToMany @LRS private Collection<Subscriber> subscribers; @ExternalValues({"true=1", "false=2"}) @Type(int.class) private boolean weekly; ... }
OpenJPA has extended the JPA 2.0 schema to include elements and attributes corresponding to OpenJPA extended metadata and mapping annotations. The schema are contained in 2 files: extendable-orm.xsd and openjpa-orm.xsd. The extendable-orm.xsd file provides copies of some of the JPA 2.0 schema elements with additional schema to make it extendable. The openjpa-orm.xsd file extends the extendable-orm.xsd with OpenJPA specific elements and attributes representing OpenJPA annotations. Currently, only a subset of annotations have actually been implemented, and some of those have been partially tested. The current status can be found by comments in the openjpa-orm.xsd schema file.
In order to use the OpenJPA extensions in your mapping file you must include the namespaces for these 2 new schemas as well as for the schema for JPA 2.0, as shown in the following example:
Example 6.5. OpenJPA Schema Extensions
<entity-mappings xmlns="http://openjpa.apache.org/ns/orm/extendable" xmlns:openjpa="http://openjpa.apache.org/ns/orm" xmlns:orm="http://java.sun.com/xml/ns/persistence/orm" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" version="2.0"> <entity class="org.apache.openjpa.persistence.jdbc.annotations.MultiColumnVersionPC" metadata-complete="true"> <table name="MCV"/> <attributes> <id name="id"> <orm:generated-value/> </id> <basic name="id"/> <basic name="name"/> </attributes> <openjpa:entity version-strategy="version-numbers"> <openjpa:version-columns> <openjpa:version-column name="v1"/> <openjpa:version-column name="v2"/> <openjpa:version-column name="v3" column-definition="FLOAT" scale="3" precision="10"/> </openjpa:version-columns> </openjpa:entity> </entity> </entity-mappings>