org.hibernate.loader.plan.spi

Interface LoadPlan

public interface LoadPlan

Describes a plan for performing a load of results.

Generally speaking there are 3 forms of load plans:<ul> <li> LoadPlan.Disposition.ENTITY_LOADER - An entity load plan for handling get/load handling. This form will typically have a single return (of type EntityReturn) defined by getReturns(), possibly defining fetches. </li> <li> LoadPlan.Disposition.COLLECTION_INITIALIZER - A collection initializer, used to load the contents of a collection. This form will typically have a single return (of type CollectionReturn) defined by getReturns(), possibly defining fetches </li> <li> LoadPlan.Disposition.MIXED - A query load plan which can contain multiple returns of mixed type (though all implementing Return). Again, may possibly define fetches. </li> </ul> <p/> todo : would also like to see "call back" style access for handling "subsequent actions" such as…<ul> <li>follow-on locking</li> <li>join fetch conversions to subselect fetches</li> </ul>

Nested Class Summary

Nested Classes

Modifier and Type	Interface and Description
static class	LoadPlan.Disposition Enumerated possibilities for describing the disposition of this LoadPlan.

Method Summary

Modifier and Type	Method and Description
boolean	areLazyAttributesForceFetched() Does this load plan indicate that lazy attributes are to be force fetched? <p/> Here we are talking about laziness in regards to the legacy bytecode enhancement which adds support for partial selects of an entity’s state (e.g., skip loading a lob initially, wait until/if it is needed) <p/> This one would effect the SQL that needs to get generated as well as how the result set would be read.
LoadPlan.Disposition	getDisposition() What is the disposition of this LoadPlan, in terms of its returns.
QuerySpaces	getQuerySpaces() Gets the QuerySpaces for the load plan, which contains a QuerySpace reference for each non-scalar return and for each entity, collection, and composite FetchSource.
List<? extends Return>	getReturns() Get the returns indicated by this LoadPlan.<ul> <li> A LoadPlan.Disposition.ENTITY_LOADER LoadPlan would have just a single Return of type EntityReturn.
boolean	hasAnyScalarReturns() Convenient form of checking getReturns() for scalar root returns.

Method Detail

getDisposition

LoadPlan.Disposition getDisposition()

What is the disposition of this LoadPlan, in terms of its returns.

Returns:

The LoadPlan’s disposition

getReturns

List<? extends Return> getReturns()

Get the returns indicated by this LoadPlan.<ul> <li> A LoadPlan.Disposition.ENTITY_LOADER LoadPlan would have just a single Return of type EntityReturn. </li> <li> A LoadPlan.Disposition.COLLECTION_INITIALIZER LoadPlan would have just a single Return of type CollectionReturn. </li> <li> A LoadPlan.Disposition.MIXED LoadPlan would contain a mix of EntityReturn and ScalarReturn elements, but no CollectionReturn. </li> </ul> <p/> When generating SQL, the Returns provide columns/formulas used in the "select clause".

Returns:

The Returns for this LoadPlan.

See Also:

LoadPlan.Disposition

getQuerySpaces

QuerySpaces getQuerySpaces()

Gets the QuerySpaces for the load plan, which contains a QuerySpace reference for each non-scalar return and for each entity, collection, and composite FetchSource. <p/> When generating SQL, the query spaces provide data for the "from clause" including joins.

Returns:

The QuerySpaces

areLazyAttributesForceFetched

boolean areLazyAttributesForceFetched()

Does this load plan indicate that lazy attributes are to be force fetched? <p/> Here we are talking about laziness in regards to the legacy bytecode enhancement which adds support for partial selects of an entity’s state (e.g., skip loading a lob initially, wait until/if it is needed) <p/> This one would effect the SQL that needs to get generated as well as how the result set would be read. Therefore we make this part of the LoadPlan contract. <p/> NOTE that currently this is only relevant for HQL loaders when the HQL has specified the FETCH ALL PROPERTIES key-phrase. In all other cases, this returns false.

Returns:

Whether or not to

hasAnyScalarReturns

boolean hasAnyScalarReturns()

Convenient form of checking getReturns() for scalar root returns.

Returns:

true if getReturns() contained any scalar returns; false otherwise.