com.bea.ld.dsmediator.update
Class DataServiceMediator

java.lang.Object
  com.bea.ld.dsmediator.update.Mediator
      com.bea.ld.dsmediator.update.DataServiceMediator

public class DataServiceMediator

extends com.bea.ld.dsmediator.update.Mediator

Constructor Summary

DataServiceMediator()
should not be called directly

Method Summary

void	executeUpdatePlan(UpdatePlan updatePlan) For each unique data source instance within update plan, submit will invoked for inserted, updated or deleted.
void	flushCachedMetaData() flush the cached decomposition map and data source meta data internal use only, invoked in development mode
com.bea.ld.dsmediator.update.DataServiceMetaData	getDataServiceMetaData(java.lang.String dsName) lookup DataObject meta data by ds name
UpdatePlan	getUpdatePlan(commonj.sdo.DataGraph sdo) For a given view SDO object, decompose it and return a update plan Update plan is a sequence of nodes decomposed from original View SDO object.
java.util.Properties[]	submit(commonj.sdo.DataGraph dataGraph, java.lang.String dsName) First, the change summary for the data graph will be analyzed using the metadata for the associated SDO and an update plan will be produced.
java.util.Properties[]	submit(java.lang.String[] dgStrings, java.lang.String dsName) Deserialize the serialized SDO into a DataGraph, then call the method submit.
java.util.Properties[]	submit(java.lang.String dgString, java.lang.String dsName) Deprecated.

Methods inherited from class com.bea.ld.dsmediator.update.Mediator

getInstance, setMediatorClassName

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

DataServiceMediator

public DataServiceMediator()

should not be called directly

Method Detail

submit

public java.util.Properties[] submit(java.lang.String dgString,
                                     java.lang.String dsName)

Deprecated.

Submit a serialized SDO for update/insert/delete to the given DataService (logical or physical).

Parameters:

dgString - serialized SDO

dsName - data service this submit is targeted for

Returns:

Primary keys. If adds of top level data objects have occurred that have primary keys, then after a successful submit, the values of the primary keys for the inserted tuples will be returned as an array of java Properties, (xpath name/value in string) pairs. This only applies to the primary keys of the top level data object. Nested data objects that have their primary keys computed do not return their values. Returning the top level pkeys of inserted tuples, allows the developer to refetch tuples based on their new primary keys if desired. If one had an array of Customer with primary key CustID in your SDO and inserted two customers, then the submit would return an array of two Properties with the name being CustID, relative to the Customer type and the value in string format being the new primary key value for each inserted Customer.

submit

public java.util.Properties[] submit(java.lang.String[] dgStrings,
                                     java.lang.String dsName)

Deserialize the serialized SDO into a DataGraph, then call the method submit.

Parameters:

dgStrings - array of serialized SDOs to update

dsName - DataService to execute the SDO update against

Returns:

Primary keys. If adds of top level data objects have occurred that have primary keys, then after a successful submit, the values of the primary keys for the inserted tuples will be returned as an array of java Properties, (xpath name/value in string) pairs. This only applies to the primary keys of the top level data object. Nested data objects that have their primary keys computed do not return their values. Returning the top level pkeys of inserted tuples, allows the developer to refetch tuples based on their new primary keys if desired. If one had an array of Customer with primary key CustID in your SDO and inserted two customers, then the submit would return an array of two Properties with the name being CustID, relative to the Customer type and the value in string format being the new primary key value for each inserted Customer.

submit

public java.util.Properties[] submit(commonj.sdo.DataGraph dataGraph,
                                     java.lang.String dsName)

First, the change summary for the data graph will be analyzed using the metadata for the associated SDO and an update plan will be produced. The update plan will be used to apply the changes in the SDO for the data sources used in the view - view update decomposition. Second, any custom update override methods #performUpdate will be invoked for each altered dataServiceMetaData object in the update plan if the update override exists. If there is no update override or the update override returns "true" to continue, submit will be called on each dataServiceMetaData instances.

Parameters:

dataGraph - data graph being submitted for update

dsName - DataService to execute the SDO update against

Returns:

Primary keys. If adds of top level data objects have occurred that have primary keys, then after a successful submit, the values of the primary keys for the inserted tuples will be returned as an array of java Properties, (xpath name/value in string) pairs. This only applies to the primary keys of the top level data object. Nested data objects that have their primary keys computed do not return their values. Returning the top level pkeys of inserted tuples, allows the developer to refetch tuples based on their new primary keys if desired. If one had an array of Customer with primary key CustID in your SDO and inserted two customers, then the submit would return an array of two Properties with the name being CustID, relative to the Customer type and the value in string format being the new primary key value for each inserted Customer.

getUpdatePlan

public UpdatePlan getUpdatePlan(commonj.sdo.DataGraph sdo)

For a given view SDO object, decompose it and return a update plan Update plan is a sequence of nodes decomposed from original View SDO object. Each node in sequence can either be a SDO instance only or a SDO with a sub-update plan, which is another sequence of nodes. in View SDO, if a parent DataObject get updated, at least partially due to the changes on the DataObjects nested inside it, then SDO instances corresponding to those children DataObjects will be addChild into sub-update plan. Update plan contains inserted, updated, deleted and unchanged SDO instances as well. Update plan is constructed from a LinkedHashMap tree, independent SDO instances are addChild in the same level and update sequence on their source dataobjects is preserved

Parameters:

sdo -

Returns:

executeUpdatePlan

public void executeUpdatePlan(UpdatePlan updatePlan)

For each unique data source instance within update plan, submit will invoked for inserted, updated or deleted.

Parameters:

updatePlan -

getDataServiceMetaData

public com.bea.ld.dsmediator.update.DataServiceMetaData getDataServiceMetaData(java.lang.String dsName)

lookup DataObject meta data by ds name

Parameters:

dsName - ds file name

Returns:

internal data structure used by mediator that describes the dataobject

flushCachedMetaData

public void flushCachedMetaData()

flush the cached decomposition map and data source meta data internal use only, invoked in development mode