Package org.apache.jackrabbit.oak.plugins.document.check

Class AsyncNodeStateProcessor

java.lang.Object

org.apache.jackrabbit.oak.plugins.document.check.AsyncDocumentProcessor

org.apache.jackrabbit.oak.plugins.document.check.AsyncNodeStateProcessor

All Implemented Interfaces:

DocumentProcessor

Direct Known Subclasses:

NodeCounter, ReferenceCheck

public abstract class AsyncNodeStateProcessor
extends AsyncDocumentProcessor

A DocumentProcessor that processes NodeStates.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
protected class	AsyncNodeStateProcessor.NodeStateTask

Field Summary

Fields

Modifier and Type	Field	Description
protected RevisionVector	headRevision
protected DocumentNodeStore	ns
protected NodeState	uuidIndex

Constructor Summary

Constructors

Modifier	Constructor	Description
protected	AsyncNodeStateProcessor(DocumentNodeStore ns, RevisionVector headRevision, ExecutorService executorService)

Method Summary

Modifier and Type	Method	Description
protected Optional<Callable<Void>>	createTask(@NotNull NodeDocument document, @NotNull BlockingQueue<Result> results)
protected @Nullable NodeState	getNodeByUUID(@NotNull String uuid, @NotNull AtomicReference<String> resolvedPath)	Utility method that resolves he uuid into a NodeState with the help of the UUID index.
protected boolean	process(NodeDocument doc)	Decide early whether a NodeDocument should be processed or not.
protected abstract void	runTask(@NotNull Path path, @Nullable NodeState state, @NotNull Consumer<Result> resultConsumer)	Responsibility of the subclass to implement the processor logic.

Methods inherited from class org.apache.jackrabbit.oak.plugins.document.check.AsyncDocumentProcessor

processDocument

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.apache.jackrabbit.oak.plugins.document.check.DocumentProcessor

end

Field Detail

ns

protected final DocumentNodeStore ns

headRevision

protected final RevisionVector headRevision

uuidIndex

protected final NodeState uuidIndex

Constructor Detail

AsyncNodeStateProcessor

protected AsyncNodeStateProcessor(DocumentNodeStore ns,
                                  RevisionVector headRevision,
                                  ExecutorService executorService)

Method Detail

process

protected boolean process(NodeDocument doc)

Decide early whether a NodeDocument should be processed or not. This implementation returns true if the document not a split document, otherwise false. This method can be overridden by subclasses.

Parameters:

doc - the document to process.

Returns:

whether the document should be processed or ignored.

getNodeByUUID

@Nullable
protected final @Nullable NodeState getNodeByUUID(@NotNull
                                                  @NotNull String uuid,
                                                  @NotNull
                                                  @NotNull AtomicReference<String> resolvedPath)

Utility method that resolves he uuid into a NodeState with the help of the UUID index.

Parameters:

uuid - the UUID to resolve.

resolvedPath - will be set to the resolved path if available.

Returns:

the NodeState with the given UUID or null if it cannot be resolved or doesn't exist.

createTask

protected final Optional<Callable<Void>> createTask(@NotNull
                                                    @NotNull NodeDocument document,
                                                    @NotNull
                                                    @NotNull BlockingQueue<Result> results)

Specified by:

createTask in class AsyncDocumentProcessor

runTask

protected abstract void runTask(@NotNull
                                @NotNull Path path,
                                @Nullable
                                @Nullable NodeState state,
                                @NotNull
                                @NotNull Consumer<Result> resultConsumer)

Responsibility of the subclass to implement the processor logic. This method will run as a task with an executor service.

Parameters:

path - the path of the NodeState to process.

state - the NodeState or null if the node does not exist at this path. This may happen for nodes that have been deleted but not yet garbage collected.

resultConsumer - consumes the results of this task.