org.apache.jackrabbit.core.journal

Interface Journal

All Known Implementing Classes:

AbstractJournal, DatabaseJournal, FileJournal, JNDIDatabaseJournal, MemoryJournal, MSSqlDatabaseJournal, OracleDatabaseJournal

public interface Journal

Generic journal interface.

Method Summary

Methods

Modifier and Type	Method and Description
void	close() Close this journal.
InstanceRevision	getInstanceRevision() Gets the instance that manages the cluster node's local revision.
RecordProducer	getProducer(String identifier) Return the record producer for a given identifier.
RecordIterator	getRecords() Return an iterator over all available records in the journal.
RecordIterator	getRecords(long startRevision) Return an iterator over all records after the specified revision.
void	init(String id, NamespaceResolver resolver) Initialize journal.
void	register(RecordConsumer consumer) Register a record consumer.
void	sync(boolean startup) Synchronize contents from journal.
boolean	unregister(RecordConsumer consumer) Unregister a record processor.

Method Detail

init

void init(String id,
        NamespaceResolver resolver)
          throws JournalException

Initialize journal.

Parameters:

id - id this journal should use to write its own records

resolver - resolver used when reading/writing records

Throws:

JournalException - if an error occurs

register

void register(RecordConsumer consumer)
              throws JournalException

Register a record consumer.

Parameters:

consumer - record consumer

Throws:

JournalException - if an error occurs

unregister

boolean unregister(RecordConsumer consumer)

Unregister a record processor.

Parameters:

consumer - record processor to unregister

Returns:

true if the consumer was previously registered; false otherwise

sync

void sync(boolean startup)
          throws JournalException

Synchronize contents from journal. This will compare the journal's revision with the revisions of all registered consumers and invoke their RecordConsumer.consume(org.apache.jackrabbit.core.journal.Record) method when their identifier matches the one found in the records. The startup flag allow for a separate treatment of the initial sync when the cluster nodes starts up. This might be needed for example when there are a lot of old revisions in a database.

Parameters:

startup - indicates if the cluster node is syncing on startup or does a normal sync.

Throws:

JournalException - if an error occurs

getProducer

RecordProducer getProducer(String identifier)
                           throws JournalException

Return the record producer for a given identifier.

Parameters:

identifier - identifier

Returns:

the record producer for a given identifier.

Throws:

JournalException - if an error occurs

close

void close()

Close this journal. This should release any resources still held by this journal.

getInstanceRevision

InstanceRevision getInstanceRevision()
                                     throws JournalException

Gets the instance that manages the cluster node's local revision.

Returns:

the InstanceRevision manager

Throws:

JournalException - on error

getRecords

RecordIterator getRecords(long startRevision)
                          throws JournalException

Return an iterator over all records after the specified revision.

Parameters:

startRevision - start point (exlusive)

Returns:

an iterator over all records after the specified revision.

Throws:

JournalException - if an error occurs

getRecords

RecordIterator getRecords()
                          throws JournalException

Return an iterator over all available records in the journal.

Returns:

an iterator over all records.

Throws:

JournalException - if an error occurs