public interface DataMovementManagerextends Closeable
DataMovementManager is the starting point for getting new
instances of QueryBatcher, WriteBatcher and RowBatcher, configured
with a DatabaseClient and ForestConfiguration. On instantiation, it
will immediately call readForestConfig to obtain the
ForestConfiguration from which it can create host-specific
DatabaseClient instances for each applicable host. Applicable hosts
are those with forests for the database specified in the
DatabaseClient. If not specified, the database is the default
database for the port in the DatabaseClient. Call dataMovementMangaer.release() when
you're done with your DataMovementManager instance to free
resources associated with those host-specific DatabaseClient
instances.
Identify whether the DataMovementManager
connects directly to each MarkLogic host with a forest for the
database or whether the DataMovementManager uses a gateway such as
a load balancer to communicate with the MarkLogic hosts.
Calls release() on all host-specific
DatabaseClient instances (but not on the DatabaseClient instance
used to create this DataMovementManager instance).
Calls release() on all host-specific
DatabaseClient instances (but not on the DatabaseClient instance
used to create this DataMovementManager instance).
Begins job tracking on the WriteBatcher. Calling
startJob is not required on a WriteBatcher if you don't intend to
ever call stopJob or look at the JobReport.
Parameters:
batcher - the WriteBatcher instance which has to
be started
Get a snapshot report of the state of the job
when the call is made. Depends on some pre-installed listeners
attached to the job to collect the details.
Parameters:
ticket - the JobTicket for which the report has to
be generated
Immediately cancel all queued tasks, prevent new
tasks from being added to the queue, and begin to free all
resources. In-process tasks cannot be cancelled because they talk
to the REST server which does not have such a mechanism. This does
not block until all in-process tasks are finished, for that use
QueryBatcher.awaitCompletion() or
WriteBatcher.awaitCompletion().
Immediately cancel all queued tasks, prevent new
tasks from being added to the queue, and begin to free all
resources. In-process tasks cannot be cancelled because they talk
to the REST server which does not have such a mechanism. This does
not block until all in-process tasks are finished, for that use
QueryBatcher.awaitCompletion() or
WriteBatcher.awaitCompletion().
Create a new QueryBatcher instance configured to retrieve uris
from this Iterator. This form enables the uris (actually any
String) to come from any source, whereas the other form requires
the uris to come from the results of a query. This form is helpful
when deleting documents when one cannot set the server's merge
timestamp. For more discussion, see QueryBatcher.
The Iterator needn't be thread-safe as it is only iterated from
one thread.
Create a new RowBatcher instance to export all
of the rows from a view in batches.
You pass a sample handle (that is, an adapter for the Java class
that should store a batch of retrieved rows). The handle must
implement the ContentHandle
and StructuredReadHandle
interfaces.
Type Parameters:
T - the Java class that stores a batch of
retrieved roles
Parameters:
rowsHandle - a sample handle for storing a batch
of rows
Identify whether the DataMovementManager
connects directly to each MarkLogic host with a forest for the
database or whether the DataMovementManager uses a gateway such as
a load balancer to communicate with the MarkLogic hosts.
Returns:
the connection type
Copyright (c) 2010-2025 Progress
Software Corporation and/or its subsidiaries or affiliates. All
Rights Reserved.