Page History

@PublicEvolving
public interface TwoPhaseCommittingSink<InputT, CommT> extends Sink<InputT> {
[..]
    PrecommittingSinkWriter<InputT, CommT> createWriter(InitContext context) throws IOException;
    
    @Deprecated
    Committer<CommT> createCommitter() throws IOException;

    /**
     * Creates a {@link Committer} that permanently makes the previously written data visible
     * through {@link Committer#commit(Collection)}.
     *
     * @param context The context information for the committer initialization.
     * @return A committer for the two-phase commit protocol.
     * @throws IOException for any failure during creation.
     */
    default Committer<CommT> createCommitter(CommitterInitContext context) throws IOException {
       return createCommitter();
    }

    /** Returns the serializer of the committable type. */
    SimpleVersionedSerializer<CommT> getCommittableSerializer();
[..]
   /** The interface exposes some runtime info for creating a {@link Committer}. */
    @PublicEvolving
    interface CommitterInitContext {
        /**
         * The first checkpoint id when an application is started and not recovered from a
         * previously taken checkpoint or savepoint.
         */
        long INITIAL_CHECKPOINT_ID = 1;

        /** @return The id of task where the committer is running. */
        int getSubtaskId();

        /** @return The number of parallel committer tasks. */
        int getNumberOfParallelSubtasks();

        /**
         * Gets the attempt number of this parallel subtask. First attempt is numbered 0.
         *
         * @return Attempt number of the subtask.
         */
        int getAttemptNumber();

        /** @return The metric group this committer belongs to. */
        SinkCommitterMetricGroup metricGroup();

        /**
         * Returns id of the restored checkpoint, if state was restored from the snapshot of a
         * previous execution.
         */
        OptionalLong getRestoredCheckpointId();

        /**
         * The ID of the current job. Note that Job ID can change in particular upon manual restart.
         * The returned ID should NOT be used for any job management tasks.
         */
        JobID getJobId();
    }
}