Storm-源码分析- bolt (backtype.storm.task)

Bolt关键的接口为execute, 
Tuple的真正处理逻辑, 通过OutputCollector.emit发出新的tuples, 调用ack或fail处理的tuple

/**
 * An IBolt represents a component that takes tuples as input and produces tuples
 * as output. An IBolt can do everything from filtering to joining to functions
 * to aggregations. It does not have to process a tuple immediately and may
 * hold onto tuples to process later.
 *
 * <p>A bolt's lifecycle is as follows:</p>
 *
 * <p>IBolt object created on client machine. The IBolt is serialized into the topology
 * (using Java serialization) and submitted to the master machine of the cluster (Nimbus).
 * Nimbus then launches workers which deserialize the object, call prepare on it, and then
 * start processing tuples.</p>
 *
 * <p>If you want to parameterize an IBolt, you should set the parameter's through its
 * constructor and save the parameterization state as instance variables (which will
 * then get serialized and shipped to every task executing this bolt across the cluster).</p>
 *
 * <p>When defining bolts in Java, you should use the IRichBolt interface which adds
 * necessary methods for using the Java TopologyBuilder API.</p>
 */
public interface IBolt extends Serializable {
    /**
     * Called when a task for this component is initialized within a worker on the cluster.
     * It provides the bolt with the environment in which the bolt executes.
     *
     * <p>This includes the:</p>
     *
     * @param stormConf The Storm configuration for this bolt. This is the configuration provided to the topology merged in with cluster configuration on this machine.
     * @param context This object can be used to get information about this task's place within the topology, including the task id and component id of this task, input and output information, etc.
     * @param collector The collector is used to emit tuples from this bolt. Tuples can be emitted at any time, including the prepare and cleanup methods. The collector is thread-safe and should be saved as an instance variable of this bolt object.
     */
    void prepare(Map stormConf, TopologyContext context, OutputCollector collector);

    /**
     * Process a single tuple of input. The Tuple object contains metadata on it
     * about which component/stream/task it came from. The values of the Tuple can
     * be accessed using Tuple#getValue. The IBolt does not have to process the Tuple
     * immediately. It is perfectly fine to hang onto a tuple and process it later
     * (for instance, to do an aggregation or join).
     *
     * <p>Tuples should be emitted using the OutputCollector provided through the prepare method.
     * It is required that all input tuples are acked or failed at some point using the OutputCollector.
     * Otherwise, Storm will be unable to determine when tuples coming off the spouts
     * have been completed.</p>
     *
     * <p>For the common case of acking an input tuple at the end of the execute method,
     * see IBasicBolt which automates this.</p>
     *
     * @param input The input tuple to be processed.
     */
    void execute(Tuple input);

    /**
     * Called when an IBolt is going to be shutdown. There is no guarentee that cleanup
     * will be called, because the supervisor kill -9's worker processes on the cluster.
     *
     * <p>The one context where cleanup is guaranteed to be called is when a topology
     * is killed when running Storm in local mode.</p>
     */
    void cleanup();
}

首先OutputCollector, 主要是emit和emitDirect接口

List<Integer> emit(String streamId, Tuple anchor, List<Object> tuple)

emit, 3个参数, 发送到的streamid, anchors(来源tuples), tuple(values list) 
        如果streamid为空, 则发送到默认stream, Utils.DEFAULT_STREAM_ID 
        如果anchors为空, 则为unanchored tuple 
        1个返回值, 最终发送到的task ids

对比一下SpoutOutputCollector中的emit, 参数变化, 没有message-id, 多了anchors

而在在Bolt中, ack和fail接口在IOutputCollector中, 用于在execute中完成对上一级某tuple的处理和emit, 调用ack或fail

而在Spout中, ack和fail接口在ISpout中, 用于spout收到ack或fail tuple时调用

/**
 * This output collector exposes the API for emitting tuples from an IRichBolt.
 * This is the core API for emitting tuples. For a simpler API, and a more restricted
 * form of stream processing, see IBasicBolt and BasicOutputCollector.
 */
public class OutputCollector implements IOutputCollector {
    private IOutputCollector _delegate;

    public OutputCollector(IOutputCollector delegate) {
        _delegate = delegate;
    }

    /**
     * Emits a new tuple to a specific stream with a single anchor. The emitted values must be
     * immutable.
     *
     * @param streamId the stream to emit to
     * @param anchor the tuple to anchor to
     * @param tuple the new output tuple from this bolt
     * @return the list of task ids that this new tuple was sent to
     */
    public List<Integer> emit(String streamId, Tuple anchor, List<Object> tuple) {
        return emit(streamId, Arrays.asList(anchor), tuple);
    }

    /**
     * Emits a tuple directly to the specified task id on the specified stream.
     * If the target bolt does not subscribe to this bolt using a direct grouping,
     * the tuple will not be sent. If the specified output stream is not declared
     * as direct, or the target bolt subscribes with a non-direct grouping,
     * an error will occur at runtime. The emitted values must be
     * immutable.
     *
     * @param taskId the taskId to send the new tuple to
     * @param streamId the stream to send the tuple on. It must be declared as a direct stream in the topology definition.
     * @param anchor the tuple to anchor to
     * @param tuple the new output tuple from this bolt
     */
    public void emitDirect(int taskId, String streamId, Tuple anchor, List<Object> tuple) {
        emitDirect(taskId, streamId, Arrays.asList(anchor), tuple);
    }

    @Override
    public List<Integer> emit(String streamId, Collection<Tuple> anchors, List<Object> tuple) {
        return _delegate.emit(streamId, anchors, tuple);
    }

    @Override
    public void emitDirect(int taskId, String streamId, Collection<Tuple> anchors, List<Object> tuple) {
        _delegate.emitDirect(taskId, streamId, anchors, tuple);
    }

    @Override
    public void ack(Tuple input) {
        _delegate.ack(input);
    }

    @Override
    public void fail(Tuple input) {
        _delegate.fail(input);
    }

    @Override
    public void reportError(Throwable error) {
        _delegate.reportError(error);
    }
}

本文章摘自博客园，原文发布日期：2013-08-05