风暴维基上有关于三叉戟状态的好文档。你的问题的简单答案是 urlToTweeters 是可以从中查询的状态对象。我假设上面的陈述来自三叉戟教程，复制如下：

TridentState urlToTweeters = topology.newStaticState(getUrlToTweetersState());
TridentState tweetersToFollowers = topology.newStaticState(getTweeterToFollowersState());
topology.newDRPCStream("reach")
  .stateQuery(urlToTweeters, new Fields("args"), new MapGet(), new Fields("tweeters")).each(new Fields("tweeters"), new ExpandList(), new Fields("tweeter"))
  /* At this point we have the tweeters for each url passed in args */
  .shuffle()        
  .stateQuery(tweetersToFollowers, new Fields("tweeter"), new MapGet(), new Fields("followers"))
  .parallelismHint(200)
  .each(new Fields("followers"), new ExpandList(), new Fields("follower"))
  .groupBy(new Fields("follower"))
  .aggregate(new One(), new Fields("one"))
  .parallelismHint(20)
  .aggregate(new Count(), new Fields("reach"));

在这个例子中， urlToTweeters 将存储到tweeter和drpc的urlMap reach 在下一行定义的查询（将url作为其参数）最终将产生reach。但是在途中（用注解内联标记），你会看到每个url的tweeter流，也就是说，查询的结果 urlToTweeters .

我希望回答永远不会太迟，至少别人会觉得我的答案有用：）
所以， topology.newStaticState() 是trident对可查询数据存储的抽象。的参数 newStaticState() 应该是基于方法契约的 storm.trident.state.StateFactory . 反过来，工厂应该实施 makeState() 方法返回的示例 storm.trident.state.State . 但是，如果您计划查询您的状态，则应返回istance storm.trident.state.map.ReadOnlyMapState 相反，因为 storm.trident.state.State 没有用于查询实际数据源的方法（如果您尝试使用除 ReadOnlyMapState ).
所以，让我们试试吧！
虚拟状态实现：

public static class ExampleStaticState implements ReadOnlyMapState<String> {

    private final Map<String, String> dataSourceStub;

    public ExampleStaticState() {
        dataSourceStub = new HashMap<>();
        dataSourceStub.put("tuple-00", "Trident");
        dataSourceStub.put("tuple-01", "definitely");
        dataSourceStub.put("tuple-02", "lacks");
        dataSourceStub.put("tuple-03", "documentation");
    }

    @Override
    public List<String> multiGet(List<List<Object>> keys) {

        System.out.println("DEBUG: MultiGet, keys is " + keys);

        List<String> result = new ArrayList<>();

        for (List<Object> inputTuple : keys) {
            result.add(dataSourceStub.get(inputTuple.get(0)));
        }

        return result;
    }

    @Override
    public void beginCommit(Long txid) {
        // never gets executed...
        System.out.println("DEBUG: Begin commit, txid=" + txid);
    }

    @Override
    public void commit(Long txid) {
        // never gets executed...
        System.out.println("DEBUG: Commit, txid=" + txid);
    }
}

工厂：

public static class ExampleStaticStateFactory implements StateFactory {
    @Override
    public State makeState(Map conf, IMetricsContext metrics, int partitionIndex, int numPartitions) {
        return new ExampleStaticState();
    }
}

一个简单的 psvm （又名 public static void main ):

public static void main(String... args) {
    TridentTopology tridentTopology = new TridentTopology();
    FeederBatchSpout spout = new FeederBatchSpout(Arrays.asList(new String[]{
            "foo"
    }));
    TridentState state = tridentTopology.newStaticState(new ExampleStaticStateFactory());
    tridentTopology
            .newStream("spout", spout)
            .stateQuery(state, new Fields("foo"), new MapGet(), new Fields("bar"))
            .each(new Fields("foo", "bar"), new Debug())
            ;

    Config conf = new Config();
    conf.setNumWorkers(6);

    LocalCluster localCluster = new LocalCluster();
    localCluster.submitTopology("tridentTopology", conf, tridentTopology.build());

    spout.feed(Arrays.asList(new Values[]{
            new Values("tuple-00"),
            new Values("tuple-01"),
            new Values("tuple-02"),
            new Values("tuple-03")
    }));

    localCluster.shutdown();
}

最后，输出：

DEBUG: MultiGet, keys is [[tuple-00], [tuple-01], [tuple-02], [tuple-03]]
DEBUG: [tuple-00, Trident]
DEBUG: [tuple-01, definitely]
DEBUG: [tuple-02, lacks]
DEBUG: [tuple-03, documentation]

您可以看到，statequery（）从输入批处理中获取值，并将它们Map到“数据存储”中找到的值。
潜水再深一点，你就可以看看 MapGet 类（其示例用于在拓扑内部查询的对象）并在其中查找以下内容：

public class MapGet extends BaseQueryFunction<ReadOnlyMapState, Object> {
    @Override
    public List<Object> batchRetrieve(ReadOnlyMapState map, List<TridentTuple> keys) {
        return map.multiGet((List) keys);
    }    

    @Override
    public void execute(TridentTuple tuple, Object result, TridentCollector collector) {
        collector.emit(new Values(result));
    }    
}

所以在引擎盖下它只会叫 multiGet() 你的方法 ReadOnlyMapState 实现，然后发出在数据存储中找到的值，将它们添加到已经存在的元组中。您可以（尽管这可能不是最好的做法）创建自己的实现 BaseQueryFunction<ReadOnlyMapState, Object> 做更复杂的事。