提交示例代码:
1 | public static void main(String[] args) throws Exception { |
2 | TopologyBuilder builder = new TopologyBuilder(); |
3 | builder.setSpout("random", new RandomWordSpout(), 2); |
4 | builder.setBolt("transfer", new TransferBolt(), 4).shuffleGrouping("random"); |
5 | builder.setBolt("writer", new WriterBolt(), 4).fieldsGrouping("transfer", new Fields("word")); |
6 | Config conf = new Config(); |
7 | conf.setNumWorkers(4);// 设置启动4个Worker |
8 | conf.setNumAckers(1); // 设置一个ack线程 |
9 | conf.setDebug(true); // 设置打印所有发送的消息及系统消息 |
10 | StormSubmitter.submitTopology("test", conf, builder.createTopology()); |
11 | } |
1、构建 TopologyBuilder 对象 builder,主要用于对各个组件(bolt、spout)进行配置,TopologyBuilder主要属性字段定义如下:
1 | public class TopologyBuilder { |
2 | |
3 | // 所提交Topolog中所有的bolt将放入到_bolts中 |
4 | private Map<String, IRichBolt> _bolts = new HashMap<String, IRichBolt>(); |
5 | |
6 | // 所提交Topolog中所有的spout将放入到_spouts中 |
7 | private Map<String, IRichSpout> _spouts = new HashMap<String, IRichSpout>(); |
8 | |
9 | // 所提交Topolog中所有的spout和bolt都将放入_commons中 |
10 | private Map<String, ComponentCommon> _commons = new HashMap<String, ComponentCommon>(); |
11 | |
12 | .................................... |
13 | } |
14 |
2、以上提交代码中第三行,配置了一个id值为random,IRichSpout对象为RandomWordSpout,而并行度为2(两个线程里面跑两个任务)的spout, setSpout函数实现源码如下:
1 | public SpoutDeclarer setSpout(String id, IRichSpout spout, Number parallelism_hint) { |
2 | |
3 | validateUnusedId(id); |
4 | |
5 | initCommon(id, spout, parallelism_hint); |
6 | |
7 | _spouts.put(id, spout); |
8 | |
9 | return new SpoutGetter(id); |
10 | |
11 | } |
12 |
validateUnusedId:检测输入的id是不是唯一,若已经存在将抛出异常;
initCommon:构建ComponentCommon对象并进行相应的初始化,最后放入到_commons(以上TopologyBuilder中定义的Map);
initCommon函数实现源码:
1 | private void initCommon(String id, IComponent component, Number parallelism) { |
2 | |
3 | ComponentCommon common = new ComponentCommon(); |
4 | |
5 | // 设置消息流的来源及分组方式 |
6 | |
7 | common.set_inputs(new HashMap<GlobalStreamId, Grouping>()); |
8 | |
9 | if(parallelism!=null) |
10 | |
11 | // 设置并行度 |
12 | common.set_parallelism_hint(parallelism.intValue()); |
13 | |
14 | Map conf = component.getComponentConfiguration(); |
15 | |
16 | if(conf!=null) |
17 | |
18 | // 设置组件的配置参数 |
19 | common.set_json_conf(JSONValue.toJSONString(conf)); |
20 | |
21 | _commons.put(id, common); |
22 | } |
在ComponentCommon中主要对以下四个属性字段进行设置:
GlobalStreamId:确定消息来源,其中componentId表示所属组件,streamId为消息流的标识符;
Grouping:确定消息分组方式;
private Map<GlobalStreamId,Grouping> inputs;
StreamInfo表示输出的字段列表及是否为直接流
private Map<String,StreamInfo> streams;
private int parallelism_hint; // 设置并行度
private String json_conf; // 其它配置参数设置(必须为JSON格式)
3、SpoutGetter实现源码:
1 | protected class SpoutGetter extends ConfigGetter<SpoutDeclarer> implements SpoutDeclarer { |
2 | |
3 | public SpoutGetter(String id) { |
4 | super(id); |
5 | } |
6 | } |
ConfigGetter、SpoutGetter的实现都是在TopologyBuilder中, ConfigGetter作用:设置程序中的配置项,覆盖默认的配置项,且配置项的格式为为JSON(本质上是改变对应ComponentCommon对象中json_conf的值);
4、提交示例代码中的第四行定义了一个id为transfer,IRichSpout对象为TransferBolt,并行度为4的bolt
setBolt实现源码:
1 | public BoltDeclarer setBolt(String id, IRichBolt bolt, Number parallelism_hint) { |
2 | |
3 | validateUnusedId(id); |
4 | |
5 | initCommon(id, bolt, parallelism_hint); |
6 | |
7 | _bolts.put(id, bolt); |
8 | |
9 | return new BoltGetter(id); |
10 | |
11 | } |
12 |
设置Bolt的函数与设置Spout函数的实现唯一的区别在返回结果;
BoltGetter实现部分源码:
1 | protected class BoltGetter extends ConfigGetter<BoltDeclarer> implements BoltDeclarer { |
2 | |
3 | private String _boltId; |
4 | |
5 | public BoltGetter(String boltId) { |
6 | |
7 | super(boltId); |
8 | |
9 | _boltId = boltId; |
10 | |
11 | } |
12 | |
13 | public BoltDeclarer shuffleGrouping(String componentId) { |
14 | |
15 | return shuffleGrouping(componentId, Utils.DEFAULT_STREAM_ID); |
16 | |
17 | } |
18 | |
19 | public BoltDeclarer fieldsGrouping(String componentId, Fields fields) { |
20 | |
21 | return fieldsGrouping(componentId, Utils.DEFAULT_STREAM_ID, fields); |
22 | |
23 | } |
24 | |
25 | public BoltDeclarer fieldsGrouping(String componentId, String streamId, Fields fields) { |
26 | |
27 | return grouping(componentId, streamId, Grouping.fields(fields.toList())); |
28 | |
29 | } |
30 | |
31 | public BoltDeclarer shuffleGrouping(String componentId, String streamId) { |
32 | |
33 | return grouping(componentId, streamId, Grouping.shuffle(new NullStruct())); |
34 | |
35 | } |
36 | |
37 | private BoltDeclarer grouping(String componentId, String streamId, Grouping grouping) { |
38 | |
39 | _commons.get(_boltId).put_to_inputs(new GlobalStreamId(componentId, streamId), grouping); |
40 | |
41 | return this; |
42 | |
43 | } |
44 | |
45 | ......................................... |
46 | |
47 | } |
48 |
BoltGetter继承至ConfigGetter并实现了BoltDeclarer接口,并重载了BoltDeclarer(InputDeclarer)中各种分组方式(如:fieldsGrouping、shuffleGrouping),分组方式的实现本质上是在_commons中通过对用的boltId找到对应的ComponentCommon对象,对inputs属性进行设置;
5、通过以上几步完成了bolt与spout的配置(对应提交示例代码中的2~5行),6~9行是对运行环境的配置,10行用于向集群提交执行任务,builder.createTopology用于构建StormTopology对象,createTopology实现源码如下:
1 | public StormTopology createTopology() { |
2 | |
3 | Map<String, Bolt> boltSpecs = new HashMap<String, Bolt>(); |
4 | |
5 | Map<String, SpoutSpec> spoutSpecs = new HashMap<String, SpoutSpec>(); |
6 | |
7 | for(String boltId: _bolts.keySet()) { |
8 | |
9 | IRichBolt bolt = _bolts.get(boltId); |
10 | |
11 | ComponentCommon common = getComponentCommon(boltId, bolt); |
12 | |
13 | boltSpecs.put(boltId, new Bolt(ComponentObject.serialized_java(Utils.serialize(bolt)), common)); |
14 | |
15 | } |
16 | |
17 | for(String spoutId: _spouts.keySet()) { |
18 | |
19 | IRichSpout spout = _spouts.get(spoutId); |
20 | |
21 | ComponentCommon common = getComponentCommon(spoutId, spout); |
22 | |
23 | spoutSpecs.put(spoutId, new SpoutSpec(ComponentObject.serialized_java(Utils.serialize(spout)), common)); |
24 | |
25 | } |
26 | |
27 | return new StormTopology(spoutSpecs,boltSpecs,new HashMap<String, StateSpoutSpec>()); |
28 | |
29 | } |
30 | |
31 |
以上源码实现中主要做了两件事:
- 通过boltId从_bolts中获取到对应的bolt对象,再通过getComponentCommon方法设置对应ComponentCommon对象的streams(输出的字段列表及是否为直接流)属性值,最后将bolt和common一起 放入到boltSpecs集合中。
- 通过spoutId从_spouts中获取到对应的spout对象,再通过getComponentCommon方法设置对应ComponentCommon对象的streams(输出的字段列表及是否为直接流)属性值,最后将spout和common一起 放入到boltSpecs集合中。
- 通过以上两步使所设置的所有组件都封装到StormTopology对象中,最后提交的到集群中运行。