Storm整合Hbase

先丢个官网链接
本章博客依赖官网
第一步，先放依赖：
//为了防止冲突，我们用exclusion将hadoopClient和hadoopAuth排除

<dependency>
      <groupId>org.apache.storm</groupId>
      <artifactId>storm-hbase</artifactId>
      <version>${storm.version}</version>
      <exclusions>
        <exclusion>
          <groupId>com.googla.guava</groupId>
          <artifactId>guava</artifactId>
        </exclusion>
      </exclusions>
    </dependency>
    <dependency>
      <groupId>com.google.guava</groupId>
      <artifactId>guava</artifactId>
      <version>16.0.1</version>
    </dependency>

同时也要在上面的hadoop-client上添加排除代码

<dependency>
      <groupId>org.apache.hadoop</groupId>
      <artifactId>hadoop-client</artifactId>
      <version>${hadoop-client}</version>
      <exclusions>
        <exclusion>
          <groupId>com.googla.guava</groupId>
          <artifactId>guava</artifactId>
        </exclusion>
      </exclusions>
    </dependency>

启动habse，先添加一张表，表名"wc",列族“cf”

 create "wc","cf"

在这里我们只需要编写一个spout接收数据，和一个bolt进行wordcount数据即可，逻辑与之前Storm整合Redis一致，这边直接贴代码
编写第一个spout

public static class DataSourceSpout extends BaseRichSpout {
		//由于需要发送数据，所以定义一个SpoutOutputCollector 
        private SpoutOutputCollector collector;
        @Override
        public void open(Map conf, TopologyContext context, SpoutOutputCollector collector) {
        	//初始化SpoutOutputCollector 
            this.collector = collector;
        }
		
		//这边我们自定义一个数组
        public static final String[] words = new String[]{"apple","banana","orange","strawberry",};

        @Override
        public void nextTuple() {
        	//我们通过随机读取数组的方式生成数据进行测试
            Random random = new Random();
            String word = words[random.nextInt(words.length)];
            //这边为了防止刷屏，我们设定sleep一秒
            Utils.sleep(1000);
            //发送数据
            this.collector.emit(new Values(word));
        }

        @Override
        public void declareOutputFields(OutputFieldsDeclarer declarer) {
        	//声明输出字段，和上面发送的数据相对应
            declarer.declare(new Fields("word"));

        }
    }

编写下一个bolt

public static class CountWords extends BaseRichBolt{
		//由于需要发送数据，所以定义一个SpoutOutputCollector 
        private OutputCollector collector;

        @Override
        public void prepare(Map stormConf, TopologyContext context, OutputCollector collector) {
        	//初始化SpoutOutputCollector 
            this.collector = collector;
        }
		
		//定义一个map存放数据
        Map<String,Integer> map = new HashMap<>();
        @Override
        public void execute(Tuple input) {
        	//抓取上面发送过来的数据
            String word = input.getStringByField("word");
            Integer i = map.get(word);
            if (i == null){
                i = 0;
            }
            i++;

            map.put(word,i);
            //这边输出遗传字符串方便测试
            System.out.println("emit : "+word + "   "+map.get(word));
            //将word和count发送出去
            this.collector.emit(new Values(word,map.get(word)));
        }


        @Override
        public void declareOutputFields(OutputFieldsDeclarer declarer) {
        	//注意这边发送的字段名必须与数据库的字段名一致
            declarer.declare(new Fields("word","count"));
        }
    }

重点在于main方法的编写：
结合官网：

public static void main(String[] args) {
		//创建一个config,用于存放hbase连接信息
        Config config = new Config();
        //这边新建一个map
        Map<String, Object> map = new HashMap<>();
        //这边网map添加两个数据，这个数据可以从hbaser-site.xml里去找
        map.put("hbase.rootdi","hdfs://192.168.0.133:8020/hbase");
        map.put("hbase.zookeeper.quorum","192.168.0.133:2181");
        //将map放到config中去
        config.put("hbase.conf",map);

		//这边是官网参考代码
        SimpleHBaseMapper mapper = new SimpleHBaseMapper()
        		//设置rowKey
                .withRowKeyField("word")
                //设置字段，名称必须与上面发送的相对应
                .withColumnFields(new Fields("word"))
                .withCounterFields(new Fields("count"))
                //设置列族
                .withColumnFamily("cf");
		//这边第一个参数是表名
        HBaseBolt hbase = new HBaseBolt("wc", mapper)
        		//这边必须设置ConfigKey,将上面map里的放进去就行
                .withConfigKey("hbase.conf");


        TopologyBuilder builder = new TopologyBuilder();
        builder.setSpout("DataSourceSpout",new DataSourceSpout());
        builder.setBolt("CountWords",new CountWords()).shuffleGrouping("DataSourceSpout");
        builder.setBolt("hbase",hbase).shuffleGrouping("CountWords");


        LocalCluster cluster = new LocalCluster();
        cluster.submitTopology("LocalWCStormHbaseTop",config,builder.createTopology());
    }

运行查看hbase

hbase里数据一直在跟新，这样我们的Storm整合Hbase就完成了