Storm-Crawler与Cassandra的集成并在Gradle中构建拓扑

dependencies {
    // Storm-Crawler依赖
    implementation 'com.digitalpebble.stormcrawler:storm-crawler-core:1.17'
    // Cassandra依赖
    implementation 'com.datastax.cassandra:cassandra-driver-core:4.13.0'
}

import com.datastax.driver.core.Cluster;
import com.datastax.driver.core.Session;
import com.digitalpebble.stormcrawler.bolt.JSoupParserBolt;
import com.digitalpebble.stormcrawler.bolt.URLFilterBolt;
import com.digitalpebble.stormcrawler.spout.FileSpout;
import org.apache.storm.Config;
import org.apache.storm.LocalCluster;
import org.apache.storm.topology.TopologyBuilder;

public class MyCrawlerTopology {

    public static void main(String[] args) {
        // 创建TopologyBuilder
        TopologyBuilder builder = new TopologyBuilder();

        // 配置Cassandra连接信息
        Cluster cluster = Cluster.builder()
                .addContactPoint("127.0.0.1")
                .build();
        Session session = cluster.connect("mykeyspace");

        // 添加Spout和Bolt
        builder.setSpout("spout", new FileSpout());
        builder.setBolt("filter", new URLFilterBolt()).shuffleGrouping("spout");
        builder.setBolt("parser", new JSoupParserBolt()).shuffleGrouping("filter");

        // 创建本地集群
        LocalCluster cluster = new LocalCluster();

        // 提交拓扑
        cluster.submitTopology("my-crawler-topology", new Config(), builder.createTopology());

        // 关闭连接
        session.close();
        cluster.close();
    }
}