[大数据]第四章 Flink实战之WordCount基础入门

1、Flink开发环境搭建

1.1、创建Maven项目

• 1、选择“File” -->“New”–>“Project”
  
• 2、选择 Maven，设置JDK版本，选择maven项目的模板
  

org.apache.maven.archetypes:maven-archetype-quickstart
#代表普通的maven项目面板

• 3、设置Groupid和Artifactid
  

Groupid:公司名称
Artifactid:项目模块

• 4、检查Maven环境，选择”Next“
  

主要选择maven路径和settings路径

• 5、检查项目名和工作空间，选择finish
  
• 6、等待项目创建，下载资源，创建完成后目录结构如下
  

1.2、pom文件配置

<?xml version="1.0" encoding="UTF-8"?>
<project xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
         xmlns="http://maven.apache.org/POM/4.0.0"
         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
    <modelVersion>4.0.0</modelVersion>

    <groupId>com.hainiu</groupId>
    <artifactId>hainiuflink</artifactId>
    <version>1.0</version>
    
    <properties>
        <java.version>1.8</java.version>
        <scala.version>2.11</scala.version>
        <flink.version>1.9.3</flink.version>
        <parquet.version>1.10.0</parquet.version>
        <hadoop.version>2.7.3</hadoop.version>
        <fastjson.version>1.2.72</fastjson.version>
        <redis.version>2.9.0</redis.version>
        <mysql.version>5.1.35</mysql.version>
        <log4j.version>1.2.17</log4j.version>
        <slf4j.version>1.7.7</slf4j.version>
        <maven.compiler.source>1.8</maven.compiler.source>
        <maven.compiler.target>1.8</maven.compiler.target>
        <maven.compiler.compilerVersion>1.8</maven.compiler.compilerVersion>
        <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
        <project.build.scope>compile</project.build.scope>
<!--        <project.build.scope>provided</project.build.scope>-->
        <mainClass>com.hainiu.Driver</mainClass>
    </properties>

    <dependencies>
        <dependency>
            <groupId>org.slf4j</groupId>
            <artifactId>slf4j-log4j12</artifactId>
            <version>${slf4j.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>

        <dependency>
            <groupId>log4j</groupId>
            <artifactId>log4j</artifactId>
            <version>${log4j.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <!-- flink的hadoop兼容 -->
        <dependency>
            <groupId>org.apache.hadoop</groupId>
            <artifactId>hadoop-client</artifactId>
            <version>${hadoop.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <!-- flink的hadoop兼容 -->
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-hadoop-compatibility_${scala.version}</artifactId>
            <version>${flink.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <!-- flink的java的api -->
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-java</artifactId>
            <version>${flink.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <!-- flink streaming的java的api -->
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-streaming-java_${scala.version}</artifactId>
            <version>${flink.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <!-- flink的scala的api -->
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-scala_${scala.version}</artifactId>
            <version>${flink.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <!-- flink streaming的scala的api -->
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-streaming-scala_${scala.version}</artifactId>
            <version>${flink.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <!-- flink运行时的webUI -->
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-runtime-web_${scala.version}</artifactId>
            <version>${flink.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <!-- 使用rocksdb保存flink的state -->
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-statebackend-rocksdb_${scala.version}</artifactId>
            <version>${flink.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <!-- flink操作hbase -->
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-hbase_${scala.version}</artifactId>
            <version>${flink.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <!-- flink操作es -->
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-connector-elasticsearch5_${scala.version}</artifactId>
            <version>${flink.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <!-- flink 的kafka -->
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-connector-kafka-0.10_${scala.version}</artifactId>
            <version>${flink.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <!-- flink 写文件到HDFS -->
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-connector-filesystem_${scala.version}</artifactId>
            <version>${flink.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <!-- mysql连接驱动 -->
        <dependency>
            <groupId>mysql</groupId>
            <artifactId>mysql-connector-java</artifactId>
            <version>${mysql.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <!-- redis连接 -->
        <dependency>
            <groupId>redis.clients</groupId>
            <artifactId>jedis</artifactId>
            <version>${redis.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <!-- flink操作parquet文件格式 -->
        <dependency>
            <groupId>org.apache.parquet</groupId>
            <artifactId>parquet-avro</artifactId>
            <version>${parquet.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <dependency>
            <groupId>org.apache.parquet</groupId>
            <artifactId>parquet-hadoop</artifactId>
            <version>${parquet.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-parquet_${scala.version}</artifactId>
            <version>${flink.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
        <!-- json操作 -->
        <dependency>
            <groupId>com.alibaba</groupId>
            <artifactId>fastjson</artifactId>
            <version>${fastjson.version}</version>
            <scope>${project.build.scope}</scope>
        </dependency>
    </dependencies>

    <build>
        <resources>
            <resource>
                <directory>src/main/resources</directory>
            </resource>
        </resources>
        <plugins>
            <plugin>
                <groupId>org.apache.maven.plugins</groupId>
                <artifactId>maven-assembly-plugin</artifactId>
                <configuration>
                    <descriptors>
                        <descriptor>src/assembly/assembly.xml</descriptor>
                    </descriptors>
                    <archive>
                        <manifest>
                            <mainClass>${mainClass}</mainClass>
                        </manifest>
                    </archive>
                </configuration>
                <executions>
                    <execution>
                        <id>make-assembly</id>
                        <phase>package</phase>
                        <goals>
                            <goal>single</goal>
                        </goals>
                    </execution>
                </executions>
            </plugin>
            <plugin>
                <groupId>org.apache.maven.plugins</groupId>
                <artifactId>maven-surefire-plugin</artifactId>
                <version>2.12</version>
                <configuration>
                    <skip>true</skip>
                    <forkMode>once</forkMode>
                    <excludes>
                        <exclude>**/**</exclude>
                    </excludes>
                </configuration>
            </plugin>
            <plugin>
                <groupId>org.apache.maven.plugins</groupId>
                <artifactId>maven-compiler-plugin</artifactId>
                <version>3.1</version>
                <configuration>
                    <source>${java.version}</source>
                    <target>${java.version}</target>
                    <encoding>${project.build.sourceEncoding}</encoding>
                </configuration>
            </plugin>
        </plugins>
    </build>

</project>

2、Flink开发流程

2.1、获取执行环境

2.1.1、批处理环境获取

ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();

2.1.2、流处理环境获取

StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();

2.1.3、自动模式

• Flink.12开始，流批一体，自动获取模式 - AUTOMATIC()

env.setRuntimeMode(RuntimeExecutionMode.AUTOMATIC);

2.2、创建/加载数据源

2.1.2、从本地文件加载数据源 - 批处理

String path = "H:\\flink_demo\\flink_test\\src\\main\\resources\\wordcount.txt";
DataSet<String> inputDataSet = env.readTextFile(path);

2.1.2、从scoket中读取数据 - 流处理

DataStreamSource<String> elementsSource= env.socketTextStream("10.50.40.131", 9999);

• scoket：

2.1.3、从文件中动态读取数据 - 流处理

DataStream<String> lines = env.readTextFile("file:///path");

2.3、数据转换处理

• 对数据加工转换：扁平化 + 分组 + 聚合

DataSet<Tuple2<String, Integer>> resultDataSet = inputDataSet.flatMap(new MyFlatMapFunction())
    .groupBy(0) // (word, 1) -> 0 表示 word
    .sum(1);

2.4、数据输出/存储

• 输出位置：输出到文件，输出到控制台，输出到MQ，输出到DB,输出到scoket…

//print()它会对流中的每个元素都调用 toString() 方法。
resultDataSet.print();

2.5、计算模型

• 获取执行环境（execution environment）
• 加载/创建初始数据集
• 对数据集进行各种转换操作（生成新的数据集）
• 指定将计算的结果放到何处去
• 触发APP执行

3、实战案例

3.1、批处理开发样例

import org.apache.flink.api.common.functions.FlatMapFunction;
import org.apache.flink.api.java.DataSet;
import org.apache.flink.api.java.ExecutionEnvironment;
import org.apache.flink.api.java.tuple.Tuple2;
import org.apache.flink.util.Collector;

public class WordCount {
    public static void main(String[] args) throws Exception {
        // 1、创建执行环境
        ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();

        // 2、读取数据
        String path = "H:\\flink_demo\\flink_test\\src\\main\\resources\\wordcount.txt";
        // DataSet -> Operator -> DataSource
        DataSet<String> inputDataSet = env.readTextFile(path);

        // 3、扁平化 + 分组 + sum
        DataSet<Tuple2<String, Integer>> resultDataSet = inputDataSet.flatMap(new MyFlatMapFunction())
                .groupBy(0) // (word, 1) -> 0 表示 word
                .sum(1);

        resultDataSet.print();
    }

    public static class MyFlatMapFunction implements FlatMapFunction<String, Tuple2<String, Integer>> {

        @Override
        public void flatMap(String input, Collector<Tuple2<String, Integer>> collector) throws Exception {
            String[] words = input.split(" ");
            for (String word : words) {
                collector.collect(new Tuple2<>(word, 1));
            }
        }
    }
}

• 运行结果
  

3.2、流处理开发样例

import org.apache.flink.api.common.functions.FlatMapFunction;
import org.apache.flink.api.java.tuple.Tuple2;
import org.apache.flink.streaming.api.datastream.DataStreamSource;
import org.apache.flink.streaming.api.datastream.SingleOutputStreamOperator;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import org.apache.flink.util.Collector;

public class StreamCount {
    public static void main(String[] args) throws Exception {
        // 1、创建执行环境
        StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();

        // 2、读取 文件数据 数据
        DataStreamSource<String> inputDataStream = env.readTextFile("H:\\flink_demo\\flink_test\\src\\main\\resources\\wordcount.txt");

        // 3、计算
        SingleOutputStreamOperator<Tuple2<String, Integer>> resultDataStream = inputDataStream.flatMap(new FlatMapFunction<String, Tuple2<String, Integer>>() {
            @Override
            public void flatMap(String input, Collector<Tuple2<String, Integer>> collector) throws Exception {
                String[] words = input.split(" ");
                for (String word : words) {
                    collector.collect(new Tuple2<>(word, 1));
                }
            }
        }).keyBy(0)
                .sum(1);

        // 4、输出
        resultDataStream.print();

        // 5、启动 env
        env.execute();
    }
}

• 运行结果

统计文本:Flink
输出:Flink,1

统计文本:增加Spark
输出:Flink,2 Spark,1

统计文本:新增python
输出:Flink,3 Spark,2 python,1

3.3、流/批不同

流式处理的结果:是不断刷新的，在这个例子中，数据时一行一行进入处理任务的，进来一批处理一批，没有数据就处于等待状态。

4、Flink在Yarn部署

4.1、代码打包

import org.apache.flink.api.common.functions.FlatMapFunction;
import org.apache.flink.api.java.tuple.Tuple2;
import org.apache.flink.api.java.utils.ParameterTool;
import org.apache.flink.streaming.api.datastream.DataStreamSource;
import org.apache.flink.streaming.api.datastream.SingleOutputStreamOperator;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import org.apache.flink.util.Collector;

public class StreamCount {
    public static void main(String[] args) throws Exception {
        // 1、创建执行环境
        StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();

        // 2、读取 文件数据 数据
        //DataStreamSource<String> inputDataStream = env.readTextFile("H:\\flink_demo\\flink_test\\src\\main\\resources\\wordcount.txt");
        //2.2、用parameter tool工具从程序启动参数中提取配置项
        ParameterTool parameterTool = ParameterTool.fromArgs(args);
        String host = parameterTool.get("host");
        int port = parameterTool.getInt("port");
        DataStreamSource<String> inputDataStream = env.socketTextStream(host,port);
        
        // 3、计算
        SingleOutputStreamOperator<Tuple2<String, Integer>> resultDataStream = inputDataStream.flatMap(new FlatMapFunction<String, Tuple2<String, Integer>>() {
            @Override
            public void flatMap(String input, Collector<Tuple2<String, Integer>> collector) throws Exception {
                String[] words = input.split(" ");
                for (String word : words) {
                    collector.collect(new Tuple2<>(word, 1));
                }
            }
        }).keyBy(0)
                .sum(1);

        // 4、输出
        resultDataStream.print().setParallelism(1);

        //打印执行计划图
        System.out.println(env.getExecutionPlan());
        // 5、启动 env
        env.execute();
    }
}

• 先编译后打包
  

4.2、通过UI界面部署到集群

（1）将jar包上传到flink集群

• Flink-ui界面 -》submit new job -》Add New
  
  （2）配置参数

entry class:入口类
program Arguments:入口参数
parallelism:并行度,作业提交job时,如果代码没有配,以全局并行度为准,没有以此并性都为准,没有,以全局并行度为准
savapoint path:保存点,从之前存盘的地方启动

（3）运行结果

未成功，待解决

5、疑难杂症

5.1、maven的Plugin文件爆红

• 解决方式 - 设置maven为本地maven库

New -> Settings -> maven

6.2、Flink Web UI截面提交时显示Server reponse message

• 解决方式

vim {flink_home}/log