Spark-sql数据的加载与保存

3.1 读取和保存文件

SparkSQL读取和保存的文件一般为三种，JSON文件、CSV文件和列式存储的文件，同时可以通过添加参数，来识别不同的存储和压缩格式。

3.1.1 CSV文件

1）代码实现

package com.atguigu.sparksql;

import com.atguigu.sparksql.Bean.User;

import org.apache.spark.SparkConf;

import org.apache.spark.api.java.function.MapFunction;

import org.apache.spark.sql.*;

public class Test06_CSV {

    public static void main(String[] args) throws ClassNotFoundException {

        //1. 创建配置对象

        SparkConf conf = new SparkConf().setAppName("sparksql").setMaster("local[*]");

        //2. 获取sparkSession

        SparkSession spark = SparkSession.builder().config(conf).getOrCreate();

        //3. 编写代码

        DataFrameReader reader = spark.read();

        // 添加参数  读取csv

        Dataset<Row> userDS = reader

                .option("header", "true")//默认为false 不读取列名

                .option("sep",",") // 默认为, 列的分割

                // 不需要写压缩格式  自适应

                .csv("input/user.csv");

        userDS.show();

        // 转换为user的ds

        // 直接转换类型会报错  csv读取的数据都是string

//        Dataset<User> userDS1 = userDS.as(Encoders.bean(User.class));

        userDS.printSchema();

        Dataset<User> userDS1 = userDS.map(new MapFunction<Row, User>() {

            @Override

            public User call(Row value) throws Exception {

                return new User(Long.valueOf(value.getString(0)), value.getString(1));

            }

        }, Encoders.bean(User.class));

        userDS1.show();

        // 写出为csv文件

        DataFrameWriter<User> writer = userDS1.write();

        writer.option("seq",";")

                .option("header","true")

//                .option("compression","gzip")// 压缩格式

                // 写出模式

                // append 追加

                // Ignore 忽略本次写出

                // Overwrite 覆盖写

                // ErrorIfExists 如果存在报错

                .mode(SaveMode.Append)

                .csv("output");

        //4. 关闭sparkSession

        spark.close();

    }

}

3.1.2 JSON文件

package com.atguigu.sparksql;

import com.atguigu.sparksql.Bean.User;

import org.apache.spark.SparkConf;

import org.apache.spark.sql.*;

public class Test07_JSON {

    public static void main(String[] args) {

        //1. 创建配置对象

        SparkConf conf = new SparkConf().setAppName("sparksql").setMaster("local[*]");

        //2. 获取sparkSession

        SparkSession spark = SparkSession.builder().config(conf).getOrCreate();

        //3. 编写代码

        Dataset<Row> json = spark.read().json("input/user.json");

        // json数据可以读取数据的数据类型

        Dataset<User> userDS = json.as(Encoders.bean(User.class));

        userDS.show();

       

        // 读取别的类型的数据也能写出为json

        DataFrameWriter<User> writer = userDS.write();

        writer.json("output1");

        //4. 关闭sparkSession

        spark.close();

    }

}

3.1.3 Parquet文件

列式存储的数据自带列分割。

package com.atguigu.sparksql;

import org.apache.spark.SparkConf;

import org.apache.spark.sql.Dataset;

import org.apache.spark.sql.Row;

import org.apache.spark.sql.SparkSession;

public class Test08_Parquet {

    public static void main(String[] args) {

        //1. 创建配置对象

        SparkConf conf = new SparkConf().setAppName("sparksql").setMaster("local[*]");

        //2. 获取sparkSession

        SparkSession spark = SparkSession.builder().config(conf).getOrCreate();

        //3. 编写代码

        Dataset<Row> json = spark.read().json("input/user.json");

        // 写出默认使用snappy压缩

//        json.write().parquet("output");

        // 读取parquet 自带解析  能够识别列名

        Dataset<Row> parquet = spark.read().parquet("output");

        parquet.printSchema();

        //4. 关闭sparkSession

        spark.close();

    }

}

3.2 与MySQL交互

1）导入依赖

<!-- 早期MySQL版本 -->

<dependency>

    <groupId>mysql</groupId>

    <artifactId>mysql-connector-java</artifactId>

    <version>5.1.27</version>

</dependency>

<!-- 5.8(8.0)MySQL版本 -->

<dependency>

    <groupId>mysql</groupId>

    <artifactId>mysql-connector-java</artifactId>

    <version>8.0.18</version>

</dependency>

2）从MySQL读数据

package com.atguigu.sparksql;

import org.apache.spark.SparkConf;

import org.apache.spark.sql.Dataset;

import org.apache.spark.sql.Row;

import org.apache.spark.sql.SparkSession;

import java.util.Properties;

public class Test09_Table {

    public static void main(String[] args) {

        //1. 创建配置对象

        SparkConf conf = new SparkConf().setAppName("sparksql").setMaster("local[*]");

        //2. 获取sparkSession

        SparkSession spark = SparkSession.builder().config(conf).getOrCreate();

        //3. 编写代码

        Dataset<Row> json = spark.read().json("input/user.json");

        // 添加参数

        Properties properties = new Properties();

        properties.setProperty("user","root");

        properties.setProperty("password","000000");

//        json.write()

//                // 写出模式针对于表格追加覆盖

//                .mode(SaveMode.Append)

//                .jdbc("jdbc:mysql://hadoop102:3306","gmall.testInfo",properties);

        Dataset<Row> jdbc = spark.read().jdbc("jdbc:mysql://hadoop102:3306/gmall?useSSL=false&useUnicode=true&characterEncoding=UTF-8&allowPublicKeyRetrieval=true", "test_info", properties);

        jdbc.show();

        //4. 关闭sparkSession

        spark.close();

    }

}

3.3 与Hive交互

SparkSQL可以采用内嵌Hive（spark开箱即用的hive），也可以采用外部Hive。企业开发中，通常采用外部Hive。

3.3.1 Linux中的交互

1）添加MySQL连接驱动到spark-yarn的jars目录

[atguigu@hadoop102 spark-yarn]$ cp /opt/software/mysql-connector-java-5.1.27-bin.jar /opt/module/spark-yarn/jars

2）添加hive-site.xml文件到spark-yarn的conf目录

[atguigu@hadoop102 spark-yarn]$ cp /opt/module/hive/conf/hive-site.xml /opt/module/spark-yarn/conf

3）启动spark-sql的客户端即可

[atguigu@hadoop102 spark-yarn]$  bin/spark-sql --master yarn

spark-sql (default)> show tables;

3.3.2 IDEA中的交互

1）添加依赖

<dependencies>

    <dependency>

       <groupId>org.apache.spark</groupId>

       <artifactId>spark-sql_2.12</artifactId>

       <version>3.3.1</version>

    </dependency>

    <dependency>

       <groupId>mysql</groupId>

       <artifactId>mysql-connector-java</artifactId>

       <version>5.1.27</version>

    </dependency>

    <dependency>

       <groupId>org.apache.spark</groupId>

       <artifactId>spark-hive_2.12</artifactId>

       <version>3.3.1</version>

    </dependency>

    <dependency>

       <groupId>org.projectlombok</groupId>

       <artifactId>lombok</artifactId>

       <version>1.18.22</version>

    </dependency>

</dependencies>

2）拷贝hive-site.xml到resources目录（如果需要操作Hadoop，需要拷贝hdfs-site.xml、core-site.xml、yarn-site.xml）

3）代码实现

package com.atguigu.sparksql;

import org.apache.spark.SparkConf;

import org.apache.spark.sql.SparkSession;

public class Test10_Hive {

    public static void main(String[] args) {

        System.setProperty("HADOOP_USER_NAME","atguigu");

        //1. 创建配置对象

        SparkConf conf = new SparkConf().setAppName("sparksql").setMaster("local[*]");

        //2. 获取sparkSession

        SparkSession spark = SparkSession.builder()

                .enableHiveSupport()// 添加hive支持

                .config(conf).getOrCreate();

        //3. 编写代码

        spark.sql("show tables").show();

        spark.sql("create table user_info(name String,age bigint)");

        spark.sql("insert into table user_info values('zhangsan',10)");

        spark.sql("select * from user_info").show();

        //4. 关闭sparkSession

        spark.close();

    }

}