当前位置：首页 > news >正文

Flink DataSet API

news 来源：原创 2025/7/6 13:50:38

文章目录

- - DataSet Sources
  - DataSet Transformation
  - DataSet Sink
  - 序列化器
  - 样例一：读 csv 文件生成 csv 文件
  - 样例二：读 starrocks 写 starrocks
  - 样例三：DataSet、Table Sql 处理后写入 StarRocks
  - `DataSet<Row>` 遍历
  - 遇到的坑

分类：

Source：数据源创建初始数据集，例如来自文件或 Java 集合。
Transformation：数据转换将一个或多个 DataSet 转换为新的 DataSet
Sink：将计算结果存储或返回

DataSet Sources

ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();// 从本地文件系统读
DataSet<String> localLines = env.readTextFile("file:///path/to/my/textfile");// 读取HDFS文件
DataSet<String> hdfsLines = env.readTextFile("hdfs://nnHost:nnPort/path/to/my/textfile");// 读取CSV文件
DataSet<Tuple3<Integer, String, Double>> csvInput = env.readCsvFile("hdfs:///the/CSV/file").types(Integer.class, String.class, Double.class);// 读取CSV文件中的部分
DataSet<Tuple2<String, Double>> csvInput = env.readCsvFile("hdfs:///the/CSV/file").includeFields("10010").types(String.class, Double.class);// 读取CSV映射为一个java类
DataSet<Person>> csvInput = env.readCsvFile("hdfs:///the/CSV/file").pojoType(Person.class, "name", "age", "zipcode");// 读取一个指定位置序列化好的文件
DataSet<Tuple2<IntWritable, Text>> tuples =env.readSequenceFile(IntWritable.class, Text.class, "hdfs://nnHost:nnPort/path/to/file");// 从输入字符创建
DataSet<String> value = env.fromElements("Foo", "bar", "foobar", "fubar");// 创建一个数字序列
DataSet<Long> numbers = env.generateSequence(1, 10000000);// 从关系型数据库读取
DataSet<Tuple2<String, Integer> dbData =
env.createInput(JDBCInputFormat.buildJDBCInputFormat()
.setDrivername("org.apache.derby.jdbc.EmbeddedDriver")
.setDBUrl("jdbc:derby:memory:persons")
.setQuery("select name, age from persons")
.setRowTypeInfo(new RowTypeInfo(BasicTypeInfo.STRING_TYPE_INFO, BasicTypeInfo.INT_TYPE_INFO))
.finish());

DataSet Transformation

可参考：
Flink从入门到放弃(入门篇3)-DataSetAPI
Flink的DataSet基本算子总结

DataSet Sink

// text data
DataSet<String> textData = // [...]// write DataSet to a file on the local file system
textData.writeAsText("file:///my/result/on/localFS");// write DataSet to a file on a HDFS with a namenode running at nnHost:nnPort
textData.writeAsText("hdfs://nnHost:nnPort/my/result/on/localFS");// write DataSet to a file and overwrite the file if it exists
textData.writeAsText("file:///my/result/on/localFS", WriteMode.OVERWRITE);// tuples as lines with pipe as the separator "a|b|c"
DataSet<Tuple3<String, Integer, Double>> values = // [...]
values.writeAsCsv("file:///path/to/the/result/file", "\n", "|");// this writes tuples in the text formatting "(a, b, c)", rather than as CSV lines
values.writeAsText("file:///path/to/the/result/file");// this writes values as strings using a user-defined TextFormatter object
values.writeAsFormattedText("file:///path/to/the/result/file",new TextFormatter<Tuple2<Integer, Integer>>() {public String format (Tuple2<Integer, Integer> value) {return value.f1 + " - " + value.f0;}});

使用自定义输出格式：

DataSet<Tuple3<String, Integer, Double>> myResult = [...]// write Tuple DataSet to a relational database
myResult.output(// build and configure OutputFormatJDBCOutputFormat.buildJDBCOutputFormat().setDrivername("org.apache.derby.jdbc.EmbeddedDriver").setDBUrl("jdbc:derby:memory:persons").setQuery("insert into persons (name, age, height) values (?,?,?)").finish());

序列化器

Flink 自带了针对诸如 int，long，String 等标准类型的序列化器
针对 Flink 无法实现序列化的数据类型，我们可以交给 Avro 和 Kryo

使用方法：

ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();使用avro序列化：env.getConfig().enableForceAvro();
使用kryo序列化：env.getConfig().enableForceKryo();
使用自定义序列化：env.getConfig().addDefaultKryoSerializer(Class<?> type, Class<? extends Serializer<?>> serializerClass)

样例一：读 csv 文件生成 csv 文件

参考：（3）Flink学习- Table API & SQL编程

        <dependency><groupId>org.apache.flink</groupId><artifactId>flink-java</artifactId><version>1.9.1</version></dependency><dependency><groupId>org.apache.flink</groupId><artifactId>flink-streaming-java_2.11</artifactId><version>1.9.1</version></dependency><!--使用Java编程语言支持DataStream / DataSet API的Table＆SQL API--><dependency><groupId>org.apache.flink</groupId><artifactId>flink-table-api-java-bridge_2.11</artifactId><version>1.9.1</version><!--<scope>provided</scope>--></dependency><!--表程序规划器和运行时--><dependency><groupId>org.apache.flink</groupId><artifactId>flink-table-planner_2.11</artifactId><version>1.9.1</version><!--<scope>provided</scope>--></dependency><dependency><groupId>org.apache.flink</groupId><artifactId>flink-jdbc_2.11</artifactId><version>1.9.1</version><!--<scope>provided</scope>--></dependency><dependency><groupId>org.projectlombok</groupId><artifactId>lombok</artifactId><version>1.16.18</version></dependency><dependency><groupId>mysql</groupId><artifactId>mysql-connector-java</artifactId><version>5.1.49</version></dependency>

import lombok.Data;
import org.apache.flink.api.common.typeinfo.TypeInformation;
import org.apache.flink.api.common.typeinfo.Types;
import org.apache.flink.api.java.DataSet;
import org.apache.flink.api.java.ExecutionEnvironment;
import org.apache.flink.core.fs.FileSystem;
import org.apache.flink.table.api.Table;
import org.apache.flink.table.api.java.BatchTableEnvironment;
import org.apache.flink.table.sinks.CsvTableSink;public class SQLWordCount {public static void main(String[] args) throws Exception {// 1、获取执行环境 ExecutionEnvironment （批处理用这个对象）ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();BatchTableEnvironment bTableEnv = BatchTableEnvironment.create(env);
//        DataSet<WC> input = env.fromElements(
//                WC.of("hello", 1),
//                WC.of("hqs", 1),
//                WC.of("world", 1),
//                WC.of("hello", 1)
//        );// 注册数据集
//        tEnv.registerDataSet("WordCount", input, "word, frequency");// 2、加载数据源到 DataSetDataSet<Student> csv = env.readCsvFile("D:\\tmp\\data.csv").ignoreFirstLine().pojoType(Student.class, "name", "age");// 3、将DataSet装换为TableTable students = bTableEnv.fromDataSet(csv);bTableEnv.registerTable("student", students);// 4、注册student表Table result = bTableEnv.sqlQuery("select name,age from student");result.printSchema();DataSet<Student> dset = bTableEnv.toDataSet(result, Student.class);// DataSet<Row> dset = bTableEnv.toDataSet(result, Row.class);System.out.println("count-->" + dset.count());dset.print();// 5、sink输出CsvTableSink sink1 = new CsvTableSink("D:\\tmp\\result.csv", ",", 1, FileSystem.WriteMode.OVERWRITE);String[] fieldNames = {"name", "age"};TypeInformation[] fieldTypes = {Types.STRING, Types.INT};bTableEnv.registerTableSink("CsvOutPutTable", fieldNames, fieldTypes, sink1);result.insertInto("CsvOutPutTable");env.execute("SQL-Batch");}@Datapublic static class Student {private String name;private int age;}
}

准备测试文件 data.csv：

name,age
zhangsan,23
lisi,43
wangwu,12

运行程序后会生成 D:\\tmp\\result.csv 文件。

样例二：读 starrocks 写 starrocks

import org.apache.flink.api.common.typeinfo.TypeInformation;
import org.apache.flink.api.common.typeinfo.Types;
import org.apache.flink.api.java.DataSet;
import org.apache.flink.api.java.ExecutionEnvironment;
import org.apache.flink.api.java.io.jdbc.JDBCInputFormat;
import org.apache.flink.api.java.io.jdbc.JDBCOutputFormat;
import org.apache.flink.api.java.operators.DataSource;
import org.apache.flink.api.java.typeutils.RowTypeInfo;
import org.apache.flink.types.Row;public class SQLWordCount {public static void main(String[] args) throws Exception {TypeInformation[] fieldTypes = {Types.STRING, Types.INT};RowTypeInfo rowTypeInfo = new RowTypeInfo(fieldTypes);JDBCInputFormat jdbcInputFormat = JDBCInputFormat.buildJDBCInputFormat().setDrivername("com.mysql.jdbc.Driver").setDBUrl("jdbc:mysql://192.168.xx.xx:9030/dwd?characterEncoding=utf8").setUsername("root").setPassword("").setQuery("select * from student").setRowTypeInfo(rowTypeInfo).finish();final ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();// 方式一DataSource s = env.createInput(jdbcInputFormat);s.output(JDBCOutputFormat.buildJDBCOutputFormat().setDrivername("com.mysql.jdbc.Driver").setDBUrl("jdbc:mysql://192.168.xx.xx:9030/dwd?characterEncoding=utf8").setUsername("root").setPassword("").setQuery("insert into student values(?, ?)").finish());// 方式二
//        DataSet<Row> dataSource = env.createInput(jdbcInputFormat);
//
//        dataSource.output(JDBCOutputFormat.buildJDBCOutputFormat()
//                .setDrivername("com.mysql.jdbc.Driver")
//                .setDBUrl("jdbc:mysql://192.168.xx.xx:9030/dwd?characterEncoding=utf8")
//                .setUsername("root").setPassword("")
//                .setQuery("insert into student values(?, ?)")
//                .finish()
//        );env.execute("SQL-Batch");}
}

数据准备：

CREATE TABLE student (name STRING,age INT
) ENGINE=OLAP 
DUPLICATE KEY(`name`)
DISTRIBUTED BY RANDOM
PROPERTIES (
"compression" = "LZ4",
"fast_schema_evolution" = "false",
"replicated_storage" = "true",
"replication_num" = "1"
);insert into student values('zhangsan', 23);

参考：
flink 读取mysql源 JDBCInputFormat、自定义数据源
flink1.10中三种数据处理方式的连接器说明
flink读写MySQL的两种方式

注意：如果运行 java -cp flink-app-1.0-SNAPSHOT-jar-with-dependencies.jar com.xiaoqiang.app.SQLWordCount 时报错：Exception in thread "main" com.typesafe.config.ConfigException$UnresolvedSubstitution: reference.conf @ jar:file:flink-app-1.0-SNAPSHOT-jar-with-dependencies.jar!/reference.conf: 875: Could not resolve substitution to a value: ${akka.stream.materializer}

解决：报错：Flink Could not resolve substitution to a value: ${akka.stream.materializer}

    <build><plugins><!-- Java Compiler --><plugin><groupId>org.apache.maven.plugins</groupId><artifactId>maven-compiler-plugin</artifactId><version>3.1</version><configuration><source>1.8</source><target>1.8</target></configuration></plugin><plugin><groupId>org.apache.maven.plugins</groupId><artifactId>maven-shade-plugin</artifactId><version>2.3</version><executions><execution><phase>package</phase><goals><goal>shade</goal></goals><configuration><transformers><!--<transformer implementation="org.apache.maven.plugins.shade.resource.ManifestResourceTransformer"><mainClass>flink.KafkaDemo1</mainClass></transformer>--><transformer implementation="org.apache.maven.plugins.shade.resource.AppendingTransformer"><resource>reference.conf</resource></transformer></transformers></configuration></execution></executions></plugin></plugins></build>

样例三：DataSet、Table Sql 处理后写入 StarRocks

import org.apache.flink.api.common.typeinfo.TypeInformation;
import org.apache.flink.api.common.typeinfo.Types;
import org.apache.flink.api.java.DataSet;
import org.apache.flink.api.java.ExecutionEnvironment;
import org.apache.flink.api.java.io.jdbc.JDBCInputFormat;
import org.apache.flink.api.java.io.jdbc.JDBCOutputFormat;
import org.apache.flink.api.java.typeutils.RowTypeInfo;
import org.apache.flink.table.api.Table;
import org.apache.flink.table.api.java.BatchTableEnvironment;
import org.apache.flink.types.Row;public class SQLWordCount {public static void main(String[] args) throws Exception {TypeInformation[] fieldTypes = {Types.STRING, Types.INT};RowTypeInfo rowTypeInfo = new RowTypeInfo(fieldTypes);JDBCInputFormat jdbcInputFormat = JDBCInputFormat.buildJDBCInputFormat().setDrivername("com.mysql.jdbc.Driver").setDBUrl("jdbc:mysql://192.168.xx.xx:9030/dwd?characterEncoding=utf8").setUsername("root").setPassword("").setQuery("select * from student").setRowTypeInfo(rowTypeInfo).finish();ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();BatchTableEnvironment bTableEnv = BatchTableEnvironment.create(env);DataSet<Row> dataSource = env.createInput(jdbcInputFormat);dataSource.print();Table students = bTableEnv.fromDataSet(dataSource);bTableEnv.registerTable("student", students);Table result = bTableEnv.sqlQuery("select name, age from (select f0 as name, f1 as age from student) group by name, age");result.printSchema();DataSet<Row> dset = bTableEnv.toDataSet(result, Row.class);dset.output(JDBCOutputFormat.buildJDBCOutputFormat().setDrivername("com.mysql.jdbc.Driver").setDBUrl("jdbc:mysql://192.168.xx.xx:9030/dwd?characterEncoding=utf8").setUsername("root").setPassword("").setQuery("insert into student values(?, ?)").finish());env.execute("SQL-Batch");}
}

`DataSet<Row>` 遍历

try {dataSet.map(new MapFunction<Row, String>() {@Overridepublic String map(Row value) throws Exception {// 在这里处理每一行的数据float dateNum = (float) value.getField(2);float dateAge = (float) value.getField(3);return "dataSet 遍历完成";}}).print();
} catch (Exception e) {throw new RuntimeException(e);
}

如果你需要转换每一行为多行输出，可以使用 FlatMapFunction：

rows.flatMap(new FlatMapFunction<Row, Row>() {@Overridepublic void flatMap(Row value, Collector<Row> out) throws Exception {// 转换逻辑// 例如，复制原行并输出out.collect(value);// 如果需要多行输出，可以再次调用out.collect(...)}
}).collect();

请注意，.collect() 方法用于强制执行计算，并获取结果到客户端。在生产代码中，你可能想要将结果输出到文件或者发送到外部系统，而不是仅仅收集到客户端。

遇到的坑

坑1：Bang equal '!=' is not allowed under the current SQL conformance level
解决：将 sql 中的 != 修改为 <>

坑2：java.lang.RuntimeException: No new data sinks have been defined since the last execution. The last execution refers to the latest call to 'execute()', 'count()', 'collect()', or 'print()'.
解释：在最后一行代码 env.execute() 执行的时候，没有新的数据接收器被定义，对于 Flink 批处理而前一行代码 result.print() 已经触发了代码的执行和输出，所以再执行 env.execute()，就是多余的了，因此报了上面的异常。
解决方法：去掉最后一行代码 env.execute(); 就可以了。