Spring Data JPA Batch Insertion

Posted 2021-01-14 Bigben

tags:

篇首语：本文由小常识网(cha138.com)小编为大家整理，主要介绍了Spring Data JPA Batch Insertion相关的知识，希望对你有一定的参考价值。

转自：https://www.jeejava.com/spring-data-jpa-batch-insertion/

Spring Data JPA Batch Insertion will show you how we can insert a large dataset into a database at once using Spring Data JPA. For this tutorial we will create a Spring Boot project in Eclipse. We will also see how Spring @Transactional annotation works. Spring transaction required in order to rollback the inserted data at any point if your application fails for any reason.

Sometimes we need to insert or update large number of records in the database. It’s not a good idea to insert multiple records into database one by one in a traditional approach. It will hit the application’s performance. Spring provides batch operations with the help of JpaRepository or CrudRepository, it inserts or updates records into database in one shot. You can also use JDBC API to insert multiple records or batch insertion into database but here we will use Spring JPA’s built-in functionality to get benefits of Spring API.

By default Spring does not save your data into database when you call save() method with multiple entities (a list of objects) passed as argument, hence you have to save entities one by one, which is time consuming and performance gets affected. For this there are few properties that need to be configured to let Spring Data JPA work on batch insertion into database. We will see how these properties set during creation of database configuration class below.

You may also like to read:

Batch Insert using Spring JdbcTemplate

Transaction Management in Spring

Hibernate UserType using Spring Data JPA

Spring EnableEncryptableProperties with Jasypt

Spring Data JPA Entity Auditing using EntityListeners

Spring Data JPA Entity Graph

Spring Data JPA CRUD Example

Prerequisites

The following configurations are required in order to run the application

Eclipse
JDK 1.8
Have gradle installed and configured
Spring dependencies in build.gradle script

Now we will see the below steps how to create a gradle based spring project in Eclipse to work on example Spring Data JPA Batch Insertion.

Creating and setting up the project

First you need to setup the gradle based project in Eclipse and we have to make sure using the below build script we will be able to build the blank project.

In this file notice we have applied required plugins and added required dependencies, such as spring-boot-starter-web, spring-boot-starter-data-jpa and oracle jdbc driver to interact with Java and database API.

Once you create below file, please try to build the project, you should be able to build the blank project.

buildscript {
	ext {
	   springBootVersion = ‘1.5.9.RELEASE‘
    }
    repositories {
		mavenLocal()
        mavenCentral()
    }
    dependencies {
        classpath("org.springframework.boot:spring-boot-gradle-plugin:${springBootVersion}")
    }
}

apply plugin: ‘java‘
apply plugin: ‘org.springframework.boot‘

sourceCompatibility = 1.8
targetCompatibility = 1.8

repositories {
    mavenLocal()
    mavenCentral()
}

dependencies {
	compile(‘org.springframework.boot:spring-boot-starter-web‘)
	compile("org.springframework.boot:spring-boot-starter-data-jpa")
	runtime("com.oracle.jdbc:ojdbc7:12.1.0.2")
}

Creating application.properties file under classpath

Below is the application.properties file under classpath directory src/main/resources and you need to define database credentials to establish connection with database.

Also if you do not want server to run on default port then you may want to specify the server port using server.port key.

Here in Spring Data JPA Batch Insertion example, I am going to use Oracle database but you may use any database as per your requirements.

#datasource
spring.datasource.driverClassName=oracle.jdbc.driver.OracleDriver
spring.datasource.hibernate.dialect=org.hibernate.dialect.Oracle12cDialect
spring.datasource.url=jdbc:Oracle:thin:@//<host>:<port>/<service name>
spring.datasource.username=<username>
spring.datasource.password=<password>

server.port=9999

#disable schema generation from Hibernate
spring.jpa.hibernate.ddl-auto=none

Creating database configuration class

Below is the configuration class that will be used to define various database related beans such as DataSource, EntityManagerFactory etc.

As we know JPA is a specification or Interface and someone has to provide its own implementation, so here we are using Hibernate as an implementation of JPA API.

As we have application.properties file in classpath, so we don’t need to load the properties file.

We have let Spring know where our Spring Data JPA Repository interfaces using the annotation @EnableJpaRepositories and we have also let Spring know where to look for Entity classes using the setter method factory.setPackagesToScan(“com.jeejava.entity”).

By default Spring does not work when you want to insert multiple records or entities using save() method of JpaRepository or CrudRepository and that’s why you need to set few properties into

LocalContainerEntityManagerFactoryBean as shown below in entityManagerFactory() method. You can change the batch size, here I have put 500.

By default Spring transaction works out of the box so you may not need to annotate the configuration class with @EnableTransactionManagement.

package com.jeejava.config;

import javax.persistence.EntityManagerFactory;
import javax.sql.DataSource;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.core.env.Environment;
import org.springframework.data.jpa.repository.config.EnableJpaRepositories;
import org.springframework.jdbc.datasource.DriverManagerDataSource;
import org.springframework.orm.jpa.LocalContainerEntityManagerFactoryBean;
import org.springframework.orm.jpa.vendor.Database;
import org.springframework.orm.jpa.vendor.HibernateJpaVendorAdapter;

@Configuration
@EnableJpaRepositories(basePackages = "com.jeejava.repository")
public class DatabaseConfig {

	@Autowired
	private Environment environment;

	@Bean
	public DataSource dataSource() {
		DriverManagerDataSource ds = new DriverManagerDataSource();
		ds.setDriverClassName(environment.getRequiredProperty("spring.datasource.driverClassName"));
		ds.setUrl(environment.getRequiredProperty("spring.datasource.url"));
		ds.setUsername(environment.getRequiredProperty("spring.datasource.username"));
		ds.setPassword(environment.getRequiredProperty("spring.datasource.password"));
		return ds;
	}

	@Bean
	public EntityManagerFactory entityManagerFactory(DataSource dataSource) {
		HibernateJpaVendorAdapter vendorAdapter = new HibernateJpaVendorAdapter();
		vendorAdapter.setDatabase(Database.ORACLE);
		LocalContainerEntityManagerFactoryBean factory = new LocalContainerEntityManagerFactoryBean();
		
		//Use these properties to let spring work on batch insertion
		Properties jpaProperties = new Properties();
		jpaProperties.put("hibernate.jdbc.batch_size", 500);
		jpaProperties.put("hibernate.order_inserts", true);
		jpaProperties.put("hibernate.order_updates", true);
		lemfb.setJpaProperties(jpaProperties);
		
		factory.setJpaVendorAdapter(vendorAdapter);
		factory.setPackagesToScan("com.jeejava.entity");
		factory.setDataSource(dataSource);
		factory.afterPropertiesSet();
		return factory.getObject();
	}
}

Creating entity class

This is the entity class that maps Java object to database table. This entity class represents a single row in database table. When you want to save multiple rows in database table then you pass a list of entity objects to JpaRepository or CrudRepository’s save() method in order to save multiple entities or objects and this basically happens through Spring Data JPA Batch Insertion configuration. We also save single object using the same save() method.

package com.jeejava.entity;

import java.io.Serializable;
import javax.persistence.Column;
import javax.persistence.Entity;
import javax.persistence.Id;
import javax.persistence.Table;

@Entity
@Table(name = "EMPLOYEE")
public class Employee implements Serializable {
	
	private static final long serialVersionUID = 1L;
	
	@Id
	@Column(name = "EMPLOYEE_ID")
	private Integer empId;
	
	@Column(name = "EMPLOYEE_NAME")
	private String empName;
	//getters and setters
}

Spring Data JPA Repository

Here is the Spring Data JPA Repository interface. Here JpaRepository takes two parameters Employee object, i.e., entity object and primary key, i.e., Integer. You may have Long, String or any other class object as a primary key as well.

package com.jeejava.repository;

import org.springframework.data.jpa.repository.JpaRepository;
import com.jeejava.entity.Employee;

public interface EmployeeRepository extends JpaRepository<Employee, Integer> {
}

Creating Service class

This is the service class that interacts with data layer as well as controller layer and acts as a mediator between them. This class generally handles all business logic.

In this service we will see the example on Spring Data JPA Batch Insertion. Here notice how I am determining when to insert into database.

Here we iterate through list of employee objects and add to temporary Employee array list. Once we find counter equals to batch size(500) then we save those entity objects and at the same time we also clear the temp list because we don’t need those records in temp list any more.

Notice we have used @Transactional annotation in order to support Spring’s transaction management to rollback database insertion at any point of failures.

package com.jeejava.service;

import java.util.List;
import javax.annotation.Resource;
import org.springframework.stereotype.Service;
import com.jeejava.entity.Employee;
import com.jeejava.repository.EmployeeRepository;

@Service
public class EmployeeService {
	@Resource
	private EmployeeRepository employeeRepository;
	@Transactional
	public void saveEmployees(List<Employee> employees) {
		int size = employees.size();
		int counter = 0;
		
		List<Employee> temp = new ArrayList<>();
		
		for (Employee emp : employees) {
			temp.add(emp);
			
			if ((counter + 1) % 500 == 0 || (counter + 1) == size) {
				employeeRepository.save(temp);
				temp.clear();
			}
			counter++;
		}
	}
}

Spring REST Controller

The Spring REST Controller class is resposible for handling requests and responses from clients. This holds all the REST services end-points. Using these end-points we would be able to get the JSON response.

Here we have only one end-point called /employees/save that saves a list of employees into database when you hit the URL http://localhost:9999/employees/save from REST client or Postman with a list of employee objects in JSON format as a body parameter.

package com.jeejava.controller;

import java.util.List;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.http.HttpStatus;
import org.springframework.http.ResponseEntity;
import org.springframework.web.bind.annotation.GetMapping;
import org.springframework.web.bind.annotation.RestController;
import com.jeejava.entity.Employee;
import com.jeejava.service.EmployeeService;

@RestController
public class EmployeeRestController {

	@Autowired
	private EmployeeService employeeService;

	@PostMapping("/employees/save")
	public ResponseEntity<Void> saveEmployees(@RequestBody List<Employee> employees) {
		employeeService.saveEmployees(employees);
		return new ResponseEntity<Void>(HttpStatus.OK);
	}
}

Here is the application main class that is enough to start up the application in Spring Boot.

package com.jeejava.application;

import org.springframework.boot.SpringApplication;
import org.springframework.boot.autoconfigure.SpringBootApplication;

@SpringBootApplication(scanBasePackages = "com.jeejava")
public class Application {
	
	public static void main(String[] args) {
		SpringApplication.run(Application.class, args);
	}
}

Once you run the above main class and application gets start up, hit the URL http://localhost:9999/employees/save from REST client or Postman with list of employee objects as json and you should get the JSON response with OK.