/gs-batch-processing

Creating a Batch Service :: Learn how to create a basic batch-driven solution.

Primary LanguageJava

tags projects
spring-batch

This guide walks you through the process of creating a basic batch-driven solution.

What you’ll build

You’ll build a service that imports data from a CSV spreadsheet, transforms it with custom code, and stores the final results in a database.

What you’ll need

Typically your customer or a business analyst supplies a spreadsheet. In this case, you make it up.

src/main/resources/sample-data.csv

link:initial/src/main/resources/sample-data.csv[role=include]

This spreadsheet contains a first name and a last name on each row, separated by a comma. This is a fairly common pattern that Spring handles out-of-the-box, as you will see.

Next, you write a SQL script to create a table to store the data.

src/main/resources/schema-all.sql

link:initial/src/main/resources/schema-all.sql[role=include]
Note
Spring Boot runs schema-@@platform@@.sql automatically during startup. -all is the default for all platforms.

Create a business class

Now that you see the format of data inputs and outputs, you write code to represent a row of data.

src/main/java/hello/Person.java

link:complete/src/main/java/hello/Person.java[role=include]

You can instantiate the Person class either with first and last name through a constructor, or by setting the properties.

Create an intermediate processor

A common paradigm in batch processing is to ingest data, transform it, and then pipe it out somewhere else. Here you write a simple transformer that converts the names to uppercase.

src/main/java/hello/PersonItemProcessor.java

link:complete/src/main/java/hello/PersonItemProcessor.java[role=include]

PersonItemProcessor implements Spring Batch’s ItemProcessor interface. This makes it easy to wire the code into a batch job that you define further down in this guide. According to the interface, you receive an incoming Person object, after which you transform it to an upper-cased Person.

Note
There is no requirement that the input and output types be the same. In fact, after one source of data is read, sometimes the application’s data flow needs a different data type.

Put together a batch job

Now you put together the actual batch job. Spring Batch provides many utility classes that reduce the need to write custom code. Instead, you can focus on the business logic.

src/main/java/hello/BatchConfiguration.java

link:complete/src/main/java/hello/BatchConfiguration.java[role=include]

For starters, the @EnableBatchProcessing annotation adds many critical beans that support jobs and saves you a lot of leg work. This example uses a memory-based database (provided by @EnableBatchProcessing), meaning that when it’s done, the data is gone.

Break it down:

src/main/java/hello/BatchConfiguration.java

link:/complete/src/main/java/hello/BatchConfiguration.java[role=include]

. The first chunk of code defines the input, processor, and output. - reader() creates an ItemReader. It looks for a file called sample-data.csv and parses each line item with enough information to turn it into a Person. - processor() creates an instance of our PersonItemProcessor you defined earlier, meant to uppercase the data. - write(DataSource) creates an ItemWriter. This one is aimed at a JDBC destination and automatically gets a copy of the dataSource created by @EnableBatchProcessing. It includes the SQL statement needed to insert a single Person driven by Java bean properties.

The next chunk focuses on the actual job configuration.

src/main/java/hello/BatchConfiguration.java

link:/complete/src/main/java/hello/BatchConfiguration.java[role=include]

. The first method defines the job and the second one defines a single step. Jobs are built from steps, where each step can involve a reader, a processor, and a writer.

In this job definition, you need an incrementer because jobs use a database to maintain execution state. You then list each step, of which this job has only one step. The job ends, and the Java API produces a perfectly configured job.

In the step definition, you define how much data to write at a time. In this case, it writes up to ten records at a time. Next, you configure the reader, processor, and writer using the injected bits from earlier.

Note
chunk() is prefixed <Person,Person> because it’s a generic method. This represents the input and output types of each "chunk" of processing, and lines up with ItemReader<Person> and ItemWriter<Person>.

Make the application executable

Although batch processing can be embedded in web apps and WAR files, the simpler approach demonstrated below creates a standalone application. You package everything in a single, executable JAR file, driven by a good old Java main() method.

src/main/java/hello/Application.java

link:complete/src/main/java/hello/Application.java[role=include]

The main() method defers to the SpringApplication helper class, providing Application.class as an argument to its run() method. This tells Spring to read the annotation metadata from Application and to manage it as a component in the Spring application context.

The @ComponentScan annotation tells Spring to search recursively through the hello package and its children for classes marked directly or indirectly with Spring’s @Component annotation. This directive ensures that Spring finds and registers BatchConfiguration, because it is marked with @Configuration, which in turn is a kind of @Component annotation.

The @EnableAutoConfiguration annotation switches on reasonable default behaviors based on the content of your classpath. For example, it looks for any class that implements the CommandLineRunner interface and invokes its run() method. In this case, it runs the demo code for this guide.

For demonstration purposes, there is code to create a JdbcTemplate, query the database, and print out the names of people the batch job inserts.

The job prints out a line for each person that gets transformed. After the job runs, you can also see the output from querying the database.

Converting (firstName: Jill, lastName: Doe) into (firstName: JILL, lastName: DOE)
Converting (firstName: Joe, lastName: Doe) into (firstName: JOE, lastName: DOE)
Converting (firstName: Justin, lastName: Doe) into (firstName: JUSTIN, lastName: DOE)
Converting (firstName: Jane, lastName: Doe) into (firstName: JANE, lastName: DOE)
Converting (firstName: John, lastName: Doe) into (firstName: JOHN, lastName: DOE)
Found <firstName: JILL, lastName: DOE> in the database.
Found <firstName: JOE, lastName: DOE> in the database.
Found <firstName: JUSTIN, lastName: DOE> in the database.
Found <firstName: JANE, lastName: DOE> in the database.
Found <firstName: JOHN, lastName: DOE> in the database.

Summary

Congratulations! You built a batch job that ingested data from a spreadsheet, processed it, and wrote it to a database.