Provides Spring-batch JobRepository and related metadata persistence in Neo4J.
Currently, the support of NoSQL databases in Spring-Batch is limited to ItemReader/ItemWriter for Neo4j and Mongo.
This means that to persist state of Spring-Batch job you would need to have an RDMBS database even if you don't need it as there is no JobRepository
persistence adapter available.
See these open tickets:
This project aims to solve this problem for Neo4j. It offers ready to use Spring Batch metadata persistence adapter for the Neo4j database. This adapter allows Spring Batch to store its metadata directly in Neo4j so that no RDBMS/SQL database is needed. It is built on top of Spring-Data for entity and repository abstraction and on top of MapStruct to map Spring Batch data structures to Spring-Data Neo4j entities. This way, the amount of boilerplate code to write was kept at the minimum level.
repositories {
jcenter()
}
dependencies {
implementation 'com.github.valb3r.springbatch.adapters:neo4j-adapter:0.1.4'
}
Add jcenter
repository:
<repositories>
<repository>
<snapshots>
<enabled>false</enabled>
</snapshots>
<id>jcenter</id>
<name>bintray</name>
<url>https://jcenter.bintray.com</url>
</repository>
</repositories>
Import neo4j-adapter
:
<dependency>
<groupId>com.github.valb3r.springbatch.adapters</groupId>
<artifactId>neo4j-adapter</artifactId>
<version>0.1.4</version>
</dependency>
See Bintray
repository: https://bintray.com/valb3r/maven/neo4j-adapter.
All examples can be found inside examples folder.
This is the example of how to enable Spring-Batch Neo4j adapter
@EnableBatchProcessing
@SpringBootApplication(
// Spring-Batch includes this by default, disabling them
exclude = {
DataSourceAutoConfiguration.class,
DataSourceTransactionManagerAutoConfiguration.class
}
)
@EnableSpringBatchNeo4jAdapter
public class ExampleApplication {
public static void main(String[] args) {
SpringApplication.run(ExampleApplication.class);
}
}
After you have enabled Neo4j adapter with @EnableSpringBatchNeo4jAdapter
you can use Spring-Batch as if it was
RDBMS database. See this snippet for example:
Example:Execute simple batch job
@Slf4j
@Service
@RequiredArgsConstructor
public class SimpleJobService {
@Getter
private final AtomicReference<String> result = new AtomicReference<>();
private final JobRepository jobRepository;
private final JobBuilderFactory jobBuilderFactory;
private final StepBuilderFactory stepBuilderFactory;
@SneakyThrows
public void runSimpleJob() {
val job = jobBuilderFactory.get("FOO")
.start(stepBuilderFactory.get("ONE").tasklet((a, b) -> {
log.info("STEP ONE!");
return null;
}).build())
.next(stepBuilderFactory.get("TWO").tasklet((a, b) -> {
log.info("STEP TWO!");
result.set("Step TWO DONE");
return null;
}).build())
.build();
val exec = jobRepository.createJobExecution("Test one", new JobParameters());
job.execute(exec);
}
}
You can see how to configure the adapter and how to use it in 'real' project here:
- https://github.com/valb3r/time-tracker/ in particular worker submodule.
The adapter will reuse spring.data.neo4j
database connection properties if you provide them in
i.e. application.yml
.
For example:
spring:
# Run neo4j with docker:
# docker run --rm -d --publish=7474:7474 --publish=7687:7687 --volume=$HOME/neo4j/data:/data -e NEO4J_AUTH=neo4j/docker neo4j/neo4j-experimental:4.0.0-rc01
data:
neo4j:
uri: bolt://localhost:7687
username: neo4j
password: docker
open-in-view: false
use-native-types: true
All repositories are provided by:
@EnableNeo4jRepositories(
basePackages = {
"com.github.valb3r.springbatch.adapters.neo4j.ogm.repository"
}
)
This way, all of them are configured using the same Spring-Data configuration as you have for your own Neo4j-repositories.