在 Groovy 中使用 JavaConfig stepScope() 会导致 NullPointerException

Posted 2023-02-27

【中文标题】在 Groovy 中使用 JavaConfig stepScope() 会导致 NullPointerException

【英文标题】：Using JavaConfig stepScope() in Groovy causes NullPointerException

【发布时间】：2018-12-17 07:40:16

【问题描述】：

在找到an accepted answer from a Spring Batch dev here 以及下面附带的JavaConfig 代码之后，我仍然对如何使用stepScope() 感到有些困惑。我试图在下面限定 multiResourceItemReader 的范围，但只是将 stepScope() 的 bean 定义添加到文件的顶部或底部会导致此错误：

Exception encountered during context initialization - cancelling refresh attempt: 
org.springframework.beans.factory.BeanCreationException: Error creating bean with name 'readFiles' defined in class path resource [com/onlinephotosubmission/csvImporter/service/BatchJobService.class]: 
    Bean instantiation via factory method failed; 
nested exception is org.springframework.beans.BeanInstantiationException: 
    Failed to instantiate [java.lang.Object]: Factory method 'readFiles' threw exception; 
nested exception is java.lang.NullPointerException: 
    Cannot invoke method get() on null object

我所知道的是 stepScope() 必须在一个 @Configuration 文件中，除此之外，我完全不知道需要做什么。

---

BatchJobService.groovy

@Configuration
@EnableBatchProcessing
class BatchJobService 

@Autowired
JobBuilderFactory jobBuilderFactory
@Autowired
StepBuilderFactory stepBuilderFactory
@Autowired
JobLauncher jobLauncher
@Autowired
AfterJobListener afterJobListener

//  Set values from properties file

@Value('$input.directory:file:inputs/*.csv')
Resource[] resources

@Value('$report.directory:output')
String reportsDir

@Value('$completed.directory:completed')
String completedDir

@Value('$report.name.prepend:people')
String prependName

@Value('$timestamp.format:dd_MM_yyyy_HH_mm_ss')
String timestampFormat

//  End set properties
@Bean
StepScope stepScope() 
    final StepScope stepScope = new StepScope()
    stepScope.setAutoProxy(true)
    return stepScope


@Bean
Job readFiles() 
    return jobBuilderFactory
            .get("readFiles")
            .incrementer(new RunIdIncrementer())
            .flow(step1())
            .end()
            .listener(afterJobListener)
            .build()


@Bean
Step step1() 
    return stepBuilderFactory
            .get("step1")
    //NOTE: may need to adjust chunk size larger (say 1000 to take all transacions at once)
    // or smaller (say 1 to take each transaction individually).
    // Bigger is usually better, though.
            .<Person, Person>chunk(1000)
            .reader(multiResourceItemReader())
            .processor(modifier())
            .writer(writer())
            .build()


@Bean
MultiResourceItemReader<Person> multiResourceItemReader() 
    MultiResourceItemReader<Person> resourceItemReader = new MultiResourceItemReader<Person>()
    resourceItemReader.setResources(resources)
    resourceItemReader.setDelegate(reader())
    return resourceItemReader


@Bean
FlatFileItemReader<Person> reader() 
    FlatFileItemReader<Person> reader = new FlatFileItemReader<Person>()
    reader.setLinesToSkip(1)    //skips header line
    reader.setLineMapper(new DefaultLineMapper()
    
        setLineTokenizer(new DelimitedLineTokenizer(",")
        
            setNames(["email", "identifier"] as String[])
        )
        setFieldSetMapper(new BeanWrapperFieldSetMapper<Person>()   // BeanWrapperFieldSetMapper maps the line token values to a POJO directly by name
        
            setTargetType(Person.class)
        )
    )

    return reader



@Bean
PersonItemProcessor modifier()
    return new PersonItemProcessor()



@Bean
FlatFileItemWriter<Person> writer() 
    FlatFileItemWriter<Person> writer = new FlatFileItemWriter<>()
    writer.setAppendAllowed(true)
    writer.setResource(new FileSystemResource(reportsDir + "/" + prependName + getTime() + ".csv"))
    writer.setLineAggregator(new DelimitedLineAggregator<Person>()
    
        setDelimiter(",")
        setFieldExtractor(new BeanWrapperFieldExtractor<Person>()
        
            setNames(["status", "email", "identifier"] as String[])
        )
    )
    return writer

【参考方案1】：

@EnableBatchProcessing 自动导入StepScope，因此您无需在应用程序上下文中将其声明为 bean。当 XML 和 Java Config 混合使用时，您链接到的问题就会发生。在您的情况下，我只看到 Java Config，所以问题不应该发生。

我正在尝试确定下面的 multiResourceItemReader 的范围

我只知道 stepScope() 必须在 @Configuration 文件中，除此之外，我完全不知道需要做什么。

仅仅声明step作用域是不够的，还需要在bean定义上加上@StepScope注解。

您可以在此处的参考文档中找到有关 StepScope 的更多详细信息：https://docs.spring.io/spring-batch/4.0.x/reference/html/step.html#step-scope