Spring Data R2DBC保存含空Employee列表的Department报错解决方案咨询

保存带有空员工列表的Department对象时，抛出以下错误：

java.lang.IllegalArgumentException: Unsupported array type: io.spring.graphql.employeedemo2.model.Employee

代码片段 #

Department.class

@Data
@NoArgsConstructor
@Table(name = "department_tbl")
public class Department {

    @Id
    @Column("department_name")
    private int id;

    @Column("department_name")
    private String name;

    private List<Employee> employees;

    public Department(int id, String name, List<Employee> employees){
        this.id = id;
        this.name = name;
        this.employees = new ArrayList<>();
    }
}

AddDepartmentInput.class

@Data
@NoArgsConstructor
@AllArgsConstructor
public class AddDepartmentInput {

    private int id;
    private String name;

}

GraphQL Mutation方法

@MutationMapping
public Mono<Department> createDepartment(@Argument AddDepartmentInput departmentInput) {
    return Mono.just(departmentInput)
        .map(dptInput -> {
            Department department = new Department();
            department.setId(dptInput.getId());
            department.setName(dptInput.getName());
            department.setEmployees(new ArrayList<>()); // 空列表初始化
            return department;
        })
        .flatMap(department -> departmentRepository.save(department));
}

schema.graphqls

type Employee {
    id: ID!
    name: String
    salary: Float
    departmentId: ID!
}

type Department {
    id: ID!
    name: String
    employees: [Employee]
}

type Mutation {
    createDepartment(departmentInput: AddDepartmentInput ):Department
}

input AddDepartmentInput {
    id:ID!
    name:String
}

背景 #

• 使用Spring Data R2DBC进行数据持久化
• 希望允许初始创建无关联员工的部门

核心问题 #

如何通过Spring Data R2DBC有效保存带有空employees列表的Department对象？处理模型中的空列表有哪些推荐方案？

报错原因 #

Spring Data R2DBC会默认把实体类属性映射到数据库表的列，但List<Employee> employees是关联集合，department_tbl表中并没有对应列。当给这个属性赋值为空列表时，R2DBC会尝试将其转换为数据库支持的类型，而Employee是自定义对象无法直接转换，因此抛出类型不支持的错误。

推荐方案 #

1. 用@Transient标记关联集合（最推荐）

给employees属性添加@Transient注解，让Spring Data R2DBC在持久化时忽略该属性：

@Data
@NoArgsConstructor
@Table(name = "department_tbl")
public class Department {

    @Id
    @Column("department_id") // 注意：原代码此处Column注解写错，应对应主键列名而非部门名称列
    private int id;

    @Column("department_name")
    private String name;

    @Transient
    private List<Employee> employees;

    public Department(int id, String name, List<Employee> employees){
        this.id = id;
        this.name = name;
        this.employees = new ArrayList<>();
    }
}

之后若要获取部门关联的员工，可在DepartmentRepository中定义自定义查询：

interface DepartmentRepository extends ReactiveCrudRepository<Department, Integer> {
    @Query("SELECT e.* FROM employee_tbl e WHERE e.department_id = :departmentId")
    Flux<Employee> findEmployeesByDepartmentId(Integer departmentId);
}

在GraphQL的字段解析器中，查询Department的employees字段时调用该方法获取关联数据即可。

2. 临时将列表设为null（不推荐）

如果不想修改实体类，可将空列表初始化改为赋值null：

department.setEmployees(null);

这种方式能让R2DBC忽略该属性，但空列表和null的业务语义不同，后续处理员工列表时需要额外判空，不够优雅。

3. 配置R2DBC关联关系（适合复杂场景）

若需要维护双向关联，可使用Spring Data R2DBC的关联注解：

• 在Employee类中添加@ManyToOne关联Department：

@Data
@NoArgsConstructor
@Table(name = "employee_tbl")
public class Employee {
    @Id
    private int id;
    private String name;
    private Float salary;
    @Column("department_id")
    private Integer departmentId;

    @ManyToOne
    @JoinColumn(name = "department_id", insertable = false, updatable = false)
    private Department department;
}

• 在Department类中添加@OneToMany关联Employee：

@OneToMany(mappedBy = "department")
private List<Employee> employees;

注意：Spring Data R2DBC的关联支持相对有限，@OneToMany默认懒加载，且依赖正确的表结构，适合复杂业务场景，简单场景优先选方案1。

内容的提问来源于stack exchange，提问作者Thilak Sparrow