Spring data jpa 复杂动态查询、筛选、排序方式总结（结合上一篇博客排除空值，完美！）

四.继承JpaSpecificationExecutor接口进行复杂查询

spring data jpa 通过创建方法名来做查询，只能做简单的查询，那如果我们要做复杂一些的查询呢，多条件分页怎么办，这里，spring data jpa为我们提供了JpaSpecificationExecutor接口，只要简单实现toPredicate方法就可以实现复杂的查询

参考:https://www.cnblogs.com/happyday56/p/4661839.html

1.首先让我们的接口继承于JpaSpecificationExecutor

public interface TaskDao extends JpaSpecificationExecutor<Task>{
}

2.JpaSpecificationExecutor提供了以下接口

public interface JpaSpecificationExecutor<T> {
 
    T findOne(Specification<T> spec);
 
    List<T> findAll(Specification<T> spec);
 
    Page<T> findAll(Specification<T> spec, Pageable pageable);
 
    List<T> findAll(Specification<T> spec, Sort sort);
 
    long count(Specification<T> spec);
}
 
//其中Specification就是需要我们传入查询方法的参数，它是一个接口
 
 
public interface Specification<T> {
 
    Predicate toPredicate(Root<T> root, CriteriaQuery<?> query, CriteriaBuilder cb);
}

提供唯一的一个方法toPredicate，我们只要按照JPA 2.0 criteria api写好查询条件就可以了，关于JPA 2.0 criteria api的介绍和使用，欢迎参考 
http://blog.youkuaiyun.com/dracotianlong/article/details/28445725 

http://developer.51cto.com/art/200911/162722.htm

3.接下来我们在service bean

@Service
public class TaskService {
 
    @Autowired TaskDao taskDao ;
 
 
    /**
     * 复杂查询测试
     * @param page
     * @param size
     * @return
     */
    public Page<Task> findBySepc(int page, int size){
 
        PageRequest pageReq = this.buildPageRequest(page, size);
        Page<Task> tasks = this.taskDao.findAll(new MySpec(), pageReq);
        //传入了new MySpec() 既下面定义的匿名内部类 其中定义了查询条件
        return tasks;
 
    }
 
     /**
      * 建立分页排序请求 
      * @param page
      * @param size
      * @return
      */
     private PageRequest buildPageRequest(int page, int size) {
           Sort sort = new Sort(Direction.DESC,"createTime");
           return new PageRequest(page,size, sort);
     }
 
    /**
     * 建立查询条件
     */
    private class MySpec implements Specification<Task>{
 
        @Override
        public Predicate toPredicate(Root<Task> root, CriteriaQuery<?> query, CriteriaBuilder cb) {
 
     //1.混合条件查询
          Path<String> exp1 = root.get("taskName");
            Path<Date>  exp2 = root.get("createTime");
            Path<String> exp3 = root.get("taskDetail");
            Predicate predicate = cb.and(cb.like(exp1, "%taskName%"),cb.lessThan(exp2, new Date()));
            return cb.or(predicate,cb.equal(exp3, "kkk"));
 
           /* 类似的sql语句为:
            Hibernate: 
                select
                    count(task0_.id) as col_0_0_ 
                from
                    tb_task task0_ 
                where
                    (
                        task0_.task_name like ?
                    ) 
                    and task0_.create_time<? 
                    or task0_.task_detail=?
            */
 
    //2.多表查询
        Join<Task,Project> join = root.join("project", JoinType.INNER);
            Path<String> exp4 = join.get("projectName");
            return cb.like(exp4, "%projectName%");
 
           /* Hibernate: 
            select
                count(task0_.id) as col_0_0_ 
            from
                tb_task task0_ 
            inner join
                tb_project project1_ 
                    on task0_.project_id=project1_.id 
            where
                project1_.project_name like ?*/ 
           return null ;  
        }
    }
}

4.实体类task代码如下

@Entity
@Table(name = "tb_task")
public class Task {
 
    private Long id ;
    private String taskName ;
    private Date createTime ;
    private Project project;
    private String taskDetail ;
 
    @Id
    @GeneratedValue(strategy = GenerationType.IDENTITY)
    public Long getId() {
        return id;
    }
    public void setId(Long id) {
        this.id = id;
    }
 
    @Column(name = "task_name")
    public String getTaskName() {
        return taskName;
    }
    public void setTaskName(String taskName) {
        this.taskName = taskName;
    }
 
    @Column(name = "create_time")
    @DateTimeFormat(pattern = "yyyy-MM-dd hh:mm:ss")
    public Date getCreateTime() {
        return createTime;
    }
    public void setCreateTime(Date createTime) {
        this.createTime = createTime;
    }
 
 
    @Column(name = "task_detail")
    public String getTaskDetail() {
        return taskDetail;
    }
    public void setTaskDetail(String taskDetail) {
        this.taskDetail = taskDetail;
    }
 
    @ManyToOne(fetch = FetchType.LAZY)
    @JoinColumn(name = "project_id")
    public Project getProject() {
        return project;
    }
    public void setProject(Project project) {
        this.project = project;
    }
 
}

通过重写toPredicate方法，返回一个查询 Predicate，spring data jpa会帮我们进行查询。

也许你觉得，每次都要写一个类来实现Specification很麻烦，那或许你可以这么写

public class TaskSpec {
 
    public static Specification<Task> method1(){
 
        return new Specification<Task>(){
            @Override
            public Predicate toPredicate(Root<Task> root, CriteriaQuery<?> query, CriteriaBuilder cb) {
                return null;
            }
 
        };
    }
 
    public static Specification<Task> method2(){
 
        return new Specification<Task>(){
            @Override
            public Predicate toPredicate(Root<Task> root, CriteriaQuery<?> query, CriteriaBuilder cb) {
                return null;
            }
 
        };
    }
 
}

那么用的时候，我们就这么用

Page<Task> tasks = this.taskDao.findAll(TaskSpec.method1(), pageReq);

 

最后，结合上一篇博客排除空值：https://blog.youkuaiyun.com/HD243608836/article/details/104562241

 

 

~~~~~~~~~~~~~~~~~~~~~~~~~~~

节选自：https://blog.youkuaiyun.com/qq_30054997/article/details/79420141