Spring Data JPA复杂动态查询
Spring Data JPA其诸多优点给我们的工作带来了很多便利,但对于接触不久的同学来说,有些情况让我们头疼。一些复杂的查询,比如涉及到聚合函数、动态多条件等,着实有些棘手。在不够了解的情况下,觉得Spring Data JPA在这方面不太人性化,有时候我们干脆使用原生sql粗暴的来解决这类查询问题。但这与Spring Data JPA的初衷是相悖的,在不断的学习中,慢慢发现Spring Data JPA的冰山一角,不禁感叹于它的强大。现将一些所得,与大家分享。
先看示例的数据表:
一张简单的学生信息表:包含住址、姓名、学校、电话、出生日期、分数等信息。
应用场景:查询某几个学校出生日期在某个范围内的学生信息,按学校分组统计,并计算各学校人数、总分数、第一名姓名及分数,另外还涉及到排序和筛选。
实现方式:
EntityManager:EntityManager是JPA中用于增删改查的接口,它的作用相当于一座桥梁,连接内存中的java对象和数据库的数据存储。可以用getCriteriaBuilder()的方式获取CriteriaBuilder对象。
CriteriaBuilder接口:用于构造标准查询、复合条件、表达式、排序等。可以通过createQuery的方式获取CriteriaQuery实例。
CriteriaQuery接口:代表一个specific的顶层查询对象,它包含着查询的各个部分,比如:select 、from、where、group by、order by。
Root接口:代表Criteria查询的根对象,定义了实体类型,能为将来导航获得想要的结果,它与SQL查询中的FROM子句类似。
实现代码:
//通过注解@PersistenceContext注入的方式来获得EntityManager对象
@PersistenceContext
private EntityManager entityManager;
public void multiQueryStudent (StudentParam studentParam) { //studentParam:自定义的查询参数体
List<String> schoolList = studentParam.getSchoolList(); //查询条件:学校List
String startDate = studentParam.getStartDate(); //查询条件:出生日期-起
String endDate = studentParam.getEndDate(); //查询条件:出生日期-止
CriteriaBuilder cb = entityManager.getCriteriaBuilder();
//StudentScoreSum指定了查询结果返回至自定义对象
CriteriaQuery<StudentScoreSum> query = cb.createQuery(StudentScoreSum.class);
Root<StudentEntity> root = query.from(StudentEntity.class);
Path<String> schoolPath = root.get("school");
Path<Integer> scorePath = root.get("score");
Path<String> namePath = root.get("name");
Path<String> birthdayPath = root.get("birthday");
//拼接where条件
List<Predicate> predicateList = new ArrayList<Predicate>();
if (schoolList != null && schoolList.size() > 0) {
CriteriaBuilder.In<String> in = cb.in(schoolPath);
for (String school : schoolList) {
in.value(school);
}
predicateList.add(in);
}
if (startDate != null && !"".equals(startDate)) {
predicateList.add(cb.greaterThan(birthdayPath, startDate));
}
if (endDate != null && !"".equals(endDate)) {
predicateList.add(cb.lessThan(birthdayPath, endDate));
}
Predicate[] predicates = new Predicate[predicateList.size()];
predicates = predicateList.toArray(predicates);
//加上where条件
query.where(predicates);
//指定查询项,select后面的东西
query.multiselect(schoolPath, cb.count(root).as(Integer.class), cb.sum(scorePath), namePath, cb.max(scorePath));
//按学校分组
query.groupBy(schoolPath);
//排序
query.orderBy(cb.desc(cb.max(scorePath)));
//筛选第一名成绩大于80分的
query.having(cb.greaterThan(cb.max(scorePath), 80));
TypedQuery<StudentScoreSum> q = entityManager.createQuery(query);
List<StudentScoreSum> result = q.getResultList();
for (StudentScoreSum studentScoreSum : result) {
//打印查询结果
System.out.println(studentScoreSum.toString());
}
}
自定义的查询参数体:
public class StudentParam {
private String name; //姓名
private String address; //住址
private List<String> schoolList; //学校
private String startDate; //出生日期-起
private String endDate; //出生日期-止
//省略get、set方法
}
自定义的查询结果实体类:
public class StudentScoreSum {
private String school;
private Integer count; //学生人数
private Long scoreSum; //总分数
private String name; //第一名姓名
private Integer score; //第一名分数
//省略了set、get方法
public StudentScoreSum(String school, Integer count, Long scoreSum, String name, Integer score) {
this.school = school;
this.count = count;
this.scoreSum = scoreSum;
this.name = name;
this.score = score;
}
@Override
public String toString() {
return "StudentScoreSum{" +
"school='" + school + '\'' +
", count=" + count +
", scoreSum=" + scoreSum +
", name='" + name + '\'' +
", score=" + score +
'}';
}
}
Entity实体类:
@Entity
@Table(name = "t_student")
public class StudentEntity {
@Id
private String id;
private String name;
private String address;
private String tel;
private String school;
private String birthday;
private Integer score;
//省略set、get方法。
}
查询参数赋值及查询接口调用(本人用的Junit测试方法):
@Test
public void multiQueryStudentTest() {
StudentParam studentParam = new StudentParam();
List<String> schoolList = new ArrayList<String>();
schoolList.add("上大附中");
schoolList.add("实验中学");
schoolList.add("山大附中");
schoolList.add("育英中学");
schoolList.add("西城中学");
schoolList.add("青大附中");
studentParam.setSchoolList(schoolList);
studentParam.setStartDate("2001-01-01");
studentParam.setEndDate("2003-01-01");
studentService.multiQueryStudent(studentParam);
}
执行结果如下:
Hibernate:
select
studentent0_.school as col_0_0_,
cast(count(studentent0_.id) as signed) as col_1_0_,
sum(studentent0_.score) as col_2_0_,
studentent0_.name as col_3_0_,
max(studentent0_.score) as col_4_0_
from
t_student studentent0_
where
(
studentent0_.school in (
? , ? , ? , ? , ? , ?
)
)
and studentent0_.birthday>?
and studentent0_.birthday<?
group by
studentent0_.school
having
max(studentent0_.score)>80
order by
max(studentent0_.score) desc
StudentScoreSum{school='青大附中', count=4, scoreSum=286, name='李四', score=99}
StudentScoreSum{school='西城中学', count=2, scoreSum=196, name='郎芬', score=98}
StudentScoreSum{school='山大附中', count=3, scoreSum=254, name='杜吉祥', score=89}
StudentScoreSum{school='育英中学', count=1, scoreSum=89, name='陆大安', score=89}
StudentScoreSum{school='上大附中', count=1, scoreSum=81, name='李天山', score=81}
Process finished with exit code 0
上半部分是Hibernate输出的自动生成的sql语句,自己不需要写一句sql,都是自动生成的,下半部分是查询结果。越了解,越佩服Spring Data JPA的强大,另外,常用的,像:avg()、exists、distinct、between等等,都可以支持。
最后,希望我的分享能给需要的朋友带来一点参考或帮助,不足之处请斧正。
推荐阅读