HQL的语言

HQL: 是Hibernate Query Language的缩写

1、HQL查询

特点：

与SQL相似，SQL中的语法基本上都可以直接使用
SQL查询的是表和表中的列而HQL查询的是对象或者对象中的属性
HQL的关键字不区分大小写，但是类名与属性名是区分大小写的
当查询的是类对象时，SELECT可以省略.或者使用别名查询

1)简单的查询

hql = "FROM Employee";
hql = "FROM Employee AS e"; // 使用别名
hql = "FROM Employee e"; // 使用别名，as关键字可省略

2)带上过滤条件的（可以使用别名）：Where

hql = "FROM Employee WHERE id<10";
hql = "FROM Employee e WHERE e.id<10";
hql = "FROM Employee e WHERE e.id<10 AND e.id>5";

3)带上排序条件的：Order By

hql = "FROM Employee e WHERE e.id<10 ORDER BY e.name";
hql = "FROM Employee e WHERE e.id<10 ORDER BY e.name DESC";
hql = "FROM Employee e WHERE e.id<10 ORDER BY e.name DESC, id ASC";

4)指定select子句（不可以使用select *）

hql = "SELECT e FROM Employee e"; // 相当于"FROM Employee e" 
hql = "SELECT e.name FROM Employee e"; // 只查询一个列，返回的集合的元素类型就是这个属性的类型
hql = "SELECT e.id,e.name FROM Employee e"; // 查询多个列，返回的集合的元素类型是Object数组
hql = "SELECT new Employee(e.id,e.name) FROM Employee e"; // 可以使用new语法，指定把查询出的部分属性封装到对象中

5)执行查询，获得结果（list、uniqueResult、分页）

Query query = session.createQuery("FROM Employee e WHERE id<3");
query.setFirstResult(0);
query.setMaxResults(10);
List list = query.list(); // 查询的结果是一个List集合
Employee employee = (Employee) query.uniqueResult();// 查询的结果是唯一的一个结果，当结果有多个，就会抛异常

2、其他操作

1）聚集函数：count(), max(), min(), avg(), sum()

hql = "SELECT COUNT(*) FROM Employee"; // 返回的结果是Long型的
hql = "SELECT min(id) FROM Employee"; // 返回的结果是id属性的类型

2)分组: Group By ... Having

hql = "SELECT e.name,COUNT(e.id) FROM Employee e GROUP BY e.name";
hql = "SELECT e.name,COUNT(e.id) FROM Employee e GROUP BY e.name HAVING count(e.id)>1";
hql = "SELECT e.name,COUNT(e.id) FROM Employee e WHERE id<9 GROUP BY e.name HAVING count(e.id)>1";

hql = "SELECT e.name,COUNT(e.id) " + //
 "FROM Employee e " + //
 "WHERE id<9 " + //
 "GROUP BY e.name " + //
"HAVING count(e.id)>1 " + //
"ORDER BY count(e.id) ASC";

 hql = "SELECT e.name,COUNT(e.id) AS c " + //
"FROM Employee e " + //
 "WHERE id<9 " + //
 "GROUP BY e.name " + //
 "HAVING count(e.id)>1 " + // 在having子句中不能使用列别名
"ORDER BY c ASC"; // 在orderby子句中可以使用列别名

3)连接查询 / HQL是面向对象的查询

// >> 内连接（inner关键字可以省略）
hql = "SELECT e.id,e.name,d.name FROM Employee e JOIN e.department d";
hql = "SELECT e.id,e.name,d.name FROM Employee e INNER JOIN e.department d";


// >> 左外连接（outer关键字可以省略）
 hql = "SELECT e.id,e.name,d.name FROM Employee e LEFT OUTER JOIN e.department d";


// >> 右外连接（outer关键字可以省略）
hql = "SELECT e.id,e.name,d.name FROM Employee e RIGHT JOIN e.department d";


// 可以使用更方便的方法
hql = "SELECT e.id,e.name,e.department.name FROM Employee e";

4)查询时使用参数

/ >> 方式一：使用'?'占位
hql = "FROM Employee e WHERE id BETWEEN ? AND ?";
List list = session.createQuery(hql)//
.setParameter(0, 5)// 设置参数，第1个参数的索引为0。
.setParameter(1, 15)//
.list();

// >> 方式二：使用变量名
 hql = "FROM Employee e WHERE id BETWEEN :idMin AND :idMax";
 List list = session.createQuery(hql)//
.setParameter("idMax", 15)//
.setParameter("idMin", 5)//
.list();

// 当参数是集合时，一定要使用setParameterList()设置参数值
hql = "FROM Employee e WHERE id IN (:ids)";
List list = session.createQuery(hql)//
.setParameterList("ids", new Object[] { 1, 2, 3, 5, 8, 100 })//
.list();

5)使用命名查询,需要在配置文件中配置查询的HQL语句，如下：

<query name="queryByIdRange">
            <![CDATA[FROM Employee e WHERE e.id >= :idMin AND e.id <= :idMax]]>
</query>

Query query = session.getNamedQuery("queryByIdRange"); query.setParameter("idMin", 3);
query.setParameter("idMax", 10);
List list = query.list();

6）update,注意不会通知Session缓存，需要手动刷新session.refresh(employee)

UPDATE Employee e SET e.name=? WHERE id>15

7）DELETE,注意不会通知Session缓存，需要手动刷新session.refresh(employee)

DELETE FROM Employee e WHERE id>15

update和delete调用的是executeUpdate() API

QBC: Query By Criteria是一种更面向对象的查询，很多的条件什么的都是通过方法添加的。