1.准备两个表
表a:
结构:
mysql> desc a;
+-------+-------------+------+-----+---------+-------+
| Field | Type | Null | Key | Default | Extra |
+-------+-------------+------+-----+---------+-------+
| name | varchar(40) | NO | PRI | NULL | |
| age | int(11) | YES | | NULL | |
+-------+-------------+------+-----+---------+-------+
数据
表b:
结构
mysql> desc b;
+-------+-------------+------+-----+---------+-------+
| Field | Type | Null | Key | Default | Extra |
+-------+-------------+------+-----+---------+-------+
| nameB | varchar(40) | YES | | NULL | |
| ageB | int(11) | YES | | NULL | |
+-------+-------------+------+-----+---------+-------+
数据:
2.进行连接查询测试:
(1)交叉连接(笛卡尔积) cross join
mysql> select * from a,b; #第一种
+------+------+-------+------+
| name | age | nameB | ageB |
+------+------+-------+------+
| A1 | 1 | B1 | 1 |
| A2 | 2 | B1 | 1 |
| A1 | 1 | B2 | 22 |
| A2 | 2 | B2 | 22 |
+------+------+-------+------+
4 rows in set (0.00 sec) mysql> select * from a cross join b; #第二种
+------+------+-------+------+
| name | age | nameB | ageB |
+------+------+-------+------+
| A1 | 1 | B1 | 1 |
| A2 | 2 | B1 | 1 |
| A1 | 1 | B2 | 22 |
| A2 | 2 | B2 | 22 |
+------+------+-------+------+
4 rows in set (0.00 sec) mysql> select a.*,b.* from a cross join b; #第二种的又一个写法
+------+------+-------+------+
| name | age | nameB | ageB |
+------+------+-------+------+
| A1 | 1 | B1 | 1 |
| A2 | 2 | B1 | 1 |
| A1 | 1 | B2 | 22 |
| A2 | 2 | B2 | 22 |
+------+------+-------+------+
4 rows in set (0.00 sec)
(2)内连接 join 或 inner join(在笛卡尔积的基础上过滤)
- 显示内连接
(1)不带条件的内连接
mysql> select a.*,b.* from a inner join b on a.age=b.ageb; #第一种 inner join
+------+------+-------+------+
| name | age | nameB | ageB |
+------+------+-------+------+
| A1 | 1 | B1 | 1 |
+------+------+-------+------+
1 row in set (0.00 sec)
mysql> select a.*,b.* from a join b on a.age=b.ageb; #第二种 join (默认是inner join)
+------+------+-------+------+
| name | age | nameB | ageB |
+------+------+-------+------+
| A1 | 1 | B1 | 1 |
+------+------+-------+------+
1 row in set (0.00 sec)
三个表的显示内连接:
SELECT
a.*,
b.*,
c.*
FROM exampaper a
INNER JOIN bigquestion b
INNER JOIN exampaperquestion c
ON a.paperId = b.paperId
AND b.bigQuertionId = c.bigQuertionId
四个表的显示内连接:
SELECT
train.trainingSchemaName,
train.majorName,
train.createTime,
tc.*, course.*, type.*
FROM
trainschemeinfo train
JOIN train_course tc ON train.trainingSchemeID = tc.trainningSchemeID
INNER JOIN t_course_base_info course ON tc.courseID = course.courseId
INNER JOIN coursetypeinfo type ON tc.typeNum = type.typeNum
WHERE
tc.trainningSchemeID = '661ecb064b164d1ea133956f89beddb7'
与之等价的隐士内连接:
SELECT
train.trainingSchemaName,
train.majorName,
train.createTime,
tc.*, course.*, type.*
FROM
trainschemeinfo train,
train_course tc,
t_course_base_info course,
coursetypeinfo type
WHERE
train.trainingSchemeID = tc.trainningSchemeID
AND tc.courseID = course.courseId
AND tc.typeNum = type.typeNum
AND tc.trainningSchemeID = '661ecb064b164d1ea133956f89beddb7'
(2)显示内连接带条件
mysql> select a.*,b.* from a join b on a.age=b.ageb having a.name='A1'; #having从查出的数据中挑选满足条件的元祖
+------+------+-------+------+
| name | age | nameB | ageB |
+------+------+-------+------+
| A1 | 1 | B1 | 1 |
+------+------+-------+------+
1 row in set (0.00 sec)
mysql> select a.*,b.* from a join b on a.age=b.ageb where a.name='A1'; #where查询满足条件的元素
+------+------+-------+------+
| name | age | nameB | ageB |
+------+------+-------+------+
| A1 | 1 | B1 | 1 |
+------+------+-------+------+
1 row in set (0.00 sec)
- 隐士内连接:
mysql> select * from a,b where a.age=b.ageb;
+------+------+-------+------+
| name | age | nameB | ageB |
+------+------+-------+------+
| A1 | 1 | B1 | 1 |
+------+------+-------+------+
1 row in set (0.00 sec) mysql> select * from a,b where a.age=b.ageb and a.name='A1';
+------+------+-------+------+
| name | age | nameB | ageB |
+------+------+-------+------+
| A1 | 1 | B1 | 1 |
+------+------+-------+------+
1 row in set (0.00 sec) mysql> select * from a,b where a.age=b.ageb having a.name='A1';
+------+------+-------+------+
| name | age | nameB | ageB |
+------+------+-------+------+
| A1 | 1 | B1 | 1 |
+------+------+-------+------+
1 row in set (0.00 sec)
where是从本地磁盘查询满足条件的元素，having是从查出的数据中挑选满足条件的元素。执行权限 where>sum()..聚合函数>having
(3)左外连接:(拿左边的匹配右边的，没有找到右边的为null)
mysql> select * from a left join b on a.age = b.ageb; #第一种 left join
+------+------+-------+------+
| name | age | nameB | ageB |
+------+------+-------+------+
| A1 | 1 | B1 | 1 |
| A2 | 2 | NULL | NULL |
+------+------+-------+------+
2 rows in set (0.00 sec) mysql> select * from a left outer join b on a.age = b.ageb; #第二种 left outer join
+------+------+-------+------+
| name | age | nameB | ageB |
+------+------+-------+------+
| A1 | 1 | B1 | 1 |
| A2 | 2 | NULL | NULL |
+------+------+-------+------+
2 rows in set (0.00 sec)
(4)右外连接:(拿右边的匹配左边的，没有找到左边的为null)
mysql> select * from a right join b on a.age = b.ageb; #第一种 right join
+------+------+-------+------+
| name | age | nameB | ageB |
+------+------+-------+------+
| A1 | 1 | B1 | 1 |
| NULL | NULL | B2 | 22 |
+------+------+-------+------+
2 rows in set (0.00 sec) mysql> select * from a right outer join b on a.age = b.ageb; #第二种 right outer join
+------+------+-------+------+
| name | age | nameB | ageB |
+------+------+-------+------+
| A1 | 1 | B1 | 1 |
| NULL | NULL | B2 | 22 |
+------+------+-------+------+
2 rows in set (0.00 sec)
3.Union 和 union all 取两个表的并集测试
修改b表，加一条和a表重复的数据
b表数据:
a表数据:
(1) union: 自动去掉重复元素
mysql> select * from a union select * from b;
+------+------+
| name | age |
+------+------+
| A1 | 1 |
| A2 | 2 |
| B1 | 1 |
| B2 | 22 |
+------+------+
4 rows in set (0.00 sec)
总结:
union:联合的意思，即把两次或多次查询结果合并起来。
要求：两次查询的列数必须一致
推荐：列的类型可以不一样，但推荐查询的每一列，想对应的类型以一样
可以来自多张表的数据：多次sql语句取出的列名可以不一致，此时以第一个sql语句的列名为准。
如果不同的语句中取出的行，有完全相同(这里表示的是每个列的值都相同)，那么union会将相同的行合并，最终只保留一行。也可以这样理解，union会去掉重复的行。
如果不想去掉重复的行，可以使用union all。
如果子句中有order by,limit，需用括号()包起来。推荐放到所有子句之后，即对最终合并的结果来排序或筛选，可以对union之后的数据进行排序和分页等操作。
例如:采用union合并的多个表的数据的SQL
需求是:为了显示学院、专业、班级树，但是这些信息不在一个月表，而且班级表中有专业编号，专业表中有学院编号。思路就是:分别从三个表中获取数据，然后采用union进行合并数据。
SELECT
classID AS departNum,
className AS departName,
"class" AS departType,
(SELECT
majorID
FROM t_major_base_info
WHERE majorID = class.majorID) AS updepartNum
FROM t_class_base_info class
UNION
SELECT majorID AS departNum,
majorName AS departName,
"major" AS departType,
(SELECT
collegeID FROM t_college_base_info
WHERE collegeID=major.collegeID) AS updepartNum
FROM t_major_base_info major
UNION
SELECT collegeId AS departNum,
collegeName AS departName,
"college" AS departType,
"000" AS updepartNum
FROM t_college_base_info
(2) union all 保留重复元素
UNION ALL 命令和 UNION 命令几乎是等效的，不过 UNION ALL 命令会列出所有的值。
mysql> select * from a union all select * from b;
+------+------+
| name | age |
+------+------+
| A1 | 1 |
| A2 | 2 |
| B1 | 1 |
| B2 | 22 |
| A1 | 1 |
+------+------+
5 rows in set (0.00 sec)
总结:
UNION 用于合并两个或多个 SELECT 语句的结果集，并消去表中任何重复行。
UNION 内部的 SELECT 语句必须拥有相同数量的列，列也必须拥有相似的数据类型。
同时，每条 SELECT 语句中的列的顺序必须相同.
默认地，UNION 操作符选取不同的值。如果允许重复的值，请使用 UNION ALL。
当 ALL 随 UNION 一起使用时（即 UNION ALL），不消除重复行
注意：
1、UNION 结果集中的列名总是等于第一个 SELECT 语句中的列名
2、UNION 内部的 SELECT 语句必须拥有相同数量的列。列也必须拥有相似的数据类型。同时，每条 SELECT 语句中的列的顺序必须相同
补充:今天在项目种用到了从多个表种查询数据把并且分页，实现过程是将所有表的名字传到dao，然后遍历表名字，一起添加条件(前提是需要查询的列名字相同)：
public List<Map<String, Object>> getAllData(List<String> tableNames) {
List<String> sqls = new ArrayList();
StringBuilder sb = null;
for (String tableName : tableNames) {
sb = new StringBuilder();
sb.append("select id,name from ");
sb.append(tableName);
sb.append(" where 1=1");
sb.append(" and name = 'zhangsan'");
sqls.add(sb.toString());
}
String sqlFinally = StringUtils.join(sqls, " union ");
sqlFinally += "order by name limit 5,5";
System.out.println(sqlFinally);
/*Session session = getSessionFactory().openSession();
SQLQuery sqlQuery = session.createSQLQuery(sqlFinally);
sqlQuery.setResultTransformer(Transformers.ALIAS_TO_ENTITY_MAP);
return sqlQuery.list();*/
return null;
}
测试：
public static void main(String[] args) {
GroupDaoImpl g = new GroupDaoImpl();
List tableNames = new ArrayList();
tableNames.add("t1");
tableNames.add("t2");
tableNames.add("t3");
g.getAllData(tableNames);
}
结果:
select id,name from t1 where 1=1 and name = 'zhangsan' union select id,name from t2 where 1=1 and name = 'zhangsan' union select id,name from t3 where 1=1 and name = 'zhangsan' order by name limit 5,5