有些时候,sql语句where条件中,需要一些安全判断,例如按某一条件查询时如果传入的参数是空,此时查询出的结果很可能是空的,也许我们需要参数为空时,是查出全部的信息。使用Oracle的序列、mysql的函数生成Id。这时我们可以使用动态sql。下文均采用mysql语法和函数(例如字符串链接函数CONCAT)。
selectKey 标签
在insert语句中,在Oracle经常使用序列、在MySQL中使用函数来自动生成插入表的主键,而且需要方法能返回这个生成主键。使用myBatis的selectKey标签可以实现这个效果。下面例子,使用mysql数据库自定义函数nextval('student'),用来生成一个key,并把他设置到传入的实体类中的studentId属性上。所以在执行完此方法后,边可以通过这个实体类获取生成的key。
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
|
<!-- 插入学生 自动主键-->
< insert id = "createStudentAutoKey" parameterType = "liming.student.manager.data.model.StudentEntity" keyProperty = "studentId" >
< selectKey keyProperty = "studentId" resultType = "String" order = "BEFORE" >
select nextval('student')
</ selectKey >
INSERT INTO STUDENT_TBL(STUDENT_ID,
STUDENT_NAME,
STUDENT_SEX,
STUDENT_BIRTHDAY,
STUDENT_PHOTO,
CLASS_ID,
PLACE_ID)
VALUES (#{studentId},
#{studentName},
#{studentSex},
#{studentBirthday},
#{studentPhoto, javaType=byte[], jdbcType=BLOB, typeHandler=org.apache.ibatis.type.BlobTypeHandler},
#{classId},
#{placeId})
</ insert >
|
调用接口方法,和获取自动生成key
1
2
3
4
5
6
7
8
|
StudentEntity entity = new StudentEntity();
entity.setStudentName( "黎明你好" );
entity.setStudentSex( 1 );
entity.setStudentBirthday(DateUtil.parse( "1985-05-28" ));
entity.setClassId( "20000001" );
entity.setPlaceId( "70000001" );
this .dynamicSqlMapper.createStudentAutoKey(entity);
System.out.println( "新增学生ID: " + entity.getStudentId());
|
selectKey语句属性配置细节:
属性 | 描述 | 取值 |
keyProperty | selectKey 语句生成结果需要设置的属性。 | |
resultType | 生成结果类型,MyBatis 允许使用基本的数据类型,包括String 、int类型。 | |
order |
1:BEFORE,会先选择主键,然后设置keyProperty,再执行insert语句; 2:AFTER,就先运行insert 语句再运行selectKey 语句。 |
BEFORE AFTER |
statementType | MyBatis 支持STATEMENT,PREPARED和CALLABLE 的语句形式, 对应Statement ,PreparedStatement 和CallableStatement 响应 |
STATEMENT PREPARED CALLABLE |
if标签
if标签可用在许多类型的sql语句中,我们以查询为例。首先看一个很普通的查询:
1
2
3
4
5
|
<!-- 查询学生list,like姓名 -->
< select id = "getStudentListLikeName" parameterType = "StudentEntity" resultMap = "studentResultMap" >
SELECT * from STUDENT_TBL ST
WHERE ST.STUDENT_NAME LIKE CONCAT(CONCAT('%', #{studentName}),'%')
</ select >
|
但是此时如果studentName或studentSex为null,此语句很可能报错或查询结果为空。此时我们使用if动态sql语句先进行判断,如果值为null或等于空字符串,我们就不进行此条件的判断,增加灵活性。
参数为实体类StudentEntity。将实体类中所有的属性均进行判断,如果不为空则执行判断条件。
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
|
<!-- 2 if(判断参数) - 将实体类不为空的属性作为where条件 -->
< select id = "getStudentList_if" resultMap = "resultMap_studentEntity" parameterType = "liming.student.manager.data.model.StudentEntity" >
SELECT ST.STUDENT_ID,
ST.STUDENT_NAME,
ST.STUDENT_SEX,
ST.STUDENT_BIRTHDAY,
ST.STUDENT_PHOTO,
ST.CLASS_ID,
ST.PLACE_ID
FROM STUDENT_TBL ST
WHERE
< if test = "studentName !=null " >
ST.STUDENT_NAME LIKE CONCAT(CONCAT('%', #{studentName, jdbcType=VARCHAR}),'%')
</ if >
< if test = "studentSex != null and studentSex != '' " >
AND ST.STUDENT_SEX = #{studentSex, jdbcType=INTEGER}
</ if >
< if test = "studentBirthday != null " >
AND ST.STUDENT_BIRTHDAY = #{studentBirthday, jdbcType=DATE}
</ if >
< if test = "classId != null and classId!= '' " >
AND ST.CLASS_ID = #{classId, jdbcType=VARCHAR}
</ if >
< if test = "classEntity != null and classEntity.classId !=null and classEntity.classId !=' ' " >
AND ST.CLASS_ID = #{classEntity.classId, jdbcType=VARCHAR}
</ if >
< if test = "placeId != null and placeId != '' " >
AND ST.PLACE_ID = #{placeId, jdbcType=VARCHAR}
</ if >
< if test = "placeEntity != null and placeEntity.placeId != null and placeEntity.placeId != '' " >
AND ST.PLACE_ID = #{placeEntity.placeId, jdbcType=VARCHAR}
</ if >
< if test = "studentId != null and studentId != '' " >
AND ST.STUDENT_ID = #{studentId, jdbcType=VARCHAR}
</ if >
</ select >
|
使用时比较灵活, new一个这样的实体类,我们需要限制那个条件,只需要附上相应的值就会where这个条件,相反不去赋值就可以不在where中判断。
1
2
3
4
5
6
7
8
9
10
11
12
13
|
public void select_test_2_1() {
StudentEntity entity = new StudentEntity();
entity.setStudentName( "" );
entity.setStudentSex( 1 );
entity.setStudentBirthday(DateUtil.parse( "1985-05-28" ));
entity.setClassId( "20000001" );
//entity.setPlaceId("70000001");
List<StudentEntity> list = this .dynamicSqlMapper.getStudentList_if(entity);
for (StudentEntity e : list) {
System.out.println(e.toString());
}
}
|
if + where 的条件判断
当where中的条件使用的if标签较多时,这样的组合可能会导致错误。我们以在3.1中的查询语句为例子,当java代码按如下方法调用时:
1
2
3
4
5
6
7
8
9
10
|
@Test
public void select_test_2_1() {
StudentEntity entity = new StudentEntity();
entity.setStudentName( null );
entity.setStudentSex( 1 );
List<StudentEntity> list = this .dynamicSqlMapper.getStudentList_if(entity);
for (StudentEntity e : list) {
System.out.println(e.toString());
}
}
|
如果上面例子,参数studentName为null,将不会进行STUDENT_NAME列的判断,则会直接导“WHERE AND”关键字多余的错误SQL。
这时我们可以使用where动态语句来解决。这个“where”标签会知道如果它包含的标签中有返回值的话,它就插入一个‘where'。此外,如果标签返回的内容是以AND 或OR 开头的,则它会剔除掉。
上面例子修改为:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
|
<!-- 3 select - where/if(判断参数) - 将实体类不为空的属性作为where条件 -->
< select id = "getStudentList_whereIf" resultMap = "resultMap_studentEntity" parameterType = "liming.student.manager.data.model.StudentEntity" >
SELECT ST.STUDENT_ID,
ST.STUDENT_NAME,
ST.STUDENT_SEX,
ST.STUDENT_BIRTHDAY,
ST.STUDENT_PHOTO,
ST.CLASS_ID,
ST.PLACE_ID
FROM STUDENT_TBL ST
< where >
< if test = "studentName !=null " >
ST.STUDENT_NAME LIKE CONCAT(CONCAT('%', #{studentName, jdbcType=VARCHAR}),'%')
</ if >
< if test = "studentSex != null and studentSex != '' " >
AND ST.STUDENT_SEX = #{studentSex, jdbcType=INTEGER}
</ if >
< if test = "studentBirthday != null " >
AND ST.STUDENT_BIRTHDAY = #{studentBirthday, jdbcType=DATE}
</ if >
< if test = "classId != null and classId!= '' " >
AND ST.CLASS_ID = #{classId, jdbcType=VARCHAR}
</ if >
< if test = "classEntity != null and classEntity.classId !=null and classEntity.classId !=' ' " >
AND ST.CLASS_ID = #{classEntity.classId, jdbcType=VARCHAR}
</ if >
< if test = "placeId != null and placeId != '' " >
AND ST.PLACE_ID = #{placeId, jdbcType=VARCHAR}
</ if >
< if test = "placeEntity != null and placeEntity.placeId != null and placeEntity.placeId != '' " >
AND ST.PLACE_ID = #{placeEntity.placeId, jdbcType=VARCHAR}
</ if >
< if test = "studentId != null and studentId != '' " >
AND ST.STUDENT_ID = #{studentId, jdbcType=VARCHAR}
</ if >
</ where >
</ select >
|
if + set 的更新语句
当update语句中没有使用if标签时,如果有一个参数为null,都会导致错误。
当在update语句中使用if标签时,如果前面的if没有执行,则或导致逗号多余错误。使用set标签可以将动态的配置SET 关键字,和剔除追加到条件末尾的任何不相关的逗号。
使用if+set标签修改后,如果某项为null则不进行更新,而是保持数据库原值。如下示例:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
|
<!-- 4 if/set(判断参数) - 将实体类不为空的属性更新 -->
< update id = "updateStudent_if_set" parameterType = "liming.student.manager.data.model.StudentEntity" >
UPDATE STUDENT_TBL
< set >
< if test = "studentName != null and studentName != '' " >
STUDENT_TBL.STUDENT_NAME = #{studentName},
</ if >
< if test = "studentSex != null and studentSex != '' " >
STUDENT_TBL.STUDENT_SEX = #{studentSex},
</ if >
< if test = "studentBirthday != null " >
STUDENT_TBL.STUDENT_BIRTHDAY = #{studentBirthday},
</ if >
< if test = "studentPhoto != null " >
STUDENT_TBL.STUDENT_PHOTO = #{studentPhoto, javaType=byte[], jdbcType=BLOB, typeHandler=org.apache.ibatis.type.BlobTypeHandler},
</ if >
< if test = "classId != '' " >
STUDENT_TBL.CLASS_ID = #{classId}
</ if >
< if test = "placeId != '' " >
STUDENT_TBL.PLACE_ID = #{placeId}
</ if >
</ set >
WHERE STUDENT_TBL.STUDENT_ID = #{studentId};
</ update >
|
if + trim代替where/set标签
trim是更灵活的去处多余关键字的标签,他可以实践where和set的效果。
trim代替where
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
|
<!-- 5.1 if/trim代替where(判断参数) - 将实体类不为空的属性作为where条件 -->
< select id = "getStudentList_if_trim" resultMap = "resultMap_studentEntity" >
SELECT ST.STUDENT_ID,
ST.STUDENT_NAME,
ST.STUDENT_SEX,
ST.STUDENT_BIRTHDAY,
ST.STUDENT_PHOTO,
ST.CLASS_ID,
ST.PLACE_ID
FROM STUDENT_TBL ST
< trim prefix = "WHERE" prefixOverrides = "AND|OR" >
< if test = "studentName !=null " >
ST.STUDENT_NAME LIKE CONCAT(CONCAT('%', #{studentName, jdbcType=VARCHAR}),'%')
</ if >
< if test = "studentSex != null and studentSex != '' " >
AND ST.STUDENT_SEX = #{studentSex, jdbcType=INTEGER}
</ if >
< if test = "studentBirthday != null " >
AND ST.STUDENT_BIRTHDAY = #{studentBirthday, jdbcType=DATE}
</ if >
< if test = "classId != null and classId!= '' " >
AND ST.CLASS_ID = #{classId, jdbcType=VARCHAR}
</ if >
< if test = "classEntity != null and classEntity.classId !=null and classEntity.classId !=' ' " >
AND ST.CLASS_ID = #{classEntity.classId, jdbcType=VARCHAR}
</ if >
< if test = "placeId != null and placeId != '' " >
AND ST.PLACE_ID = #{placeId, jdbcType=VARCHAR}
</ if >
< if test = "placeEntity != null and placeEntity.placeId != null and placeEntity.placeId != '' " >
AND ST.PLACE_ID = #{placeEntity.placeId, jdbcType=VARCHAR}
</ if >
< if test = "studentId != null and studentId != '' " >
AND ST.STUDENT_ID = #{studentId, jdbcType=VARCHAR}
</ if >
</ trim >
</ select >
|
trim代替set
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
|
<!-- 5.2 if/trim代替set(判断参数) - 将实体类不为空的属性更新 -->
< update id = "updateStudent_if_trim" parameterType = "liming.student.manager.data.model.StudentEntity" >
UPDATE STUDENT_TBL
< trim prefix = "SET" suffixOverrides = "," >
< if test = "studentName != null and studentName != '' " >
STUDENT_TBL.STUDENT_NAME = #{studentName},
</ if >
< if test = "studentSex != null and studentSex != '' " >
STUDENT_TBL.STUDENT_SEX = #{studentSex},
</ if >
< if test = "studentBirthday != null " >
STUDENT_TBL.STUDENT_BIRTHDAY = #{studentBirthday},
</ if >
< if test = "studentPhoto != null " >
STUDENT_TBL.STUDENT_PHOTO = #{studentPhoto, javaType=byte[], jdbcType=BLOB, typeHandler=org.apache.ibatis.type.BlobTypeHandler},
</ if >
< if test = "classId != '' " >
STUDENT_TBL.CLASS_ID = #{classId},
</ if >
< if test = "placeId != '' " >
STUDENT_TBL.PLACE_ID = #{placeId}
</ if >
</ trim >
WHERE STUDENT_TBL.STUDENT_ID = #{studentId}
</ update >
|
choose (when, otherwise)
有时候我们并不想应用所有的条件,而只是想从多个选项中选择一个。而使用if标签时,只要test中的表达式为true,就会执行if标签中的条件。MyBatis提供了choose 元素。if标签是与(and)的关系,而choose比傲天是或(or)的关系。
choose标签是按顺序判断其内部when标签中的test条件出否成立,如果有一个成立,则choose结束。当choose中所有when的条件都不满则时,则执行otherwise中的sql。类似于Java 的switch 语句,choose为switch,when为case,otherwise则为default。
例如下面例子,同样把所有可以限制的条件都写上,方面使用。choose会从上到下选择一个when标签的test为true的sql执行。安全考虑,我们使用where将choose包起来,放置关键字多于错误。
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
|
<!-- 6 choose(判断参数) - 按顺序将实体类第一个不为空的属性作为where条件 -->
< select id = "getStudentList_choose" resultMap = "resultMap_studentEntity" parameterType = "liming.student.manager.data.model.StudentEntity" >
SELECT ST.STUDENT_ID,
ST.STUDENT_NAME,
ST.STUDENT_SEX,
ST.STUDENT_BIRTHDAY,
ST.STUDENT_PHOTO,
ST.CLASS_ID,
ST.PLACE_ID
FROM STUDENT_TBL ST
< where >
< choose >
< when test = "studentName !=null " >
ST.STUDENT_NAME LIKE CONCAT(CONCAT('%', #{studentName, jdbcType=VARCHAR}),'%')
</ when >
< when test = "studentSex != null and studentSex != '' " >
AND ST.STUDENT_SEX = #{studentSex, jdbcType=INTEGER}
</ when >
< when test = "studentBirthday != null " >
AND ST.STUDENT_BIRTHDAY = #{studentBirthday, jdbcType=DATE}
</ when >
< when test = "classId != null and classId!= '' " >
AND ST.CLASS_ID = #{classId, jdbcType=VARCHAR}
</ when >
< when test = "classEntity != null and classEntity.classId !=null and classEntity.classId !=' ' " >
AND ST.CLASS_ID = #{classEntity.classId, jdbcType=VARCHAR}
</ when >
< when test = "placeId != null and placeId != '' " >
AND ST.PLACE_ID = #{placeId, jdbcType=VARCHAR}
</ when >
< when test = "placeEntity != null and placeEntity.placeId != null and placeEntity.placeId != '' " >
AND ST.PLACE_ID = #{placeEntity.placeId, jdbcType=VARCHAR}
</ when >
< when test = "studentId != null and studentId != '' " >
AND ST.STUDENT_ID = #{studentId, jdbcType=VARCHAR}
</ when >
< otherwise >
</ otherwise >
</ choose >
</ where >
</ select >
|
foreach
对于动态SQL 非常必须的,主是要迭代一个集合,通常是用于IN 条件。List 实例将使用“list”做为键,数组实例以“array” 做为键。
foreach元素是非常强大的,它允许你指定一个集合,声明集合项和索引变量,它们可以用在元素体内。它也允许你指定开放和关闭的字符串,在迭代之间放置分隔符。这个元素是很智能的,它不会偶然地附加多余的分隔符。
注意:你可以传递一个List实例或者数组作为参数对象传给MyBatis。当你这么做的时候,MyBatis会自动将它包装在一个Map中,用名称在作为键。List实例将会以“list”作为键,而数组实例将会以“array”作为键。
这个部分是对关于XML配置文件和XML映射文件的而讨论的。下一部分将详细讨论Java API,所以你可以得到你已经创建的最有效的映射。
1.参数为array示例的写法
接口的方法声明:
1
|
public List<StudentEntity> getStudentListByClassIds_foreach_array(String[] classIds);
|
动态SQL语句:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
|
<!— 7.1 foreach(循环array参数) - 作为where中in的条件 -->
< select id = "getStudentListByClassIds_foreach_array" resultMap = "resultMap_studentEntity" >
SELECT ST.STUDENT_ID,
ST.STUDENT_NAME,
ST.STUDENT_SEX,
ST.STUDENT_BIRTHDAY,
ST.STUDENT_PHOTO,
ST.CLASS_ID,
ST.PLACE_ID
FROM STUDENT_TBL ST
WHERE ST.CLASS_ID IN
< foreach collection = "array" item = "classIds" open = "(" separator = "," close = ")" >
#{classIds}
</ foreach >
</ select >
|
测试代码,查询学生中,在20000001、20000002这两个班级的学生:
1
2
3
4
5
6
7
8
|
@Test
public void test7_foreach() {
String[] classIds = { "20000001" , "20000002" };
List<StudentEntity> list = this .dynamicSqlMapper.getStudentListByClassIds_foreach_array(classIds);
for (StudentEntity e : list) {
System.out.println(e.toString());
}
}
|
2.参数为list示例的写法
接口的方法声明:
1
|
public List<StudentEntity> getStudentListByClassIds_foreach_list(List<String> classIdList);
|
动态SQL语句:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
|
<!-- 7.2 foreach(循环List<String>参数) - 作为where中in的条件 -->
< select id = "getStudentListByClassIds_foreach_list" resultMap = "resultMap_studentEntity" >
SELECT ST.STUDENT_ID,
ST.STUDENT_NAME,
ST.STUDENT_SEX,
ST.STUDENT_BIRTHDAY,
ST.STUDENT_PHOTO,
ST.CLASS_ID,
ST.PLACE_ID
FROM STUDENT_TBL ST
WHERE ST.CLASS_ID IN
< foreach collection = "list" item = "classIdList" open = "(" separator = "," close = ")" >
#{classIdList}
</ foreach >
</ select >
|
测试代码,查询学生中,在20000001、20000002这两个班级的学生:
1
2
3
4
5
6
7
8
9
10
|
@Test
public void test7_2_foreach() {
ArrayList<String> classIdList = new ArrayList<String>();
classIdList.add( "20000001" );
classIdList.add( "20000002" );
List<StudentEntity> list = this .dynamicSqlMapper.getStudentListByClassIds_foreach_list(classIdList);
for (StudentEntity e : list) {
System.out.println(e.toString());
}
}
|
3.自己把参数封装成Map的类型
1
2
3
4
5
6
|
< select id = "dynamicForeach3Test" resultType = "Blog" >
select * from t_blog where title like "%"#{title}"%" and id in
< foreach collection = "ids" index = "index" item = "item" open = "(" separator = "," close = ")" >
#{item}
</ foreach >
</ select >
|
上述collection的值为ids,是传入的参数Map的key,对应的Mapper代码:
1
|
public List<Blog> dynamicForeach3Test(Map<String, Object> params);
|
对应测试代码:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
|
@Test
public void dynamicForeach3Test() {
SqlSession session = Util.getSqlSessionFactory().openSession();
BlogMapper blogMapper = session.getMapper(BlogMapper. class );
final List<Integer> ids = new ArrayList<Integer>();
ids.add( 1 );
ids.add( 2 );
ids.add( 3 );
ids.add( 6 );
ids.add( 7 );
ids.add( 9 );
Map<String, Object> params = new HashMap<String, Object>();
params.put( "ids" , ids);
params.put( "title" , "中国" );
List<Blog> blogs = blogMapper.dynamicForeach3Test(params);
for (Blog blog : blogs)
System.out.println(blog);
session.close();
}
|