druid多数据源配置+Datasurce动态切换方式

时间:2021-12-07 22:00:53

druid多数据源配置+Datasurce动态切换

AbstractRoutingDataSource 数据源动态切换

spring 使用AbstractRoutingDataSource自定义动态数据源时的事务处理, 需要继承spring的AbstractRoutingDataSource定义自己的动态数据源,可以根据需要动态的切换不同数据库的数据源,使用起来非常方便。

?
1
2
3
4
5
6
7
8
9
10
11
public class ChooseDataSource extends AbstractRoutingDataSource {
 /**
  * 获取与数据源相关的key
  * 此key是Map<String,DataSource> resolvedDataSources 中与数据源绑定的key值
  * 在通过determineTargetDataSource获取目标数据源时使用
  */
 @Override
 protected Object determineCurrentLookupKey() {
  return RouteHolder.getRouteKey();
 }
}

通过容器RouteHolder存储当前线程使用的数据源的key

?
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
/**
 * 保存当前线程数据源的key
 */
public class RouteHolder {
 private static ThreadLocal<String> routeKey = new ThreadLocal<String>();
 
 /**
  * 获取当前线程的数据源路由的key
  * @return
  */
 public static String getRouteKey()
 {
  String key = routeKey.get();
  return key;
 }
 /**
  * 绑定当前线程数据源路由的key
  * 在使用完成之后,必须调用removeRouteKey()方法删除
  * @param key
  */
 public static void  setRouteKey(String key)
 {
  routeKey.set(key);
 }
 
 /**
  * 删除与当前线程绑定的数据源路由的key
  */
 public static void removeRouteKey()
 {
  routeKey.remove();
 }
}

使用spring 的aop编程在业务逻辑方法运行前将当前方法使用数据源的key从业务逻辑方法上自定义注解@DataSource中解析数据源key并添加到RouteHolder中

?
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
/**
 * 执行dao方法之前的切面
 * 获取datasource对象之前往RouteHolder中指定当前线程数据源路由的key
 *
 */
public class DataSourceAspect {
    /**
     * 在dao层方法之前获取datasource对象之前在切面中指定当前线程数据源路由的key
     */
    public void before(JoinPoint point)
    {
        Object target = point.getTarget();
        String method = point.getSignature().getName();
        Class<?>[] classz = target.getClass().getInterfaces();
        Class<?>[] parameterTypes = ((MethodSignature) point.getSignature())
                .getMethod().getParameterTypes();
        try {
            if(classz != null && classz.length > 0) {
            Method m = classz[0].getMethod(method, parameterTypes);
            if (m != null && m.isAnnotationPresent(DataSource.class)) {
                DataSource data = m.getAnnotation(DataSource.class);
                RouteHolder.setRouteKey(data.value());
                }
            }
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
}

解释:

DataSourceAspect 这个切面类,应该针对的被代理类应该是service的实现类(serviceImpl),因为dao层用的mybatis只有一个dao层的接口,所以放在service上做处理比较好。

druid多数据源配置+Datasurce动态切换方式

业务逻辑方法

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
@Named("userService")
public class UserService
{
 @Inject
 private UserDao userDao;
 
 @DataSource("master")
 @Transactional(propagation=Propagation.REQUIRED)
 public void updatePasswd(int userid,String passwd)
 {
  User user = new User();
  user.setUserid(userid);
  user.setPassword(passwd);
  userDao.updatePassword(user);
 }
 @DataSource("slave")
 @Transactional(propagation=Propagation.REQUIRED)
 public User getUser(int userid)
 {
  User user = userDao.getUserById(userid);
  System.out.println("username------:"+user.getUsername());
  return user;
 }
}

注解类

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
import java.lang.annotation.ElementType;
import java.lang.annotation.Target;
import java.lang.annotation.Retention;
import java.lang.annotation.RetentionPolicy;
/**
 * RUNTIME
 * 编译器将把注释记录在类文件中,在运行时 VM 将保留注释,因此可以反射性地读取。
 * @author jiangxm
 *
 */
@Retention(RetentionPolicy.RUNTIME)
@Target(ElementType.METHOD)
public @interface DataSource {
    String value();
}

spring的配置文件

?
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
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
<bean id="dataSource" class="com.westone.datasource.DbRouteDataSource">
  <property name="targetDataSources">
    <map>
       <!-- write -->
       <entry key="master" value-ref="master"></entry>
       <!-- read -->
       <entry key="slave" value-ref="slave"></entry>
    </map>
  </property>
 </bean>
 
 <bean id="master" class="org.apache.commons.dbcp.BasicDataSource"
        <property name="driverClassName" value="${jdbc.driverclass}" /> 
        <property name="url" value="${jdbc.masterurl}" /> 
        <property name="username" value="${jdbc.username}" /> 
        <property name="password" value="${jdbc.password}" /> 
        <property name="maxActive" value="${jdbc.maxActive}"></property
        <property name="maxIdle" value="${jdbc.maxIdle}"></property
        <property name="maxWait" value="${jdbc.maxWait}"></property
    </bean
    
    <bean id="slave" class="org.apache.commons.dbcp.BasicDataSource"
        <property name="driverClassName" value="${jdbc.driverclass}" /> 
        <property name="url" value="${jdbc.slaveurl}" /> 
        <property name="username" value="${jdbc.username}" /> 
        <property name="password" value="${jdbc.password}" /> 
        <property name="maxActive" value="${jdbc.maxActive}"></property
        <property name="maxIdle" value="${jdbc.maxIdle}"></property
        <property name="maxWait" value="${jdbc.maxWait}"></property
    </bean>   
      
    <bean id="sqlSessionTemplate" class="org.mybatis.spring.SqlSessionTemplate"
        <constructor-arg index="0" ref="sqlSessionFactory" />
    </bean>
    <bean id="sqlSessionFactory" class="org.mybatis.spring.SqlSessionFactoryBean"
        <property name="configLocation" value="classpath:config/mybatis/mybatis.cfg.xml"></property
        <property name="dataSource" ref="dataSource" /> 
    </bean>
    
    <!--  配置mapper的映射扫描器 根据包中定义的接口自动生成dao的实现类-->
    <bean class="org.mybatis.spring.mapper.MapperScannerConfigurer">
      <property name="basePackage" value="com.westone.dao"></property>
    </bean>  
   
   <!-- 为业务逻辑层的方法解析@DataSource注解  为当前线程的routeholder注入数据源key -->
    <bean id="aspectBean" class="com.westone.datasource.aspect.DataSourceAspect"></bean>
    
    <aop:config>
     <aop:aspect id="dataSourceAspect" ref="aspectBean">
         <aop:pointcut id="dataSourcePoint" expression="execution(public * com.westone.service.*.*(..))" />
         <aop:before method="beforeDaoMethod" pointcut-ref="dataSourcePoint"/>
     </aop:aspect>
    </aop:config>  
   
    <!-- 事务管理器配置 -->
    <bean id="transactionManager" class="org.springframework.jdbc.datasource.DataSourceTransactionManager"
        <property name="dataSource" ref="dataSource" /> 
    </bean>
 <!-- 开启事务注解驱动    在业务逻辑层上使用@Transactional 注解 为业务逻辑层管理事务-->
    <tx:annotation-driven  transaction-manager="transactionManager"/>

事务管理配置一定要配置在往RouteHolder中注入数据源key之前 否则会报

Could not open JDBC Connection for transaction; nested exception is java.lang.IllegalStateException: Cannot determine target DataSource for lookup key [null] 找不到数据源错误。

由此就可以根据方法上的@DataSource(“master”) 注解配置不同的数据源key 使用动态数据源。

解释:

?
1
java.lang.reflect.Method.getAnnotation(Class annotationClass)

参数:

annotationClass - Class对象对相应的注解类型,比如Datasource.class 。

返回值:

如果存在于此元素,则返回该元素注解指定的注解对象,否则返回为null

例子

?
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
import java.lang.annotation.Annotation;
import java.lang.annotation.Retention;
import java.lang.annotation.RetentionPolicy;
import java.lang.reflect.Method;
public class MethodDemo {
   public static void main(String[] args) {
      Method[] methods = SampleClass.class.getMethods();
      Annotation annotation = methods[0].getAnnotation(CustomAnnotation.class);
      if(annotation instanceof CustomAnnotation){
         CustomAnnotation customAnnotation = (CustomAnnotation) annotation;
         System.out.println("name: " + customAnnotation.name());
         System.out.println("value: " + customAnnotation.value());
      }
   }
}
@CustomAnnotation(name="SampleClass",  value = "Sample Class Annotation")
class SampleClass {
   private String sampleField;
   @CustomAnnotation(name="getSampleMethod",  value = "Sample Method Annotation")
   public String getSampleField() {
      return sampleField;
   }
   public void setSampleField(String sampleField) {
      this.sampleField = sampleField;
   }
}
@Retention(RetentionPolicy.RUNTIME)
@interface CustomAnnotation {
   public String name();
   public String value();
}

编译并运行上面的程序,这将产生以下结果

-name: getSampleMethod

value: Sample Method Annotation

?
1
getInterfaces()

能够获得这个对象所实现的接口

配置多数据源并实现Druid自动切换

Spring Boot配置多数据源

配置yml文件

这里并没有对spring.datasource配置数据源,因为增加新数据源后,系统会覆盖由spring.datasource自动配置的内容。

这里自定义了两个数据源spring.datasource.cmmi和spring.datasource.zentao

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
spring:
  datasource:
    type: com.alibaba.druid.pool.DruidDataSource
    base:
      type: com.alibaba.druid.pool.DruidDataSource
      driver-class-name: com.mysql.cj.jdbc.Driver
      initialize: true #指定初始化数据源,是否用data.sql来初始化,默认: true
      name: cmmi
      url: jdbc:mysql://127.0.0.1:3306/cmmi?useUnicode=true&characterEncoding=utf-8&useSSL=false&useJDBCCompliantTimezoneShift=true&useLegacyDatetimeCode=false&serverTimezone=UTC&zeroDateTimeBehavior=convertToNull
      username: root
      password: root
    zentao:
      type: com.alibaba.druid.pool.DruidDataSource
      driver-class-name: com.mysql.cj.jdbc.Driver
      initialize: true
      name: zentaopro
      url: jdbc:mysql://127.0.0.1:3306/zentaopro?useUnicode=true&characterEncoding=utf-8&useSSL=false&useJDBCCompliantTimezoneShift=true&useLegacyDatetimeCode=false&serverTimezone=UTC&zeroDateTimeBehavior=convertToNull
      username: root
      password: root

主数据源配置

注意,配置类需要对DataSource、DataSourceTransactionManager、SqlSessionFactory 、SqlSessionTemplate四个数据项进行配置;DataSource类型需要引入javax.sql.DataSource;当系统中有多个数据源时,必须有一个数据源为主数据源,使用@Primary修饰。

@MapperScan对指定dao包建立映射,确保在多个数据源下,自动选择合适的数据源,而在service层里不需要做特殊说明。

?
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
@Configuration
@MapperScan(basePackages = "cmmi.dao.base", sqlSessionTemplateRef = "baseSqlSessionTemplate")
public class BaseDataSourceConfig {
    @Bean(name = "baseDataSource")
    @ConfigurationProperties(prefix = "spring.datasource.base")
    @Primary
    public DataSource setDataSource() {
        return DataSourceBuilder.create().build();
    }
    @Bean(name = "baseTransactionManager")
    @Primary
    public DataSourceTransactionManager setTransactionManager(@Qualifier("baseDataSource") DataSource dataSource) {
        return new DruidDataSource();
    }
    @Bean(name = "baseSqlSessionFactory")
    @Primary
    public SqlSessionFactory setSqlSessionFactory(@Qualifier("baseDataSource") DataSource dataSource) throws Exception {
        SqlSessionFactoryBean bean = new SqlSessionFactoryBean();
        bean.setDataSource(dataSource);
        bean.setMapperLocations(new PathMatchingResourcePatternResolver().getResources("classpath:mapper/base/*.xml"));
        return bean.getObject();
    }
    @Bean(name = "baseSqlSessionTemplate")
    @Primary
    public SqlSessionTemplate setSqlSessionTemplate(@Qualifier("baseSqlSessionFactory") SqlSessionFactory sqlSessionFactory) throws Exception {
        return new SqlSessionTemplate(sqlSessionFactory);
    }
}

从数据源配置

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
@Configuration
@MapperScan(basePackages = "cmmi.dao.zentao", sqlSessionTemplateRef = "zentaoSqlSessionTemplate")
public class ZentaoDataSourceConfig {
    @Bean(name = "zentaoDataSource")
    @ConfigurationProperties(prefix = "spring.datasource.zentao")
    public DataSource setDataSource() {
        return new DruidDataSource();
    }
    @Bean(name = "zentaoTransactionManager")
    public DataSourceTransactionManager setTransactionManager(@Qualifier("zentaoDataSource") DataSource dataSource) {
        return new DataSourceTransactionManager(dataSource);
    }
    @Bean(name = "zentaoSqlSessionFactory")
    public SqlSessionFactory setSqlSessionFactory(@Qualifier("zentaoDataSource") DataSource dataSource) throws Exception {
        SqlSessionFactoryBean bean = new SqlSessionFactoryBean();
        bean.setDataSource(dataSource);
        bean.setMapperLocations(new PathMatchingResourcePatternResolver().getResources("classpath:mapper/zentao/*.xml"));
        return bean.getObject();
    }
    @Bean(name = "zentaoSqlSessionTemplate")
    public SqlSessionTemplate setSqlSessionTemplate(@Qualifier("zentaoSqlSessionFactory") SqlSessionFactory sqlSessionFactory) throws Exception {
        return new SqlSessionTemplate(sqlSessionFactory);
    }
}

使用dao

这里只需要正常使用dao就可以了,spring会根据数据源配置的映射自动选择相应数据源,而不需要在service做特殊说明。

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
@Service
public class TestService {
    private final ZtUserMapper ztUserMapper;
    private final LevelDic levelDic;
    @Autowired
    public TestService(ZtUserMapper ztUserMapper, LevelDic levelDic) {
        this.ztUserMapper = ztUserMapper;
        this.levelDic = levelDic;
    }
    public void test() {
        ztUserMapper.selectByPrimaryKey(1);
        levelDic.setDicId(new Integer(1).byteValue());
    }
}

日志

o.a.c.c.C.[Tomcat].[localhost].[/cmmi] : Initializing Spring FrameworkServlet ‘dispatcherServlet'
o.s.web.servlet.DispatcherServlet : FrameworkServlet ‘dispatcherServlet': initialization started
o.s.web.servlet.DispatcherServlet : FrameworkServlet ‘dispatcherServlet': initialization completed in 23 ms
com.alibaba.druid.pool.DruidDataSource : {dataSource-1,cmmi} inited
com.alibaba.druid.pool.DruidDataSource : {dataSource-2,zentaopro} inited

以上为个人经验,希望能给大家一个参考,也希望大家多多支持服务器之家。

原文链接:https://blog.csdn.net/fragrant_no1/article/details/87803041