在实际开发过程中，同一个项目很可能需要多个数据源进行支持，所以今天我把自己整合Mybatis多数据源的过程和大家分享下。

添加所需依赖

<!--mysql连接驱动 -->
        <dependency>
            <groupId>mysql</groupId>
            <artifactId>mysql-connector-java</artifactId>
            <scope>runtime</scope>
        </dependency>

        <!--通用mapper -->
        <dependency>
            <groupId>tk.mybatis</groupId>
            <artifactId>mapper-spring-boot-starter</artifactId>
            <version>1.1.0</version>
        </dependency>
        
        <!--druid数据源 -->
        <dependency>
            <groupId>com.alibaba</groupId>
            <artifactId>druid</artifactId>
            <version>1.0.25</version>
        </dependency>

这里为了测试方便引用了Mybatis通用mapper插件，只适合对单表的增删改进行操作，避免了xml的配置和手写sql，特别方便。大家也可以利用常规方式进行测试，这里不做过多解释。在application.properties配置文件进行相关配置。我们配置了两个数据源，一个作为主数据源，另一个作为从数据源。

# master 主数据源配置
master.datasource.url=jdbc:mysql://localhost/master?useUnicode=true&characterEncoding=utf8
master.datasource.username=root
master.datasource.password=root
master.datasource.driverClassName=com.mysql.jdbc.Driver

# cluster 从数据源配置
cluster.datasource.url=jdbc:mysql://localhost/cluster?useUnicode=true&characterEncoding=utf8
cluster.datasource.username=root
cluster.datasource.password=root
cluster.datasource.driverClassName=com.mysql.jdbc.Driver

#连接池的配置信息
## 初始化大小，最小，最大
spring.druid.initialSize=5
spring.druid.minIdle=5
spring.druid.maxActive=20
## 配置获取连接等待超时的时间
spring.druid.maxWait=60000
# 配置间隔多久才进行一次检测，检测需要关闭的空闲连接，单位是毫秒
spring.druid.timeBetweenEvictionRunsMillis=60000
# 配置一个连接在池中最小生存的时间，单位是毫秒
spring.druid.minEvictableIdleTimeMillis=300000
spring.druid.validationQuery=SELECT 1 FROM DUAL
spring.druid.testWhileIdle=true
spring.druid.testOnBorrow=false
spring.druid.testOnReturn=false
spring.druid.poolPreparedStatements=true
spring.druid.maxPoolPreparedStatementPerConnectionSize=20
# 配置监控统计拦截的filters，去掉后监控界面sql无法统计，'wall'用于防火墙
spring.druid.filters=stat,wall,log4j
# 通过connectProperties属性来打开mergeSql功能；慢SQL记录
spring.druid.connectionProperties=druid.stat.mergeSql=true;druid.stat.slowSqlMillis=5000

主数据源的配置文件：

@Configuration
// 扫描 Mapper
@MapperScan(basePackages = MasterDataSourceConfig.MASTER_PACKAGE, sqlSessionFactoryRef = "masterSqlSessionFactory")
public class MasterDataSourceConfig {

    //扫描从数据源mapper接口所在的包
    static final String MASTER_PACKAGE = "cn.jj.mapper.master";
    //扫描从数据源xml文件
    static final String MAPPER_LOCATION = "classpath:mapper/master/*.xml";

    @Value("${master.datasource.url}")
    private String url;

    @Value("${master.datasource.username}")
    private String user;

    @Value("${master.datasource.password}")
    private String password;

    @Value("${master.datasource.driverClassName}")
    private String driverClass;

    @Bean(name = "masterDataSource")
    @Primary
    public DataSource masterDataSource() {
        DruidDataSource dataSource = new DruidDataSource();
        dataSource.setDriverClassName(driverClass);
        dataSource.setUrl(url);
        dataSource.setUsername(user);
        dataSource.setPassword(password);
        return dataSource;
    }

    @Bean(name = "masterTransactionManager")
    @Primary
    public DataSourceTransactionManager masterTransactionManager() {
        return new DataSourceTransactionManager(masterDataSource());
    }

    @Bean(name = "masterSqlSessionFactory")
    @Primary
    public SqlSessionFactory masterSqlSessionFactory(@Qualifier("masterDataSource") DataSource masterDataSource)
            throws Exception {
        final SqlSessionFactoryBean sessionFactory = new SqlSessionFactoryBean();
        sessionFactory.setDataSource(masterDataSource);
        sessionFactory.setMapperLocations(new PathMatchingResourcePatternResolver()
                .getResources(MasterDataSourceConfig.MAPPER_LOCATION));
        return sessionFactory.getObject();
    }
}

从数据源配置文件：

@Configuration
//扫描 Mapper 
@MapperScan(basePackages = ClusterDataSourceConfig.CLASTER_PACKAGE, sqlSessionFactoryRef = "clusterSqlSessionFactory")
public class ClusterDataSourceConfig {

    //扫描从数据源mapper接口所在的包
    static final String CLASTER_PACKAGE = "cn.jj.mapper.cluster";
    //扫描从数据源xml文件
    static final String MAPPER_LOCATION = "classpath:mapper/cluster/*.xml";

    @Value("${cluster.datasource.url}")
    private String url;

    @Value("${cluster.datasource.username}")
    private String user;

    @Value("${cluster.datasource.password}")
    private String password;

    @Value("${cluster.datasource.driverClassName}")
    private String driverClass;

    @Bean(name = "clusterDataSource")
    public DataSource clusterDataSource() {
        DruidDataSource dataSource = new DruidDataSource();
        dataSource.setDriverClassName(driverClass);
        dataSource.setUrl(url);
        dataSource.setUsername(user);
        dataSource.setPassword(password);
        return dataSource;
    }

    @Bean(name = "clusterTransactionManager")
    public DataSourceTransactionManager clusterTransactionManager() {
        return new DataSourceTransactionManager(clusterDataSource());
    }

    @Bean(name = "clusterSqlSessionFactory")
    public SqlSessionFactory clusterSqlSessionFactory(@Qualifier("clusterDataSource") DataSource clusterDataSource)
            throws Exception {
        final SqlSessionFactoryBean sessionFactory = new SqlSessionFactoryBean();
        sessionFactory.setDataSource(clusterDataSource);
        sessionFactory.setMapperLocations(new PathMatchingResourcePatternResolver()
                .getResources(ClusterDataSourceConfig.MAPPER_LOCATION));
        return sessionFactory.getObject();
    }
}

项目结构：

到这里Mybatis多数据源已经配置好，需要注意的是，主，从数据源一定要配置正确，特别是以下位置一定要和工程目录对应正确。

数据库数据信息

主数据库：

从数据库：

编写测试类测试

@RunWith(SpringRunner.class)
@SpringBootTest
public class MapperTest2 {

    @Autowired
    private CityService cityService;

    @Autowired
    private UserService userService;

    // 查询主表
    @Test
    public void selectCity() {
        List<City> city = cityService.selectAll();
        System.out.println("主表信息如下:");
        for (City city2 : city) {
            System.out.println(
                    "id:"+ city2.getId() + "\t" + "name:" + city2.getName() + "\t" + "state:" + city2.getState());
        }
    }

    // 查询从表
    @Test
    public void selectUser() {
        try {
            List<User> user = userService.searchAll();
            System.out.println("从表信息如下:");
            for (User user2 : user) {
                System.out.println("id:"+user2.getId() + "\t" + user2.getGender() + "\t" + user2.getName());
            }
        } catch (Exception e) {
            e.printStackTrace();
        }
    }

}

控制台输出信息如下：

spring boot 整合Mabatis 配置多数据源完成。

（完）

参考文章
Spring Boot整合MyBatis