背景

最近讓我做一個大數據的系統，分析了一下，麻煩的地方就是多數據源切換抽取數據?？紤]到可以跨服務器跨數據庫抽數，再整理數據，就配置了這個動態數據源的解決方案。在此分享給大家。

實現方案

數據庫配置文件

我們項目使用的是yml形式的配置文件，采用的是hikari的數據庫連接池。第一步我們自然是配置多個數據庫源頭。我們找到spring的datasource，在下方配置三個數據源。

spring: application: name: dynamicDatasource datasource: test1: driver-class-name: com.mysql.jdbc.Driver url: jdbc:mysql://127.0.0.1:3306/test1?serverTimezone=GMT%2B8&characterEncoding=utf-8&useSSL=false username: root password: 123456 test2: driver-class-name: com.mysql.jdbc.Driver url: jdbc:mysql://127.0.0.1:3306/test2?serverTimezone=GMT%2B8&characterEncoding=utf-8&useSSL=false username: root password: 123456 test3: driver-class-name: com.mysql.jdbc.Driver url: jdbc:mysql://127.0.0.1:3306/test3?serverTimezone=GMT%2B8&characterEncoding=utf-8&useSSL=false username: root password: 123456 hikari: leak-detection-threshold: 2000

定義數據源實體類

我們可以建立個datasourceBean文件夾專門管理數據源的實體類。

我們這里要建立三個實體類。分別對應test1，test2，test3

@Configurationpublic class Test1DataSourceBean { @Value('${spring.datasource.test1.driver-class-name}') private String test1Driver; @Value('${spring.datasource.test1.url}') private String test1Url; @Value('${spring.datasource.test1.username}') private String test1Username; @Value('${spring.datasource.test1.password}') private String test1Password; @Bean(name='test1DataSource') public DataSource test1DataSource() throws Exception{ HikariDataSource dataSource = new HikariDataSource(); dataSource.setDriverClassName(test1Driver); dataSource.setJdbcUrl(test1Url); dataSource.setUsername(test1Username); dataSource.setPassword(test1Password); return dataSource; }}

@Configurationpublic class Test2DataSourceBean { @Value('${spring.datasource.test2.driver-class-name}') private String test2Driver; @Value('${spring.datasource.test2.url}') private String test2Url; @Value('${spring.datasource.test2.username}') private String test2Username; @Value('${spring.datasource.test2.password}') private String test2Password; @Bean(name='test2DataSource') public DataSource test2DataSource() throws Exception{ HikariDataSource dataSource = new HikariDataSource(); dataSource.setDriverClassName(test2Driver); dataSource.setJdbcUrl(test2Url); dataSource.setUsername(test2Username); dataSource.setPassword(test2Password); return dataSource; }}

@Configurationpublic class Test3DataSourceBean { @Value('${spring.datasource.test3.driver-class-name}') private String test3Driver; @Value('${spring.datasource.test3.url}') private String test3Url; @Value('${spring.datasource.test3.username}') private String test3Username; @Value('${spring.datasource.test3.password}') private String test3Password; @Bean(name='test3DataSource') public DataSource test3DataSource() throws Exception{ HikariDataSource dataSource = new HikariDataSource(); dataSource.setDriverClassName(test3Driver); dataSource.setJdbcUrl(test3Url); dataSource.setUsername(test3Username); dataSource.setPassword(test3Password); return dataSource; }}

定義一個枚舉類管理數據源

public enum DatabaseType { test1('test1', 'test1'), test2('test2', 'test2'), test3('test3','test3'); private String name; private String value; DatabaseType(String name, String value){ this.name = name; this.value = value; } public String getName(){ return name; } public String getValue(){ return value; }}

定義一個線程安全的數據源容器

public class DatabaseContextHolder { private static final ThreadLocal<DatabaseType> contextHolder = new ThreadLocal<>(); public static void setDatabaseType(DatabaseType type){ contextHolder.set(type); } public static DatabaseType getDatabaseType(){ return contextHolder.get(); }}

定義動態數據源

public class DynamicDataSource extends AbstractRoutingDataSource{ protected Object determineCurrentLookupKey() { return DatabaseContextHolder.getDatabaseType(); }}

mybatis配置類

網上的很多文章配置出來都會產生數據源循環依賴的問題，這里解決了這個問題。

@Configuration@MapperScan(basePackages='cn.test.jichi', sqlSessionFactoryRef='sessionFactory')public class MybatisConfig { /** * @Description:設置動態數據源 */ @Bean(name='dynamicDataSource') @Primary public DynamicDataSource DataSource( @Qualifier('test1DataSource') DataSource test1DataSource, @Qualifier('test2DataSource') DataSource test2DataSource, @Qualifier('test3DataSource') DataSource test3DataSource){ Map<Object, Object> targetDataSource = new HashMap<>(); targetDataSource.put(DatabaseType.test1, test1DataSource); targetDataSource.put(DatabaseType.test2, test2DataSource); targetDataSource.put(DatabaseType.test3, test3DataSource); DynamicDataSource dataSource = new DynamicDataSource(); dataSource.setTargetDataSources(targetDataSource); dataSource.setDefaultTargetDataSource(test1DataSource); return dataSource; } /** * @Description:根據動態數據源創建sessionFactory */ @Bean(name='sessionFactory') public SqlSessionFactory sessionFactory( @Qualifier('test1DataSource') DataSource test1DataSource, @Qualifier('test2DataSource') DataSource test2DataSource, @Qualifier('test3DataSource') DataSource test3DataSource) throws Exception{ SqlSessionFactoryBean sessionFactoryBean = new SqlSessionFactoryBean(); //構造方法，解決動態數據源循環依賴問題。 sessionFactoryBean.setDataSource(this.DataSource(test1DataSource,test2DataSource, test3DataSource)); return sessionFactoryBean.getObject(); }}

示例

public void testDymnaicDatasource(){ //不切換數據源默認是自己的。 System.out.println('-----默認數據源'); DemoEntity totalCount = demoMapper.getTotalCount(); String nameCount1 = totalCount.getNameCount(); String ageCount2 = totalCount.getAgeCount(); System.out.println('nameCount:'+nameCount1); System.out.println('ageCount:'+ageCount2); //數據源切換為branch System.out.println('-----數據源為test2'); DynamicDataSourceUtils.chooseBranchDataSource(); Integer nameCount = demoMapper.getNameCount(); Integer ageCount = demoMapper.getAgeCount(); System.out.println('nameCount:'+nameCount); System.out.println('ageCount:'+ageCount); //數據源為basic System.out.println('-----數據源為test3'); DynamicDataSourceUtils.chooseBasicDataSource(); Integer ageCount1 = demoMapper.getAgeCount(); System.out.println('ageCount:'+ageCount1); }

總結

至此實現了多數據源的動態切換?？梢栽谕粋€方法里面進行操作多個數據源。

以上就是本文的全部內容，希望對大家的學習有所幫助，也希望大家多多支持好吧啦網。