1. Spring与MyBatis整合全景解析作为Java生态中最主流的两个框架Spring和MyBatis的组合堪称企业级开发的黄金搭档。我在过去五年中参与过17个采用这种技术栈的项目从电商系统到金融交易平台这套组合拳始终表现出色。不同于Hibernate的全自动ORMMyBatis提供的半自动化SQL控制能力让开发者既能享受对象映射的便利又能精准控制查询逻辑。Spring 5.3版本后对MyBatis的整合支持更加完善通过mybatis-spring-boot-starter组件我们现在可以在10分钟内完成基础集成。但真正要发挥这对组合的威力需要深入理解以下几个核心交互点SqlSessionFactoryBean的生命周期管理Mapper接口的代理机制事务管理器的协同工作类型处理器的扩展方式2. 环境搭建与基础配置2.1 依赖管理关键点使用Gradle构建时除了声明基础依赖还需要特别注意各组件版本匹配dependencies { implementation org.mybatis.spring.boot:mybatis-spring-boot-starter:3.0.3 implementation com.zaxxer:HikariCP:5.1.0 // 推荐连接池 runtimeOnly mysql:mysql-connector-java:8.0.33 annotationProcessor org.projectlombok:lombok:1.18.30 }重要提示MyBatis Starter 3.x版本要求Spring Boot 3.0如果仍在使用Spring Boot 2.x系列需锁定mybatis-spring-boot-starter 2.3.1版本2.2 数据源配置进阶在application.yml中配置Hikari连接池时这些参数对性能影响显著spring: datasource: hikari: maximum-pool-size: 20 minimum-idle: 5 connection-timeout: 30000 idle-timeout: 600000 max-lifetime: 1800000 connection-test-query: SELECT 1实测发现连接池大小设置为CPU核心数的2~3倍时TPS每秒事务数表现最佳。对于8核服务器我通常设置maximum-pool-size为16-24之间。3. MyBatis核心组件深度配置3.1 SqlSessionFactory定制化通过Java Config方式配置SqlSessionFactory时可以注入各种自定义组件Bean public SqlSessionFactory sqlSessionFactory(DataSource dataSource) throws Exception { SqlSessionFactoryBean factoryBean new SqlSessionFactoryBean(); factoryBean.setDataSource(dataSource); // 配置驼峰命名转换 org.apache.ibatis.session.Configuration configuration new Configuration(); configuration.setMapUnderscoreToCamelCase(true); factoryBean.setConfiguration(configuration); // 添加自定义类型处理器 factoryBean.setTypeHandlers(new AddressTypeHandler()); // XML映射文件位置 factoryBean.setMapperLocations(new PathMatchingResourcePatternResolver() .getResources(classpath*:mapper/**/*.xml)); return factoryBean.getObject(); }3.2 动态SQL编写技巧在mapper.xml中编写复杂查询时这些技巧能提升可维护性select idsearchUsers resultTypeUser SELECT * FROM users where if testname ! null and name ! AND name LIKE CONCAT(%, #{name}, %) /if if teststatusList ! null and statusList.size() 0 AND status IN foreach collectionstatusList itemstatus open( separator, close) #{status} /foreach /if if testminCreateTime ! null AND create_time #{minCreateTime} /if /where ORDER BY choose when testorderBy namename/when otherwiseid/otherwise /choose /select避坑指南XML中的特殊符号如, 必须使用转义字符例如表示表示。但在CDATA区块内可以直接使用原符号4. 高级特性实战4.1 插件开发与拦截器实现分页插件示例Intercepts(Signature(type StatementHandler.class, method prepare, args {Connection.class, Integer.class})) public class PaginationInterceptor implements Interceptor { private static final ThreadLocalPageInfo PAGE_INFO new ThreadLocal(); Override public Object intercept(Invocation invocation) throws Throwable { StatementHandler handler (StatementHandler) invocation.getTarget(); MetaObject metaObject SystemMetaObject.forObject(handler); PageInfo pageInfo PAGE_INFO.get(); if (pageInfo ! null) { String originalSql (String) metaObject.getValue(delegate.boundSql.sql); String limitSql originalSql LIMIT pageInfo.getOffset() , pageInfo.getPageSize(); metaObject.setValue(delegate.boundSql.sql, limitSql); } return invocation.proceed(); } public static void startPage(int pageNum, int pageSize) { PAGE_INFO.set(new PageInfo(pageNum, pageSize)); } public static void clearPage() { PAGE_INFO.remove(); } }使用时在Service层调用public ListUser getUsersByPage(int pageNum, int pageSize) { try { PaginationInterceptor.startPage(pageNum, pageSize); return userMapper.selectAll(); } finally { PaginationInterceptor.clearPage(); } }4.2 多数据源动态切换在金融系统中经常需要访问多个数据库通过AbstractRoutingDataSource实现public class DynamicDataSource extends AbstractRoutingDataSource { private static final ThreadLocalString CONTEXT new ThreadLocal(); public static void setDataSource(String name) { CONTEXT.set(name); } public static void clear() { CONTEXT.remove(); } Override protected Object determineCurrentLookupKey() { return CONTEXT.get(); } }配置类中定义多个数据源Configuration public class DataSourceConfig { Bean Primary public DataSource dynamicDataSource( Qualifier(masterDataSource) DataSource master, Qualifier(slaveDataSource) DataSource slave) { MapObject, Object targetDataSources new HashMap(); targetDataSources.put(master, master); targetDataSources.put(slave, slave); DynamicDataSource dataSource new DynamicDataSource(); dataSource.setTargetDataSources(targetDataSources); dataSource.setDefaultTargetDataSource(master); return dataSource; } }5. 性能优化实战5.1 二级缓存配置陷阱在MyBatis配置中启用二级缓存settings setting namecacheEnabled valuetrue/ /settings然后在Mapper接口上添加注解CacheNamespace(implementation MybatisRedisCache.class, eviction MybatisRedisCache.class) public interface UserMapper { // ... }血泪教训二级缓存默认使用内存存储集群环境下会导致数据不一致。必须实现自定义缓存如RedisCache并妥善处理序列化5.2 批量操作优化使用BatchExecutor大幅提升插入效率Autowired private SqlSessionTemplate sqlSessionTemplate; public void batchInsert(ListUser users) { sqlSessionTemplate.execute(SqlSession - { UserMapper mapper sqlSession.getMapper(UserMapper.class); for (int i 0; i users.size(); i) { mapper.insert(users.get(i)); if (i % 500 0 || i users.size() - 1) { sqlSession.flushStatements(); } } return null; }); }实测对比逐条插入10000条记录耗时12.8秒批量方式仅需1.3秒6. 疑难问题排查指南6.1 SQL未执行问题当发现Mapper方法调用后SQL未执行按以下步骤排查检查方法是否为public动态代理要求确认方法没有final修饰符验证MapperScan是否扫描到对应包查看MyBatis日志级别是否设置为DEBUG6.2 参数绑定异常遇到There is no getter for property named...错误时检查Param注解是否遗漏确认参数类型与SQL中的#{}匹配复杂对象要确保有正确的getter方法6.3 事务失效场景Spring事务不生效的常见原因方法访问权限不是public方法被同类中其他方法调用自调用问题异常类型非RuntimeException且未指定rollbackFor数据源未配置事务管理器7. 监控与诊断工具7.1 使用Arthas监控SQL安装Arthas后通过以下命令抓取MyBatis生成的SQL# 监控指定Mapper接口 watch com.example.mapper.UserMapper * {params,returnObj,throwExp} -x 3 # 查看SQL执行情况 profiler start --event sql profiler stop7.2 日志配置技巧在logback-spring.xml中精确控制MyBatis日志logger nameorg.mybatis levelDEBUG/ logger namejava.sql.Connection levelDEBUG/ logger namejava.sql.Statement levelDEBUG/ logger namejava.sql.PreparedStatement levelTRACE/配合p6spy可以获得更完整的SQL日志# application.properties spring.datasource.driver-class-namecom.p6spy.engine.spy.P6SpyDriver spring.datasource.urljdbc:p6spy:mysql://localhost:3306/db8. 现代MyBatis生态整合8.1 MyBatis-Plus实战基础CRUD操作示例Service public class UserServiceImpl extends ServiceImplUserMapper, User implements UserService { public PageUser queryByCondition(QueryCondition cond) { return lambdaQuery() .eq(cond.getStatus() ! null, User::getStatus, cond.getStatus()) .like(StringUtils.isNotBlank(cond.getName()), User::getName, cond.getName()) .page(Page.of(cond.getPageNum(), cond.getPageSize())); } }8.2 结合Spring AI构建智能查询通过自定义拦截器实现自然语言转SQLpublic class NL2SqlInterceptor implements Interceptor { Autowired private AiTextClient aiClient; Override public Object intercept(Invocation invocation) throws Throwable { BoundSql boundSql ((StatementHandler)invocation.getTarget()) .getBoundSql(); String userQuery (String) boundSql.getParameterObject(); String generatedSql aiClient.generateSql(userQuery); resetSql(invocation, generatedSql); return invocation.proceed(); } private void resetSql(Invocation invocation, String sql) { // 反射修改SQL内容 } }这种模式适合构建面向业务人员的零代码查询系统