1. MyBatis Plus在Spring Boot生态中的定位MyBatis Plus简称MP作为MyBatis的增强工具在Spring Boot项目中扮演着数据访问层的加速器角色。它并非传统意义上的中间件而是通过封装常用CRUD操作、提供条件构造器等特性显著简化了持久层开发。与原生MyBatis相比MP在保持灵活性的同时将单表操作的代码量减少了约70%。这种设计哲学与Spring Boot约定优于配置的理念高度契合。在实际项目结构中MP通常作为spring-boot-starter-data-jpa的替代方案出现。其核心价值体现在三个方面首先通过BaseMapper接口提供开箱即用的通用方法其次QueryWrapper/LambdaQueryWrapper构建类型安全的查询条件最后代码生成器可自动生成Entity、Mapper、Service层基础代码。这三个特性共同构成了MP在Spring Boot技术栈中的差异化优势。提示虽然MP能极大提升开发效率但复杂多表关联查询场景仍需配合XML映射文件使用这是许多初学者容易忽视的平衡点。2. 环境配置与基础集成2.1 依赖引入与版本匹配在Spring Boot 2.7.x项目中推荐使用MP 3.5.x版本。pom.xml配置示例如下dependency groupIdcom.baomidou/groupId artifactIdmybatis-plus-boot-starter/artifactId version3.5.3.1/version /dependency dependency groupIdcom.baomidou/groupId artifactIdmybatis-plus-generator/artifactId version3.5.3.1/version scopecompile/scope /dependency注意Spring Boot 3.x需要对应MP 3.5.4版本且JDK基线要求17。常见的版本冲突包括MP与MyBatis原生包版本不兼容分页插件与Spring Boot自动配置冲突多数据源场景下的事务管理器冲突2.2 基础配置项详解application.yml中必须配置的MP核心参数mybatis-plus: configuration: log-impl: org.apache.ibatis.logging.stdout.StdOutImpl # SQL日志输出 map-underscore-to-camel-case: true # 自动驼峰转换 global-config: db-config: id-type: auto # 主键策略 logic-delete-field: deleted # 逻辑删除字段 logic-not-delete-value: 0 logic-delete-value: 1 mapper-locations: classpath*:/mapper/**/*.xml # XML映射文件位置关键配置项说明log-impl建议开发环境使用StdOutImpl生产环境改用SLF4Jid-type需根据数据库类型选择AUTO(数据库自增)、INPUT(手动输入)、ASSIGN_ID(雪花ID)逻辑删除字段需在数据库表中有对应列类型建议为tinyint3. 核心功能深度解析3.1 条件构造器的实战技巧LambdaQueryWrapper是类型安全的查询构建方式典型用法// 查询年龄大于18且姓张的用户 ListUser users userMapper.selectList( new LambdaQueryWrapperUser() .gt(User::getAge, 18) .likeRight(User::getName, 张) .orderByDesc(User::getCreateTime) );复杂条件组合时的注意事项and()和or()嵌套使用时要注意括号优先级apply()方法可插入原生SQL片段但需防SQL注入动态条件应使用boolean condition参数控制// 动态条件示例 public ListUser searchUsers(String name, Integer minAge) { return userMapper.selectList(new LambdaQueryWrapperUser() .like(StringUtils.isNotBlank(name), User::getName, name) .ge(minAge ! null, User::getAge, minAge) ); }3.2 ActiveRecord模式与BaseMapper的抉择MP支持两种操作模式BaseMapper模式推荐Mapper public interface UserMapper extends BaseMapperUser {}优点符合分层架构便于扩展自定义SQLActiveRecord模式public class User extends ModelUser { // 实体类可直接操作数据库 public ListUser findAdults() { return selectList(new LambdaQueryWrapperUser().ge(age, 18)); } }优点快速原型开发 缺点破坏分层结构生产环境建议优先使用BaseMapper模式复杂查询可结合Select注解实现Select(SELECT * FROM user WHERE dept_id #{deptId} ORDER BY ${sortColumn}) ListUser selectByDept(Param(deptId) Long deptId, Param(sortColumn) String sortColumn);4. 高级特性与企业级应用4.1 插件体系实战配置MP通过插件机制扩展功能常用插件配置示例Configuration public class MybatisPlusConfig { Bean public MybatisPlusInterceptor mybatisPlusInterceptor() { MybatisPlusInterceptor interceptor new MybatisPlusInterceptor(); // 分页插件 interceptor.addInnerInterceptor(new PaginationInnerInterceptor(DbType.MYSQL)); // 乐观锁插件 interceptor.addInnerInterceptor(new OptimisticLockerInnerInterceptor()); // 动态表名插件 DynamicTableNameInnerInterceptor dynamicTableName new DynamicTableNameInnerInterceptor(); dynamicTableName.setTableNameHandler((sql, tableName) - { return t_ tableName; // 动态添加前缀 }); interceptor.addInnerInterceptor(dynamicTableName); return interceptor; } }企业级项目中特别有用的插件多租户插件实现SAAS系统的数据隔离数据权限插件根据角色动态过滤数据SQL性能分析插件拦截执行超过1秒的SQL4.2 代码生成器的深度定制AutoGenerator是提升效率的利器推荐配置模板FastAutoGenerator.create(jdbc:mysql://localhost:3306/demo, root, 123456) .globalConfig(builder - { builder.author(dev) // 作者 .outputDir(System.getProperty(user.dir) /src/main/java) // 输出目录 .enableSwagger() // 开启swagger .dateType(DateType.TIME_PACK) // 使用Java8时间类型 .commentDate(yyyy-MM-dd); }) .packageConfig(builder - { builder.parent(com.example) // 父包名 .moduleName(system) // 模块名 .entity(entity) // 实体类包 .service(service) // service包 .serviceImpl(service.impl) .mapper(mapper) .xml(mapper.xml) .controller(controller); }) .strategyConfig(builder - { builder.addInclude(user, role) // 包含的表 .addTablePrefix(t_) // 过滤表前缀 .entityBuilder() .enableLombok() // 使用lombok .logicDeleteColumnName(deleted) // 逻辑删除字段 .versionColumnName(version) // 乐观锁字段 .controllerBuilder() .enableRestStyle(); // 生成RestController }) .templateEngine(new FreemarkerTemplateEngine()) // 使用Freemarker .execute();生成代码后的二次开发建议自定义Service接口继承IService重写Page分页方法支持VO转换添加DTO/VO目录进行对象隔离5. 性能优化与生产实践5.1 批处理操作的最佳实践对比不同批处理方式的性能差异测试数据量10万条操作方式耗时(ms)内存峰值(MB)循环单条插入28500512MP的saveBatch4200680手动拼接批量SQL850350JDBC批处理模式620300推荐方案// JDBC批处理事务 Transactional public void batchInsert(ListUser users) { String sql INSERT INTO user(name,age) VALUES(?,?); jdbcTemplate.batchUpdate(sql, new BatchPreparedStatementSetter() { Override public void setValues(PreparedStatement ps, int i) throws SQLException { ps.setString(1, users.get(i).getName()); ps.setInt(2, users.get(i).getAge()); } Override public int getBatchSize() { return users.size(); } }); }5.2 多数据源集成方案Spring Boot MP Druid多数据源配置要点主数据源配置Primary Bean(masterDataSource) ConfigurationProperties(prefix spring.datasource.master) public DataSource masterDataSource() { return DruidDataSourceBuilder.create().build(); }从数据源配置Bean(slaveDataSource) ConfigurationProperties(prefix spring.datasource.slave) public DataSource slaveDataSource() { return DruidDataSourceBuilder.create().build(); }动态数据源路由public class DynamicDataSource extends AbstractRoutingDataSource { Override protected Object determineCurrentLookupKey() { return DataSourceContextHolder.getDataSourceType(); } }MP配置适配Bean public MybatisPlusInterceptor mybatisPlusInterceptor() { MybatisPlusInterceptor interceptor new MybatisPlusInterceptor(); interceptor.addInnerInterceptor(new DynamicDataSourceInterceptor()); // 自定义拦截器 return interceptor; }常见坑点事务注解需指定事务管理器分页插件需要在每个SqlSessionFactory单独配置跨数据源事务需要引入JTA解决方案6. 监控与问题排查6.1 SQL监控方案推荐组合使用以下工具P6Spy打印完整可执行的SQLspring: datasource: driver-class-name: com.p6spy.engine.spy.P6SpyDriver url: jdbc:p6spy:mysql://localhost:3306/demoDruid监控Bean public ServletRegistrationBeanStatViewServlet druidServlet() { ServletRegistrationBeanStatViewServlet reg new ServletRegistrationBean(); reg.setServlet(new StatViewServlet()); reg.addUrlMappings(/druid/*); reg.addInitParameter(loginUsername, admin); reg.addInitParameter(loginPassword, admin); return reg; }PrometheusGranfa通过micrometer暴露指标Bean public MeterRegistryCustomizerMeterRegistry metricsCommonTags() { return registry - registry.config().commonTags(application, demo-app); }6.2 典型异常处理批量插入重复主键try { userService.saveBatch(users); } catch (DuplicateKeyException e) { // 转为单条插入模式 users.forEach(user - { try { userService.save(user); } catch (DuplicateKeyException ignore) {} }); }乐观锁冲突Transactional public void updateWithLock(Long id) { User user userMapper.selectById(id); user.setName(newName); int rows userMapper.updateById(user); if (rows 0) { throw new OptimisticLockException(数据已被修改请刷新重试); } }SQL注入防护禁止直接拼接${}参数使用QueryWrapper的条件构造方法自定义防注入过滤器public class SqlInjectionFilter implements Filter { Override public void doFilter(ServletRequest request, ServletResponse response, FilterChain chain) { String param request.getParameter(query); if (StringUtils.containsSqlInjection(param)) { throw new IllegalArgumentException(非法参数); } chain.doFilter(request, response); } }7. 现代架构中的演进方向7.1 响应式编程适配Spring WebFlux R2DBC集成方案Repository public interface UserReactiveRepository extends ReactiveCrudRepositoryUser, Long, ReactiveMybatisRepositoryUser { Query(SELECT * FROM user WHERE age :age) FluxUser findByAgeGreaterThan(int age); } Service public class UserReactiveService { private final UserReactiveRepository repository; public FluxUser getAdults() { return repository.findByAgeGreaterThan(18) .timeout(Duration.ofSeconds(1)) .onErrorResume(e - repository.findAll()); } }7.2 云原生实践Kubernetes部署时的特殊配置健康检查端点management: endpoint: health: probes: enabled: true health: db: enabled: true validation-query: SELECT 1动态配置中心集成RefreshScope Configuration public class DynamicDataSourceConfig { Value(${datasource.url}) private String url; // 数据源bean定义 }分布式缓存策略Cacheable(cacheNames users, key #id) public User getUserById(Long id) { return userMapper.selectById(id); } CacheEvict(cacheNames users, key #user.id) public void updateUser(User user) { userMapper.updateById(user); }在微服务架构中建议将MP与Spring Cloud Alibaba组件结合通过Seata处理分布式事务使用Sentinel保护数据库访问Nacos配置中心管理数据源参数8. 项目实战构建企业级权限系统8.1 数据库设计关键点CREATE TABLE sys_user ( id BIGINT NOT NULL COMMENT 主键, username VARCHAR(64) NOT NULL COMMENT 账号, password VARCHAR(128) NOT NULL COMMENT 密码, dept_id BIGINT COMMENT 部门ID, create_time DATETIME DEFAULT CURRENT_TIMESTAMP, update_time DATETIME ON UPDATE CURRENT_TIMESTAMP, deleted TINYINT DEFAULT 0 COMMENT 逻辑删除, version INT DEFAULT 0 COMMENT 乐观锁, PRIMARY KEY (id), UNIQUE KEY uk_username (username) ) ENGINEInnoDB DEFAULT CHARSETutf8mb4 COMMENT用户表; CREATE TABLE sys_role ( id BIGINT NOT NULL, role_name VARCHAR(64) NOT NULL, role_code VARCHAR(64) NOT NULL, PRIMARY KEY (id) ) ENGINEInnoDB DEFAULT CHARSETutf8mb4 COMMENT角色表; CREATE TABLE sys_user_role ( id BIGINT NOT NULL, user_id BIGINT NOT NULL, role_id BIGINT NOT NULL, PRIMARY KEY (id), KEY idx_user_id (user_id) ) ENGINEInnoDB DEFAULT CHARSETutf8mb4 COMMENT用户角色关联表;8.2 核心业务实现用户-角色-权限查询使用MP的连表查询public ListUserWithRoles getUserWithRoles(Long userId) { return userMapper.selectJoinList( new LambdaQueryWrapperUser() .eq(User::getId, userId) .leftJoin(SysUserRole.class, SysUserRole::getUserId, User::getId) .leftJoin(SysRole.class, SysRole::getId, SysUserRole::getRoleId) .select(User.class, info - !info.getColumn().equals(password)) .select(SysRole::getRoleName, SysRole::getRoleCode) ); }数据权限过滤实现public void addDataPermissionFilter(QueryWrapper? wrapper) { LoginUser loginUser SecurityUtils.getLoginUser(); if (loginUser.isAdmin()) { return; } String deptIds loginUser.getDataScope(); // 获取数据权限范围 wrapper.apply(dept_id IN ({0}), deptIds); // 或者使用MP的多租户插件 TenantLineInnerInterceptor tenantInterceptor new TenantLineInnerInterceptor(); tenantInterceptor.setTenantLineHandler(new TenantLineHandler() { Override public Expression getTenantId() { return new LongValue(loginUser.getTenantId()); } Override public boolean ignoreTable(String tableName) { return !sys_user.equals(tableName); } }); }审计日志自动记录Bean public MybatisPlusInterceptor mybatisPlusInterceptor() { MybatisPlusInterceptor interceptor new MybatisPlusInterceptor(); interceptor.addInnerInterceptor(new DataChangeRecorderInnerInterceptor() .insertUpdateColumns(create_by, create_time) .updateColumns(update_by, update_time)); return interceptor; }9. 常见问题解决方案9.1 复杂查询优化策略当遇到MP无法满足的复杂查询时推荐以下解决方案XML映射文件注解混合模式Select(script SELECT u.*, d.dept_name FROM sys_user u LEFT JOIN sys_dept d ON u.dept_id d.id where if testname ! nullAND u.name LIKE CONCAT(#{name}, %)/if if testdeptId ! nullAND u.dept_id #{deptId}/if /where ORDER BY ${sortField} ${sortOrder} /script) ListUserDeptVO selectUserDeptList(Param(name) String name, Param(deptId) Long deptId, Param(sortField) String sortField, Param(sortOrder) String sortOrder);QueryDSL集成方案Repository public interface UserQueryDslRepository extends JpaRepositoryUser, Long, QuerydslPredicateExecutorUser {} public ListUser complexQuery(String name, Integer minAge) { QUser qUser QUser.user; BooleanBuilder builder new BooleanBuilder(); if (StringUtils.isNotBlank(name)) { builder.and(qUser.name.contains(name)); } if (minAge ! null) { builder.and(qUser.age.goe(minAge)); } return userQueryDslRepository.findAll(builder); }存储过程调用Select(CALL sp_get_user_by_dept(#{deptId})) Options(statementType StatementType.CALLABLE) ListUser callUserByDeptProc(Long deptId);9.2 分布式ID生成方案对比方案优点缺点适用场景数据库自增简单可靠分库分表困难单机小流量系统UUID全局唯一无序影响性能临时数据标识雪花算法趋势递增、性能好时钟回拨问题分布式系统(默认方案)Redis生成高性能依赖Redis可用性高并发场景号段模式减少DB访问需要维护号段状态中大型系统MP默认雪花算法配置示例Bean public IdentifierGenerator idGenerator() { return new DefaultIdentifierGenerator() { Override public Number nextId(Object entity) { // 自定义ID生成逻辑 if (entity instanceof Order) { return redisTemplate.opsForValue().increment(order_id_seq); } return super.nextId(entity); } }; }10. 技术演进与生态整合10.1 与Spring Boot 3.x的适配要点JDK17基础环境要求Jakarta EE 9命名空间变化!-- 旧版 -- dependency groupIdjavax.servlet/groupId artifactIdjavax.servlet-api/artifactId /dependency !-- 新版 -- dependency groupIdjakarta.servlet/groupId artifactIdjakarta.servlet-api/artifactId /dependencyGraalVM原生镜像支持native-image -H:IncludeResourcesmapper/.*.xml \ -H:Namemyapp \ -jar target/myapp.jar10.2 与前端框架的协同开发Vue3 MP后端典型交互模式分页查询参数对接GetMapping(/users) public PageResultUserVO listUsers( RequestParam(required false) String name, RequestParam(defaultValue 1) Integer pageNo, RequestParam(defaultValue 10) Integer pageSize) { PageUser page new Page(pageNo, pageSize); LambdaQueryWrapperUser wrapper Wrappers.lambdaQuery(User.class) .like(StringUtils.isNotBlank(name), User::getName, name); IPageUserVO voPage userMapper.selectPage(page, wrapper) .convert(user - { UserVO vo new UserVO(); BeanUtils.copyProperties(user, vo); vo.setStatusLabel(UserStatus.of(user.getStatus()).getLabel()); return vo; }); return PageResult.success(voPage); }参数校验统一处理PostMapping(/users) public Result saveUser(Valid RequestBody UserDTO dto) { User user new User(); BeanUtils.copyProperties(dto, user); userMapper.insert(user); return Result.success(); } // DTO示例 public class UserDTO { NotBlank(message 用户名不能为空) Length(max 64, message 用户名最长64字符) private String username; Pattern(regexp ^(?.*[A-Za-z])(?.*\\d)[A-Za-z\\d]{8,}$, message 密码需包含字母和数字且至少8位) private String password; }文件导入导出集成PostMapping(/users/import) public Result importExcel(RequestParam MultipartFile file) { ListUser users EasyExcel.read(file.getInputStream()) .head(User.class) .sheet() .doReadSync(); userService.saveBatch(users); return Result.success(); } GetMapping(/users/export) public void exportExcel(HttpServletResponse response) { response.setContentType(application/vnd.openxmlformats-officedocument.spreadsheetml.sheet); response.setHeader(Content-Disposition, attachment;filenameusers.xlsx); ListUser users userMapper.selectList(null); EasyExcel.write(response.getOutputStream(), User.class) .sheet(用户数据) .doWrite(users); }11. 安全加固方案11.1 敏感数据保护字段加解密处理TableField(typeHandler EncryptTypeHandler.class) private String mobile; public class EncryptTypeHandler extends BaseTypeHandlerString { private final Encryptor encryptor new Sm4Encryptor(密钥); Override public void setNonNullParameter(PreparedStatement ps, int i, String parameter, JdbcType jdbcType) { ps.setString(i, encryptor.encrypt(parameter)); } Override public String getNullableResult(ResultSet rs, String columnName) { return encryptor.decrypt(rs.getString(columnName)); } }SQL注入防护使用#{}语法而非${}启用MP的非法SQL拦截插件interceptor.addInnerInterceptor(new IllegalSQLInnerInterceptor());自定义关键词过滤器Component public class SqlInjectionFilter implements Filter { private static final SetString BLACKLIST Set.of(drop, truncate, --); Override public void doFilter(ServletRequest request, ServletResponse response, FilterChain chain) throws IOException, ServletException { HttpServletRequest req (HttpServletRequest) request; for (String param : req.getParameterMap().keySet()) { String value req.getParameter(param); if (BLACKLIST.stream().anyMatch(value::containsIgnoreCase)) { throw new IllegalArgumentException(检测到非法参数); } } chain.doFilter(request, response); } }11.2 审计日志实现实体类注解Data public class AuditLog { TableId(type IdType.ASSIGN_ID) private Long id; private String operation; private String params; private String ip; private Long userId; TableField(fill FieldFill.INSERT) private LocalDateTime createTime; }AOP切面记录Aspect Component public class LogAspect { Autowired private AuditLogMapper logMapper; Around(annotation(operLog)) public Object around(ProceedingJoinPoint joinPoint, OperLog operLog) throws Throwable { long start System.currentTimeMillis(); Object result joinPoint.proceed(); long time System.currentTimeMillis() - start; AuditLog log new AuditLog(); log.setOperation(operLog.value()); log.setParams(JsonUtils.toJson(joinPoint.getArgs())); log.setUserId(SecurityUtils.getUserId()); log.setIp(IpUtils.getIpAddr()); logMapper.insert(log); return result; } }配合ELK实现日志分析logging: file: name: logs/app.log logstash: enabled: true host: localhost port: 5044 queue-size: 102412. 测试策略与质量保障12.1 单元测试最佳实践内存数据库测试配置DataJpaTest AutoConfigureTestDatabase(replace Replace.NONE) Import(MybatisPlusConfig.class) public class UserMapperTest { Autowired private UserMapper userMapper; Test Sql(/test-data.sql) void testSelectByDept() { ListUser users userMapper.selectList( new LambdaQueryWrapperUser() .eq(User::getDeptId, 1)); assertEquals(2, users.size()); } }事务回滚测试SpringBootTest Transactional Rollback public class UserServiceTest { Autowired private UserService userService; Test void testSaveUser() { User user new User(); user.setUsername(test); boolean saved userService.save(user); assertTrue(saved); assertNotNull(user.getId()); } }性能基准测试Test RepeatedTest(10) void testBatchInsertPerformance() { ListUser users IntStream.range(0, 1000) .mapToObj(i - new User(user i, i % 50)) .collect(Collectors.toList()); long start System.currentTimeMillis(); userService.saveBatch(users); long cost System.currentTimeMillis() - start; assertTrue(cost 2000, 批处理应在2秒内完成); }12.2 集成测试方案Testcontainers数据库测试Testcontainers SpringBootTest public class IntegrationTest { Container static MySQLContainer? mysql new MySQLContainer(mysql:8.0); DynamicPropertySource static void registerPgProperties(DynamicPropertyRegistry registry) { registry.add(spring.datasource.url, mysql::getJdbcUrl); registry.add(spring.datasource.username, mysql::getUsername); registry.add(spring.datasource.password, mysql::getPassword); } Test void testWithRealDatabase() { // 测试代码 } }API契约测试SpringBootTest(webEnvironment WebEnvironment.RANDOM_PORT) AutoConfigureWireMock(port 0) public class ContractTest { Autowired private TestRestTemplate restTemplate; Test void getUserContract() { ResponseEntityUser response restTemplate.getForEntity( /api/users/1, User.class); assertThat(response.getStatusCode()).isEqualTo(HttpStatus.OK); assertThat(response.getBody()) .hasFieldOrPropertyWithValue(username, admin) .hasFieldOrProperty(createTime); } }混沌工程测试SpringBootTest public class ChaosTest { Autowired private UserService userService; Test void testWithNetworkLatency() { ChaosMesh.injectLatency(1000); // 模拟1秒网络延迟 long start System.currentTimeMillis(); userService.getById(1L); long cost System.currentTimeMillis() - start; assertTrue(cost 1000, 应体现延迟影响); ChaosMesh.clear(); } }13. 部署与运维实战13.1 多环境配置管理Profile差异化配置# application-dev.yml mybatis-plus: mapper-locations: classpath*:/mapper/**/*.xml configuration: log-impl: org.apache.ibatis.logging.stdout.StdOutImpl # application-prod.yml mybatis-plus: mapper-locations: classpath*:/mapper/**/*.xml configuration: log-impl: org.apache.ibatis.logging.slf4j.Slf4jImplKubernetes ConfigMap挂载apiVersion: v1 kind: ConfigMap metadata: name: app-config data: application-prod.yml: | mybatis-plus: configuration: cache-enabled: false --- apiVersion: apps/v1 kind: Deployment spec: template: spec: containers: - name: app volumeMounts: - name: config mountPath: /config volumes: - name: config configMap: name: app-config动态配置刷新RefreshScope RestController public class ConfigController { Value(${mybatis-plus.configuration.cache-enabled}) private Boolean cacheEnabled; GetMapping(/config) public String getConfig() { return Cache enabled: cacheEnabled; } }13.2 健康检查与监控Spring Boot Actuator集成management: endpoints: web: exposure: include: health,info,metrics health: db: enabled: true diskspace: enabled: true metrics: export: prometheus: enabled: true自定义健康指标Component public class MybatisHealthIndicator implements HealthIndicator { Autowired private SqlSessionFactory sqlSessionFactory; Override public Health health() { try (SqlSession session sqlSessionFactory.openSession()) { session.selectOne(SELECT 1); return Health.up().build(); } catch (Exception e) { return Health.down(e).build(); } } }Grafana监控看板配置-- 慢SQL统计 SELECT uri as endpoint, count(*) as calls, avg(duration) as avg_time, max(duration) as max_time FROM http_server_requests WHERE uri LIKE %/api/% AND duration 1000 GROUP BY uri ORDER BY max_time DESC14. 遗留系统迁移策略14.1 从原生MyBatis迁移渐进式迁移步骤第一步引入MP依赖保留原有MyBatis配置第二步将Mapper接口改为继承BaseMapper第三步逐步替换XML中的简单CRUD为MP方法第四步复杂SQL保持XML方式简单查询改用Wrapper共存配置示例Configuration MapperScan(com.old.mapper) public class OldMybatisConfig { Bean public SqlSessionFactory oldSqlSessionFactory(DataSource dataSource) throws Exception { SqlSessionFactoryBean factory new SqlSessionFactoryBean(); factory.setDataSource(dataSource); factory.setMapperLocations(new PathMatchingResourcePatternResolver() .getResources(classpath*:old-mapper/*.xml)); return factory.getObject(); } } Configuration MapperScan(com.new.mapper) public class MybatisPlusConfig { Bean public MybatisPlusInterceptor mybatisPlusInterceptor() { MybatisPlusInterceptor interceptor new MybatisPlusInterceptor(); interceptor.addInnerInterceptor(new PaginationInnerInterceptor()); return interceptor; } }14.2 从JPA迁移主要差异点处理JPA的Entity→ MP的TableTableFieldJpaRepository方法 →BaseMapper方法对照findAll()→selectList(null)save()→insert()/updateById()deleteById()→deleteById()混合使用方案Repository public interface UserRepository extends JpaRepositoryUser, Long, BaseMapperUser { // JPA方式 Query(SELECT u FROM User u WHERE u.name LIKE %
1. MyBatis Plus在Spring Boot生态中的定位MyBatis Plus简称MP作为MyBatis的增强工具在Spring Boot项目中扮演着数据访问层的加速器角色。它并非传统意义上的中间件而是通过封装常用CRUD操作、提供条件构造器等特性显著简化了持久层开发。与原生MyBatis相比MP在保持灵活性的同时将单表操作的代码量减少了约70%。这种设计哲学与Spring Boot约定优于配置的理念高度契合。在实际项目结构中MP通常作为spring-boot-starter-data-jpa的替代方案出现。其核心价值体现在三个方面首先通过BaseMapper接口提供开箱即用的通用方法其次QueryWrapper/LambdaQueryWrapper构建类型安全的查询条件最后代码生成器可自动生成Entity、Mapper、Service层基础代码。这三个特性共同构成了MP在Spring Boot技术栈中的差异化优势。提示虽然MP能极大提升开发效率但复杂多表关联查询场景仍需配合XML映射文件使用这是许多初学者容易忽视的平衡点。2. 环境配置与基础集成2.1 依赖引入与版本匹配在Spring Boot 2.7.x项目中推荐使用MP 3.5.x版本。pom.xml配置示例如下dependency groupIdcom.baomidou/groupId artifactIdmybatis-plus-boot-starter/artifactId version3.5.3.1/version /dependency dependency groupIdcom.baomidou/groupId artifactIdmybatis-plus-generator/artifactId version3.5.3.1/version scopecompile/scope /dependency注意Spring Boot 3.x需要对应MP 3.5.4版本且JDK基线要求17。常见的版本冲突包括MP与MyBatis原生包版本不兼容分页插件与Spring Boot自动配置冲突多数据源场景下的事务管理器冲突2.2 基础配置项详解application.yml中必须配置的MP核心参数mybatis-plus: configuration: log-impl: org.apache.ibatis.logging.stdout.StdOutImpl # SQL日志输出 map-underscore-to-camel-case: true # 自动驼峰转换 global-config: db-config: id-type: auto # 主键策略 logic-delete-field: deleted # 逻辑删除字段 logic-not-delete-value: 0 logic-delete-value: 1 mapper-locations: classpath*:/mapper/**/*.xml # XML映射文件位置关键配置项说明log-impl建议开发环境使用StdOutImpl生产环境改用SLF4Jid-type需根据数据库类型选择AUTO(数据库自增)、INPUT(手动输入)、ASSIGN_ID(雪花ID)逻辑删除字段需在数据库表中有对应列类型建议为tinyint3. 核心功能深度解析3.1 条件构造器的实战技巧LambdaQueryWrapper是类型安全的查询构建方式典型用法// 查询年龄大于18且姓张的用户 ListUser users userMapper.selectList( new LambdaQueryWrapperUser() .gt(User::getAge, 18) .likeRight(User::getName, 张) .orderByDesc(User::getCreateTime) );复杂条件组合时的注意事项and()和or()嵌套使用时要注意括号优先级apply()方法可插入原生SQL片段但需防SQL注入动态条件应使用boolean condition参数控制// 动态条件示例 public ListUser searchUsers(String name, Integer minAge) { return userMapper.selectList(new LambdaQueryWrapperUser() .like(StringUtils.isNotBlank(name), User::getName, name) .ge(minAge ! null, User::getAge, minAge) ); }3.2 ActiveRecord模式与BaseMapper的抉择MP支持两种操作模式BaseMapper模式推荐Mapper public interface UserMapper extends BaseMapperUser {}优点符合分层架构便于扩展自定义SQLActiveRecord模式public class User extends ModelUser { // 实体类可直接操作数据库 public ListUser findAdults() { return selectList(new LambdaQueryWrapperUser().ge(age, 18)); } }优点快速原型开发 缺点破坏分层结构生产环境建议优先使用BaseMapper模式复杂查询可结合Select注解实现Select(SELECT * FROM user WHERE dept_id #{deptId} ORDER BY ${sortColumn}) ListUser selectByDept(Param(deptId) Long deptId, Param(sortColumn) String sortColumn);4. 高级特性与企业级应用4.1 插件体系实战配置MP通过插件机制扩展功能常用插件配置示例Configuration public class MybatisPlusConfig { Bean public MybatisPlusInterceptor mybatisPlusInterceptor() { MybatisPlusInterceptor interceptor new MybatisPlusInterceptor(); // 分页插件 interceptor.addInnerInterceptor(new PaginationInnerInterceptor(DbType.MYSQL)); // 乐观锁插件 interceptor.addInnerInterceptor(new OptimisticLockerInnerInterceptor()); // 动态表名插件 DynamicTableNameInnerInterceptor dynamicTableName new DynamicTableNameInnerInterceptor(); dynamicTableName.setTableNameHandler((sql, tableName) - { return t_ tableName; // 动态添加前缀 }); interceptor.addInnerInterceptor(dynamicTableName); return interceptor; } }企业级项目中特别有用的插件多租户插件实现SAAS系统的数据隔离数据权限插件根据角色动态过滤数据SQL性能分析插件拦截执行超过1秒的SQL4.2 代码生成器的深度定制AutoGenerator是提升效率的利器推荐配置模板FastAutoGenerator.create(jdbc:mysql://localhost:3306/demo, root, 123456) .globalConfig(builder - { builder.author(dev) // 作者 .outputDir(System.getProperty(user.dir) /src/main/java) // 输出目录 .enableSwagger() // 开启swagger .dateType(DateType.TIME_PACK) // 使用Java8时间类型 .commentDate(yyyy-MM-dd); }) .packageConfig(builder - { builder.parent(com.example) // 父包名 .moduleName(system) // 模块名 .entity(entity) // 实体类包 .service(service) // service包 .serviceImpl(service.impl) .mapper(mapper) .xml(mapper.xml) .controller(controller); }) .strategyConfig(builder - { builder.addInclude(user, role) // 包含的表 .addTablePrefix(t_) // 过滤表前缀 .entityBuilder() .enableLombok() // 使用lombok .logicDeleteColumnName(deleted) // 逻辑删除字段 .versionColumnName(version) // 乐观锁字段 .controllerBuilder() .enableRestStyle(); // 生成RestController }) .templateEngine(new FreemarkerTemplateEngine()) // 使用Freemarker .execute();生成代码后的二次开发建议自定义Service接口继承IService重写Page分页方法支持VO转换添加DTO/VO目录进行对象隔离5. 性能优化与生产实践5.1 批处理操作的最佳实践对比不同批处理方式的性能差异测试数据量10万条操作方式耗时(ms)内存峰值(MB)循环单条插入28500512MP的saveBatch4200680手动拼接批量SQL850350JDBC批处理模式620300推荐方案// JDBC批处理事务 Transactional public void batchInsert(ListUser users) { String sql INSERT INTO user(name,age) VALUES(?,?); jdbcTemplate.batchUpdate(sql, new BatchPreparedStatementSetter() { Override public void setValues(PreparedStatement ps, int i) throws SQLException { ps.setString(1, users.get(i).getName()); ps.setInt(2, users.get(i).getAge()); } Override public int getBatchSize() { return users.size(); } }); }5.2 多数据源集成方案Spring Boot MP Druid多数据源配置要点主数据源配置Primary Bean(masterDataSource) ConfigurationProperties(prefix spring.datasource.master) public DataSource masterDataSource() { return DruidDataSourceBuilder.create().build(); }从数据源配置Bean(slaveDataSource) ConfigurationProperties(prefix spring.datasource.slave) public DataSource slaveDataSource() { return DruidDataSourceBuilder.create().build(); }动态数据源路由public class DynamicDataSource extends AbstractRoutingDataSource { Override protected Object determineCurrentLookupKey() { return DataSourceContextHolder.getDataSourceType(); } }MP配置适配Bean public MybatisPlusInterceptor mybatisPlusInterceptor() { MybatisPlusInterceptor interceptor new MybatisPlusInterceptor(); interceptor.addInnerInterceptor(new DynamicDataSourceInterceptor()); // 自定义拦截器 return interceptor; }常见坑点事务注解需指定事务管理器分页插件需要在每个SqlSessionFactory单独配置跨数据源事务需要引入JTA解决方案6. 监控与问题排查6.1 SQL监控方案推荐组合使用以下工具P6Spy打印完整可执行的SQLspring: datasource: driver-class-name: com.p6spy.engine.spy.P6SpyDriver url: jdbc:p6spy:mysql://localhost:3306/demoDruid监控Bean public ServletRegistrationBeanStatViewServlet druidServlet() { ServletRegistrationBeanStatViewServlet reg new ServletRegistrationBean(); reg.setServlet(new StatViewServlet()); reg.addUrlMappings(/druid/*); reg.addInitParameter(loginUsername, admin); reg.addInitParameter(loginPassword, admin); return reg; }PrometheusGranfa通过micrometer暴露指标Bean public MeterRegistryCustomizerMeterRegistry metricsCommonTags() { return registry - registry.config().commonTags(application, demo-app); }6.2 典型异常处理批量插入重复主键try { userService.saveBatch(users); } catch (DuplicateKeyException e) { // 转为单条插入模式 users.forEach(user - { try { userService.save(user); } catch (DuplicateKeyException ignore) {} }); }乐观锁冲突Transactional public void updateWithLock(Long id) { User user userMapper.selectById(id); user.setName(newName); int rows userMapper.updateById(user); if (rows 0) { throw new OptimisticLockException(数据已被修改请刷新重试); } }SQL注入防护禁止直接拼接${}参数使用QueryWrapper的条件构造方法自定义防注入过滤器public class SqlInjectionFilter implements Filter { Override public void doFilter(ServletRequest request, ServletResponse response, FilterChain chain) { String param request.getParameter(query); if (StringUtils.containsSqlInjection(param)) { throw new IllegalArgumentException(非法参数); } chain.doFilter(request, response); } }7. 现代架构中的演进方向7.1 响应式编程适配Spring WebFlux R2DBC集成方案Repository public interface UserReactiveRepository extends ReactiveCrudRepositoryUser, Long, ReactiveMybatisRepositoryUser { Query(SELECT * FROM user WHERE age :age) FluxUser findByAgeGreaterThan(int age); } Service public class UserReactiveService { private final UserReactiveRepository repository; public FluxUser getAdults() { return repository.findByAgeGreaterThan(18) .timeout(Duration.ofSeconds(1)) .onErrorResume(e - repository.findAll()); } }7.2 云原生实践Kubernetes部署时的特殊配置健康检查端点management: endpoint: health: probes: enabled: true health: db: enabled: true validation-query: SELECT 1动态配置中心集成RefreshScope Configuration public class DynamicDataSourceConfig { Value(${datasource.url}) private String url; // 数据源bean定义 }分布式缓存策略Cacheable(cacheNames users, key #id) public User getUserById(Long id) { return userMapper.selectById(id); } CacheEvict(cacheNames users, key #user.id) public void updateUser(User user) { userMapper.updateById(user); }在微服务架构中建议将MP与Spring Cloud Alibaba组件结合通过Seata处理分布式事务使用Sentinel保护数据库访问Nacos配置中心管理数据源参数8. 项目实战构建企业级权限系统8.1 数据库设计关键点CREATE TABLE sys_user ( id BIGINT NOT NULL COMMENT 主键, username VARCHAR(64) NOT NULL COMMENT 账号, password VARCHAR(128) NOT NULL COMMENT 密码, dept_id BIGINT COMMENT 部门ID, create_time DATETIME DEFAULT CURRENT_TIMESTAMP, update_time DATETIME ON UPDATE CURRENT_TIMESTAMP, deleted TINYINT DEFAULT 0 COMMENT 逻辑删除, version INT DEFAULT 0 COMMENT 乐观锁, PRIMARY KEY (id), UNIQUE KEY uk_username (username) ) ENGINEInnoDB DEFAULT CHARSETutf8mb4 COMMENT用户表; CREATE TABLE sys_role ( id BIGINT NOT NULL, role_name VARCHAR(64) NOT NULL, role_code VARCHAR(64) NOT NULL, PRIMARY KEY (id) ) ENGINEInnoDB DEFAULT CHARSETutf8mb4 COMMENT角色表; CREATE TABLE sys_user_role ( id BIGINT NOT NULL, user_id BIGINT NOT NULL, role_id BIGINT NOT NULL, PRIMARY KEY (id), KEY idx_user_id (user_id) ) ENGINEInnoDB DEFAULT CHARSETutf8mb4 COMMENT用户角色关联表;8.2 核心业务实现用户-角色-权限查询使用MP的连表查询public ListUserWithRoles getUserWithRoles(Long userId) { return userMapper.selectJoinList( new LambdaQueryWrapperUser() .eq(User::getId, userId) .leftJoin(SysUserRole.class, SysUserRole::getUserId, User::getId) .leftJoin(SysRole.class, SysRole::getId, SysUserRole::getRoleId) .select(User.class, info - !info.getColumn().equals(password)) .select(SysRole::getRoleName, SysRole::getRoleCode) ); }数据权限过滤实现public void addDataPermissionFilter(QueryWrapper? wrapper) { LoginUser loginUser SecurityUtils.getLoginUser(); if (loginUser.isAdmin()) { return; } String deptIds loginUser.getDataScope(); // 获取数据权限范围 wrapper.apply(dept_id IN ({0}), deptIds); // 或者使用MP的多租户插件 TenantLineInnerInterceptor tenantInterceptor new TenantLineInnerInterceptor(); tenantInterceptor.setTenantLineHandler(new TenantLineHandler() { Override public Expression getTenantId() { return new LongValue(loginUser.getTenantId()); } Override public boolean ignoreTable(String tableName) { return !sys_user.equals(tableName); } }); }审计日志自动记录Bean public MybatisPlusInterceptor mybatisPlusInterceptor() { MybatisPlusInterceptor interceptor new MybatisPlusInterceptor(); interceptor.addInnerInterceptor(new DataChangeRecorderInnerInterceptor() .insertUpdateColumns(create_by, create_time) .updateColumns(update_by, update_time)); return interceptor; }9. 常见问题解决方案9.1 复杂查询优化策略当遇到MP无法满足的复杂查询时推荐以下解决方案XML映射文件注解混合模式Select(script SELECT u.*, d.dept_name FROM sys_user u LEFT JOIN sys_dept d ON u.dept_id d.id where if testname ! nullAND u.name LIKE CONCAT(#{name}, %)/if if testdeptId ! nullAND u.dept_id #{deptId}/if /where ORDER BY ${sortField} ${sortOrder} /script) ListUserDeptVO selectUserDeptList(Param(name) String name, Param(deptId) Long deptId, Param(sortField) String sortField, Param(sortOrder) String sortOrder);QueryDSL集成方案Repository public interface UserQueryDslRepository extends JpaRepositoryUser, Long, QuerydslPredicateExecutorUser {} public ListUser complexQuery(String name, Integer minAge) { QUser qUser QUser.user; BooleanBuilder builder new BooleanBuilder(); if (StringUtils.isNotBlank(name)) { builder.and(qUser.name.contains(name)); } if (minAge ! null) { builder.and(qUser.age.goe(minAge)); } return userQueryDslRepository.findAll(builder); }存储过程调用Select(CALL sp_get_user_by_dept(#{deptId})) Options(statementType StatementType.CALLABLE) ListUser callUserByDeptProc(Long deptId);9.2 分布式ID生成方案对比方案优点缺点适用场景数据库自增简单可靠分库分表困难单机小流量系统UUID全局唯一无序影响性能临时数据标识雪花算法趋势递增、性能好时钟回拨问题分布式系统(默认方案)Redis生成高性能依赖Redis可用性高并发场景号段模式减少DB访问需要维护号段状态中大型系统MP默认雪花算法配置示例Bean public IdentifierGenerator idGenerator() { return new DefaultIdentifierGenerator() { Override public Number nextId(Object entity) { // 自定义ID生成逻辑 if (entity instanceof Order) { return redisTemplate.opsForValue().increment(order_id_seq); } return super.nextId(entity); } }; }10. 技术演进与生态整合10.1 与Spring Boot 3.x的适配要点JDK17基础环境要求Jakarta EE 9命名空间变化!-- 旧版 -- dependency groupIdjavax.servlet/groupId artifactIdjavax.servlet-api/artifactId /dependency !-- 新版 -- dependency groupIdjakarta.servlet/groupId artifactIdjakarta.servlet-api/artifactId /dependencyGraalVM原生镜像支持native-image -H:IncludeResourcesmapper/.*.xml \ -H:Namemyapp \ -jar target/myapp.jar10.2 与前端框架的协同开发Vue3 MP后端典型交互模式分页查询参数对接GetMapping(/users) public PageResultUserVO listUsers( RequestParam(required false) String name, RequestParam(defaultValue 1) Integer pageNo, RequestParam(defaultValue 10) Integer pageSize) { PageUser page new Page(pageNo, pageSize); LambdaQueryWrapperUser wrapper Wrappers.lambdaQuery(User.class) .like(StringUtils.isNotBlank(name), User::getName, name); IPageUserVO voPage userMapper.selectPage(page, wrapper) .convert(user - { UserVO vo new UserVO(); BeanUtils.copyProperties(user, vo); vo.setStatusLabel(UserStatus.of(user.getStatus()).getLabel()); return vo; }); return PageResult.success(voPage); }参数校验统一处理PostMapping(/users) public Result saveUser(Valid RequestBody UserDTO dto) { User user new User(); BeanUtils.copyProperties(dto, user); userMapper.insert(user); return Result.success(); } // DTO示例 public class UserDTO { NotBlank(message 用户名不能为空) Length(max 64, message 用户名最长64字符) private String username; Pattern(regexp ^(?.*[A-Za-z])(?.*\\d)[A-Za-z\\d]{8,}$, message 密码需包含字母和数字且至少8位) private String password; }文件导入导出集成PostMapping(/users/import) public Result importExcel(RequestParam MultipartFile file) { ListUser users EasyExcel.read(file.getInputStream()) .head(User.class) .sheet() .doReadSync(); userService.saveBatch(users); return Result.success(); } GetMapping(/users/export) public void exportExcel(HttpServletResponse response) { response.setContentType(application/vnd.openxmlformats-officedocument.spreadsheetml.sheet); response.setHeader(Content-Disposition, attachment;filenameusers.xlsx); ListUser users userMapper.selectList(null); EasyExcel.write(response.getOutputStream(), User.class) .sheet(用户数据) .doWrite(users); }11. 安全加固方案11.1 敏感数据保护字段加解密处理TableField(typeHandler EncryptTypeHandler.class) private String mobile; public class EncryptTypeHandler extends BaseTypeHandlerString { private final Encryptor encryptor new Sm4Encryptor(密钥); Override public void setNonNullParameter(PreparedStatement ps, int i, String parameter, JdbcType jdbcType) { ps.setString(i, encryptor.encrypt(parameter)); } Override public String getNullableResult(ResultSet rs, String columnName) { return encryptor.decrypt(rs.getString(columnName)); } }SQL注入防护使用#{}语法而非${}启用MP的非法SQL拦截插件interceptor.addInnerInterceptor(new IllegalSQLInnerInterceptor());自定义关键词过滤器Component public class SqlInjectionFilter implements Filter { private static final SetString BLACKLIST Set.of(drop, truncate, --); Override public void doFilter(ServletRequest request, ServletResponse response, FilterChain chain) throws IOException, ServletException { HttpServletRequest req (HttpServletRequest) request; for (String param : req.getParameterMap().keySet()) { String value req.getParameter(param); if (BLACKLIST.stream().anyMatch(value::containsIgnoreCase)) { throw new IllegalArgumentException(检测到非法参数); } } chain.doFilter(request, response); } }11.2 审计日志实现实体类注解Data public class AuditLog { TableId(type IdType.ASSIGN_ID) private Long id; private String operation; private String params; private String ip; private Long userId; TableField(fill FieldFill.INSERT) private LocalDateTime createTime; }AOP切面记录Aspect Component public class LogAspect { Autowired private AuditLogMapper logMapper; Around(annotation(operLog)) public Object around(ProceedingJoinPoint joinPoint, OperLog operLog) throws Throwable { long start System.currentTimeMillis(); Object result joinPoint.proceed(); long time System.currentTimeMillis() - start; AuditLog log new AuditLog(); log.setOperation(operLog.value()); log.setParams(JsonUtils.toJson(joinPoint.getArgs())); log.setUserId(SecurityUtils.getUserId()); log.setIp(IpUtils.getIpAddr()); logMapper.insert(log); return result; } }配合ELK实现日志分析logging: file: name: logs/app.log logstash: enabled: true host: localhost port: 5044 queue-size: 102412. 测试策略与质量保障12.1 单元测试最佳实践内存数据库测试配置DataJpaTest AutoConfigureTestDatabase(replace Replace.NONE) Import(MybatisPlusConfig.class) public class UserMapperTest { Autowired private UserMapper userMapper; Test Sql(/test-data.sql) void testSelectByDept() { ListUser users userMapper.selectList( new LambdaQueryWrapperUser() .eq(User::getDeptId, 1)); assertEquals(2, users.size()); } }事务回滚测试SpringBootTest Transactional Rollback public class UserServiceTest { Autowired private UserService userService; Test void testSaveUser() { User user new User(); user.setUsername(test); boolean saved userService.save(user); assertTrue(saved); assertNotNull(user.getId()); } }性能基准测试Test RepeatedTest(10) void testBatchInsertPerformance() { ListUser users IntStream.range(0, 1000) .mapToObj(i - new User(user i, i % 50)) .collect(Collectors.toList()); long start System.currentTimeMillis(); userService.saveBatch(users); long cost System.currentTimeMillis() - start; assertTrue(cost 2000, 批处理应在2秒内完成); }12.2 集成测试方案Testcontainers数据库测试Testcontainers SpringBootTest public class IntegrationTest { Container static MySQLContainer? mysql new MySQLContainer(mysql:8.0); DynamicPropertySource static void registerPgProperties(DynamicPropertyRegistry registry) { registry.add(spring.datasource.url, mysql::getJdbcUrl); registry.add(spring.datasource.username, mysql::getUsername); registry.add(spring.datasource.password, mysql::getPassword); } Test void testWithRealDatabase() { // 测试代码 } }API契约测试SpringBootTest(webEnvironment WebEnvironment.RANDOM_PORT) AutoConfigureWireMock(port 0) public class ContractTest { Autowired private TestRestTemplate restTemplate; Test void getUserContract() { ResponseEntityUser response restTemplate.getForEntity( /api/users/1, User.class); assertThat(response.getStatusCode()).isEqualTo(HttpStatus.OK); assertThat(response.getBody()) .hasFieldOrPropertyWithValue(username, admin) .hasFieldOrProperty(createTime); } }混沌工程测试SpringBootTest public class ChaosTest { Autowired private UserService userService; Test void testWithNetworkLatency() { ChaosMesh.injectLatency(1000); // 模拟1秒网络延迟 long start System.currentTimeMillis(); userService.getById(1L); long cost System.currentTimeMillis() - start; assertTrue(cost 1000, 应体现延迟影响); ChaosMesh.clear(); } }13. 部署与运维实战13.1 多环境配置管理Profile差异化配置# application-dev.yml mybatis-plus: mapper-locations: classpath*:/mapper/**/*.xml configuration: log-impl: org.apache.ibatis.logging.stdout.StdOutImpl # application-prod.yml mybatis-plus: mapper-locations: classpath*:/mapper/**/*.xml configuration: log-impl: org.apache.ibatis.logging.slf4j.Slf4jImplKubernetes ConfigMap挂载apiVersion: v1 kind: ConfigMap metadata: name: app-config data: application-prod.yml: | mybatis-plus: configuration: cache-enabled: false --- apiVersion: apps/v1 kind: Deployment spec: template: spec: containers: - name: app volumeMounts: - name: config mountPath: /config volumes: - name: config configMap: name: app-config动态配置刷新RefreshScope RestController public class ConfigController { Value(${mybatis-plus.configuration.cache-enabled}) private Boolean cacheEnabled; GetMapping(/config) public String getConfig() { return Cache enabled: cacheEnabled; } }13.2 健康检查与监控Spring Boot Actuator集成management: endpoints: web: exposure: include: health,info,metrics health: db: enabled: true diskspace: enabled: true metrics: export: prometheus: enabled: true自定义健康指标Component public class MybatisHealthIndicator implements HealthIndicator { Autowired private SqlSessionFactory sqlSessionFactory; Override public Health health() { try (SqlSession session sqlSessionFactory.openSession()) { session.selectOne(SELECT 1); return Health.up().build(); } catch (Exception e) { return Health.down(e).build(); } } }Grafana监控看板配置-- 慢SQL统计 SELECT uri as endpoint, count(*) as calls, avg(duration) as avg_time, max(duration) as max_time FROM http_server_requests WHERE uri LIKE %/api/% AND duration 1000 GROUP BY uri ORDER BY max_time DESC14. 遗留系统迁移策略14.1 从原生MyBatis迁移渐进式迁移步骤第一步引入MP依赖保留原有MyBatis配置第二步将Mapper接口改为继承BaseMapper第三步逐步替换XML中的简单CRUD为MP方法第四步复杂SQL保持XML方式简单查询改用Wrapper共存配置示例Configuration MapperScan(com.old.mapper) public class OldMybatisConfig { Bean public SqlSessionFactory oldSqlSessionFactory(DataSource dataSource) throws Exception { SqlSessionFactoryBean factory new SqlSessionFactoryBean(); factory.setDataSource(dataSource); factory.setMapperLocations(new PathMatchingResourcePatternResolver() .getResources(classpath*:old-mapper/*.xml)); return factory.getObject(); } } Configuration MapperScan(com.new.mapper) public class MybatisPlusConfig { Bean public MybatisPlusInterceptor mybatisPlusInterceptor() { MybatisPlusInterceptor interceptor new MybatisPlusInterceptor(); interceptor.addInnerInterceptor(new PaginationInnerInterceptor()); return interceptor; } }14.2 从JPA迁移主要差异点处理JPA的Entity→ MP的TableTableFieldJpaRepository方法 →BaseMapper方法对照findAll()→selectList(null)save()→insert()/updateById()deleteById()→deleteById()混合使用方案Repository public interface UserRepository extends JpaRepositoryUser, Long, BaseMapperUser { // JPA方式 Query(SELECT u FROM User u WHERE u.name LIKE %
