mysql拦截器实现crud_常用 crud 的思考和设计

Crud 项目介绍简化单表的CRUD基本代码.

项目地址

为什么使用比如学生管理系统.表设计有 课程表t_classes等等...在管理系统中我们需要添加课程的时候需要做一次 controller 、 service 、 redis 、 dao 这几类操作. 每多一个表格都需要做这一批操作.

一般有新增、修改、删除、根据 id 查询.当使用了这个项目后通过标记一些注解即可获得上述的功能.

假设现在有表格字段类型idint

namevarchar

常规实现创建 controller 类

创建 service 类

创建 一些验证方法

组合起来@RestController

@RequestMapping("/demo")

public class ProjectDemoController {

@Autowired

private ProjectDemoMapper projectDemoMapper;

@PostMapping("/add")

public ResultVO add(

@RequestBody ProjectDemo req

) {

int i = projectDemoMapper.insertSelective(req);

if (i > 0) {

return ResultVO.success();

}

else {

return ResultVO.failed();

}

}

}省略验证方法, 省略service 编写. 正常应该分一分.

现在 只有数据库层面的交互代码我们可能还需要一个redis上的操作

(hash数据类型) 那么这部分代码还需要在写一次@RestController

@RequestMapping("/demo")

public class ProjectDemoController {

@Autowired

private ProjectDemoMapper projectDemoMapper;

@Autowired

private StringRedisTemplate stringRedisTemplate;

@PostMapping("/add")

public ResultVO add(

@RequestBody ProjectDemo req

) {

int i = projectDemoMapper.insertSelective(req);

if (i > 0) {

stringRedisTemplate.opsForHash()

.put("demo", String.valueOf(req.getId()), JSON.toJSONString(req));

return ResultVO.success();

}

else {

return ResultVO.failed();

}

}

}这部分代码出现的次数可能会随着实体类的增多而增多. 每当一个表出现上述的 CRUD 都需要在编辑以便代码. 很是繁琐 ， 对此提出了一个设想.

crud 实现

省略 controller 的编写路径说明: /rest+ @CrudController#uri + add\editor\del\byId

定义一个 实体对象@CrudController(uri = "/project/demo", idType = Integer.class)

public class ProjectInt extends AbsEntity implements Serializable {

private String name;

public String getName() {

return name;

}

public void setName(String name) {

this.name = name;

}

}定义一个mapper@Mapper

@CacheKey(type = ProjectInt.class,key = "asdc")

public interface ProjectIntMapper extends A {

@Override

@Insert("INSERT INTO `project_int`(`name`) VALUES (#{name} ) ")

@Options(useGeneratedKeys = true, keyProperty = "id", keyColumn = "id")

int insertSelective(ProjectInt record);

@Override

@Select("select * from project_int where id = #{id,javaType=INTEGER} ")

ProjectInt selectByPrimaryKey(@Param("id") Integer integer);

@Override

@Delete("DELETE FROM `dest`.`project_int` WHERE `id` = #{id} ")

int deleteByPrimaryKey(@Param("id") Integer integer);

@Override

@Update("UPDATE `dest`.`project_int` SET `name` = #{name} WHERE `id`= #{id} ")

int updateByPrimaryKeySelective(ProjectInt record);

}注册 servlet@Bean

ServletRegistrationBean myServletRegistration() {

ServletRegistrationBean srb = new ServletRegistrationBean();

srb.setServlet(new OcaServlet());

srb.setUrlMappings(Arrays.asList("/rest/*"));

return srb;

}权限验证:在使用了两个servlet后通过拦截器验证会有问题，请使用 filter 进行权限验证

自定义入参验证自定义参数验证是不可避免的. 在这里提出一个接口com.github.huifer.crud.ctr.validated.ValidatedInterface

entityClass 填写对应的数据库对象即可@Service

public class ProjectIntValidated implements ValidatedInterface {

Gson gson = new Gson();

public Class> entityClass() {

return ProjectInt.class;

}

public void validateDelete(ProjectInt projectInt) {

System.out.println(gson.toJson(projectInt));

}

public void validateAdd(ProjectInt projectInt) {

System.out.println(gson.toJson(projectInt));

}

public void validateById(ProjectInt projectInt) {

System.out.println(gson.toJson(projectInt));

}

public void validateEditor(ProjectInt projectInt) {

System.out.println(gson.toJson(projectInt));

}

}如果验证不通过可以在这个方法中直接抛出异常.

省略 mybatis + redis 编写有时我们需要直接调用普通的 crud 不需要从 controller 进行调用 。在这个前提下我们可以使用

需要在mapper 和 db 实体类进行标记public class IssuesEntity implements BaseEntity {

private Integer id;

private String newTitle;

private Date date;

}

@Mapper

@CacheKey(key = "issues", type = IssuesEntity.class)

public interface IssuesMapper extends A {

@Insert(" insert into issue(new_title)values(#{newTitle,jdbcType=VARCHAR})")

@Options(useGeneratedKeys = true, keyProperty = "id", keyColumn = "id")

int insertSelective(IssuesEntity record);

@Select("select id as id , new_title as newTitle from issue where id = #{integer} ")

IssuesEntity selectByPrimaryKey(Integer integer);

@Override

@Update("UPDATE `issue` SET `new_title` = #{newTitle} WHERE `id` = #{id} ")

int updateByPrimaryKeySelective(IssuesEntity record);

@Override

@Delete("delete from issue where id = #{integer}")

int deleteByPrimaryKey(Integer integer);

}调用层代码@Autowired

private CrudFacade> crudFacade;

@Test

void testInsert() {

IssuesEntity issuesEntity = new IssuesEntity();

issuesEntity.setNewTitle("mybatis_test");

crudFacade.insert(issuesEntity);

}使用后直接具有 db 操作和 redis 的操作

省略 redis 编写有时对象可能直接存储在 redis 中而不是存储 db 。 在这个前提下可以使用类标记@CacheKey(key = "tt", type = IssuesEntity.class, idMethod = "ooo")

public class IssuesEntity implements BaseEntity {

private Integer id;

private String newTitle;

private Date date;

}调用层代码@Autowired

private CrudEntityFacade crudEntityFacade;

@Test

void testInsert() {

IssuesEntity issuesEntity = new IssuesEntity();

issuesEntity.setNewTitle("insert");

issuesEntity.setDate(new Date());

crudEntityFacade.insert(issuesEntity);

}

自定义 json 序列化规则目前只支持 gson 和 jackjson 两种gson 自定实现: com.github.huifer.crud.common.conf.json.GsonConfigSetting

jackjson 自定义实现: com.github.huifer.crud.common.conf.json.JackJsonConfigSetting

注意

目前 redis 仅支持 hash 操作