springboot自定义注解这么用，轻松捕获系统操作日志（上）

通常为了监控系统，我们希望将请求的入参和出参记录到数据库中，已备后查。除了在每个方法里面加日志处理代码，手动保存到数据库，还有其他的办法吗？

今天就给大伙介绍一个注解 @Log2DB

什么？没这个注解？是的，这个注解是自定义的，那用这个注解能干什么呢？怎么写一个自己的注解呢？今天我们一起来探讨下。

首先，我们随便找个注解，看下它的结构@RestController

@Target({ElementType.TYPE})
@Retention(RetentionPolicy.RUNTIME)
@Documented
@Controller
@ResponseBody
public @interface RestController {
    @AliasFor(
        annotation = Controller.class
    )
    String value() default "";
}

是个interface，但是又加了一个@符号，那好吧，我们也搞一个这个。

@Target({ElementType.METHOD, ElementType.TYPE})//注解作用的位置，ElementType.METHOD表示该注解仅能作用于方法上
@Retention(RetentionPolicy.RUNTIME)//注解的生命周期，表示注解会被保留到什么阶段，可以选择编译阶段SOURCE、类加载阶段CLASS，或运行阶段RUNTIME
@Documented//注解信息会被添加到Java文档中
public @interface Log2DB {
    String name() default "接口";
    boolean enabled() default true;//是否启用，默认是
}

注解是加进去了，怎么用呢？

@Api("testController相关操作")
@Slf4j
@RestController
@RequestMapping("test")
public class TestController {

    @Log2DB(name = "获取用户")
    @GetMapping("get")
    @ApiOperation("获取用户1")
    public String getUser(@RequestParam("userId")String userId){
        log.info("get is called");
        return "get is called";
    }

    @Log2DB(name = "获取用户")
    @PostMapping("post")
    @ApiOperation("保存用户1")
    public String postUser(@RequestBody Map request){
        log.info("post is called");
        return "post is called";
    }
}

这就加上去了，但不是说好了保存到数据库吗？我们用面向切面把@Log2DB这个注解拦截下来，定义一个aspect

@Aspect
@Configuration
@Slf4j
public class Log2DBAspect {
   //拦截哪个包下的类
   //拦截哪个注解com.example.lesson20.annotation.Log2DB
   @Pointcut("execution(public * *(..)) && @annotation(com.example.lesson20.annotation.Log2DB)")
   public void logs(){
      log.info("@Pointcut");
   }

   @Around("logs()")
   public Object doLog(ProceedingJoinPoint joinPoint) throws Throwable {
      log.info("@Around");
      MethodSignature signature = (MethodSignature) joinPoint.getSignature();
        Log2DB logAnnotation = signature.getMethod().getAnnotation(Log2DB.class);
        //如果不需要记录到数据库，继续业务调用并返回
        if(!logAnnotation.enabled()){
           return joinPoint.proceed();
        }

        //定义需要从request和response中获取的信息 begin
      String name = "",//请求的服务名称
            url="",//请求服务地址
            method="",//请求方法 get post put
            clientIp="",//客户端ip
            browser="",//浏览器
            clazz_method="",//请求的类和方法名
            headers="", //head头信息
            request="";//请求body入参
      Object response=null;//返回参数
      Date bgn,end;//请求开始结束时间
      Boolean success = true;//是否成功
      long cost = 0;//耗时 毫秒
      //定义需要从request和response中获取的信息 end

      ServletRequestAttributes attributes = (ServletRequestAttributes) RequestContextHolder.getRequestAttributes();
      if(attributes==null){
         return joinPoint.proceed();
      }
      HttpServletRequest httpServletRequest = attributes.getRequest();

      url = httpServletRequest.getRequestURL().toString();
      method = httpServletRequest.getMethod();
      clientIp =getClientIp(httpServletRequest);
      clazz_method =joinPoint.getSignature().getDeclaringTypeName()+"."+joinPoint.getSignature().getName();

      name = logAnnotation.name();

      //获取url后面的参数串
      StringBuffer params = new StringBuffer();
      for (String item : httpServletRequest.getParameterMap().keySet()) {
         params.append(item).append("=").append(httpServletRequest.getParameter(item)).append("&");
      }
      String tmp = params.toString();
      if(StringUtils.hasText(tmp)){
         tmp = tmp.substring(0,tmp.length()-1);
         url = url+"?"+tmp;
      }

      //获取header头信息
      List<Map<String,String>> headerList = new ArrayList<>();
      Enumeration<String> headerNames = httpServletRequest.getHeaderNames();
      while (headerNames.hasMoreElements()) {
         String key = headerNames.nextElement();
         String value = httpServletRequest.getHeader(key);
         Map<String,String> map = new HashMap<String,String>();
         map.put(key,value);
         headerList.add(map);

         if("user-agent".equals(key.toLowerCase())){
            browser = value;
         }
      }
      headers = new ObjectMapper().writeValueAsString(headerList);

      request = new ObjectMapper().writeValueAsString(joinPoint.getArgs());

      //记录起始时间
      bgn = new Date();
      /** 执行目标方法 */
      try{
         response= joinPoint.proceed();
         log.info("response={}",response==null?"": new ObjectMapper().writeValueAsString(response));
      }
      catch(Exception e){
         success=false;
         response = e.getMessage();
         log.error("errorMessage: {}", e.getMessage());
         e.printStackTrace();
         throw e;
      }
      finally{
         end = new Date();
         /** 记录操作时间 */
         cost = (end.getTime() - bgn.getTime());

         SimpleDateFormat format = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss:SSS");
         log.info("name={}",name);
         log.info("url={}",url);
         log.info("method={}",method);
         log.info("ip={}",clientIp);
         log.info("browser={}",browser);
         log.info("class_method={}",clazz_method);
         log.info("header={}",headers);
         log.info("request={}",request);
         log.info("response={}",response);
         log.info("success={}",success);
         log.info("bgn={}",format.format(bgn));
         log.info("end={}",format.format(end));
         log.info("cost={}", cost);

         //TODO: 保存到数据库 这里大家伙自行sql保存到数据库，我就不参合了
         //不过我建议这个地方可以考虑先放到mq 消息队列中，然后慢慢消费，不要占用mysql的资源。
         // 或者日志就存储到mongodb中也不失为一个好办法
      }
      return response;
   }

   @Before("logs()")
   public void doBefore(JoinPoint point){
      log.info("@Before");
   }

   @After("logs()")
   public void doAfter(){
      log.info("@After");
   }

   //在doAfter之后执行，主要用户记录程序执行后的返回值
   @AfterReturning(returning="object",pointcut="logs()")
   public void doAfterReturning(Object object){
      log.info("@AfterReturning");
   }

   //获取客户端ip
   private String getClientIp(HttpServletRequest request) {
      String ip = request.getHeader("x-forwarded-for");
      if (ip == null || ip.length() == 0 || "unknown".equalsIgnoreCase(ip)) {
         ip = request.getHeader("Proxy-Client-IP");
      }
      if (ip == null || ip.length() == 0 || "unknown".equalsIgnoreCase(ip)) {
         ip = request.getHeader("WL-Proxy-Client-IP");
      }
      if (ip == null || ip.length() == 0 || "unknown".equalsIgnoreCase(ip)) {
         ip = request.getRemoteAddr();
      }
      return ip;
   }
}

类有点长，我就不给大家一一解释了，里面注释都很详细，大家直接copy paste就可以用了，最后我们来验证下是不是这些参数都拿到了，postman请求一次，看看后台打印日志

@Around
@Before
post is called
@AfterReturning
@After
response="post is called"
name=保存用户1
url=http://localhost:8080/test/post?userId=abc
method=POST
ip=0:0:0:0:0:0:0:1
browser=PostmanRuntime/7.26.8
class_method=com.example.lesson20.controller.TestController.postUser
header=[{"content-type":"application/json"},{"user-agent":"PostmanRuntime/7.26.8"},{"accept":"*/*"},{"cache-control":"no-cache"},{"postman-token":"b3f49862-8b4e-4c7a-b817-3ff4f4e19336"},{"host":"localhost:8080"},{"accept-encoding":"gzip, deflate, br"},{"connection":"keep-alive"},{"content-length":"47"}]
request=[{"username":"abc","password":123}]
response=post is called
success=true
bgn=2021-07-03 10:00:25:183
end=2021-07-03 10:00:25:214
cost=31

嗯，是不错。但是方法的中文名字在Log2DB和ApiOperation中都要写一次，有点麻烦。下一期我们一起来解决这个问题。