[英]Expressjs - Handling Errors With Middleware
I'm building an API (using Expressjs v4) and typically I've dealt with errors within the route rather than using middleware. 我正在构建一个API(使用Expressjs v4),通常我处理的是路由中的错误,而不是使用中间件。 For example:
例如:
router.use('/', function(req, res, next) {
...
if (err)
return res.status(500).send({type: "serverError", message: "Something has gone wrong on our end."});
}
I now realise that middleware is the "way to go." 我现在意识到中间件是“前进的道路”。 I've seen the rather limited documentation on the Expressjs site here: http://expressjs.com/guide/error-handling.html but still unsure of a few things.
我已经在Expressjs网站上看到了相当有限的文档: http ://expressjs.com/guide/error-handling.html,但仍然不确定。
I've added in the server.js
: 我在
server.js
添加了:
function errorHandler(err, req, res, next) {
}
but how do I supposed to handle the different types of errors (400,404,500 etc)? 但是我应该如何处理不同类型的错误(400,404,500等)?
I'm finding myself writing 3 lines of code each time an error occurs: 我发现自己每次发生错误时都会编写3行代码:
//A route
var err = new Error();
err.status = 404;
return next(err);
and I can access the status using: 我可以使用以下方式访问状态:
function errorHandler(err, req, res, next) {
console.log(err.status);
if(err.status == "400")
//do something
else
//etc etc
}
Surely there's an easier way than this? 当然有比这更简单的方法吗? Am I missing something?
我想念什么吗?
You should create your own Error type that allows you to provide all the information necessary to act on the error. 您应该创建自己的错误类型,该类型允许您提供执行错误所需的所有信息。
var util = require('util');
function HttpError(message, statusCode){
this.message = message;
this.statusCode = statusCode;
this.stack = (new Error()).stack;
}
util.inherits(Error, HttpError);
module.exports = HttpError;
Then include your new error object in your code and use it like 然后在代码中包含新的错误对象,并像使用它一样
next(new HttpError('Not Found', 404));
Or you could go crazy and implement an error type for each response code pre-populating the statusCode
part. 或者,您可能会发疯并为每个预填充
statusCode
部分的响应代码实现错误类型。
Ref: What's a good way to extend Error in JavaScript? 参考: 在JavaScript中扩展Error的好方法是什么?
Instead manually creating error , you can delegate that error like below. 代替手动创建error ,您可以像下面这样委派该错误。
return next(err);
And your error will go deep down all the routes defined until it find routes with below signature. 您的错误将深入所有定义的路由,直到找到具有以下签名的路由。
app.use(function (err, req, res, next) {
});
You can see the err argument in above route. 您可以在上述路线中看到err参数。
Ideally you can use below two methods for both DEVELOPMENT & PRODUCTION environment. 理想情况下,您可以在开发和生产环境中使用以下两种方法。
if (process.env.NODE_ENV === 'development') {
app.use(function (err, req, res, next) {
res.status(err.status || 500);
logger.log('info', err.message + " expected URL was " + req.url);
res.status(err.status).send(err.status, {
message: err.message,
error : err
});
});
}
app.use(function (err, req, res, next) {
res.status(err.status || 500);
logger.log('error', err.message + " expected URL was " + req.url);
res.status(err.status).send(err.status, {
message: err.message,
error : {}
});
});
You can capture actual populated error object in there. 您可以在其中捕获实际填充的错误对象。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.