使用Node.js测试URL

Question

Assuming I have an array of URLs and I want to ensure each URL is working I have created the following code. However only the last URL in the array is getting tested. How can I ensure each url returns a 200 response code? To be clear these are all remote addresses I am testing and they point to decently sized PDFs.

Updated based on @lukas.pukenis's response. The results are similar, only a few files are actually checked.

function check(l) {
    console.log(l);

    http.get(l, function(res) {
        if (res.statusCode != 200) {
            console.log(res.statusCode + ' on '+l);
        } else {
            console.log('success on ' + l);
        }
    });
}

for (link in fileLinks) {
  check(fileLinks[link]);
}

This code outputs:

http://somesite.com/somefile1.pdf
http://somesite.com/somefile2.pdf
http://somesite.com/somefile3.pdf
...
all the rest of them
...
http://somesite.com/somefile99.pdf
success on http://somesite.com/somefile1.pdf
success on http://somesite.com/somefile2.pdf
404 on http://somesite.com/somefile5.pdf
success on http://somesite.com/somefile7.pdf

Answer 1

This is because your loop rewrites l variable each time with var l = fileLinks[link];

so l has a value of the last value of array. In order to preserve unique l value you need to store it somewhere. Better - function. Like this:

function check(l) {
  var req = http.get(l, function(res) {
    if (res.statusCode != 200) {
      console.log(res.statusCode + ' on '+l);
    } else {
      console.log('success on ' + l);
    }
  }

  req.on('close', function() {
    console.log('Request done');
  });

for (link in fileLinks) {
  var l = fileLinks[link];
  check(l);
}

Having a function is no magic here. It just preserves your local values in memory for each function call, so l is unique each time it needs to be.

Answer 2

The for expression shouldn't be used with arrays. Replace the for loop for something like this:

fileLinks.forEach(function(item){
  check(item);
});

When doing this many outgoing requests, you may want to increase the maxSockets to something greater than 5, the default , otherwise you might get unexpected behavior. Do this after you require('http') :

http.globalAgent.maxSockets = 150;

Also, when you stick your console.log outside the callback function, it's not going to be displayed at the same time the response comes back from the server. It's redundant anyway. Here is a complete working example:

var http = require('http');
var url = require('url');

function check(l) {
  var u = url.parse(l);

  var opts = {
    host: u.host,
    path: u.path,
    agent: false // prevents pooling behavior
  };

    http.get(opts, function(res) {
        if (res.statusCode != 200) {
            console.log(res.statusCode + ' on '+l);
        } else {
            console.log('success on ' + l);
        }
    });
}

fileLinks = ['http://www.google.com','http://www.google.com'];

fileLinks.forEach(function(item){
  check(item);
});