How does node-http-proxy parse the target url?

Question

I ran into a problem and I feel that node-http-proxy is changing my target links. I got a few examples below.

I am using express as my server and using Metaweather API .

The problem is that I was able to get data from the endpoints below https://www.metaweather.com/api/location/2487956/ https://www.metaweather.com/api/location/2487956/2013/4/30/

But when I try to call the API from https://www.metaweather.com/api/location/search/?lattlong=36.96,-122.02

It fails with status code 500 which I lead me thinking that node-http-proxy added some values after 122.02 as it was not closed with /

server.js

const express = require("express");
const next = require("next");
const dev = process.env.NODE_ENV !== "production";
const app = next({ dev });
const handle = app.getRequestHandler();

const httpProxy = require("http-proxy");

const proxyOptions = {
  changeOrigin: true
};

const apiProxy = httpProxy.createProxyServer(proxyOptions);

const apiUrl =
  "https://www.metaweather.com/api/location/search/?lattlong=36.96,-122.02";

/*
https://www.metaweather.com/api/location/search/?lattlong=36.96,-122.02 - failed with 500
https://www.metaweather.com/api/location/2487956/ - passed
https://www.metaweather.com/api/location/2487956/2013/4/30/ - passed
*/

app
  .prepare()
  .then(() => {
    const server = express();

    server.use("/api", (req, res) => {
      console.log("Going to call this API " + apiUrl);
      apiProxy.web(req, res, { target: apiUrl });
    });

    server.get("*", (req, res) => {
      return handle(req, res);
    });

    server.listen(3000, err => {
      if (err) throw err;
      console.log("> Ready on http://localhost:3000");
    });
  })
  .catch(ex => {
    console.error(ex.stack);
    process.exit(1);
  });

Thanks for looking into this question.

Answer 1

I have reproduced where this is happening in node-http-proxy.

In common.js there is a function called urlJoin which is appending the req.url to the end of the target url.

I'm not exactly sure what the intent is, but it's a start.

Here's my test:

const urlJoin = function() {
  //
  // We do not want to mess with the query string. All we want to touch is the path.
  //
var args = Array.prototype.slice.call(arguments),
    lastIndex = args.length - 1,
    last = args[lastIndex],
    lastSegs = last.split('?'),
    retSegs;

args[lastIndex] = lastSegs.shift();

//
// Join all strings, but remove empty strings so we don't get extra slashes from
// joining e.g. ['', 'am']
//
retSegs = [
  args.filter(Boolean).join('/')
      .replace(/\/+/g, '/')
      .replace('http:/', 'http://')
      .replace('https:/', 'https://')
];

// Only join the query string if it exists so we don't have trailing a '?'
// on every request

// Handle case where there could be multiple ? in the URL.
retSegs.push.apply(retSegs, lastSegs);

return retSegs.join('?')
};

let path = urlJoin('/api/location/search/?lattlong=36.96,-122.02', '/');

console.log(path);
//                 /api/location/search/?lattlong=36.96,-122.02/