HEAD requests with Node v5 have issues with gzipped content and fail.

[czardoz]

This script reproduces the issue:

    var req = require('request');

    req({
        url: 'http://google.com',
        method: 'HEAD',
        headers: {},
        followAllRedirects: false,
        followRedirect: true,
        jar: true,
        timeout: 15000,
        gzip: true,
        rejectUnauthorized: false,
        strictSSL: true
    }, function (err, res, body) {
        if (err) {
            console.error(err.stack || err);
        }
    });

Steps to reproduce:

aniket [/tmp/request-bug] -> node -v
v5.1.0
aniket [/tmp/request-bug] -> npm -v
3.3.12
aniket [/tmp/request-bug] -> cat node_modules/request/package.json| grep version
  "version": "2.69.0"
aniket [/tmp/request-bug] -> node reproduce.js 
Error: unexpected end of file
    at Zlib._handle.onerror (zlib.js:363:17)
aniket [/tmp/request-bug] -> 

[nathanmarks]

@czardoz FWIW I had the same issue with some HEAD requests and it is because of followAllRedirects. Unless you set it to true, it won't follow HEAD redirects (IE non www to www, or http to https).

edit: Nvm, I see the issue with more URLs now too.

[czardoz]

Is this issue reproducible for others?

@simov Is there a label for a "confirmed" bug? I can consistently reproduce it with Node v5.

[nathanmarks]

@czardoz Not sure if this helps you, but I was doing HEAD requests to avoid downloading entire responses.

I ended up switching to GET requests, streaming the response and aborting the request early.

[czardoz]

@nathanmarks While that's a viable workaround, I use this library to run user-specified HTTP requests, so I cannot really change the request method :)

[simov]

Fixed here #2176

Let me know what do you think.