使用 mocha/chai 来确保 REST API 提供文件？

Question

I am wanting to validate that a call to one of my REST API's end-point is serving up a file, but I am not sure how to go about it and I am not seeing any examples on this? I did look at the documentation , but this didn't help me much.

The server side code essentially does (in Express):

handleRetrieveContent(req, res, next) {
   const filepaht = '...';
   res.sendFile(filepath)
}

and the test case:

it('Should get a file', (done) => {
    chai.request(url)
        .get('/api/exercise/1?token=' + token)
        .end(function(err, res) {
            if (err) { done(err); }
            res.should.have.status(200);
            // Not sure what the test here should be?
            res.should.be.json;
            // TODO get access to saved file and do tests on it                
        });
});     

I am essentially wanting to do the following tests:

• ensure the response is a file
• ensure the file is of valid content (checksum test)

Any help would be appreciated.

Answer 1

The solution provided was based on further experimenting and an answer provided in https://github.com/chaijs/chai-http/issues/126 - note code assumes ES6 (tested with Node 6.7.0).

const chai = require('chai');
const chaiHttp = require('chai-http');
const md5 = require('md5');
const expect = chai.expect;

const binaryParser = function (res, cb) {
    res.setEncoding('binary');
    res.data = '';
    res.on("data", function (chunk) {
        res.data += chunk;
    });
    res.on('end', function () {
        cb(null, new Buffer(res.data, 'binary'));
    });
};

it('Should get a file', (done) => {
    chai.request(url)
    .get('/api/exercise/1?token=' + token)
        .buffer()
        .parse(binaryParser)
        .end(function(err, res) {
            if (err) { done(err); }
            res.should.have.status(200);

            // Check the headers for type and size
            res.should.have.header('content-type');
            res.header['content-type'].should.be.equal('application/pdf');
            res.should.have.header('content-length');
            const size = fs.statSync(filepath).size.toString();
            res.header['content-length'].should.be.equal(size);
           
            // verify checksum                
            expect(md5(res.body)).to.equal('fa7d7e650b2cec68f302b31ba28235d8');              
        });
});

Edit: Most of this was in Read response output buffer/stream with supertest/superagent on node.js server with possible improvements