Add support for matching multipart form values

I’m not sure if this is already supported, but I can’t find any docs. I’d like to be able to match some form values from a multipart POST.

Currently, it seems that multipart POSTs are not really supported at all, unless I’m just doing something wrong.

I keep getting the error

Error: Nock: No match for request POST https://my.domain.com/some/path <hex dump of multipart data>

My code:

nock('https://my.domain.com')
    .post('/some/path')
    .reply(200, 'some content');

var form = request.post('https://my.domain.com/some/path').form();
form.append('somedata', 'foo');
form.append('file', fs.createReadSteam('somefile.txt'));

I’ve even tried using the filteringRequestBody trick to return '*', but that did not seem to help at all.

update: ignore the above stuff

I had a bug 😃. But I’d still like to be able to match on specific values in the multipart data. Is this possible?

[lakenen]

@pgte I think my original question was misleading, because I was confused by the error (which was caused by a bug in my code), but I believe the issue is still valid. Here’s some clarification:

I’d like to be able to specify a matcher for form values (multipart or urlencoded). Perhaps something like:

nock('...')
    .matchForm('myField', 'some value') // value matching
    .matchForm('myOtherField', /\d+/) // or as a regexp
    .matchForm('myBinaryField', /* something to denote that a field should be binary data? */)
    .post('/some/path');

Does that sound possible? I don’t know how the other matchers work, so I really don’t know if this is feasible, but I do think it would be very useful!

[nfriedly]

BTW, you can sort of fake it right now - the body is passed in as a string… but sometimes it’s utf-8 and other times it’s hex, depending on if there’s anything binary in the contents.

However, this was good enough for me - run the regex on the body as a string, and if that fails to match, try converting it from hex, and then re-run the regexp.

That said, I too would appreciate better support here. (Ideally with content-type & filename checking for files.) Not sure I’ll be able to add it, but my hat’s off to anyone who does 🎩