Thank you for the contribution. This looks like a good start, but very much needs tests. There is an open PR that added tests for the accept method (#1373) and one more that further extends the method (#1441). Both need some rebasing or other fixes to get merged. Could I interest you in that? If so, start with the accept PRs (rebase and whatever else is needed), then rebase this PR on top and share the testing infrastructure (the acceptFileDummyInput function).

imo, per-computer filetype validation isn't too trustworthy as it can change a lot from computer to computer (e.g. we have seen .pdf with the type of application/download, www/unknown, application/x-pdf). As a result, I would prefer to stay away from #1441 and its edge cases.

Is there anything that needs to be done in #1373 aside from a squash/rebase?

No, I think that's all there is to it.

Is there any reason why I should do it over you? I am not quite sure I follow the problem =/ The attribution should still be preserved -- here is an example in jscs: jscs-dev/node-jscs@18b4ffc

If you are unfamiliar with how to grab commits from a PR outside of GitHub, here is a gist describing the details:

https://gist.github.com/piscisaureus/3342247

The problem are the 7 commits on that PR, of which at least three produce conflicts during a rebase. I figured you could help resolve those, since you'd likely end up integrating some of those changes into your own PR anyway.

Taking a shot at getting #1373 back on its feet

@jzaefferer @twolfson Please look at #6834.

This PR was opened over a year ago, I no longer work at the organization I started it from, and I no longer have time to contribute to this PR. Closing this PR -- someone can pull over the changes into another PR if they want them