From/ToFormUrlEncoded (Do not merge)

[jkarni]

Fixes #23. This isn't ready yet. Some questions:

• It turns out To/FromFormUrlEncoded don't actually expect/return url-encoded forms. Encoding and decoding happens before/after. To me this means their names should be changed to To/FromForm, but I wanted to see what others thought.
• I imagine using to/parseQueryParam in the generic instance is more correct than to/parseUrlPiece?
• The generic instance returns the underlying parse error, but not yet the key for which parsing failed. I would say doing that could be done in a separate PR. It involves changing the signature of gFromForm though, and maybe that shouldn't be considered internal.
• I'm not very familiar with GHC.Generics and generic programming, so there may be issues here.
• There should probably eventually be an options for the generic stuff.

I still need to check sum record types, squash some commits, and move the tests for encoding/decoding over.

[cdepillabout]

I guess this PR has bee stalled for a while, but is there anything I can do to help push it along?

I've wanted something like this for the last couple months, and I finally created an issue about it over on the servant repository, but I didn't realize it was already being worked on!

[jkarni]

@cdepillabout I just finished up some more tests, squashed, and cleaned up a little. Do you want to try rebasing against master? It's a pretty ugly merge/rebase!

[jkarni]

Relatedly, while moving the encoding and decoding functions over I noticed that they seem to be doing much more converting between String, ByteString and Text than is healthy. At some point we should try to improve on that.

[cdepillabout]

@jkarni Sure, I'll do the rebase onto master. You might have to give a few days though.

[cdepillabout]

@jkarni: I've rebased your code onto master. You can find the updated code here:

#31

[jkarni]

This too has been superseded by #32