Generic derivation overhaul (with configurability)

[travisbrown]

Supersedes #163 and #79. Also fixes #69, although I'm not entirely sure why. Still needs tests before it gets merged.

[travisbrown]

The build is failing because of an interaction between macro-compat and sbt-unidoc on 2.10 that I'll investigate later.

[codecov-io]

Current coverage is 77.24%

Merging #164 into master will decrease coverage by -0.26% as of 1d93c8f

Powered by Codecov. Updated on successful CI builds.

[travisbrown]

For what it's worth, these changes may speed up compilation a bit (at least they do on my machine, from over a minute for tests/test:compile to under 50 seconds).

[travisbrown]

Quick update: there's a problem here for e.g. the following nested case classes:

case class Foo(someField: Int)
case class Foos(foos: List[Foo])

In this case if you ask for a snake-case decoder for Foos the decoder it uses for each Foo won't be configured to use snake case.

I'm working on a fix that switches the inheritance relationship between Decoder and ConfiguredDecoder, and will update the PR soon.

[wennergr]

Any updates on this? Would love to see this PR move forward ;)

[sveinnfannar]

Really looking forward to seeing this released :)

[travisbrown]

@wennergr @sveinnfannar Unfortunately configurability won't be in 0.4.0, which will be out very soon (possibly today), but 0.4.0 does include an overhaul (see #247) in preparation for configurability that's bringing huge improvements in compile times and stability—see the 0.4.0-RC1 release notes, the link here, etc.

I'm hoping to have 0.5 out with configurability in the next couple of weeks, but no promises yet.

[sveinnfannar]

Thanks for the update @travisbrown!

[dvic]

Any updates on this (in light of the 0.5.x releases)? Will this land in 0.6.x?

By the way, thanks a lot for this awesome library 👍 !

[travisbrown]

Superseded (mostly) by #429.