Stackage: Why isn't newtype-generics included in the new nightly with GHC-8.4.1?

[sjakobi]

newtype-generics builds and tests cleanly with nightly-2018-03-13.

[DanBurton]

I wanted to get a nightly out there quickly, even if it is much smaller than usual. I used some rough heuristics and removed wide swaths of packages. Much too wide, but I was aware of this. We will be working on re-enabling as many packages as possible over the next few days.

[sjakobi]

Oh, ok. I'll leave this open so other package maintainers will also see your explanation.

[sjakobi]

newtype-generics builds and tests cleanly with nightly-2018-03-13.

Turns out I missed a non-zero exit code and tested with lts-10! :)

Here's how stack init fails with the new nightly:

$ stack init --resolver nightly --force
Looking for .cabal or package.yaml files to use to init the project.
Using cabal packages:
- ./

Selected resolver: nightly-2018-03-13
Selected resolver: nightly-2018-03-13
The following packages specified by flags or options are not found: persistent-sqlite yaml

Not sure if this is a stack bug or an problem with the snapshot.

[fosskers]

I'm also seeing that failure (re: persistent-sqlite).

[sjakobi]

The following packages specified by flags or options are not found: persistent-sqlite yaml

This was fixed in https://github.com/commercialhaskell/stack/pull/3916.

[DanBurton]

A stack minor release is underway, but the impatient can tell stack to upgrade itself from source.

stack upgrade --git --git-branch stable

[DanBurton]

Note that newtype-generics still hasn't made a release to hackage with bounds that allow for base-4.11. That seems to be the only reason for its exclusion at this point. I've opened an issue: https://github.com/jcristovao/newtype-generics/issues/21

[sjakobi]

Note that newtype-generics still hasn't made a release to hackage with bounds that allow for base-4.11.

Oh, that's a bit embarassing! :/

I've uploaded a new release!

[DanBurton]

It's in the build plan now; I expect it to appear in the next nightly.

[sjakobi]

Thanks a lot for your work, @DanBurton! :)