commercialhaskell / stackage

Stable Haskell package sets: vetted consistent packages from Hackage
https://www.stackage.org/
MIT License
530 stars 805 forks source link

amazonka revised to base < 4.19 ! #7417

Closed juhp closed 6 months ago

juhp commented 6 months ago

base-4.19.1.0 (changelog) is out of bounds for:

juhp commented 6 months ago

See also brendanhay/amazonka#988

juhp commented 6 months ago

Seems this was an intentional change: apparently only a few amazonka packages are currently in Stackage. Though it seems there may be new releases soon that bring amazonka to 9.8 iiuc.

So I will go ahead and disable them for now.

Thanks to @endgame for quickly responding in the upstream ticket

endgame commented 6 months ago

Just to be clear: I don't think there's a lot of benefit to having only amazonka-core and a few niche service bindings building against GHC 9.8. If I'm wrong, please let me know and I can raise bounds, but it doesn't seem that useful to Stackage.

juhp commented 6 months ago

Okay right makes sense - not sure how they got enabled, but my guess is our commenter mass-bounds tool might have re-enabled them, because "they could be": in that sense just disabling (bounding for) amazonka-core could be enough in future?

endgame commented 6 months ago

I must confess I don't understand the words in your comment. amazonka-*-2.1 (when it eventually happens) should support GHC 9.8 and hopefully beyond.

juhp commented 6 months ago
endgame commented 5 months ago

Hmm. It is generally but not universally the case that we'd want base bounds to be consistent across amazonka-*. If amazonka itself had not been broken (because its amazonka-sts and amazonka-sso dependencies were unbuildable), and core services had continued to work (e.g., S3, EC2), I wouldn't want to restrict the packages so broadly.