Meeting URL: https://meet.jit.si/StableHaskellMeetBiWeekly
Previous meeting notes
- How can we support the stable api work
- Reach out in solidarity for the effort.
- Invite to next meeting.
- Holding the token: Trevis
- Nightly builds from email
- How widespread is the desire for nightlies?
- Maintenance resources are finite
- Best effort currently; cannot promise dependeable nightly builds without more resources
- One cause of failures: patches land, based on testing on a few platforms. Nightly build fails because the patch doesn't work on some other platform/configuration. It's good to have daily knowlege of this happening; not so good to have to guarantee that every nightly build works on every platform/configuation.
- Maybe we could say
- Dependable nightlies for some platforms (the ones we use in per-MR tests); e.g. certain Linux distributions Ubuntu/Debian/Fedora; Windows; mac(?)
- Non-dependable versions on a best-effort basis, but may well be broken.
- Nightlies version of platforms
- Not wanting to add stress to team!
- How widespread is the desire for nightlies?
- GHC Steering committee Stability pre-proposal;
- Relates to: GHC proposal #601.
- Relates to:
-Wsevere
proposal - Re the
-Wsevere
proposal, Chris proposes the following tweak- New language extension:
-XWombat
implies-Werror=severe
- Then, later
GHC202x
implies-XWombat
- Why a new language extension? We could just say
GHC2024
implies-Werror=severe
. But generally GHC20x doesn't affect warning settings, so maybe this would be a change in policy? - This relates to conversations here:
- New language extension:
- None
-
Bulletin discussed previously
- Previously holding the token: Chris
-
GHC warning policy document as discussed previously
- Previously holding the token: Chris
-
Creating and maintaining a set of hlint rules to promote stability
- Previously holding the token: Trevis
- Opened issue asking if we can write rules for typeclasses.
- Template Haskell