Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Updated flake #5

Draft
wants to merge 28 commits into
base: wavewave-flake
Choose a base branch
from
Draft

Updated flake #5

wants to merge 28 commits into from

Conversation

sellout
Copy link
Member

@sellout sellout commented Dec 15, 2022

Lots of refactoring.

This also rebases on upstream master, to get rid of our downstream dependency on concat-plugin.

sellout and others added 23 commits August 11, 2022 12:55
Maybe easier to maintain than Travis?
This moves tests that depend on the plugin from concat-examples to
concat-plugin. The one shortcoming, IMO, is that it requries duplicating
the SMT flag.
Experiments with encountering FunCo

Minor cleanup

Introduce polarity parameter

Stash passing coercion arguments

Undo [Coercion] passing

Implement AppCo properly

Improve comment

use liftCoSubstWith (didn't seem too hard)

Less warnings
Following the change from Coercible to RepCat in many cases.
Co-authored-by: Greg Pfeil <greg@technomadic.org>

Explicit support for ghc 8.10.7, ghc 9.0.2, ghc 9.2.4.  Some support
for older versions has been explicitly removed.

Mostly updates to the new ghc API, with provisions for Linear Haskell.

Also, new hack to prevent ghc from inlining toCcc''.

The golden tests give slightly different results depending on the ghc
version (coercions seem to move around), so their output files are now
in different directories according to that version.
This is needed for derivatives that go to a different target category
like Syn - the crossF results from the derivative of fmapC.
From a patch by Raoul Schlotterbeck.
Just a matter of adding FlexibleContexts.
This brings FunCo to the top in many cases.
lam Let can occasionally do a better job.
Moving lam let compose moved a few subexpressions.
@sellout sellout force-pushed the updated-flake branch 6 times, most recently from 09ccc3c to 240baf5 Compare February 20, 2023 22:58
... rather than test suites, as that's what they are.

Also, no sensible way to run these in CI.
@sellout sellout force-pushed the updated-flake branch 3 times, most recently from 545e93c to b94caec Compare March 1, 2023 17:34
wavewave and others added 3 commits March 11, 2023 18:16
They really are – they don’t run tests, and are hard to use without human interaction.
Mostly because Haddock thinks commented-out guards are docs.
- add a Nix formatter;
- add a default package & devShell;
- restructure the overlays;
- move the inputs to the end;
- don’t repeat the GHC version list;
- move lib into its own file (and add more functions);
- add nixConfig;
- provide a “Haskell overlay”;
- disable GHC 9.4 and HEAD builds, as they fail;
- enable all packages now that they work with newer GHC (except for concat-hardware);
- enable `checkPhase` for all Cabal packages;
- replace string passing with a package set;
- extract "ghc902" to `defaultCompiler`;
- add a Garnix config to enable darwin builds; and
- expose executables from Cabal packages.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants