effectfully
effectfully
Currently there’s no way to use the compilation machinery from `plutus-core/plutus-core/stdlib/PlutusCore/StdLib/Type.hs` with SOPs, meaning we don’t have any heavy tests for SOPs in `plutus-core` or `untyped-plutus-core`. We should have some....
Currently the following UPLC pseudocode: ``` let y = delay e in f (\_ -> force y) + g (\_ -> force y) ``` will evaluate `e` twice if both...
`PlutusIR.Test.compileAndMaybeTypecheck` type checks the given IR term, compiles it and infers the PLC type of the result. For better test coverage it would be good to establish the relation between...
Plinth builtins don't know anything about semantics variants, meaning Plinth code evaluated as Haskell may give a different result compared to what it gives when evaluated as one of the...
On current `master` in the [`builtinListIndexing.pir.golden`](https://github.com/IntersectMBO/plutus/blob/5ee29364a7a146ee025e94a275bb01360613447f/plutus-tx-plugin/test/Budget/9.6/builtinListIndexing.pir.golden) test we have the following: ``` chooseList {data} {Unit -> Unit -> data} xs (\(ds : Unit) -> error {Unit -> data}) (\(ds :...
This issue is for discussing how the Plutus team should prioritize work. Everybody from @IntersectMBO/plutus-core is encouraged to contribute, but we are interested in opinions of external people as well...
Tests for builtins are very weak, we should do much better and try to generate random built-in functions to cover various corner cases of the evaluator. Things to do: -...
We should try using [tasty-papi](https://hackage.haskell.org/package/tasty-papi) for longitudinal deterministic benchmarking. It measures CPU instructions rather than time, so not perfect, but the determinism part is great, so overall probably a good...
We have a number of tests showing the distribution of generated values, for example ```haskell test_listDistribution :: Int -> ([()] -> Gen [[()]]) -> Int -> TestTree test_listDistribution numRuns split...
We have the wonderful nightly test suite run which upscales `hedgehog` tests by 50x (which then gets downscaled back in individual tests when 50000 would take way too long). We...