| Copyright | (C) 2021 QBayLogic B.V. |
|---|---|
| License | BSD2 (see the file LICENSE) |
| Maintainer | QBayLogic B.V. <devops@qbaylogic.com> |
| Safe Haskell | None |
| Language | Haskell2010 |
Clash.Hedgehog.Unique
Description
Random generation of unique variables and unique containers.
Synopsis
- genUnique :: forall m. MonadGen m => m Unique
- genUniqMap :: forall m k v. (MonadGen m, Uniquable k) => Range Int -> m k -> m v -> m (UniqMap v)
- sampleUniqMap :: forall m v. (Alternative m, MonadGen m, HasType v) => (v -> Bool) -> Type -> UniqMap v -> m (v, [Type])
- sampleAnyUniqMap :: forall m v. (Alternative m, MonadGen m, HasType v) => UniqMap v -> m (v, [Type])
- class Bias a where
- sampleUniqMapBiased :: forall m v. (Alternative m, MonadGen m, HasType v, Bias v) => (v -> Bool) -> Type -> UniqMap v -> m (v, [Type])
Documentation
genUniqMap :: forall m k v. (MonadGen m, Uniquable k) => Range Int -> m k -> m v -> m (UniqMap v) Source #
sampleUniqMap :: forall m v. (Alternative m, MonadGen m, HasType v) => (v -> Bool) -> Type -> UniqMap v -> m (v, [Type]) Source #
sampleAnyUniqMap :: forall m v. (Alternative m, MonadGen m, HasType v) => UniqMap v -> m (v, [Type]) Source #
Determine the bias of an item. This is used to set the weight of that item
so we can sample using the frequency generator instead of
element or choice.
Where might you want to introduce such a bias? If there is a collection of
elements where there is a likeliness that real code would use certain values
more or less, we want to be able to capture this. An obvious example of this
is the TyConMap, where without it every constructor would have an even
weighting, when in reality some (like Void# or Addr# are much less
likely to appear in code written by a Clash user).
sampleUniqMapBiased :: forall m v. (Alternative m, MonadGen m, HasType v, Bias v) => (v -> Bool) -> Type -> UniqMap v -> m (v, [Type]) Source #