Figuring out the representation of a Plutarch type

We've discussed before how Plutarch types are merely tags and don't have a direct connection to their runtime representations. It's important to be able to intuitively figure out the runtime representations from the data type declaration though. This is why most types follow certain conventions.

The representation can only be one of two categories: builtin and Scott encoded. All trivial builtin types are already defined in Plutarch: PInteger, PByteString, PString, PBool, PUnit, PBuiltinList and PBuiltinPair.

Now, let's discuss patterns of data declarations and what representation they should hint at:

• If it's an ADT that derives PlutusType with DPTStrat _ = PlutusTypeScott, then the ADTs will be scottencoded. This is what you generally want for non-trivial types that are not stored in datums or redeemers.

  e.g. PList derives PlutusType generically and is represented with Scott encoding.

• If it's an ADT that derives PlutusType with DPTStrat _ = PlutusTypeData it's data encoded. Particularly, it's a Data value - which is part of the builtin types.

  e.g. PScriptContext derives PlutusType using DPTStrat _ = PlutusTypeData

• If it's a representationally equal wrapper (think Haskell newtype) to a term containing a Plutarch type - it should have the same representation as that underlying Plutarch type.

  e.g. newtype PPositive (s :: S) = PPositive (Term s PInteger) is just represented as PInteger. This is ensured by deriving PlutusType with DPTStrat _ = PlutusTypeNewtype.