Cryptol.TypeCheck.Solver.Class

Solving class constraints.

Documentation

solveZeroInst :: Type -> Solved

Solve a Zero constraint by instance, if possible.

solveLogicInst :: Type -> Solved

Solve a Logic constraint by instance, if possible.

solveRingInst :: Type -> Solved

Solve a Ring constraint by instance, if possible.

solveFieldInst :: Type -> Solved

Solve a Field constraint by instance, if possible.

solveIntegralInst :: Type -> Solved

Solve an Integral constraint by instance, if possible.

solveRoundInst :: Type -> Solved

Solve a Round constraint by instance, if possible.

solveEqInst :: Type -> Solved

Solve Eq constraints.

solveCmpInst :: Type -> Solved

Solve Cmp constraints.

solveSignedCmpInst :: Type -> Solved

Solve SignedCmp constraints.

solveLiteralInst :: Type -> Type -> Solved

Solve Literal constraints.

solveFLiteralInst :: Type -> Type -> Type -> Type -> Solved

Solving fractional literal constraints.

solveValidFloat :: Type -> Type -> Solved

This places constraints on the floating point numbers that we can work with. This is a bit of an odd check, as it is really a limitiation of the backend, and not the language itself.

On the other hand, it helps us give sane results if one accidentally types a polymorphic float at the REPL. Hopefully, most users will stick to particular FP sizes, so this should be quite transparent.