Equations
- Lean.Meta.instInhabitedAbstractMVarsResult = { default := { paramNames := default, numMVars := default, expr := default } }
- ngen : Lean.NameGenerator
- lctx : Lean.LocalContext
- mctx : Lean.MetavarContext
- nextParamIdx : Nat
- lmap : Lean.HashMap Lean.LMVarId Lean.Level
- emap : Lean.HashMap Lean.MVarId Lean.Expr
- abstractLevels : Bool
Instances For
@[inline, reducible]
Instances For
@[always_inline]
Equations
- One or more equations did not get rendered due to their size.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- Lean.Meta.AbstractMVars.mkFreshFVarId = do let __do_lift ← Lean.Meta.AbstractMVars.mkFreshId pure { name := __do_lift }
Instances For
Abstract (current depth) metavariables occurring in e
.
The result contains
- An array of universe level parameters that replaced universe metavariables occurring in
e
. - The number of (expr) metavariables abstracted.
- And an expression of the form
fun (m_1 : A_1) ... (m_k : A_k) => e'
, wherek
equal to the number of (expr) metavariables abstracted, ande'
ise
after we replace the metavariables.
Example: given f.{?u} ?m1
where ?m1 : ?m2 Nat
, ?m2 : Type -> Type
. This function returns
{ levels := #[u], size := 2, expr := (fun (m2 : Type -> Type) (m1 : m2 Nat) => f.{u} m1) }
This API can be used to "transport" to a different metavariable context.
Given a new metavariable context, we replace the AbstractMVarsResult.levels
with
new fresh universe metavariables, and instantiate the (m_i : A_i)
in the lambda-expression
with new fresh metavariables.
If levels := false
, then level metavariables are not abstracted.
Application: we use this method to cache the results of type class resolution.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.