This PR sets up the new integrated test/bench suite. It then migrates all benchmarks and some related tests to the new suite. There's also some documentation and some linting. For now, a lot of the old tests are left alone so this PR doesn't become even larger than it already is. Eventually, all tests should be migrated to the new suite though so there isn't a confusing mix of two systems.
20 lines
743 B
Text
20 lines
743 B
Text
|
|
|
|
def tst : IO (Option Nat) := do
|
|
let x? : Option Nat ← pure none;
|
|
pure x?
|
|
|
|
def tst2 (x : Nat) : IO (Option Nat) := do
|
|
let x? : Option Nat := x;
|
|
if x?.isNone then
|
|
/-
|
|
We need the `some` because we propagate the expected type at `pure` applications.
|
|
The expected type is `IO (Option Nat)`, and we elaborate `x+1` with expected type
|
|
`Option Nat`, which forces us the elaborator (to try) to synthesize `[Add (Option Nat)]`.
|
|
If we disable expected type propagation for `pure` we can elaborate it without `some`.
|
|
The `x+1` will be elaborated without an expected type. We will infer the type
|
|
`?m Nat` for `pure (x+1)`, and coercions are used to convert it into `IO (Option Nat)`.
|
|
-/
|
|
return some (x+1)
|
|
else
|
|
return x?
|