d4b08fcf96
Unification constraints of the form
ctx |- ?m[inst:i v] == T
and
ctx |- (?m a1 ... an) == T
are delayed by elaborator because the produce case-splits.
On the other hand, the step that puts terms is head-normal form is eagerly applied.
This is a bad idea for constraints like the two above. The elaborator will put T in head normal form
before executing process_meta_app and process_meta_inst. This is just wasted work, and creates
fully unfolded terms for solvers and provers.
The new test demonstrates the problem. In this test, we mark several terms as non-opaque.
Without this commit, the produced goal is a huge term.
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
30 lines
693 B
Text
30 lines
693 B
Text
Set: pp::colors
|
||
Set: pp::unicode
|
||
Imported 'macros'
|
||
Imported 'pair'
|
||
Imported 'subtype'
|
||
Imported 'optional'
|
||
Using: subtype
|
||
Using: optional
|
||
Defined: sum_pred
|
||
Defined: sum
|
||
Proved: sum::inl_pred
|
||
Proved: sum::inr_pred
|
||
Proved: sum::inhabl
|
||
Proved: sum::inhabr
|
||
Defined: sum::inl
|
||
Defined: sum::inr
|
||
Proved: sum::inl_inj
|
||
Proved: sum::inr_inj
|
||
Proved: sum::distinct
|
||
sum3.lean:96:0: error: invalid tactic command, unexpected end of file
|
||
Proof state:
|
||
A :
|
||
(Type U),
|
||
B :
|
||
(Type U),
|
||
n :
|
||
sum A B,
|
||
pred :
|
||
(proj1 (subtype::rep n) = optional::none) ≠ (proj2 (subtype::rep n) = optional::none)
|
||
⊢ (∃ a : A, n = sum::inl a B) ∨ (∃ b : B, n = sum::inr A b)
|