I tried to figure out why one of my tactics is slow. Here is an example run for one call to the tactic:
Monad.Raw-monad._⟨$⟩_ {equivalence-relation⁺} {equality-with-J} {lzero} {lzero} {_52 h hyp xs} (_53 h hyp xs) {ℕ} {ℕ} suc (_59 h hyp xs)The tactic blocks on meta-variable
_52
.Monad.Raw-monad._⟨$⟩_ {equivalence-relation⁺} {equality-with-J} {lzero} {lzero} {λ v → _78 v} (_53 h hyp xs) {ℕ} {ℕ} suc (_59 h hyp xs)The tactic blocks on meta-variable
_78
. Note that _52
seems to have been η-expanded to λ v → _78 v
.Monad.Raw-monad._⟨$⟩_ {equivalence-relation⁺} {equality-with-J} {lzero} {lzero} {Maybe {lzero}} (Maybe.raw-monad {equivalence-relation⁺} equality-with-J {lzero}) {ℕ} {ℕ} suc (_59 h hyp xs)The tactic blocks on meta-variable
_59
. Note that this meta-variable was part of the first term mentioned above.It seems wasteful for the tactic to repeatedly reduce and traverse the input in this way. Some suggestions for improvement:
_52
, _53
and _59
. The tactic should not be restarted until all of these meta-variables have been solved.RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4