We extend Memo as to store the original performance execution data, this way, we can display it correctly, and we distinguish the case where a sentence was memoized better.
This is a fix, because before we would display the wrong data for incremental checking.
This is complementary to #686 and #689.
We cherry pick the protocol fixes from #689 as to have this PR in mergeable / testeable shape.
We extend
Memo
as to store the original performance execution data, this way, we can display it correctly, and we distinguish the case where a sentence was memoized better.This is a fix, because before we would display the wrong data for incremental checking.
This is complementary to #686 and #689.
We cherry pick the protocol fixes from #689 as to have this PR in mergeable / testeable shape.