This is super-exciting! Two questions: Are there any plans to systematically test that the new parser/transformer doesn’t produce different ASTs from the old parser? For instance, I could imagine comparing the output of -dump-parse on some large corpus of Swift code to s...