Macros have decidedly nonzero cost. Look at the discussion from the core devs surrounding this; just the implementation is a major change for the parsing and AS

Macros have decidedly nonzero cost. Look at the discussion from the core devs surrounding this; just the implementation is a major change for the parsing and AST modules with unclear performance implications. Introducing this much complexity to the language is far from "no cost" as well. I predict that if this PEP goes through, it will be a watershed moment for Python and the language will become an unrecognizable mess (more so than already) within a few years. If GvR was still in charge, I would be sure that this would be dismissed. Now, I am not so sure, and that troubles me.