I feel that MicroPython's desire to keep up with CPython's language features is coming into serious conflict with the fact that CPython's parser technology is superior (PEG parser since Python 3.9). I believe that that fact alone will increasingly and disproportionately contribute to a widening gap between MicroPython and CPython. For example, I have recently been looking at implementing two syntax features: #8479 (function keyword syntax restriction) and #8503 (decorator syntax relaxation), and really, both of these changes should have almost exclusively been grammar changes, and that is quite far from the case. I know these are a lot of moving parts and variables to consider; I just feel that way, and I am willing to believe that the symptoms I am noticing can be solved through cleaner, more modular code rather than a parser overhaul. My concern is as follows: how effectively can we keep up with CPython's syntax features while mitigating the "spaghettification" of our code?

What do you feel? Too early to tell? Chasing ghosts? Or are we indeed sitting on a timed bomb?

For reference, #7847 (comment) discusses a bit why a PEG parser wouldn't be practical for an embedded system.