Jacqueline Firth

Good points. The first point can be addressed by analyzing the pattern and seeing if it would have a different meaning under syntax parse. We could also just restrict it...

It's not that impossible. I made a [`known-not-false`](https://github.com/jackfirth/resyntax/blob/081a31d99096aacdaa8f782171254266379d6de6/default-recommendations/private/boolean.rkt#L22-L23) syntax class for this purpose. We could just add `(weak-box-value _)` to the list of recognized patterns. Then I think the rule...

Actually, looking at this more, I think your original code can be simplified further down to this: ```scheme (weak-box-value (hash-ref! element-cache key (lambda () (make-weak-box (make-cached-element style content key))))) ```...

One tricky part here is how to display a suggested fix that adds imports. A fix on line 200 with an added import might want to insert the import on...

Found another instance where this would help: in [this code](https://github.com/racket/racket/blob/837c819ce1caf734ee90f38c76b8b76eda11ac4e/racket/collects/pkg/private/addl-installs.rkt#L109-L117) after replacing the `(apply append (for/list ...))` with a single `(for*/list ...)`, the `map` can be fused into that `for*/list`.

Hmm. Yeah that's not ideal. There's two options: 1. Make `--package` install the package when it's not already installed 2. Make `--package` error with a much better error message when...

The analyze-mode v.s. fix-mode distinction being a flag is definitely something I want to avoid, as it's very unlikely they'll remain as similar as they are now over time and...

The reason I want to avoid `--fix` and `--analyze` flags is because I'm certain that eventually we're going to add features to fix-mode that don't make sense for analyze-mode. For...

So `resyntax --file foo.rkt --fix` wouldn't work? I'd rather not break the intuition that `--flag` arguments are unordered.

Yes, that would work.