Wing Hei Chan

Not trying to discourage you, but I think `~matches` already supports multiple values by supplying multiple groups, like ``` #lang rhombus/static check: values(1, 2, 3) ~matches: 1; 2; 3 ```

I think it would be better if you avoided adding in new functions *for now*; you could import what you need in the test file and leave a note there....

I think it’ll be good if we can at least merge the main part of this PR (the change to `check`). If you’re okay with that, I can cherry-pick that...

No worries, it’s okay if you’d like to further improve it! I just think that putting too many things in one PR makes it more difficult to go through.

I cleaned up and refined this PR’s commits at usaoc/rhombus-prototype@ce73a2f (and shamelessly added myself as the coauthor). If you don’t mind, this PR can reset to that commit and I’ll...

Given the current way Rhombus blocks are expanded, I think the alternative strategy you suggest will require much more significant change, because the expansion uses a trampolining macro, which can’t...

After some more thinking, I’m not sure the Racket example really corresponds to the Rhombus counterpart either. Racket `let` doesn’t work like Rhombus `let` in that the former creates a...

One thing that *is* broken: ``` #lang rhombus/and_meta block: let x = "outer" expr.macro 'm $(id :: Term)': 'block: let $id = "inner" x' m x ``` I think we...

A bigger problem, I think, is that dot providers are currently invoked with a parsed expression as the left-hand side. This means that the dot provider for methods cannot have...

Somewhat relatedly: evaluation of subexpressions in `Real.{at_least,above,below,at_most}` and `{Real,Int}.in`.