Dmitry Iv.

@evanw can we please consider [this PR](https://github.com/evanw/esbuild/pull/513)? UMD for esbuild is such a great pain and time killer atm. There's no simple clean ways to do that.

There's also [signal-struct](https://github.com/dy/signal-struct) for comparison. It's not based on proxies, essentially just seals initial passed object.

@rreusser do you have an idea for the API? I can imagine something like ```js import {Panel, Number, Color, Group} from 'control-panel' ... ```

@abbr could you please react on the PR #52? I faced that issue in [glslify-sync](https://github.com/vorg/glslify-sync), it is quite acute: ``` js setTimeout(function () { var source = glslify('./file.glsl'); }); ```...

The one by Bhaktivinoda Thakur https://www.amazon.ca/Srimad-Bhagavad-Gita-Bhaktivinoda-Hardcover-Thakura/dp/8184030118 There's also some references to check against: http://bhaktivinoda.com/the-commentaries-on-bhagavad-gita/

Thank you for having a look @josh-hadley, rescaling to 2048 units per em improved quality and fixed errors. I think to avoid such situations it's recommended to run some [linter/validator](https://adobe-type-tools.github.io/afdko/AFDKO-Overview.html#23-validation)...

Rescaling doesn't seem to have helped. Here's an example of the same error with 2048 upm. All values are integers more than 1. [wavefont.zip](https://github.com/adobe-type-tools/afdko/files/6055829/wavefont.zip)

It has something to do with points order in glif. In the original order the first point is `line`, moving that first point to the end fixes the issue. ```diff...

As well as [alpinejs](https://github.com/alpinejs/alpine/blob/main/packages/alpinejs/src/evaluator.js#L81), [petite-vue](https://github.com/vuejs/petite-vue/blob/main/src/eval.ts#L20) and any other frameworks of that kind - bundling full evaluator (like [subscript](https://github.com/dy/subscript)) would be heavy, not as performant and not much safer. I wonder...

Ok, sprae@10 supports safe-eval via justin.