Geoffrey Booth

All of the possible words are listed here: https://github.com/jashkenas/coffeescript/blob/ed6733d17746eeafca415cc261cce92e6f840ff6/src/lexer.coffee#L1225-L1264 We can’t add any more reserved keywords without breaking changes (which basically can’t happen anymore). These are mostly spoken for, and...

Hi, I wrote https://github.com/nodejs/node/issues/36396. It’s been my intention to ship an ESM CoffeeScript loader as part of this package, since this package already includes a CommonJS loader (in `coffeescript/register`). There’s...

> This seems like a good idea to me; it matches the rule of Assign. > > (Hope you don’t mind the necro.) Yes, it’s fine, do you mind rebasing...

> I would make almost the opposite proposal: keep using `var`, and push the declaration to the first assignment to the variable within the function scope. The intent of the...

I thought there was already something similar to the “two pass” approach happening in order to output the `var` line itself. As the nodes are being assembled, we keep track...

> Perhaps that could be merged first, and this issue could be left for those who want to beautify the output in some cases further? Sure, we can at least...

Is this the same as class properties, that are only in stage 2? See https://github.com/jashkenas/coffeescript/issues/4497#issuecomment-293724531

We can add support, but the compilation output would need to be Stage 4 ES (i.e. ES2017 or below). In other words, our version would need to resemble the Babel...

Here’s a good [overview of current ES proposals](https://blog.logrocket.com/interesting-ecmascript-2017-proposals-163b787cf27c). Some CoffeeScript-inspired ones on there. It covers the class-related ones that are in progress, including how they’ve been changing from stage to...

We’re a long way from the days of `with` and the other original “bad parts.” The standards groups working today seem pretty solid to me, and I think our default...