Neil Fraser

Nope. I got a fragile, but consistently reproducible test case on my machine, and was in the process of debugging it. But then got assigned other work. This bug still...

This PR would break this product since raw HTML would be printed on the user's screen.

Ok, that's a real problem. Babel (at least as run in JS-Interpreter's [Babel demo](https://neil.fraser.name/software/JS-Interpreter/demos/babel.html)) is not providing the required polyfills. That's completely against the purpose it is intended to serve....

I agree with the approach you outline. It's the same as that used by String's .replace function which has native code wrapped by a polyfill: https://github.com/NeilFraser/JS-Interpreter/blob/master/interpreter.js#L1759 There are no current...

When run on https://neil.fraser.name/software/JS-Interpreter/ in both uncompressed and compressed modes, I get `changed!`. Can't recreate. Can you double-check?

Whoa, it works fine (`changed!`) on Firefox, it fails (`default`) in Chrome. Was not expecting that!

This is getting weirder and weirder. Take this screenshot from Chrome's console: Here the exact same commands are entered twice. The difference is that the lines above are copied and...

Filed a bug with Chrome console: https://issues.chromium.org/issues/327467399 Let's put this issue on ice until we can get Chrome to be consistent (one way or the other). Firefox behaves as expected....

The bug in Chromium has been fixed by the team (their comment was "Amazing bug!😂"). I've tested it using the current release of Chrome and I believe this bug has...

Be careful. This PR is untested and will break the product **if** the div's content is intended to be HTML.