Elia Schito

Confirmed on v1.0, trying to replicate this also uncovered the fact that spec coverage for irb-mode is not that great at the moment. Thanks for reporting it!

Actually `require_relative` has a slightly different behavior, it requires relative from the file in which it's used (vs. the current dir). E.g.: _`require "./foo"`_ ``` ⤑ cd /tmp/ ~ ⤑...

Yeah, sorry, I guess we all agree, I initially misunderstood your reply 😊

😞 node doesn't implemnt btoa/atob, it uses has `Buffer` class instead (https://stackoverflow.com/a/23097961). Still undecided if it's worth removing the polyfill for the browser 🤔.

_Notes from the core-team meeting of 2022-06-01_ ```rb # - ::JS generic opal/javascript API (e.g. `::JS.function`, `::JS.global`, `::JS.new`, …) # - ::Opal runtime opal/javascript API (e.g. `::Opal.bridge`, `::Opal.hash2`, …) #...

depending on the environment in which the code is run you can also try with https://github.com/evanw/node-source-map-support or other project that will sort out the mapping and backtrace for you… Another...

@Mogztter sorry for the delay on this, it's a big thing and I think we can surely pursue it, but not hastily. Looks much like the symbol tables of ruby...

@Mogztter I ran [a quick experiment](https://github.com/opal/opal/compare/elia/symbols-experiment) storing symbols in `Opal.$`, calling method doesn't work but this is how compiled code looks like: ```rb opal:elia/symbols-experiment ⤑ bin/opal -e 'def asdf; 123;...

> `#allocate` does `Object.create(JSConstructor.prototype)` This makes sense! > and then `#initialize` does `JSConstructor.call(self, ...args)`. Yes! according to the [new operator](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Operators/new#description) docs) that's what happens, and we may decide if we...

Seems that some newly supported syntax is blocked at compiler level: https://travis-ci.org/opal/opal/jobs/35417114#L173 using `bundle exec rake mspec_node` should give you a more meaningful stacktrace to identify the offending line