Are primordials safe from monkeypatching the [Symbol.species] method?

[Melab]

I already knew that certain JavaScript builtins can be affected by the value returned by the @@species method, but I was surprised to see how much it can muck with theur operation. Does anyone know if primordials are safe from manipulation of it? If not, is there a way to fix that?

[avivkeller]

CC @benjamingr

[aduh95]

Does anyone know if primordials are safe from manipulation of it?

IMO this is not the question you should ask: primordials are just a tool to let internal code access the original value of the built-ins. Internal code that uses them is guarantee to behave as defined in the ECMAScript spec – i.e. if the ES spec defines a dependency on globally mutable things, primordials inherits this.

The more interesting question is "are Node.js internals affected by users mutating things?", and when the response is "No", primordials might be a solution – but it's not a silver bullet, they have limitations (which I tried to document in primordials.md) and using them have downsides (performance, code readability).

So, if you can report a feature of Node.js that's broken when the a @@species getter is modified, it would certainly be interesting to investigate, and the solution may or may not involve using more primordials. But it's a problem that would very much require a case-by-case solution, not a general thing.

If not, is there a way to fix that?

Fixing it is not a goal – with maybe the exception of Safe… primordials, which purposefully deviate from the spec to be less reliant on globally mutable objects/methods.