False Positive - "bs" isn't a valid Stimulus controller. Did you mean "maps"?Stimulus LSP(stimulus.controller.invalid)

[marcoroth]

False Positive - "bs" isn't a valid Stimulus controller. Did you mean "maps"?Stimulus LSP(stimulus.controller.invalid)

[marcoroth]

Thanks for reporting this one, @fig!

I'm not quite sure how and if we can solve this. But I'm wondering if we could have an option to maintain an "ignorelist" of identifiers for which Stimulus LSP shouldn't report warnings.

[fig]

@marcoroth , I'm afraid I don't have a clue either. 🤷
I'm not very familiar with the internals of the JavaScript LSP space. Would it be possible to scan the internals to register "foreign" JS modules (or whatever they're called). An "ignorelist" sounds like a bit of a shotgun approach, to me.

Now I'm wondering, how does Stimulus deal with naming collisions with third party libraries?

[fig]

Ah, wait! Stimulus requires a data-controller to be declared in a parent element. Maybe this is the seam we are looking for? Stimulus-LSP should check parent nodes. Maybe it already does? Mind you, if we disregard targets outwith a related controller, that kind of negates the purpose of detecting unregistered Targets?
This is not shotgun surgery, It's a circular firing squad!

[marcoroth]

Looking for the data-controller attribute would only work if the controller is within the same file.

This might be given for most cases, but there's also the use-case where you have the controller elements in different views/partials/components.

[Gert-JanPeeters]

Instead of a standard "ignorelist", maybe we can provide the LSP an "ignorelist" on initialization?

This has the benefit that we leave the choice of what to ignore to the client and it also sets up further customization in the future. It does require some more setup...

https://microsoft.github.io/language-server-protocol/specifications/specification-3-14/#workspace_configuration
https://microsoft.github.io/language-server-protocol/specifications/specification-3-14/#workspace_didChangeConfiguration

[christopher-b]

Did anyone work out a workaround for this? Thanks!