Try (this-as v (.-props v)) and see what you get

Interop, esp with libs that control things like that, can be fiddly...probably a bug, but not sure where

(this-as v (.-props v)) => undefined 😞 But… It's only a problem in function components. There is no this.

ah right, but that might be the rub...a lot of Fulcro was written when there were only components as classes

So, likely to be a bug or two around that. Fulcro nests its own props under the raw js props, but passes only the fulcro portion through...ah, yeah, I'm hacking "this" in the hooks side, and I even left a comment in the docstring: "

You should *not* use the returned `this` directly,
as it is a placeholder.

Just to be clear, are you using defsc with use-hooks? or are you using the new raw components support with hooks (no defsc)?

defsc with user-hooks, yes

Do you know how to use deps/shadow-cljs to include Fulcro source in your compilation, so you can try hacking on it?

This sort of thing, but with only Fulcro (not all the RAD stuff): https://github.com/fulcrologic/fulcro-rad-demo#running-from-source If you could do that, then I could point you at some things to try

e.g. shadow-cljs -A:f3-dev server with that f3-dev alias in your own global deps.edn, with the correct source path to a clone of the Fulcro repo

If you use IntelliJ you can also add Fulcro as a module to IntelliJ (just hit the + in the Clojure Deps tab and point at Fulcro's deps.edn). This of course is just a convenience...editing the source of Fulcro at this point with any editor will cause shadow to recompile your project with the checked-out source

in configure-hooks-component! you could try using low-level js bind to set this to a js object with props...that's one way of hacking it, since defsc is sort of supposed to act like a class...don't love that...if you see a better way, feel free to explore

I don't have the cycles at the moment to fully analyze

Thanks, I will look into this, if I find the time to do so. For now, I circumvented the problem by using a wrapper component.

@U0CKQ19AQ btw In com.fulcrologic.fulcro.raw.components:692 there is a log/spy :info. Looks like a mistake, and it's too small for an own MR. 🙂

RAD should probably be forcing a re-initialization. There's no reason to preserve the form state in the RAD model.

:thumbsup: can submit a PR if you’d like

great

at your convenience: https://github.com/fulcrologic/fulcro-rad/pull/61

Released

are you sure you can’t use it? I think Cursive may just not recognize it

works for me

the require format is invalid there, you have a vector of deps in place that you can only put one

if we expand the reader macro, you will have something like: (:require [a] [b] [[c] [d]]) (note the [[c] [d]] is what the cljs reader macro will return)

Doh. Thank you! Onwards! 🙂 :man-facepalming: (face-palm)

Remember difference between #?( and Tha splicing variant with @ in it (I don't remember, perhaps #?@( ?)