Thank for the changes Walid !
We'll do some checks against the Google codebase and we'll let you know if we can go forward with the change.

We're investigating an issue illustrated by the following snippet:

function valueChangesWrapper<T>(control: AbstractControl<T>) {
  return control.valueChanges;
}
const fg = new FormGroup({foo: new FormControl('')});
fg.valueChanges;
const a = valueChangesWrapper(fg);

Notice that the inferred type of valueChangesWrapper inappropriately includes FormControlState. We think this is due to type unification of FormGroup.reset with AbstractControl.reset.

This is super tricky to debug/understand.

I've allowed myself to push the changes to fix the issues that were brought up by testing the fix against Google's code base.

Basically what happens is that changing the reset() signature had implicationg on the TValue generic of AbstractControl.

Hello !

Any update on this PR ? ;)

I'm not super familiar with this area,
Do I understand correctly, that this behavior is something that is implemented, but wasn't supported on the type level?
How would it disambiguate between an object that has a disabled prop, and value props, and setting up state?

TGP is "green".

@kirjs to answer your question, you'd have to do the same as you would declare such control: the state like structure must be defined in a state.

const fc = new FormControl({ value: {value: 'test', disabled: true}, disabled: true });
fc.reset({ value: {value: 'xx', disabled: false}, disabled: true })
console.log(fc.value); // {value: 'xx', disabled: false}

This PR was merged into the repository by commit 610bebf.

The changes were merged into the following branches: main

This issue has been automatically locked due to inactivity.
Please file a new issue if you are encountering a similar or related problem.

Read more about our automatic conversation locking policy.

_{This action has been performed automatically by a bot.}