Contextual helpers

[cibernox]

rendered

[cursedquail]

What happens to the existing syntax for invoking helpers, which for this example would be:

{{my-component format-date=(moment-format)}}

?

[cibernox]

that would invoke the helper which generates a string that is passed to the component, like today.

[knownasilya]

Yes please, and then deprecate {{my-helper}} so we don't have that ambiguity.. is it a value, component, or helper?

[miguelcobain]

It can be any of those, but there are priorities in its resolution. value, helper and then component, I think.

[knownasilya]

Yes, but from a developer point of view, it's ambiguous. Especially for developers new to Ember.

[miguelcobain]

A bit unrelated to the RFC, I guess, but I'm not sure how to solve that, using handlebars syntax.

[machty]

There's an inconsistency in helper lifespans that I've noticed in a few other cases that we might want to consider as part of this RFC: https://ember-twiddle.com/d869daef1f0c2c8e2808a8e60dc06ca3?numColumns=2&openFiles=templates.application.hbs%2Chelpers.count-up.js

[ming-codes]

An helper is mostly just a function. Can the problem be solved by simply yielding a function (or action)?

{{yield
  (action "capitalize-str")
}}

{{#parent-component as |capitalize-str|}}
  {{child-component helper=capitalize-str}}
{{/parent-component}}

Only down side I see is that you can't create a helper that maintain it's own state.

[cibernox]

@ming-codes I could, and it's what I'm doing for now, but there is no built in way of invoking that helper in the template, so I have to use ember-composable-helper's compute utility.

This proves that it's technically possible, but:

• Syntax is 👎
• It doesn't support stateful helpers
• It doesn't support, like in contextual components, to override passed default values

Hence, I think there is room for a feature here.

[knownasilya]

Does #432 supersede this RFC?

[cibernox]

@knownasilya maybe? I haven't seen that until now.

[chancancode]

As others have pointed out, I think the design and exploration here have been absorbed into #432. Let's move the discussion there. Thanks again @cibernox for pioneering the work in this area!