Do we know the timeframe for context receivers goi...

# language-evolution

[Marcin Wisniowski]

Do we know the timeframe for context receivers going stable?

[Javier]

Check the keep, the design is not still finished, and context receivers are almost gone, as context parameters remain with the last update.

https://github.com/Kotlin/KEEP/issues/367

[horse_badorties]

I find it difficult to get an overview about "active" proposals I have to say.
When I looked up the current state of context receivers a couple days ago I was only aware of the stuff in master .
Even there it's hard to distinguish between "active" proposals and those already stable/implemented. Some do mention their status at the top of the text, others don't. And you have to explicitly open them individually to check.
Maybe subfolders like "proposed", "experimental", "stable" would help?!

Now I've learned that there are even KEEP branches like context-parameters which adds to the complexity for me. Why not simply put the new proposal in master with an appropriate status?

Having said that - did I get that right: the context receivers proposal is dead and context parameters is the one to watch going forward??

[CLOVIS]

Having said that - did I get that right: the context receivers proposal is dead and context parameters is the one to watch going forward??

Yes. When prototyping, they found many problems with context receivers. Context parameters are a new idea to hopefully avoid these.

[Michael de Kaste]

Sigh, I knew the dangers of designing a library with an experimental feature...

[CLOVIS]

Out of curiosity, what are you using that pauses a problem with the new KEEP? In my case, I found that it requires only minor changes to my plans.

[Michael de Kaste]

Oh no, not that much either, but work is work. 😂 Looks like most, if not all of my features will keep working with minor changes. But I was on the boundary of what DID work since I've personally had the frontend compiler crash on me multiple times because I did 'weird' things with context receivers

[CLOVIS]

Ahah I'm very curious what those weird things are

[Michael de Kaste]

iirc, one of them was predefined object scopes with built-in contexts with delegation https://youtrack.jetbrains.com/issue/KT-61877/IllegalArgumentException-No-argument-for-parameter-VALUEPARAMETER-name-with-object-delegation