@cmassaro one of the tasks I mentioned during the corner

I'd like to know more about this. The composition language should already support this--at least, currying and partial functions work in our unit tests. Have you made a curried/partial function call that you expected to work, but it didn't? If so, it's a bug; if you have repro instructions, I'll be happy to take a look.

Alternatively, is this about the difficulty of creating such constructs via the frontend interface? If so, it's a larger discussion with the team :)

Maybe https://www.wikifunctions.org/wiki/Z10249 counts as an attempt?

As far as I can tell,

In T386422#10862042, @99of9 wrote:

Maybe https://www.wikifunctions.org/wiki/Z10249 counts as an attempt?

It seems like this combinator works. The other features (e.g. n-ary map and apply) are covered by other tasks. I recommend closing this. Any objections?

Another early attempt that might be related is https://www.wikifunctions.org/wiki/Z10111. I haven't looked at why it's not working.

I don't object to closing.

In T386422#11462543, @99of9 wrote:

Another early attempt that might be related is https://www.wikifunctions.org/wiki/Z10111. I haven't looked at why it's not working.

I don't object to closing.

Ooh, hahaha. Z10111 is very cute but definitely won't work. I'm looking at this:

def Z10111(Z10111K1, Z10111K2):
	return lambda a: Z10111K1(Z10111K2(a))

For that to work, one could make a custom type converter that would convert Z10111K1 into a Python callable. In theory, nothing is preventing this! Someone could absolutely make a type converter that would return an object implementing __call__ (whether a class instance, a lambda, or a function). But doing that with arbitrary Z8s wouldn't work, since the type conversion logic would essentially just be the entire orchestrator.

I'll wait a bit and see if anyone pipes up before I summarily close this.

How would one go about implementing swap then? I've tried this approach a couple times without success, and looking at the implementation for the K combinator is no help, because trying to enter a literal Z8 like it has there just summons a single, non-functional field.

Hmmm, interesting. There shouldn't be any issue from the backend's perspective. It sounds like a UI problem. Do you have screenshots? If so, I would suggest filing a separate task, tagged against WikiLambda and Abstract Wikipedia Team.

In T386422#11635865, @YoshiRulz wrote:

How would one go about implementing swap then? I've tried this approach a couple times without success, and looking at the implementation for the K combinator is no help, because trying to enter a literal Z8 like it has there just summons a single, non-functional field.

I believe support for literal Z8s in the composition editor is the subject of T394458.

In T386422#11638575, @YoshiRulz wrote:

I believe support for literal Z8s in the composition editor is the subject of T394458.

There are a different scenario: A Z8 that itself returns a different callable Z8 (for example, a function accepting a binary function like add and a value like 2, and return an on-demand unary function x=>x+2). I am not sure it is in the scope of T394458: Make it nice to use function closures and possibly anonymous functions in composition editing mode.