I've been thinking about this exact thing recently. I suppose it's because ⍸ doesn't produce scalars or unsorted vectors?

It'd certainly be more convenient if ⍸⍣¯1 was more lenient - I've found myself writing ⍸⍣¯1{⍵[⍋⍵]} a couple of times. But maybe that'd make it less correct theoretically?

As in, if a function doesn't even produce a certain output, then no input leads to that output.

@RubenVerg @rabbitgrowth Right. (F⍣¯1)r asks "What argument y can be given to F such that it yields the result r?" Anyway, often, what you really want is (⍴y)↑(⍸⍣¯1)r on a Boolean argument, and then you can just write (⍳⍴y)∊r. And if you really want the result to have variable length: (⍳⌈/r)∊r. For this reason, I've been considering monadic ⍷ as "Whence", defined as {⍸⍣¯1{⍵[⍋⍵]},⍵}

Nice name :)

Why the ravel?

@rabbitgrowth That takes care of handling a scalar as a 1-element vector, but also allows any shape collection of indices.

Maybe it should only be 1/⍵ and not ,⍵ to avoid future problems like that of ∊.

@Adám no more Type?

Sure, I've only been considering ⍷ as it is such a good fit. imo, ∊ (or ⍷) is not a particularly good fit for Type.

NARS2000 does monadic antibase, iirc, do you already repurpose that for something else?

Maybe monadic = could work for type. I currently have it in mind for a Rank function (≢⍴) but that function is pretty trivial, though = is nice considering Depth being ≡.

@RubenVerg Yeah, ⊤ is really not related other than resembling a T for Type, and I'd rather use monadic ⊤ for 2⊥⍣¯1

well, Type isn't really related to anything, is it?

maybe to one of the prototype-overfilling functions?

@Adám why specifically binary and not, say, base 10?

Binary is much more common, and it was indeed the meaning of monadic ⊥ that Iverson originally had. I don't know why it wasn't carried over to APL\360.

crazy idea, but how about a Retype function that casts between types?

They don't map to each other.

From char to num is ⎕UCS, but most numbers have no character equivalent. And how do you map from ⎕NULL to number or character? just 0 and ⎕UCS 0?

(Also, I can't imagine a use case.)

from num to char would also be Unicode, I suppose

yeah ok there's no usecase tbf

I guess re-basing an instance on another base class could be… interesting.

am I going crazy? I was sure Unicode had a rho-underbar symbol

(the idea comes from using rho-underbar for Type and Retype akin to Shape and Reshape, which I guess might be conceptually kinda similar)

It does: ϼ

I'd rather not add any more glyphs after ⍛ and ⊇ though.

@Adám ahh, it's not an APL glyph, i see

searched for apl functional symbol rho underbar and got no results

Maybe Type isn't as necessary either, as, like Rank, it is fairly simple: ⊃0⍴⊂

However, I was thinking that a Type primitive could be a bit more lenient about objects than that.

I don't think that simple functions shouldn't be added, especially now with trains

or maybe that's just me wanting to avoid parens at all costs :)

@Adám wdym?

Right, being trivially expressable in terms of other primitives is not a good criteria for exclusion. E.g. who needs + when we have -∘-…

@RubenVerg It currently NONCE ERRORs on all objects except instances of classes that have a niladic constructor. I was thinking that namespaces and classes could be their own types.

@Adám who needs - when we have ⍟⍤÷⍥*

@Adám does that mean the prototype elements of class instances are their class' niladic constructor?

@RubenVerg Exactly.

@RubenVerg No, the prototype elements of class instances are new instances of the parent class, using the niladic constructor.

yeah sorry that's what i meant

Try it online! (updated to better version)

Announcement: BAA Vector Webinar in half an hour ft. Ed Gottsman's J Viewer Demonstration — see britishaplassociation.org/webinar-schedule-2024