Is there a reason why both ß and à remove the item from the list? If we want to re-use the list with this minimum/maximum we have to duplicate it first I guess? What are the benefits of removing the item from the list after retrieving min/max? I guess for loops that get the min/max multiple times until the list is empty?

Not sure which of the three would be better: Retrieve min/max and remove it from the list like now; Pop list and push min/max (so the list is gone from the stack afterwards); or retrieve min/max without modifying the list.

So far I would prefer the second option the most with how I've been using it, but I haven't used it in loops yet.

Ignore everything I said above.. I now see Z and W.. >.>

@Mr.Xcoder Is there something to remove all items from one list? So [1,2,3,4,5,6,7,1,2,3,4,5,6,7] and [5,3,3,2,7] to [1,3,4,6,1,2,4,5,6,7]?

Not that I am aware of... But I'll try to help you golf that answer

It's so ugly now with the string replacement.. XD

Yeah it's not a particularly good fit for 05AB1E :p

@Adnan Builtin request which does what I mentioned above ^

(that is, multiset difference)

м doesn't work either >_>

Nope, tried that one before K because I forgot about K.. xD

We'll have to find a workaround for the lack of builtin I guess. One that's shorter than my ðý¹vyõ.; ;)

@KevinCruijssen This has already been changed to popping the entire value and only pushing the min/max to the stack in the rewrite.

@KevinCruijssen I am not sure how you get to that result?

Do you mean like some kind of list subtraction?

Elixir gives the following result: [1, 4, 6, 1, 2, 4, 5, 6, 7]

@Adnan I think it's "a without b's" counting occurences

in this example, remove the first "5", the first two "3"s, the first "2" and the first "7" from a

Yeah, that's what I thought, but the 3 in the result confuses me, so maybe he had something else in mind

a mistake most likely

ohh okay, well in that case it should not be a problem implementing that

I have added it on the todo list

@Adnan Ah, that 3 was indeed a mistake. Should have been [1,4,6,1,2,4,5,6,7]..

It's indeed as @Kaldo stated, a without b with correct occurrences, instead of removing all of them like K or м.

Okay, that should be no problem

👍

It's basically a built-in in Elixir :P

Yeah, that does make it a lot easier hahaha

:D

Oh, one more question, what's the purpose of §? In 05AB1E integers and strings are interchangeable, so why cast an integer to a string? I thought I would be able to use it like this: Input-type is a List (i.e. [1,2,3]), and after § this entire list becomes a string (i.e. '[1,2,3]'). But apparently it becomes ['1','2','3']. What is § used for now in that case?

I can't really think of a use case since string/integer functions are already interchangeable, so why the explicit cast?

Honestly, I have no idea why I added that

I have also never seen it being used so I'm definitely going to make that a 2-byte command

@Adnan If it isn't used anyway, could I propose it can cast a list to a string in that case? (and still a 2-byter) I can at least think of some use cases for that (although limited)

I guess that frees § for a new 1-byter command :)

Hmm, that is a good idea. I'll add that to the rewrite.

@KevinCruijssen Doesn't J already cover this need ?

@Kaldo Mostly, but not entirely. J concats all items together, so [1,2,3] becomes '123' instead of '[1,2,3]'. Not sure if there are any uses for the second though, probably not actually.. As a Java developer I sometimes cast lists to strings and use regex-replaces, but in 05AB1E there are probably shorter alternatives.

Ah well, was mainly wondering why § was there to begin with and what its uses are, but I guess it can be freed as 1-byter command now.

I remember having used § once, not sure in what context though

Another feature-request: binary minimum and binary maximum @Adnan

@Adnan @Kaldo Does either of you know a work-around for "a without b's with correct occurrences" in the current 05AB1E version? As you can see in this answer of mine I'm currently using ðý¹vyõ.;. But it's both long and ugly..

did you check the ǝ command ?

Try it online!

I haven't bothered ordering the inputs to avoid the 3 swaps

@Mr.Xcoder Binary minimum?

@Adnan Binary -> two argument. 7 3(binary min) -> 3

I know it is pair+minimum but for vectorization purposes a two-byte built-in would be far better

Oh okay, I'll add that to the rewrite as well

@Adnan yeah, I'm pretty sure I've used Ƶ in the past

and …...and ' (!)

@EriktheOutgolfer Yeah, that's because I ran all the code through the Elixir parser which parses all strings as a string before interpreting it

so it didn't count those

yeah, some commands have their own syntax

there are double quotation commands: ", “, ‘, ”, ’, •, .• (its own syntax, ends with • instead of .•)

single quotation commands: ', „, …, Ƶ

it's kind of a hell already :)

@Adnan hm, can you make the Elixir parser take the first char, and, if it's ., ž, etc., also take the second char?

I don't think I can do that

I used the parser that the 05AB1E rewrite uses and that just tries to match the correct prefix

> à and ß don’t extract but rather pop the entire element

wait, if I'm not mistaken, didn't Z and W do that?

Those push the min/max without popping

oh right, à and ß are the ones that extract, but isn't popping the same thing under a different name?

ummm...? They don't pop either

the 1 is popped there, no?

I thought you were saying that they pop the list from the stack... One version of them that does that would be useful

@Mr.Xcoder um, I think your request goes in the "Requests" section, not under "Changes and new features"

@EriktheOutgolfer I haven't edit that myself.

Perhaps Adnan added it, I haven't touched the hackmd file in a while

hm, is that something old? because it's edited in under your name

(Victor Dumbrava)

line 111, more specifically

Yeah that was edit more than a week ago, I have changed it