Vyxal

lyxal

05:10

2

Risky, 64 bytes \22+\222]+]2:+1+:++!:++2:+*+1+0+[-*-*+02]+]2{*+++**+}+0!:+*+1+0+:+*+1+!!:+++0+0+:+0+0+0!{*++1+0+]+]+]+]+]+]+]+]+]+]+]+]+]+]+]+] Try it online! This is a new language made by @Redwolf Programs. My explanation wouldn't do it justice so heres the Docs. Explanation: Takes the various...

I'm so used to Vyxal that I keep reading this answer wrong

@Ausername how's Jyxal coming along?

Razetime

no new commits yet

A username

Work in progress.

Wasif

@Ausername Jyxal?

A username

I'm three-quarters done on configuring structures

@Wasif Vyxal in Javascript.

Btw can I fix the element-counting in minilambdas?

lyxal

Sure

Transformers will take care of it anyway

A username

05:32

slaps face I forgot to do a global replace :p

lyxal

Well done.

A username

Fortunately, Vyxal strings are valid JS strings, so all I need to do is eval them :p

@Ausername Because of the way I configured modifiers like k, ø, v etc it actually fixed itself :p

A username

06:03

Also because of that, you can have things like vßß⁽ßv⁽⁽⁽ßßvvvvøm treated as one element.

A username

06:13

Can we fix string decompression with single unicode characters in the main version?

I.e. ø returns itself because there is no other character to match it to?

lyxal

@Ausername example?

A username

This returns nothing right now because there's nothing to match it to, can we change that?

I understand how ₌ works now - you dupe the stack, apply one on the stack, apply the other on the stack dupe, and append the top of the dupe to the main stack.

lyxal

@Ausername pretty much.

A username

And with ₍ basically the same thing but you pair them afterwards

lyxal

@Ausername it's literally just ₌<element><element>"

A username

06:31

Yep. I'm half copying what you've done, and half making it up as I go. Jyxal is going to end up like GCC - A lot of undefined behaviour.

Like the multiple conditions in a while loop.

lyxal

mmm wonderful

A username

That actually could turn out to be useful at some point.

I think it's starting to become slightly tacit.

32 mins ago, by A username

Also because of that, you can have things like vßß⁽ßv⁽⁽⁽ßßvvvvøm treated as one element.

A username

07:48

Jyxal's tokenisation is complete!

lyxal

08:01

Very well done

A username

A lot of stuff is just handed to the compiler tho. For example, the CHAR token is anything that matches [vß⁽]*[→←vß\⁽¨ø∆kÞ⁺]?.

lyxal

08:42

A username

09:06

Is this Jelly or something?

Also, you know when I said I had finished the tokeniser? I completely forgot about named functions :p

How do @name:variable|code; and @name:argument_list|code; work?

lyxal

Apr 24 '20 at 7:59, by Lyxal

Before Vyxal, functions were written like so:

@Ausername no, silly, that's some of my planned Transformers

A username

So just parsing the other three is fine for now?

lyxal

Which other three?.

A username

@name|code; , @name:number_of_arguments|code; and @name;?

Also with @name|code;, does it take the no arguments, the whole stack or just what it needs?

lyxal

No args

@Ausername @name:var:var:number|code;

@name:var|code;

A username

09:21

So @name:2:var|code would take two arguments to the stack, and one stored in the variable var?

Also I just realised I completely mistokenised variables 'cos I thought they were 1-character.

Tokenisation is not done by any means.

Can variables have uppercase/numeric characters?

lyxal

Letters and underscores

@Ausername correct

A username

@lyxal So uppercase but no numeric

lyxal

@Ausername correct

Upper, lower and _

A username

09:37

Ok variable tokenisation is fixed

Btw it's been a while, do you think you're going to release v2.4.0 soon?

Also what went wrong with Heroku?

Would @x:2:var:1|code; take two into the function's stack, one into the variable var, and then one more into the function's stack?

lyxal

@Ausername yes

A username

Ok

this could actually be quite useful.

A username

10:11

Just checking, °...; pushes the referenced function to the stack right?

lyxal

10:37

@Ausername correct

lyxal

12:27

₁ƛ₃₅ωkF½*∑⟇ is the new buzz

not much has changed except a single byte lol

lyxal

12:53

gist.github.com/Lyxal/4e377ca7f63d350611dbbd16846a1b1b

Also, I'm re-doing the docs (again)

λ                          = lambda
ƛ                          = lambda that automatically gets mapped
¬ any → num                = logical NOT (non-vectorising)
∧ any, any → num           = logical AND
⟑ any, any → num           = logical AND with reverse arguments
∨ any, any → num           = logical OR
⟇ any, any → num           = logical OR with reverse arguments
÷ num, num → num           = a / b # division
  num, str → [str]         = b split into a even length pieces. There may be an extra part if the string is too long.

that's all for now

o/

@Razetime @hyper-neutrino @UnrelatedString @AaronMiller @Ausername @Underslash @user @wasif what do y'all think of the potential new commands format? Should I stick to the original version or use the new one?

(i really need an easy way to ping everyone)

hyper-neutrino

make a ping bot :p

i like this

lyxal

Better?

hyper-neutrino

so would swap have signature any, any -> any, any?

lyxal

Yes

hyper-neutrino

nice

yeah i think this makes sense

lyxal

13:09

Which do you prefer? (cc everyone)

hyper-neutrino

i like this one better since it is pretty clear that the first pop is a and the second is b

and the current one doesn't really indicate the type of the push

lyxal

I'll see what the rest of vygang has to say. Thanks for the feedback!

Bye for now o/

hyper-neutrino

o/

Razetime