Vyxal

emanresu A

00:02

@user NO.

lyxal

@hyper-neutrino well just like how I'd RO TNB, it's fine to have off-topic here in small doses, so long as serious on topic isn't being disrupted

There doesn't seem much point for a room like this to have a branch off room

Because there's barely enough "off-topic" to even move

emanresu A

Yeah, although we do kinda rule Trash

Vyxal bot is afaik the #1 user in trash

lyxal

Only because of github messages

emanresu A

Yep lol

lyxal

Which are technically on topic lol

Anyhow I would say move to other rooms as you see fit

Y'all are room owners for a reason lol

I unironically trust y'all to make sound decisions about things like this

And self deprecate all you want about the above message - that won't change my belief that this project is just as much yours as it is mine

^ has always been my intention

That ownership isn't just me. That it's all of you

That everyone is equally important and in-control of Vyxal and its subsidiaries

So RO this room as you would your own

And enjoy being part of such a large and awesome team

Because y'all are a bunch of epic gamers

3

hyper-neutrino

00:33

@lyxal i appreciate the sentiment but i unironically cannot think of anything i've contributed to the vyxal code base besides one utility function and I don't even know if it still exists

:P

lyxal

@hyper-neutrino that doesn't matter. There's more than one way to contribute

Unrelated String

i think he's just talking about the community in general

lyxal

Code isn't the only thing vyxal is built on

Unrelated String

because i have done literally nothing except maybe give some feedback on builtins

hyper-neutrino

i mean, yeah, esp vyxapedia (does anyone even use that?) and the bot, but i'm just saying like

lyxal

00:34

It's built on community interactions and collaboration

hyper-neutrino

that's true

emanresu A

Vyxal is you. And me. And all of us!

3

Seggan

@hyper-neutrino vyxapedia launched me off the ground

user

01:03

@emanresuA This feels like it's from some TV show but I can't place it

hyper-neutrino

@Seggan oh, damn, nice

i might put more effort into it in that case if it's actually helpful

:P

Steffan

02:49

it's a little outdated now

!!/repos

!!/repo list

Vyxal Bot

- Vyxal: https://github.com/Vyxal/Vyxal
- Vylight: https://github.com/Vyxal/Vylight
- Vyxal-Events: https://github.com/Vyxal/Vyxal-Events
- VyxalBotSE: https://github.com/Vyxal/VyxalBotSE
- Sandbox: https://github.com/Vyxal/Sandbox
- vyxapedia: https://github.com/Vyxal/vyxapedia
- VyxalS: https://github.com/Vyxal/VyxalS
- Jyxal: https://github.com/Vyxal/Jyxal
- Philosophy: https://github.com/Vyxal/Philosophy
- Symvy: https://github.com/Vyxal/Symvy
- .github: https://github.com/Vyxal/.github
- SE-Bot: https://github.com/Vyxal/SE-Bot

lyxal

07:18

@hyper-neutrino I used your flask Web app set up for the vyxal online interpreter so that's something you've contributed

hyper-neutrino

08:00

oh right lol, pog

i'd comment on it being terrible code but if it's worked without issues for you this whole time i guess i can consider it to work lmao :P

emanresu A

Lol

user

10:41

@hyper-neutrino i copied some of it for a website I’m making and it works great

Vyxal Bot

11:39

chunkybanana opened issue #922 in Vyxal/Vyxal: Modifiers in compressed strings are somehow misparsed

lyxal

I love it when I screw up the parsing

Real fuj

*fun

Vyxal Bot

12:39

Lyxal created branch Vyxal/catch-all-literals-in-parsing

Lyxal opened PR #923 (Vyxal/Vyxal) (Vyxal:catch-all-literals-in-parsing → Vyxal:main): Make parser account for _all_ literals

Vyxal Bot

12:55

[Discord Relay]
Seggan#8111: *uses antlr*

Lyxal created branch Vyxal/make-to-base-work-with-negative-numbers

Lyxal opened PR #924 (Vyxal/Vyxal) (Vyxal:make-to-base-work-with-negative-numbers → Vyxal:main): Make τ work with negative bases

Vyxal Bot

13:17

Lyxal created branch Vyxal/add-documentation-for-string-overload-of-ṅ

Lyxal opened PR #925 (Vyxal/Vyxal) (Vyxal:add-documentation-for-string-overload-of-ṅ → Vyxal:main): Add the string overload of ṅ to the YAML

user

@VyxalBot tbh we could use ANTLR for Python Vyxal too so there's only one file that you need to use for all the implementations of Vyxal

That way, when there's a syntax change, you don't need to manually update every parser

lyxal

hmmm

if only there was a branch that did that

Seggan

stares at old antlr branch

ye i never got the time to work on it more

Vyxal Bot

Lyxal created branch Vyxal/make-Ȧ-return-a-lazylist-instead-of-generator

Lyxal opened PR #926 (Vyxal/Vyxal) (Vyxal:make-Ȧ-return-a-lazylist-instead-of-generator → Vyxal:main): Change `Ȧ` to return a LazyList

lyxal

With 26 people, you'd think they'd be able to fill that extra slot

but no

no

they just have to constantly leave it blank

can't fill it in can we?

user

13:34

lol

Seggan

13:48

i am amazed at how much easier coding a compiler with a transformed ast is

i no longer have to worry about the specifics

mostly because certain elements/structures are transformed into a sequence oof other structures

i.e. a map lambda is turned into a normal lambda + the map element

lyxal

mhm

it's almost as if that's how vyxal does it

Seggan

& is turned into load register, do stuff, store register

and the compiler doesnt even have to worry itself about it

its the transformers job

@lyxal wait am i that not creative

lyxal

at least that's how it does the lambdas

Seggan

:facepalm:

lyxal

class LambdaMapDyadic(LambdaOp):
    def __init__(self, body: list[Structure]):
        super().__init__(body, after="M", arity=2)


class LambdaMapTriadic(LambdaOp):
    def __init__(self, body: list[Structure]):
        super().__init__(body, after="M", arity=3)


class LambdaMap(LambdaOp):
    def __init__(self, body: list[Structure]):
        super().__init__(body, after="M")


class LambdaMapEager(LambdaOp):
    def __init__(self, body: list[Structure]):
        super().__init__(body, after="e")

Seggan

13:57

and i thought i had original ideas lol

lyxal

nah we're just as lazy as each other

I only did it like that because I couldn't be bothered to create all the different structures and handle them separately.

Seggan

lol jyxal still does

lyxal

you might be interested in reading the contents of my black book of vyxal

It's a notebook I used for initial planning

and has a whole bunch of thoughts/insights into stuff you could use

let me find the link

Seggan

sure, if you dont mind

user

@Seggan Great minds think alike :P

lyxal

14:03

Part 1: documentcloud.adobe.com/link/…

Part 2: documentcloud.adobe.com/link/…

Seggan

@user so true

@lyxal ty

wait they seem to be identical

lyxal

just a second

acrobat.adobe.com/link/…

That should be part 2

and it looks like I need to scan a part 3 tomorrow

because it seems that I scanned it all before I fully filled the book

Seggan

ty

lyxal

Just do note that it's planning from version 1 and earlier

So some of the characters/elements may no longer exist

Also forgive the messy handwriting in parts

Seggan

eh, as long as im not breaking any copyright laws im fine ;)

lyxal

14:11

Y'ain't

I'm reading through the physical book by phone torch light and oh boy did I do some weird stuff with the planning

You'll notice the word GFL in several places

That stands for "Getting a Feel of the Language"

Seggan

lol

@lyxal tbh my handwriting is only intelligible to me unless i try to make it clean so youre good

lyxal

@Seggan well the best way to think of elements is to test on existing challenges

You'll also notice my many failed attempts at getting v to work

That and lazylists (which were called generators at the time)

Basically, treat it as what not to do when implementing an esolang

99% of it I now consider bad

Seggan

@lyxal ive implemented it as simply a map lambda lol

lyxal

See I did that too originally

But that starts to fail once you introduce dyadic vectorisation

Or even triadic

Seggan

hmm

interesting

im thinking detect the arity of the element and set that to the lambdas arity

lyxal

14:19

Also thought of something like that

Seggan

then map/filter could steal as many args as it needs

still havent gotten to that page

lyxal

That's probably in the third part I need to do when it's not dark outside

Because it's dark outside meaning I can't scan it right now

Seggan

hmm

user

Hmm yes I forgot they don't have elctricity in Australia

Seggan

i had an idea: what if a lambdas default arity is not 1, but the arity of the first element

lyxal

14:21

@user no dumbnut I mean it's midnight

Seggan

not pertaining to v, just in general

user

@Seggan Problem is, the lambda may start with a dyad but have a triad after it, meaning it actually wants 5 inputs

lyxal

Could be a good idea, but it would have to be toggled with a flag

user

I would really like to have both the number of inputted and outputted elements recorded

Seggan

@user but with the current way, it still wants 5 input but gets 1

user

14:23

So for , it'd be [1, 0]

@Seggan True, going off of the first element would be better than the way it currently is

We could analyze the entire program before running it

Unless there are elements that take a variable number of inputs?

lyxal

There's a few that vary based on types

R for example

Seggan

jyxal's map set/get is either dyadic or triadic depending on the inputs

user

I guess those and conditional thingies would mess with that idea

Yeah nvm

Just thinking about how Factor has stack effects

Seggan

hmm reading though the notes seems you had the idea lambdas would operate on the global stack?

lyxal

I don't think so

What date?

Or heading?

Maybe I did and I don't remember

Seggan

14:33

18/7/20

2nd page

the input would be pushed to the stack and the lambda operates on the stack

quite interesting actually

lyxal

Uh about that

that stack inside the lambda?

It's local

It masks the global stack

Python lets you do that

The stack = stack.pop() takes from the parameter stack and assigns it to a local variable stack

Not my best code ngl but it worked at the time

Context is a much better way to do things

Seggan

ah

lyxal

That was before 2.6

And it's one of the reasons we don't talk about 2.5.x

No one mentions any of the 2.5 releases purely because it was all horrible and yuck and not fun

Because it was a failed code clean up with the intention of moving away from global variables

Also it intended to move everything out of one file

But it got too stressful putting out fires here and there so we just rewrote everything

Seggan

woo only 4 more methods to implement and the JVM compiler will be complete

(theres still lots of missing elements tho)

alas, ive got chores rn

lyxal

14:54

!!/'night

Vyxal Bot

@lyxal o/

Seggan