Vyxal

lyxal

01:19

stackoverflow.blog/2023/01/19/…

interesting little article

especially given that I'm considering features mentioned in the article for future 3.x releases

emanresu A

printLabel = (label: string|number) => string {
  console.log(`Please fill out ${label}`);
}

That is not valid typescript

Vyxal Bot

Lyxal approved on PR #1556 (Vyxal/Vyxal): "lgtm"

Lyxal merged PR #1556 (Vyxal/Vyxal) (Vyxal:v3-fix-fn-ctx → Vyxal:version-3): Make scoping work like Python

Lyxal deleted branch Vyxal/v3-fix-fn-ctx

mathcat

09:31

We have a new Vyxaler

emanresu A

09:47

ye

mathcat

10:01

I have been (very) passively coding in Vyxal for 1-2 years and they are already better than me.

emanresu A

"better" is subjective

mathcat

11:08

> Perhaps a Vyxal REPL deserves its own project? I'd be happy to make it, once I find time :p

Replxal?

emanresu A

Where? A vyxal repl exists

mathcat

I meant to ask "Should we move the repl code to its own repo?"

(github.com/Vyxal/Vyxal/issues/1445)

emanresu A

a) Let's be honest, Ginger's never going to get around to that
b) Meh, it's not *that* complicated. I'd guess ~50 lines.

mathcat

yeah ig

What if you hook up rich and vylight with the repl?

Most repls have some kind of colour coding.

lyxal

11:40

If a REPL was to be made, it'd be better to have more attention focused on making a decent v3 repl than a v2 repl

at least, that's what I think

mathcat

Yeah, although the debug toggle might be useful in v2 too

@mathcat wdyt, is a seperate issue needed?

lyxal

11:58

man it's been a while since I've last used stupid brick teeth

I've forgotten just how slow it can be :p

Vyxal Bot

Lyxal created branch Vyxal/add-more-v3-tests

lyxal

oh boy what fun! :p

mathcat

12:44

that looks tedious

lyxal

even better given that numbers aren't properly implemented

Vyxal Bot

12:59

Lyxal created branch Vyxal/make-decimals-work-properly-in-v3

Lyxal opened PR #1557 (Vyxal/Vyxal) (Vyxal:make-decimals-work-properly-in-v3 → Vyxal:main): Make decimals work properly in v3

Lyxal closed PR #1557 (Vyxal/Vyxal) (Vyxal:make-decimals-work-properly-in-v3 → Vyxal:main): Make decimals work properly in v3

lyxal

that's what you don't do

Vyxal Bot

Lyxal opened PR #1558 (Vyxal/Vyxal) (Vyxal:make-decimals-work-properly-in-v3 → Vyxal:version-3): Make decimals and complex literals work better

lyxal

v3 branches don't get merged into main

only v2 branches get merged into main

don't do a silly like I did

→ 4 messages moved to Trash

@VyxalBot just element tests

and just to make sure everything has tests while there's still only a handful of things that need tests

mathcat

Steffan is dominating week.golf lol

lyxal

better to go silly on it now and then incrementally add tests now rather than later

@mathcat and you're mathcat

not much new there :p

mathcat

13:12

in The Nineteenth Byte, 17 hours ago, by mathcat

ಠ_ಠ

Ginger

@emanresuA but I will get around to it

in fact I might even do it today :b

lyxal

you better get it done quick - once v3 drops, 2.x will be considered legacy :p

granted, that'll be much later this year

Ginger

eh, it's not a Ginger product if it doesn't get rewritten twice over :p

mathcat

@Ginger Breaking news: Ginger is placing something at the top of their to-do list.

lyxal

@Ginger nor a vyxal one :p

although we do it 4 times around here

once for the initial 0.x and 1.x releases, once for the 2.x<5 releases, once for 2.6.x+ and once for 3.x releases

(we don't talk about 2.5.x though)

mathcat

13:18

@lyxal wait there was a 0.x??

lyxal

@mathcat there sure was

0.0.0 through to 0.9.2

github.com/Vyxal/Vyxal/releases?page=7

bottom of the page and everything on pages after

> made • the decimal symbol instead of º

source

because decimals used to look funny!

(․): Symbolic function reference

Interesting, . was originally something like ⁽ is now

nope that's a character that looks like .

(.): Print Vy_repr(t.o.s)

there we go

that's more like what I had in mind

homoglyphs amiright

λ¬∧⟑∨⟇÷«»°\n․⍎½∆øÏÔÇæʀʁɾɽÞƈ∞⫙ß⎝⎠ !\"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~⎡⎣⨥⨪∺❝ð£¥§¦¡∂ÐřŠč√∖ẊȦȮḊĖẸṙ∑Ṡİ•\tĤ⟨⟩ƛıĲĳĴĵĶķĸĹĺĻļĽľĿŀŁłŃńŅņŇňŉŊŋŌōŎŏŐőŒœŔŕŖŗŘŚśŜŝŞşšŢţŤťŦŧŨũŪūŬŭŮůŰűŲųŴŵŶŷŸŹźŻżŽžſƀƁƂƃƄƅƆƇƊƋƌƍƎ¢≈Ωªº

That was the very first SBCS codepage in the 0.x releases

Anyhow, that's enough nostalgia for one night

!!/'night

Vyxal Bot

@lyxal o/

mathcat

14:21

@lyxal that looks so outdated lol

Ginger

15:13

okay so: Replxal

really needs a better name

(@mathcat)

wait @lyxal is v3 going to be written in Scala? because I sure don't know Scala

mathcat

@Ginger Yeah, we don't need anything else that sounds like lyxal lol

@Ginger yup

Ginger

well: frick

mathcat

I'm still trying to understand how the lexer works

Ginger

I will become much less useful after v3 I guess

petition to maintain v2 with a frozen feature-set :p

@mathcat how about Crosshatch

cause the vyxal logo is a cross

imma go with it

Vyxal Bot

15:31

GingerIndustries created a repository: Vyxal/Crosshatch

GingerIndustries created branch Crosshatch/main

mathcat

sounds good

are we doing meme-ads now :P?

Ginger

bleugh

oh yeah another thing I'm gonna do at some point is make a better logo because the current one is, shall we say, suboptimal

mathcat

gl

Vyxal Bot

GingerIndustries deleted a repository: Vyxal/Crosshatch

PyGamer0

How is v3 going? (@lyxal)

Vyxal Bot

15:48

GingerIndustries created a repository: Vyxal/Crosshatch

GingerIndustries created branch Crosshatch/main

GingerIndustries created branch Crosshatch/dev

Ginger

@lyxal you may want to set up pre-commit & co.

Vyxal Bot

GingerIndustries opened PR #1 (Vyxal/Crosshatch) (Vyxal:dev → Vyxal:main): F1rst Comm1t!

user

@PyGamer0 It's going okay, but to be honest, there aren't too many people actively working on it, just Lyxal and me, with Steffan reviewing stuff sometimes

Vyxal Bot

GingerIndustries merged PR #1 (Vyxal/Crosshatch) (Vyxal:dev → Vyxal:main): F1rst Comm1t!

user

@VyxalBot Why a whole new repo for this? Wouldn't it be more convenient to have it included in the main Vyxal repo?

Ginger

15:54

it is worthy

user

People shouldn't have to install a separate thing to get a good REPL, the pretty REPL should come with Vyxal itself

Ginger

no it shouldn't

user

Is Crosshatch going to be included with the Vyxal package somehow then?

Ginger

yeah, using python extras

so you can do pip install vyxal[repl]

or maybe we can include it by default

user

Including it by default would be best

Ginger

15:55

so you could do pip install vyxal[norepl] if you don't want it for some reason

user

Why make people install an extra thing to get a better REPL?

Unless Crosshatch is going to be a kitchen sink or something

@Ginger It's such a small amount of code that no one except a Gentoo user would care

Ginger

lol

Vyxal Bot

GingerIndustries opened issue #2 in Vyxal/Crosshatch: Get the REPL working

Ginger

wait frick that needs to be a PR

Vyxal Bot

Steffan153 approved on PR #1558 (Vyxal/Vyxal)

GingerIndustries closed issue #2 (Get the REPL working, Vyxal/Crosshatch)

GingerIndustries merged PR #1558 (Vyxal/Vyxal) (Vyxal:make-decimals-work-properly-in-v3 → Vyxal:version-3): Make decimals and complex literals work better

GingerIndustries deleted branch Vyxal/make-decimals-work-properly-in-v3

Ginger

15:58

...that was done, right? :b

Steffan

so how does installing replit help you (i just don't know what it does on gh)

@user I'm planning on doing some PRs for v3 soon, I've just been golfing too much lol

user

I mean it's not urgent

@Ginger I'm actually currently reviewing it lol but yeah lgtm

Steffan

I would've merged it myself if you'd already approved it lol

user

@anyone_interested_in_Scala Btw, check out this feature (it's likely not an efficient way to implement the number splitting, just thought you might find it cool)

hyper-neutrino

what do you need replit for and why are you requesting me and not lyxal lol

user

16:13

@Steffan ...I just realized you're Steffan, not lyxal

Your pfps look very similar

I was wondering why lyxal would say he's been planning on doing PRs when his PR literally just got merged

user

16:25

@Ginger Er, finished reviewing and it turns out it wasn't, I was going to hold it back lol

mathcat

17:02

@user I've always wanted to help, but I never get around to trying to fully understand the code.

Maybe I'll do it now.

@Steffan I can see

user

Whoa

@mathcat Yeah we're trying to document it better

If there's anything you don't understand, you can ask about it here. Feel free to ping me or lyxal (I assume he'd be okay with getting pinged too)

If you don't know where to get started, you can ask about that too

There's some docs in the documentation folder

mathcat

@user eh, we can combine those later if we want to

Ig troubleshooting will be easier if it is a fully fledged REPL

and if we do some pro-v3-stuff later

@user got it, thanks!

mathcat

18:05

There isn't a trim first and last / join first and last element in vyxal, right?

user

There's definitely trim

What is join first and last?

mathcat

ḣtJ

@user I meant remove first and last element from a list.

user

Oh

Ginger

18:41

@user fricknuts :b

sorry, itchy trigger finger

Vyxal Bot

19:06

Steffan153 commented on PR #1554 (Vyxal/Vyxal): "It doesn't look like this works if `lhs` is a list"

Ginger

19:20

what do yall think the prompt for the repl should be?

mathcat

traditional vyxal>?

mathcat

19:45

welp got ninja'd by AndrovT again

Ginger

what

also jeez the vyxal code is a mess

mathcat

2

A: Capture the Flag... with a twist

Vyxal, 12 bytes sIƛ∑÷T-;¯±∑∑ Try it Online! # 1 -> A wins # -1 -> B wins # 0 -> draw s # sort I # split into halves ƛ ; # map: ∑ # sum ÷ # push each item to stack T # triple - ...

Ginger

why tf do yall have all your main code in main.py instead of __init__.py

mathcat

@mathcat had almost exactly the same code

Ginger

I can't even open a PR to add a __version__ because there's nowhere to put it

mathcat

19:47

1 min ago, by Ginger

what

Ginger

it's only supposed to go in __init__.py, but that's empty

eh screw it

Vyxal Bot

GingerIndustries created branch Vyxal/add-a-version

GingerIndustries opened PR #1560 (Vyxal/Vyxal) (Vyxal:add-a-version → Vyxal:main): Add a __version__ attribute

mathcat

@VyxalBot lol markdown cringe

Ginger

markdown go brrr

mathcat

wait I don't think you have to hardcode it tho

Ginger

19:52

¯\_(ಠ_ಠ)_/¯

mathcat

Is there a way to fetch it from pyproject.toml?

Ginger

I'm gonna go with no, because that's not available in built distributions

there's probably a way to have poetry auto-replace it during the build process, but that feels like too much work for something this simple :b

mathcat

hmm you can probably hook it up to a github workflow

Ginger

nah, then it'd only work on GH (unless yall are already using Actions to build and publish Vyxal)

oh, you are

hmmmmmmmmmmmm

mathcat

github.com/Vyxal/Vyxal/blob/main/.github/workflows/pypi.yaml?

Ginger

19:57

yeah that

okay, it's probably possible but I'm sure as hell not gonna try to add it and then break everything :p

mathcat

I'll do some testing

Vyxal Bot

mathcat4 created branch Vyxal/workflow_test

Vyxal Bot

20:21

GingerIndustries opened issue #3 in Vyxal/Crosshatch: Have the greeting report an actual Vyxal version

GingerIndustries opened PR #4 (Vyxal/Crosshatch) (Vyxal:dev → Vyxal:main): Get the REPL working

Ginger

!!/help

Vyxal Bot

All of my commands start with @VyxalBot or !!/
- Use "ping me" or "don't ping me" to add/remove yourself from the ping list
- To evaluate Vyxal code, use "(execute|run|run code|evaluate)", followed by code, flags, and inputs inside inline code blocks (multiline code is not supported; provide multiline input in multiple code blocks)
- For a full list of commands, see https://github.com/Vyxal/VyxalBotSE/blob/master/instructions.md

Ginger

!!/ping me

Vyxal Bot

@Ginger I have put you on the ping list.

Ginger

cool

!!/repo list

that no worky?

:/

@hyper-neutrino what is the meaning of this tomfoolery

hyper-neutrino

20:33

probably a JS bot feature

do recall that there are two versions of VyxalBot

if you want to port the features you can PR them

Ginger

:|

~~you should convert the bot to use sechat, the better chat library~~

Vyxal Bot

ysthakur requested changes on PR #1560 (Vyxal/Vyxal): "I don't think this is necessary. See stackoverflow.com/a/67097076";

ysthakur created branch Crosshatch/vyxal-version

ysthakur opened PR #5 (Vyxal/Crosshatch) (Vyxal:vyxal-version → Vyxal:dev): Get Vyxal version

user

@Ginger ^ that sort of issue is one reason for putting Crosshatch into the main Vyxal repo itself. If you want to make a change to Crosshatch that requires a change in Vyxal, you now need to make one PR in Vyxal/Vyxal, wait for it to get merged and published, then make another PR in Vyxal/Crosshatch

@lyxal Btw, I forgot to mention, we should probably decide on a testing style (or different testing styles for each group of tests)

Once we sorta finalize what we want, we can make a method or something to make it easier to make new tests (no more writing given Context = Context() over and over again)

@lyxal I'm hoping Mill gets good enough soon so we won't have to use sbt anymore

@Ginger If you know Python, you shouldn't find Scala too hard :)

@mathcat The lexer and parser are two of the more complicated parts lol, you might want to start off with something easy like implementing elements

mathcat

20:55

kay

user

I mean if you were looking for somewhere to contribute, I'm not trying to pressure you into it

(okay maybe a little)

mathcat

no worries, I've been wanting to (completely) learn Scala for a while now.

mathcat

21:22

AAAH

I committed to main

what do I do

user

Repent and rebase

mathcat

what

user

git rebase -i d11292087787ad49fab0c05d55fb39ee4b693971

Then drop the last commit

You shouldn't have any merge conflicts, but if you do, fix them

Then force push

How were you even able to commit to main directly? That shouldn't be allowed

mathcat

@user okay, thanks, I'm sorry I'm bad with git

user

(that commit hash is for the commit before yours)

@mathcat It's fine, pushing to the wrong branch is a mistake we've all made

I just don't get how you were able to commit to main lol

Pretty sure I accidentally tried that once and couldn't

mathcat

21:26

I have no idea

user

Must be that SE Nitro

mathcat

yeah definitely

lyxal

21:44

@user admins can bypass branch protection on main

Don't remember why I set that

Probably should unset it

mathcat

@mathcat why u not working

lyxal

@Ginger packaging reasons

You'd have to ask pxeger, they set it up for us

@user @mathcat user is right, I'm happy to be pinged to explain how the infrastructure works

user

@Ginger IANASC but isn't it pretty common to have an empty/near-empty init.py and have all your real code elsewhere?

lyxal

@user sounds like a good idea. I'll open a discussion on the repo so formatting can be discussed

Vyxal Bot

Lyxal created a discussion in Vyxal/Vyxal: Consistent Test Case Formatting/Style

lyxal

21:54

@mathcat that was an experiment for the community ads

The style we're going with is much nicer than that

See here

@mathcat if you've ever seen things like EBNF, it's basically that but with actual regex

Extended Backus–Naur form

In computer science, extended Backus–Naur form (EBNF) is a family of metasyntax notations, any of which can be used to express a context-free grammar. EBNF is used to make a formal description of a formal language such as a computer programming language. They are extensions of the basic Backus–Naur form (BNF) metasyntax notation. The earliest EBNF was developed by Niklaus Wirth, incorporating some of the concepts (with a different syntax and notation) from Wirth syntax notation. Today, many variants of EBNF are in use. The International Organization for Standardization adopted an EBNF Standard...

It's basically saying "here is what each token is if you define it as a standardised pattern. Whenever each pattern is unambiguously matched in the source code, make a token"

Most of the regex patterns are straightforward

There's just a few like the number regex and the multi graph pattern that are a little more complex

user

@lyxal I actually meant this but yeah, code style needs a discussion too

lyxal

22:12

@user that's what I meant too

I've seen the difference between test("...") and double nested describe and it and gotten confused why I couldn't use one in the other until I realised it was library functionality :p

So consistency with the framework used would be good

user

Oh ok

Vyxal Bot

Steffan153 commented on PR #1554 (Vyxal/Vyxal): "What if somebody wants to replace a whole list in a list? I'd say put the recursive behavior into a digraph"

ysthakur created branch Vyxal/v3-more-docs

ysthakur opened PR #1562 (Vyxal/Vyxal) (Vyxal:v3-more-docs → Vyxal:version-3): Write some more docs for v3

Lyxal approved on PR #1554 (Vyxal/Vyxal): "This approval is for when the intended behaviour is reached. The overall infrastructure of the change lgtm"

lyxal

22:34

No merging it yet

It's just an approval to say that "I've looked at it I get the general idea of it and I like it."

Ginger

23:05

@user well hopefully I won't have to make any more of those changes :p

@user cool :b

@user not what I do, but maybe that's just me???

Vyxal Bot

GingerIndustries requested changes on PR #5 (Vyxal/Crosshatch): "This is clever! But keep in mind that the Vyxal version and the Crosshatch version aren't necessarily the same thing. `__version__` should be the Crosshatch version, and we should use a different constant for the Vyxal version."

Ginger

@user (at least I think that's you) ^

Transcript for