Vyxal

lyxal

2:15 AM

@AaronMiller division by 0 doesn't error though

It just returns 0

!!/run 0 0/

!!/u good

Uh oh

!!/lol u good my epic gamer

@hyper-neutrino

!!/status

@VyxalBot u good bro?

I take it that it isn't

hyper-neutrino

what's with my bot and dying so much

!!/run kH

Vyxal Bot

@hyper-neutrino Hello, World!

lyxal

!!/u good

Vyxal Bot

@lyxal I am doing well.

lyxal

!!/run 0 0/

Vyxal Bot

2:17 AM

@lyxal 0

lyxal

Intentional behaviour

Aaron Miller

@lyxal But integer division by zero does.

!!/run 0 0ḭ

Vyxal Bot

[@AaronMiller: 58730975]
0

STDERR:
integer division or modulo by zero
Most recently popped arguments: [0, 0]
Final stack: []

Aaron Miller

@lyxal you should fix the backticks on the community ad before they get locked

lyxal

@AaronMiller I don't think I special cased that

Aaron Miller

2:25 AM

Makes sense

Hi

Yo

!!/flowey quote

@Ausername Don't you have anything better to do?

A username

No

!!/flowey quote

Vyxal Bot

2:33 AM

@Ausername Is this a joke? Are you braindead? RUN INTO THE BULLETS!!!

Aaron Miller

Lmao get rekt

A username

Oh noes!

lyxal

@Ausername gist.github.com/Lyxal/…

Aaron Miller

I had high hopes for this prompt, but it died at the end: gist.github.com/AMiller42/1c85a207780dc2b3faea3f9c89ceb1b0

A username

@lyxal Did you write all of that? Because you should totally put it on Wikipedia.

Aaron Miller

2:42 AM

@Ausername It was GPT-3

A username

The first player to have at least five zeroes or five Xs in a row wins.

five

Aaron Miller

lol

A username

3:30 AM

!!/run lyxal

Vyxal Bot

@Ausername Did you forget to put backticks around your code (`code`)? Remember to escape any backticks in your code (to type `hi`, enter `\`hi\``).

A username

!!/run lyxal

→ 1 message moved to The Nineteenth Byte

Aaron Miller

4:27 AM

→ 2 messages moved to The Nineteenth Byte

lyxal

!!/switchroom

!!/switch room

!!/repo list

Vyxal Bot

lyxal

y'know, even though I think it's absolutely hilarous moving bot messages over to TNB, I ask that you respect the request made by caird

the last thing we need is stigma against us because we were too stupid

Vyxal Bot

I agree. I'd rather not get shut down by @hyper-neutrino to stop abuse *shudders*

lyxal

we already cop enough criticism based on the language itself

we don't need more

lyxal

5:35 AM

@AaronMiller done

A username

5:49 AM

Can we just close all 2.5+-related issues?

Vyxal Bot

Issue #181 (Vyxal/Vyxal) was closed by Lyxal

Issue #180 (Vyxal/Vyxal) was closed by Lyxal

lyxal

i dunno

they seem fine to me ;)

A username

;)

Idea: Nilads can be overriden with →, for example 5→⁰ ⁰ returns 5?

lyxal

that sounds like something more suited for v3

A username

Yes

Also should we group all the (on hold) feature requests into the big one?

lyxal

6:03 AM

sure

Vyxal Bot

Issue #160 (Vyxal/Vyxal) was edited by chunkybanana

Issue #190 (Vyxal/Vyxal) was closed by chunkybanana

Issue #172 (Vyxal/Vyxal) was closed by chunkybanana

Issue #145 (Vyxal/Vyxal) was closed by chunkybanana

A username

Missed one

Vyxal Bot

Issue #160 (Vyxal/Vyxal) was edited by chunkybanana

Lyxal pushed a commit to Vyxal/fb-lexer: Finished the lexer

A username

Nice job

Fizzbuzz can finally be parsed.

Vyxal Bot

PR #202 (Vyxal/Vyxal) opened by Lyxal from Vyxal:fb-lexer into Vyxal:fresh-beginnings: Create the lexer

lyxal

6:11 AM

I need at least 2 reviews on that

!!/ping everyone

Vyxal Bot

@AaronMiller @Ausername @PyGamer0 @hyper-neutrino @lyxal @math @pxeger @user ^

lyxal

tell me if you want to review the PR and I'll assign you to review it

Vyxal Bot

Issue #160 (Vyxal/Vyxal) was edited by chunkybanana

A username

made a lot tidier

Sorted stuff

lyxal

nice

@Ausername *lexed

lexer != parser

@hyper-neutrino i done finished the lexer

A username

6:31 AM

\« shouldn't occur, make sure that just happens for backticks.

github.com/Vyxal/Vyxal/pull/202/…

And you missed a lot of function-related stuff

And digraphs

ø`string` & co matter

lyxal

@Ausername that sounds more like the job of the parser

math

7:12 AM

@lyxal is there a review counter or smth?

btw 999.3 million views

nvm

I can't review it now

math

8:38 AM

@lyxal gpt-3's so clever...

what? how?

that's amazing

NO The bot's too clever:

No chance

How does it know?

Has Lyxal told the bot?

PyGamer0

@math whoa try Q: What is 51AC8?

it is gonna fail

math

wait a sec

@PyGamer0 It knows

It first said it's a programming language

PyGamer0

@math show the resutl

math

then I reloaded the tab and then it said 51AC8 is the hexadecimal representation of the ASCII character "8".

sorry

now it's even saying something else to Vyxal

lyxal

8:59 AM

GPT-3 thinks that I'm the last of an extinct intergalactic species

gist.github.com/Lyxal/4714a708e0ecf86abcddacf9993b10da

and it talked about Lyxal society structures

and then there's this

@math @PyGamer0 have you read those?

math

no

not yet

@lyxal Is there a completion preset?

in the beta

nvm

it's the no preset

Wow

guess where the Ai started

btw khanacademy is a good recommendation from gpt-3

The AI formatted it in markdown as well

because I started

temprature was too high

A username

9:22 AM

@lyxal At least group digraphs :p

lyxal

@Ausername I haven't been able to yet

A username

Why not? Just dequeue another element when you encounter them...

lyxal

No I mean I physically haven't been able to yet

I'm not at my computer

I'm on mobile

And I won't be able to for another two hours

A username

Oh

math

@lyxal enjoy your time on mobile

@Ausername uh where

A username

9:28 AM

It is painful

Also This site's meta has activity comparable to Windows Phone :p

A username

11:19 AM

!!/'night

Vyxal Bot

@Ausername o/

Vyxal Bot

12:13 PM

Lyxal pushed a commit to Vyxal/fb-lexer: Made the digraphs do good

lyxal

@Ausername the digraphs are good now

Vyxal Bot

Lyxal pushed a commit to Vyxal/master: Update issue templates

lyxal

because I wrote it 2 months ago kekw

and I never actually uploaded

Vyxal Bot

12:34 PM

Lyxal pushed a commit to Vyxal/fb-lexer: added some extra type annotation stuff

user

Yay

Vyxal Bot

12:46 PM

Lyxal pushed a commit to Vyxal/fresh-beginnings: Update README.md

user

Should we also have if name==main stuff in modules like the lexer and stuff? I requested a change on Vyxal's PR, but idk what our guidelines are/will be for that

lyxal

So how would we test the files?

In the testing workflow?

user

Yeah

A lexer_tests.py or something, maybe

Vyxal Bot

Lyxal pushed a commit to Vyxal/fb-lexer: Because we don't do if name is main in files no more

user

Noice

lyxal

12:51 PM

brb gonna go edit that into the guide code

Vyxal Bot

Lyxal pushed a commit to Vyxal/fresh-beginnings: Update code guide.md

lyxal

> This branch is 69 commits ahead, 3 commits behind master.

nice

erm @user how do I make vyxal an importable library?

because I want to create a testing file

like you did

but it keeps saying "no module called vyxal"

like bruh i've got my __init__.py

oh nbm

just needed THIS_FOLDER = os.path.dirname(os.path.abspath(__file__)) + "/.."
sys.path.insert(1, THIS_FOLDER)

user

1:08 PM

Yeah, that's very annoying :(

As long as you import test_utils or whatever it's called before importing vyxal stuff, you shouldn't need the this_folder stuff afaik

Btw are you using Python 3.7, 3.8, or 3.9? (and if it's not 3.9, does PythonAnywhere let you change to 3.9?)

lyxal

I think it's 3.8 at the max

yeah

3.8 at the amx

*max

user

Oh

lyxal

CMQ: py 3.7 or 3.8?

user

I could've sworn I saw 3.9 when I was trying to do stuff there, lemme check again

lyxal

!!/hyperping

Vyxal Bot

1:10 PM

@AaronMiller @Ausername @PyGamer0 @hyper-neutrino @lyxal @math @pxeger @user ^

user

@lyxal 3.8, hands down

lyxal

@user I literally just checked

pxeger

@lyxal this does not need a hyperping; noone in their right mind would say 3.7

user

~~Eh, I'm probably hallucinating again~~

idk if it's because Vyxal is old or what, but I can see 3.9 myself

lyxal

where is this 3.9 you speak of

user

1:14 PM

idk, but I see it right there

math

@lyxal I use 3.7 all the time

user

help.pythonanywhere.com/pages/ChangingSystemImage

@lyxal ^

math

but 3.8 is better for Vyxal I think

user

> You will not be able to switch to a system image that is older than the one it had when you originally signed up.

You'd have to make a new account or smth to use 3.9 :/

lyxal

oh

math

1:16 PM

@math because pyinstaller is not supported yet on 3.8

@pxeger eh me

no

@lyxal what's with the account name vyxal?

hyper-neutrino

3.9 introduces better static typing syntax i think, so like instead of importing List you just do list[str]

lyxal

@math we kinda have to keep the current account

because otherwise, you're looking at 600+ answers with dead links

math

oh

what's with redirecting (if we chose 3.9)?

hyper-neutrino

mass-edit them all /s (disclaimer: no)

user

@lyxal Well, you could redirect to lyxal2.pythonanywhere.com or something, but it'd be annoying

math

1:21 PM

ninja'd

heh

user

@hyper-neutrino Also pattern matching, which is amazing

hyper-neutrino

alternatively

user

@math Oops, didn't see that

hyper-neutrino

just have lyxal run old vyxal

and a new account for the remake

user

That's actually a good idea

Is vyxal3.pythonanywhere.com taken?

The username vyxal3, I mean

math

1:23 PM

vyxal is not taken, is it?

lyxal

Just have the username vyxal

I just for ir

*got it

math

yay

user

🎉

lyxal

vyxal.pythonanywhere.com

math

@lyxal github.com/Vyxal/Vyxal/pull/202#pullrequestreview-716863595?

user

1:25 PM

Time to redirect the webhooks and stuff, then?

lyxal

no

not yet

we'll actually add content to vyxal.pythonanywhere when we are ready

I only signed up tonight to get the username

user

Sounds good

@math You're Tams-Tams on GH, right? What did you mean by "who are you, actually?"?

math

@user your Se username

thanks, got it

user

Oh

I thought when you said "actually," you meant irl

math

@lyxal enum or no enum? btw how do you approve changes thingy?

lyxal

1:29 PM

@math no enum

that's overkill

user

@math Hit "View changes", then hit the green review button in the top right corner

Vyxal Bot

Lyxal pushed a commit to Vyxal/fb-lexer: added test cases

lyxal

because we do that

we have test cases in their own files

user

Thanks

The testcase functions should be named test_foo so that pytest can run them, though

I don't know if it runs the entire file

hyper-neutrino

L#175 exceeds the PEP8 maximum line length (it has 80 characters)

should i just fix that and push it, or should we use a line length of 80

lyxal

1:32 PM

L#175?

math

test file?

user

In which file?

hyper-neutrino

oh

lexer.py

didn't realize there was more than 1 file lol

lyxal

old file version probably

math

github.com/Vyxal/Vyxal/blob/fb-lexer/vyxal/lexer.py#L175?

lyxal

1:34 PM

@hyper-neutrino Finally, we're using an 80 character limit per line, with comments limited to 72 characters per line. If you're a VS Code user, you can paste "editor.rulers": [72,80] into your settings.json to have a visual indicator of the line limits.

someone didn't read the style guide ;)

hyper-neutrino

oh

i just assumed pep8's 79 lol

Vyxal Bot

Lyxal pushed a commit to Vyxal/fb-lexer: Update test_lexer.py

hyper-neutrino

i will look up how to do that for atom

lyxal

me when I have my 2 required yes reviews

merge time

hyper-neutrino

i cannot seem to figure out how kekw

Vyxal Bot

1:36 PM

PR #202 (Vyxal/Vyxal) was merged by Lyxal from Vyxal:fb-lexer into Vyxal:fresh-beginnings: Create the lexer

Lyxal pushed a commit to Vyxal/fresh-beginnings: Create the lexer (#202)

pxeger

@lyxal why limit comments to less than code?

hyper-neutrino

^

also how did you handle multibyte names for functions/variables

lyxal

@pxeger I read somewhere that it's recommended style to limit comments to 72

@hyper-neutrino upon reaching a name starting character, keeping dequeuing until a non alpha char is reached

hyper-neutrino

i suspect i can easily find a bug with this

does (ab) work

lyxal

I assumed we were leaving that for the parser

hyper-neutrino

1:39 PM

oh so we will make the parser unpack these back into individual elements?

lyxal

> Following the tokens °, @, →, and ←, Latin letters and underscores should be grouped into one token of type TokenType.NAME, whose value should be the name itself.

@hyper-neutrino only for for loops

I just followed the specs

hyper-neutrino

oh

i should have mentioned that that was far from complete after our discussion on multibyte names

Vyxal Bot

hyper-neutrino pushed a commit to Vyxal/fresh-beginnings: fix typo and line lengths for lexer specs

lyxal

@hyper-neutrino lol

I tjink i'm too tired to do naything reasonalbe now.

discusions will have to wait intil tomororw

frick

can't even spel

*spell

!!/'night

Vyxal Bot

@lyxal o/

user

1:46 PM

o/

!!/flowey quote

Vyxal Bot

@user And you know what the best part is? This is all YOUR fault!

hyper-neutrino

the problem with for loops is

if you assume there's a | and group alnums then (k`hi) will break

and if you don't, then (ak|hello) will break

user

Why does the second break?

hyper-neutrino

because k|

the solution is to scan forward until you exhaust identifier chars, then skip whitespace, and if you encounter a | next, keep the group; otherwise, backtrack and parse normally

user

Ah

hyper-neutrino

1:56 PM

this means you can no longer do things like (abc×xyz|hello) though, which we discussed would be equivalent to (abcxyz|hello)

@lyxal ^ how much do you care about keeping this behavior

there is a lot of ambiguity and weird unexpected behavior that will come up if we keep it

for example, should (abc`xyz|k`hi) be ( abc`xyz | k` h i ) or ( abc `xyz|k` h i )

Tell Lyxal I'm back ;) — null yesterday

who dis

Vyxal Bot

hyper-neutrino created branch Vyxal/hyper/parser-fix from Vyxal/fresh-beginnings

hyper-neutrino

2:19 PM

there is one minor problem with my approach

(ak`aa|hi`bye)

we scan alnums and get ak

then skip whitespace, and then next character is ` which is not |

therefore we backtrack

now we get a k` a a

and then we still get the | token

in this case, what do we do if the statement divider has a non-name on its left?

Vyxal Bot

hyper-neutrino pushed a commit to Vyxal/hyper/parser-fix: implement contextual name grouping in for for loop variable names

PR #203 (Vyxal/Vyxal) opened by hyper-neutrino from Vyxal:hyper/parser-fix into Vyxal:fresh-beginnings: implement contextual name grouping in for for loop variable names

Aaron Miller

> in for for loop

user

(grouping in) for ((for loop) variable names)?

Aaron Miller

(contextual name grouping) in for ((for loop) variable names)?

hyper-neutrino

2:41 PM

typo

fixed although the bot doesn't observe PR edits i think

Aaron Miller

Fun fact: VyLight gives up after exactly 10001 characters. The 10002nd character and beyond has no syntax highlighting.

user

...Why do you know this?

Aaron Miller

Umm... Reasons?

hyper-neutrino

thonk

Aaron Miller

I find it best not to question my infinite repository of useless information.

hyper-neutrino

2:47 PM

with the lexer we should probably rewrite vylight to work with it a) because of convenience and b) to ensure accuracy, so if the lexer changes vylight reflects that properly

Aaron Miller

3:00 PM

VyLight prioritizes the compressed string/number delimiters over the constant diagraphs when hylighting: Try it Online! Compare the first two with the other diagraphs and the other string delimiters.

hyper-neutrino

hm

Aaron Miller

3:13 PM

Hey, since when do nonexistent constants break parsing?

!!/run k‟

Vyxal Bot

@AaronMiller (output was empty)

Aaron Miller

Hmm... Should Vylight issues go under the Vylight repo, or under the Vyxal repo with the Vylight issue tag?

hyper-neutrino

if it's specific to vylight, the former

if it's a broader parsing error / inconsistency that needs to be fixed in vyxal and then addressed in vylight, probably the latter

imo

Aaron Miller

!!/issue Vylight Diagraph highlighting ignores newlines If you have a diagraph character followed by a newline, the first character on the next line gets highlighted as part of the diagraph. Try it Online! bug

Vyxal Bot

Vyxal-Bot opened issue #1 in Vyxal/Vylight: Diagraph highlighting ignores newlines

Aaron Miller

3:32 PM

!!/issue Vylight Constants and compression make highlighting no Unlike any other diagraph characters or structure delimiters, a `k` followed by a `«` or `»` character keeps the normal highlighting of the delimiter instead of giving it the highlighting of the constant diagraph. Try it Online! bug

Vyxal Bot

Vyxal-Bot opened issue #2 in Vyxal/Vylight: Constants and compression make highlighting no

Aaron Miller

Why are constant diagraphs different from other diagraphs?

Razetime

3:53 PM

k is under var_chars

we just need to add it to the digraphs constant probably

hyper-neutrino

!!/status

Vyxal Bot

@hyper-neutrino I am doing great.

hyper-neutrino

debatable

you didn't report my commit

github.com/Vyxal/Vyxal/commit/… - make the tokenizer skip comments

user

4:49 PM

@VyxalBot "Diagraph"?

Aaron Miller

@user Two-character commands: k ∆ ø Þ ¨

user

No, just the first a

Aaron Miller

oop

user

Ah, a fellow Java enjoyer, I see :P

Aaron Miller

lol

hyper-neutrino

5:27 PM

does vyxal have a replicate / repeat in-place built-in?

for example, [1, 2, 3] $ 3 gives [1, 1, 1, 2, 2, 2, 3, 3, 3]

rather than [1, 2, 3, 1, 2, 3, 1, 2, 3]

Aaron Miller

@hyper-neutrino idk, let me see if I can find anything

Aaron Miller

6:15 PM

@hyper-neutrino I don’t see exactly that, no. What’s the context?

hyper-neutrino

39

Q: Output a Pyramid (or Highway)

Given a non-empty string s, with even length, and a positive integer n, representing its height, compose a pyramid using the following rules: The pyramid should contain n non-empty lines; trailing newlines are allowed. For each 1 <= i <= n, the i-th line should contain the string with each indiv...