Vyxal

Steffan

00:50

We've got a problem - when I try to run ṆW/+3|001W, it just spins with no output, because it seems to try to output the whole list at once instead of sequentially printing.

vyxal.github.io/#WyIiLCIiLCIiLCLhuYZXLyszfDAwMVciLCIiXQ==

lyxal

yesterday, by lyxal

huh so remember how I thought mkString would be cool on infinite lists in Scala. Turns out infinite lists don't print too well using that

yesterday, by lyxal

Gonna have to roll a custom print function for VList

lyxal

02:43

!!/amilyxal

Vyxal Bot

@lyxal You are lyxal.

lyxal

Oh thank goodness

That was horrible not being me yesterday

Glad to be me again

!!/help

Vyxal Bot

I'm Vyxal Bot 2.0.0 made by Ginger and hyper-neutrino.
All of my commands start with !!/
If you want help for a specific command, do !!/help <command name>.
New to Vyxal? Try !!/info for more information.

Command list: amilyxal, blame, coffee, cookie, die, goodbye, groups, hello, help, hug, info, issue-open, maul, permissions, ping, prod, register, repo-list, run, status, sus

lyxal

!!/groups

Vyxal Bot

Sorry lyxal, I'm afraid I can't do that.

lyxal

02:47

!!/help groups

Vyxal Bot

@lyxal !!/groups <list|members> [group]: List all groups or members of a certain group.

lyxal

!!/groups list

Vyxal Bot

@lyxal All groups: admin, member, discussion

lyxal

!!/groups members discussion

Vyxal Bot

@lyxal Members of group discussion: Ginger, hyper-neutrino, Niko, Jacob, The Thonnu, Steffan, mathcat

Steffan

02:47

!!/permissions list me

Vyxal Bot

@Steffan User Steffan is a member of groups discussion, member, admin.

lyxal

!!/help ping

Vyxal Bot

@lyxal !!/ping <group> [message]: Ping all members of the group. USE WITH CAUTION!

Steffan

!!/permissions grant me member

Vyxal Bot

@Steffan User Steffan is already a member of group member.

Steffan

02:48

!!/ping cookie

Vyxal Bot

^

Steffan

!!/ping test

!!/status

Vyxal Bot

@Steffan I am doing things like making sure you don't do anything too suspicious or else I'll report you

Steffan

cookie

sus

huh

!!/ping cookie

Vyxal Bot

^

Steffan

02:49

cookie

Interesting easter egg

lyxal

!!/ping discussion In literate mode, lambda arguments without a sigil are converted to their normal form equivalent if it exists. E.g lambda add -> $add end will turn into <lambda symbol>+|#$add. Should that be a feature (meaning sigils are required for names that are also keywords) or something that should be fixed?

Vyxal Bot

@Ginger @hyper-neutrino @Niko @Jacob @The Thonnu @Steffan @mathcat ^

Steffan

!!/ping nobody

Vyxal Bot

^

Steffan

Oh it's just that nobody is in that group, so it's pinging everybody then saying ^

It should probably error

lyxal

02:52

(also I need to make map lambdas retain defined variables between branches in v3)

!!/issue open Make map lambdas retain variables between branches Something like map-lam :=x 4 add then $x n + end doesn't lose track of variable x. This means that the original value being mapped can be retained if needed

Vyxal Bot

vyxal-bot[bot] opened issue #1636 (Make map lambdas retain variables between branches) in Vyxal/Vyxal

lyxal

03:23

@steffan does the new website get the latest js version from the main repo?

Steffan

03:44

it grabs it from gh pages

lyxal

aha, good good

just wanted to know how it gets the data

mathcat

10:17

@lyxal well I don't think it's particularly useful if + isn't a valid name.

Remind me again: How do I convert literate mode to SBCS?

lyxal

If you go to vyxal.github.io/Vyxal, you can click the button that looks like a html tag

@mathcat the question is more "should lambda arguments be blindly translated from words or should there be some sort of parsing"

Because words are turned into sbcs characters if there exists a mapping

If choosing the first option, that means if you wanted to have add or any word defined as a keyword, you would need to include some sort of sigil at the front

So you'd have lambda $arg-that-is-also-keyword -> stuff end

And the $ at the front would be dropped in parsing

mathcat

hmm

lyxal

Because it would be enough to stop the argument being seen as a word

Hence it wouldn't be translated

That's a valid solution, but it may not be the most aesthetic/consistent/stylistic solution

mathcat

I would've said it depends on how hard the parsing is, but I like the bash-style keyword escaping.

lyxal

Lexing lambda arguments differently is also a valid solution, but it may not be the most fun-to-implement solution

Hence the dilemma

On one hand, I kinda like the sigil but on the other hand, it'd be nice to be able to have any argument name be consistently respected

Because you wouldn't need a sigil for names that aren't also keywords

mathcat

10:29

Checking if a mapping exists might lead to some more inconsistencies/bugs though.

But yeah, I don't see how a sigil would affect anything other than the byte count.

lyxal

Oh it doesn't impact the byte count at all

The sigil can be removed for normal form

mathcat

oh, good.

lyxal

This is purely a stylistic thing for literate mode

mathcat

I'm more inclined towards sigil, then.

lyxal

Optional or mandatory?

Because right now it's optional unless you need a keyword name

mathcat

10:38

optional, I do think mandatory sigils would make lambdas slightly more bulky.

I was thinking \, but for keywords.

lyxal

12:51

okay so turns out making map lambdas pass variables between branches isn't as easy as I thought

@AndrovT is this a valid answer to your polynomial challenge?

or does it strictly have to be a list of increasing approximations

mathcat

lmao sympy flex

AndrovT

13:21

@lyxal it looks like it is. I checked up to 100 digits and it worked

lyxal

Sympy 😎

But the input / output format is correct?

AndrovT

Yeah

lyxal

Good :p

AndrovT

Should I be more precise with what output I want?

lyxal

No, the test cases make that clear

I just wanted verification that my answer was valid because I won't be around for the next 8 or so hours to fix it lol

o/

mathcat

13:30

\o

Vyxal Bot

17:12

nayakrujul forked nayakrujul/Vyxal from Vyxal/Vyxal

Steffan

19:43

@lyxal Ah, that's what I forgot to do. I had originally tried the button with 2 2+ and it output 2 2+ so I thought the button just outputted whatever code you had :P

user

Maybe we should offer new visitors to the online interpreter cookies

Ginger

:+1:

Steffan

It gives you a cookie when you run a program that contains a cookie

I'll just add a cookie button in the bottom left that when you click it, it opens up a mini cookieclicker game

user

yesss

Steffan

Actually, it should add a new tab to the tabs

So you can click cookies when you need a break from coding

Better yet, have the tab already open

~~and it's the only tab that's not closable~~

Ginger

19:47

nah, when you run the cookie program it should trigger the minigame

user

Better yet, the interpreter starts out as just the cookie clicker game, and to the side you have options to add the Code and Output tabs if you want to take a break from gaming

Steffan

LMAO

mathcat

@Ginger ooh that'd be a cool easter egg

Steffan

Check it out guys after it builds

Ginger

@mathcat exactly

mathcat

19:57

We could open a new "cookie" tab which runs cookie clicker.

user

Ooh

The cookie is a little too big for my screen but this is awesome

Oh you can resize the tabs, nice

mathcat

COOKIES

user

@Steffan Minor bug: The cookie gets highlighted when I click on it, maybe user-select: none is needed?

mathcat

^

100 YES

user

150

mathcat

20:00

... I'm addicted plz do not remove this window

Ginger

!!/you good?

Vyxal Bot

@Ginger I am doing things like inventing a new type of instrument that creates music by moving through the air

Ginger

cool

user

!!/status

Vyxal Bot

@user I am doing code golf

user

20:02

!!/run 2 2+

Vyxal Bot

@user This command is disabled.

mathcat

@user wow

user

Totally didn't use setInterval

Vyxal Bot

ysthakur created branch cookie-no-select in Vyxal/vyxal.github.io

ysthakur opened pull request #2 (Don't allow selecting cookie in cookie clicker) in Vyxal/vyxal.github.io

ysthakur requested Steffan153's review on #2 (Don't allow selecting cookie in cookie clicker)

user

@Ginge I don't think it's necessary to add messages about review requests ^

Vyxal Bot

20:12

Steffan153 merged pull request #2 (Don't allow selecting cookie in cookie clicker) in Vyxal/vyxal.github.io

Steffan153 deleted branch cookie-no-select in Vyxal/vyxal.github.io

Steffan

Once I add the option to add back tabs you've closed, it'll probably be closed by default but you can add it from there

2

user

Is that an image? It looks very detailed for an emoji

Steffan

it's an emoji

I totally clicked that cookie 69420 times

It starts to get blurry if I scale it up enough it looks like

I guess Apple just has detailed emojis

Steffan

20:37

The current logo is really bad. I made a better one :P

mathcat

yes

Ginger

22:28

!!/ping members bot is shutting down for a bit while I transfer it to the new server

Vyxal Bot

@Ginger @Seggan @lyxal @hyper-neutrino @user @emanresu A @PyGamer0 @Niko @Jacob @The Thonnu @Steffan @mathcat ^

Ah'll be bahk.

lyxal

Uh oh who's gonna answer my amilyxal queries now?

user

22:56

@lyxal I can take over temporarily

lyxal

are you lyxal?

user

Yes

lyxal

huh

I guess so then

user

I stole your identity yesterday

lyxal

am i Lyxal?

user

22:58

No

As I said, I stole your identity

emanresu A

!!/hug

user

Now I can use all the money you saved up from doing tax fraud

@emanresuA You don't have adequate permissions to execute that command

emanresu A

!!/maul user

user

(Gimme a sec, I'm looking through the bot repo trying to find the proper responses)

@emanresuA ༼ つ ◕o◕ ༽つ

:63023446                                                                    YOU CAN RUN, BUT YOU CAN'T HIDE, USER
                                                         ___._
                                                       .'  <0>'-.._
                                                      /  /.--.____")
                                                     |   \   __.-'~
                                                     |  :  -'/
                                                    /:.  :.-'

lyxal

!!/status

user

23:02

@lyxal I am doing VyxalBot's work.

lyxal

!!/status boring

user

Number of errors: 0. That's because I'm AMAZING and WAY better than Vyxal Bot 2. It is IMPOSSIBLE for me to be boring

Also idk what the format is and I don't feel like going through the code to find out

lyxal

!!/make me a cookie

user

@lyxal No.

lyxal

!!/sudo make me a cookie

user

23:04

2

emanresu A

LMAO

lyxal

!!/issue open in VyxalBot2 Bot made moldy cookies I think this is a failure, and needs to remedied in the next patch

user

ysthakur preemptively closed issue #1657 (Bot made moldy cookies) in Vyxal/Vyxal

emanresu A

Wrong repo

lyxal

stupid bot

user

23:07

ysthakur preemptively closed issue #14 (Bot made issue in wrong repo) in Vyxal/VyxalBot2: Bot absolutely did not make the issue in wrong repo

lyxal

@user this suggests that the issue wouldn't be opened for months

meaning not only is it the wrong repo, it's also really slow

very stupid bot

user

It suggests that the bot can time travel

Because I am superior to you mortals

emanresu A

!!/issue open in Lyxal Lyxal is malfunctioning Lyxal appears to be having hallucinations

lyxal

@user you may be superior, but horribly inefficient

I want the issue opened now, not months from now

we're only up to #1636, not #1657

user

ysthakur preemptively closed issue #1 (Lyxal is malfunctioning) in Vyxal/Lyxal: [wont-fix] [bugn't]

@lyxal I experience all time at once. To me, months from now is indistinguishable from today. So you can go suck it

Ginger

23:11

!!/die

user

Believe me, I've tried :P

lyxal

@user !!/issue open in vyxalbot2 Bot overwrote an existing issue #14 already exists idiot

Ginger

ಠ_ಠ

lyxal

@lyxal speaking of which, #14 needs some PR approvals anyway lol

user

@lyxal Didn't notice that lol

@lyxal lgtm, but I find the restriction of having all statuses start with "I am doing" far too restrictive. Why not start with "I am" instead, or even better, don't have a prefix at all, let the entire message be in statuses.txt?

Ginger

23:15

@lyxal status-completed

lyxal

@user blame hyper :p

user

!!/blame

@user It was hyper's fault!

lyxal

@Ginger fixed

@Ginger bis shot autos moment

Ginger

@lyxal done and merged

user

@Ginger Could you please merge #1?

main still has only a .gitignore

Ginger

23:19

@user done :b

lyxal

21 hours ago, by lyxal

!!/ping discussion In literate mode, lambda arguments without a sigil are converted to their normal form equivalent if it exists. E.g lambda add -> $add end will turn into <lambda symbol>+|#$add. Should that be a feature (meaning sigils are required for names that are also keywords) or something that should be fixed?

Feedback on that would be appreciated

Deterministic feedback more so :p

user

I think we should force people to do lambda $add -> .... It's just way to hard to do literate lexing properly otherwise

lyxal

Mandatory for all variables or optional?

user

Mandatory

If it's optional, we'll still have to deal with the cases where they don't use a sigil

lyxal

But in those cases, it could be plausibly considering a skill issue by the programmer that isn't using the sigil

And if it's mandatory, what happens if a non-sigil name is used?

It can't error because non sigil names are fine in normal form

And everything is turned into normal form anyway

user

23:30

But we can't silently turn add into + because then the user will be like "what happened, why isn't add defined?"

So we either have to rewrite the literate lexer to take that into account or error when a lambda parameter name in literate mode doesn't have a sigil

Given that literate mode is more praclang-y anyway, I think it's fine to error

lyxal

The funny thing though is that by doing one, you can just as easily do the other

If you can enforce sigils in the litlexer, you can also make it so that words aren't translated into normal form

user

Hmm yeah

lyxal

Because they both involve figuring out where the lambda starts and where the first branch is

~~meaning both options are as painful to implement~~

user

Perhaps when we come upon an identifier like add or foo, we put it in a special token LiterateIdentifier or something, and then the parser figures that out

But then we wouldn't be able to do litLex because you can't just stringify

lyxal

The stringified version would have to come from the parser

But that doesn't seem too bad

litLex would just return a list of tokens instead of a string to be passed to the lexer oh right the lexer requires a string doesn't it

user

23:35

Well I'm thinking LiterateLexer's output would be fed into the parser directly instead of the normal lexer first

So LiterateLexer would need to be modified a bit

@lyxal There really is no good way to do this :(

lyxal

Well there is if we move away from just blindly translating to normal form

But that requires two parsers

Unless we do a little lexer post-processing

Given a lambda token, collect up until the first matching end keyword or first non-nested branch

If a branch is reached before a matching end keyword, all the things prior to the branch are kept as-is. Otherwise, everything collected is converted as normal

user

@lyxal Well no, we can just have special tokens for literate stuff and the same parser would handle them

@lyxal I suppose we could do that

lyxal

It'd either return a Group or an AlreadyCode

This post processing would happen before sbcsification

It'd only apply to normal lambdas too

Because the other types don't have arguments

So I think the non-sigil approach is viable

And I'll look into later today once I've finished doing some uni work

Because that's a thing I now have

Transcript for