The Nineteenth Byte

Unrelated String

00:17

;_;

22 hours ago, by Unrelated String

I always recognize her by the lambda, but come to think of it I keep getting you mixed up with lyxal

Oh yeah also I just happened to come out to my mom and it went great

emanresu A

00:36

🫂

Unrelated String

:D

lyxal

@UnrelatedString like I said, tax evasion vibes

Unrelated String

So true

New Posts

00:51

0

Q: Counterexampe to Shapiro inequality

Input: A positive integer n such that n is even and gerater than 12 or n is odd and greater than 23. Output: A list of non-negative integers that violates Shapiro inequality. More precisely, Let s be a list of non-negative integers such that the following code does not cause any errors (assume s ...

Ginger

01:08

@UnrelatedString :D

nice!

Unrelated String

Thanks

lyxal

01:19

mfw making an api for uni assignment, and I implement the same route twice

New Loophole Proposal

I HAvE aN APTITUDE FOR margarine. PLEASE GIvE ME TwENTy MILLIOn POUNDS OF margarine.

Ginger

01:35

yall I think the secret message is vamargarine.vwynmargarine.

lyxal

I was more going for NMP knockoff style text

because that's not New Posts

that's the New Loopholes feed

Dannyu NDos

I cannot believe it's not butter!

Ginger

@lyxal check matrix

hyper-neutrino

@UnrelatedString oo nice that's great :3 🫂

Redz

is there a difference between butter and margarine

lyxal

01:45

yes

@Ginger I already did earlier

Bbrk24

Butter is dairy, margarine is made with vegetable oils

Redz

okay makes sense

Unrelated String

@hyper-neutrino :3

Thanks!

Seggan

02:28

@user HEY

i know when its me

also ive had a black pfp for the longest

lyxal

*without changing it

hyper-neutrino

i have become transparent

Unrelated String

02:51

:D

emanresu A

03:04

@hyper-neutrino ~~what are you naming the kid~~

Sorry couldn't resist

hyper-neutrino

💀

Seggan

also i just realized theres a giant pfp next to the message input area

hyper-neutrino

wait you just noticed this—

emanresu A

^

Ginger

^

Unrelated String

03:18

^

lyxal

^

shape warrior t

05:30

Hi! I'd like to ask for feedback on my challenge in the sandbox:

https://codegolf.meta.stackexchange.com/a/26456/123885

Feedback of all sorts is welcome, but my main concern at this point is achieving a good balance regarding the flexibility of the input format. My current thoughts: I'd like it to be highly flexible -- to the point where golfers could exploit the flexibility to shave off bytes -- but not so permissive that it can be abused in a way that trivializes the challenge. In particular, I'm thinking about [this loophole](https://codegolf.meta.stackexchange.com/a/14110), and the [e…

Sandbox Posts

07:44

0

A: Sandbox for Proposed Challenges

very wip but throwing a couple ideas here so I don't forget later inspired by https://www.nytimes.com/interactive/2024/07/29/upshot/olympics-medal-table-paris.html find weights that maximize a country's placement (does this already exist?) find possible placements of a country given all possib...

Themoonisacheese

08:56

@shapewarriort this feels like it's covered by standard loopholes pretty well.

Themoonisacheese

09:07

this also feels like it'd be trivial to solve with a well-crafted regex but i haven't been able to do it with regex alone so i'll wait until you post it and do it in bash+grep

Sandbox Posts

09:52

0

A: Sandbox for Proposed Challenges

Distinguish between the previous answers answer-chaining classification restricted-source Produce a distinct value given any of the previous or the current answer as input. Challenge Write a program or function that takes in a string and outputs an integer between 1 and N, where N is the number o...

New Posts

0

Q: Is this a valid Irish word?

In Irish, most consonants are divided into broad (velarized) and slender (palatalized) variants, and the orthography marks them with neighboring vowels, which are similarly divided. This gives rise to the caol le caol agus leathan le leathan (slender with slender and broad with broad) rule – a me...

code-golf decision-problem natural-language

bb94

Can I have feedback on the following challenges?
https://codegolf.meta.stackexchange.com/questions/2140/sandbox-for-proposed-challenges/20704#20704
https://codegolf.meta.stackexchange.com/questions/2140/sandbox-for-proposed-challenges/20705#20705
https://codegolf.meta.stackexchange.com/questions/2140/sandbox-for-proposed-challenges/26461#26461

Sandbox Posts

10:42

0

A: Sandbox for Proposed Challenges

Find all 10-digit numbers that contain all the digits from 0 to 9 that can be expressed as p²+q³ for primes p and q. Leading zeroes are not allowed.

Seggan

11:06

Eyy let's goo 4d golf engine open sourced

github.com/HackerPoet/Engine4D

caird coinheringaahin g

11:29

@hyper-neutrino Dark Hyper has finally completed her revenge quest, and has returned to a light background. Something something color theory :P

noodle person

12:40

What do you think about this answer? codegolf.stackexchange.com/a/274578/108687

It measures its score in bytes using the Uiua SBCS, but adds an additional byte not regularly included in the SBCS.

Should it be counted using UTF-8, or is this an acceptable encoding?

Seggan

imo utf

lyxal

> # Currently only uses 235/256 available codepoints.

noodle person

The Uiua SBCS does not use all 256 bytes.

lyxal

it can fit

it can be scored as sbcs

Seggan

thats like encoing using ascii and using the extra 128 bytes for Я or something

noodle person

12:42

Well it can be but does this count as "language created specifically to solve the challenge"?

lyxal

@Seggan only 235 out of 256 slots have a letter allocated

ğ would be slot 236

Seggan

same with ascii

ascii only has 128/256 allocated

lyxal

I don't see what's wrong with that

noodle person

isn't it a loophole?

lyxal

so long as you have something that can actually read those bytes

@noodleperson by itself, no

noodle person

12:44

They only add that byte to the SBCS for the purposes of this challenge

lyxal

one could say it's analogous to flags

Unrelated String

@lyxal But it’s completely ad hoc…

Seggan

^

lyxal

sure, but in theory, it's a different language to Uiua

Unrelated String

Yeahhhhhhhh

Seggan

12:45

@lyxal flags are not ad hoc

noodle person

@lyxal ok but it's more like if you added -é" as a flag to solve that challenge in 0 bytes

lyxal

@Seggan depends on the language

noodle person

of course that's an extreme comparison but

Unrelated String

It’s like a milder version of the -D flag stuff in C

Seggan

its like submitting print("Я") and saying its only 10 bytes, putting Я into the 129th byte of ascii

Unrelated String

12:46

Definitely somewhere on a slippery slope towards MetaGolfScript

@Seggan THIS

lyxal

It's definitely on the edge of what's allowed

that's for sure

Unrelated String

Yep

No disputing that lol

lyxal

I'd argue that if you can make an interpreter that uses that codepage, it's valid

whether that counts as specifically made up, I don't strongly agree with that

Seggan

is Python, 10 bytes: print("Я") legal then?

lyxal

that's not python

Adám

12:48

CMQ: In JavaScript, I have {}4 giving 4, but what actually happens here?

lyxal

@Seggan that's "Python but I made a custom interpreter for it"

Unrelated String

…yeah what

lyxal

@Adám I don't like that. I don't like that at all

Unrelated String

How is that syntactically valid

Seggan

@lyxal id argue thats "made up language just for challenge"

Adám

12:50

@UnrelatedString That's my question, but it works in both FF and Chromium.

Neil

it's two statements, {} and 4; the first one does nothing and the second one evaluates to 4, which is the final result of eval("{}4")

noodle person

@Adám it's because {} makes a block in which code runs seperated by semicolons, then the second statement is the expression 4

Adám

Ah, so JS auto-injects a ; between them?

noodle person

they don't need a semicolon between them because in that context the } always ends the statement

lyxal

@Seggan if the challenge was "print the letter Я and only that", sure. But not if it was a challenge where printing Я was a small portion

noodle person

12:51

yeah ASI stackoverflow.com/questions/2846283/…

mousetail

@Seggan They simply wrote sbcs + ğ so I don't think that's legal. That's specifically made for the challenge

noodle person

I'm going to say they have to use either UTF-8 or rewrite the answer to not use that byte

mousetail

Or use a SBCS that allows some kind of escaping for characters outside the range

Like every sane SBCS should have

noodle person

Alternatively, an encoding scheme could be designed such that anything in strings is counted as UTF-8

Seggan

@lyxal dont we also define languages by their interpreters?

lyxal

12:53

@Seggan yes, but that's not a python interpreter

Seggan

therefore "i made a new interpreter" == "i made a new language" == "language specifically for the challenge"

lyxal

because it's not the official python source

@Seggan "i made a new interpreter" == "i made a new language" != "language specifically for the challenge"

Seggan

this "interpreter" was made specifically for the challenge

mousetail

In this case it's clearly specifically made for the challenge, you can't dispute that

noodle person

I'm gonna leave a comment there

Adám

12:55

I wonder if there are any tools that take some JS and adds the semicolons that would be inserted at execution time.

lyxal

@mousetail of course. I don't know how much it's MGS/"I made this up" loophole though

it's definitely a weaker case of it

Neil

@Adám eslint --fix could probably do that

mousetail

MGS and LSftC are seperate loopholes

@Adám Yes, create en eslint config file with semi as the only enabled rule then run fix

Seggan

@lyxal if we define languages by an interpreter, then this is a language made up specifically for the challenge

lyxal

It's not a creative answer, and it's boring to just stick a character like that into the SBCS. But I'm not fully convinced that it's crossed the line from "separate language that we all roll our eyes at" to "delete it because it's cheating"

mousetail

13:02

You could easily create a new SBCS that is equally short but allows escaping unicode characters

(Though I guess it wouldn't technically be an SBCS anymore)

Seggan

thats just called... utf8 lol

mousetail

UTF-8 except a different set is 1 byte

Adám

@mousetail Hm, why doesn't this complain about a missing ; between {} and 2?

mousetail

Encode 235 chars as 1 byte then 0d236 + 1 utf-8 encoded char = utf-8 encoded char

@Adám Weird, it must be a parsing error where it sees {} as a code block even though it's an object.

If I do let a={}2; it errors properly

RubenVerg

in expression position, juxtaposition is not allowed and does not trigger asi; in statement position, juxtaposition is allowed and triggers asi

mousetail

13:08

@Seggan It wouldn't quite be able to be utf-8 since UTF-8 style encoding doesn't allow 235 chars with a single byte, would need to be a slightly different encoding

RubenVerg

@mousetail what? it is a code block

noodle person

Yeah there should be a semi between the code block and 2

RubenVerg

wait no maybe i'm wrong the console interprets {}; as an object

noodle person

oh does it?

hrm

RubenVerg

i'd've thought {}; would be parsed as a code block because those are actual statemenets and therefore would have precedence over expression statements

mousetail

13:14

ESLint 100% interprets it as a code block, if I put entries in the object it errors. But in the browser ist's an object

RubenVerg

maybe it's a backwards compatibility thing, before code blocks were added {}; was an expression statement

but outside the console it doesn't even matter, you can't access the result of an expression statement, can you?

mousetail

It could have side effects

noodle person

console treates it as a code block for me

without the semicolon is a SyntaxError

RubenVerg

@mousetail what? no it's empty

if it's not empty it can be disambiguated between code block and object

@noodleperson ..or maybe not.

noodle person

I just pulled it up in ASTexplorer

mousetail

13:17

I think noodle person is right, my experiment was flawed

noodle person

thats for {}3

I think Prettier et al. don't think you should put a semicolon after a block statement I guess

RubenVerg

astexplorer also says {}; is a block, but chrome console disagrees

@noodleperson yeah just like you don't in 99% of other c-like langs

noodle person

you don't write a semicolon after a for loop body

so I guess it makes sense not to have one here

it's a little weird but makes sense

RubenVerg

in fact in astexporer {}; is a block and then EmptyStatement

@noodleperson completely makes sense to me

mousetail

Need more bracket types

noodle person

13:20

![]

Adám

JavaScript is a little weird.

mousetail

Keyboard layouts if programing language designers where in charge

Adám

@mousetail Ahem, you were saying?

mousetail

Not enough brackets

Unrelated String

@noodleperson Ohhhhh that is wild

mousetail

13:29

Cool keyboard though

Unrelated String

^

I'll always love APL keyboards just for existing

mousetail

Can you buy individual keycaps? Would be funny to just have the T with the smiley and no other special keys

noodle person

I should really pick up a little Dyalog. I've already picked up some J

Adám

@mousetail Hm, when we add new glyphs, we get individual "polyfill" key caps made. Maybe we have some.

@noodleperson I'll be happy to assist.

mousetail

You activated Adam's trap card

RubenVerg

13:34

i really want a custom made apl keyboard with the apl glyphs printed on the side but i can't find anybody who does custom keys with prints on the side

mousetail

Shouldn't be too hard to sencyl and paint yourself

Adám

@RubenVerg Wait, the image I posted is custom and has print on the side.

RubenVerg

you probably overestimate my manual skills but i can definitely try

@Adám do you know where it's made?

i remember emailng one company that mentions in their website that they only do large scale applications but got no reply, i guess they ignored me(:

or maybe they went out of business? website looked quite old

@Adám oh, no stencil here

Adám

Old photo

noodle person

13:55

@Adám Do you have any recommendations for learning resources for someone who is already familiar with array programming but not all the particulars of APL?

I currently know Uiua quite well and J well enough with assistance of the NuVoc pages.

Europe2048

Just fixed my "Draw the Ukrainian flag" solution by... reverting the edit.

noodle person

Nice

Europe2048

I also earned the Cleanup badge for this.

noodle person

Removed my comment

mousetail

16:21

Is this edit ok? codegolf.stackexchange.com/posts/74096/revisions

Seggan

17:52

@mousetail ~~arguably the revision is slander~~

Adám

18:16

@mousetail no, but I fixed it.

Seggan

@Adám i then fixed your fix :P

Adám

oops

shape warrior t

19:15

@Themoonisacheese Good to know, thanks! Also good to hear that a regex-based solution might be viable for this challenge -- definitely not an approach I was thinking about, but hopefully it indicates that there's a variety of ways to tackle this challenge.

lyxal

22:39

@mousetail ~~it's not slander if it's objectively true~~

Rust only has 547 ELO, while Java has an ELO of 711

lyxal

23:56

@lyxal okay so npm run dev doesn't work, but apparently npm start dev does

javascript ecosystems man

Transcript for