The Nineteenth Byte

oh are you down under? its like sunset over here in 'murica

Wezl 'l o - o l'

12:49 AM

... I was too slow but there was a cute picture to catch of my cat looking out the window at the twilight

Why did you feel the need to tell us that?

Ignorance is bliss, we could've been perfectly happy knowing we hadn't missed a picture of your cat

Now we're sad :(

CMP: Syntax for stack literals? ([], {}, something else?)

Wezl 'l o - o l'

Stack(a, b, c, d, e)?

I mean it depends on whether stacks are an important part of the language

That's what a normal language would do, I guess, but stacks are an essential part of the language and there's no varargs

Wezl 'l o - o l'

but all else equal, I'd choose [] over {} because tradition

OldSandboxPosts

Sandbox posts last active a week ago: Convert Klingon romanization to pIqaD [code-golf]

1:00 AM

That makes sense ig

Q: How much more to a repdigit?

0

Challenge By Wikipedia, a repdigit is a natural number composed of repeated instances of the same digit. For example, \$0\$, \$1\$, \$33\$, \$555\$, \$9999\$ are repdigits, while \$10\$, \$121\$, \$3332\$ is not. Your task is, given a number, output the smallest positive number such that the sum ...

code-golf

@user I would say [] if thats not otherwise being used in your language, otherwise I suppose {} is probably best

although you could get funky with <> if you wanted

@NewPosts Dupe hammer go brr

damn that was a fast dupeham

Yeah, I remembered it :P

1:08 AM

Got to love answering after it's closed :P

... what ... how ... ???

You should probably file a bug report lol

@des54321 Ooh, <> might be a good option, it's very distinct from [] so it's clear they're not lists

@emanresuA Just don't refresh

sometimes I forget my sandbox is sorted by score and then accidentally start reading a sandbox post from 2012

@cairdcoinheringaahing Maybe undelete it for science

And of course, Dennis outgolfs me by half :/

1:09 AM

Lol

@emanresuA It's intended behaviour

You can answer up to 4 hours after closing, so long as you don't refresh the page

Lol

Well, you have my upvote for haxxxing

@user the other funky option might be to use ||, or you could get funky with unpaired brackets like [| or something

¯\_(ツ)_/¯ I deleted the answer, given the outgolf

If it was actually a competitive answer, I'd keep it, but 6 bytes vs 11? Nah, there's nothing worth keeping

[foo,bar,baz,...| does slightly show that things can only come out of one end, but that might also be hell and a half to parse vs standard [] lists

1:11 AM

Parsing wouldn't actually be that hard as long as [ and | aren't being used for anything else (they currently aren't)

@des54321 Its not really that much more difficult to parse

I guess it would only be difficult if | could mean something inside a list literal

Parsing [| is the same as parsing []

Just with | instead of ]

but yeah i suppose my two ideas are either <> or [|

1:13 AM

Both are equally easy to parse

but yeah also if you arent using [] for anything its easy

@cairdcoinheringaahing Vyxal gets 5 ;)

Wow, impressive for having a builtin for 1# :P

Let's face it, it often comes down to "who has a specific builtin to make this challenge easier?" with closely-matched golflangs

Does Vyxal cast ints to string when checking if all equal?

1:15 AM

And Vyxal does have a few oddly specific builtins

No

Lists iirc

Base conversion?

~~Yup~~ nvm I thought you were asking if Vyxal has base conversion

IIRC digit lists, except where ranges make more sense

You can configure which one you want with a flag, I think

1:16 AM

Yeah :P

Depending on the right behaviour, 3 bytes is possible in a lang

Combine # from Jelly, with a version that finds the first such starting from n, then check if all digits are equal by stringification, then subtract n

Q: Suppose someone kills more people than they intended. Did they murder the extra people?

1:38 AM

40

If a person sets a bomb in a music festival. Based on previous attendance numbers, they expected 60,000 people to show up. 120,000 people showed up and were killed by their bomb. Legally, did they murder 60,000 people, or all 120,000? Also, suppose they set up the bomb just to kill the production...

criminal-law california murder

... what

my biggest question is why someone wanted to ask this

It's an interesting thought experiment, given the difference between manslaughter and murder

@des54321 singapore iirc

Q: Tapes, Pointers, And sub-tapes: BrainFuck+

2:11 AM

0

Tapes, Pointers, And sub-tapes: BrainFuck+ In this challenge we make an interpreter for a subset of BrainFuck+. As with BrainFuck we start with the tape [0] and pointer at index 0. BrainFuck+ Simplified: Commands + and - Update the current cell, as in BrainFuck. ++ => [2] ^ --+--++ => ...

math integer interpreter brainfuck gris

A: Sandbox for Proposed Challenges

957 rep

yay

just so you can see my rep

.

Sandbox Posts

0

Rational numbers Challenge Every rational numbers (can be written in a fraction form \$a / b\$ ( \$a\$,\$b\$ are integers)) can be represent as a periodic, repeating decimal number. For example, \$1 / 3\$ is \$0.3333333333333333...\$, so we represent it as \$0.(3)\$. Another example, \$3227 / 55...

2:46 AM

Tfw typescript lets me go number + string, number + number, but not number + (string | number) :|

Actually I guess that kinda makes sense, since the type of the result can't be inferred

im assuming number + string stringifies that number and concatenates it to the string? in which case that does make sense

Yeah :P

the opposite, string + (number | string) should probably work though

although that probably doesnt help for what youre doing

Doesn't really matter, I can just assert its type either way and it'll work :P

@des54321 Yep

3:04 AM

@emanresuA It's a better question than it seems at first

Intent-based vs. harm-based is an interesting debate to have

I guess...

E.g., should attempted murder be punished the same as a successful one? You could argue yes, because the intent was the same, and it was only incompetence or bad luck that prevented the harm.

And a fully intent-based system would only punish you for one murder if you tried to kill one person but ended up killing a hundred

Which is stupid

Sort of, but I see the logic behind it, and would lean more toward intent-based than harm-based personally

(Although a mix of the two is the only practical option)

Agreed

3:10 AM

It also has to do with the underlying purpose you're punishing people for. Is it to "give them what they deserve" (retributivism), or purely to prevent future crime (keeping dangerous people off the streets, rehabilitation as main goals)

You've convinced and confused me.

Basically, we keep punishing people and we all agree we should be, but nobody knows why

Y'know, we need more "what if the Central Powers had won" alternate history

"What if the Axis won WWII" is a cool question but a bit overdone

Alternate history: What if Lyxal had never got you to add a 69 constant to Ash?

I'd be 50 rep poorer, but since I removed the constant and gave the rep back later, that doesn't matter much :p

@RadvylfPrograms Got to say, I'm the other way around. If you harm more than intended, you deserve a stronger punishment, but, if you harm less than intended, I'm fine with a lesser punishment (tho, still a punishment)

Plus, it's a lot harder to prove intent in a court of law

3:17 AM

Yeah, that's the main problem with intent-based stuff I think.

And it's always going to be at least somewhat harm-based since you don't know to look for that intent until some harm has been done

3:31 AM

Is quine pronounced with a /kj/ or a /kw/?

^ CMQ

@Pyautogui VTC as a dupe of chat.stackexchange.com/transcript/240?m=60912822#60912822

:P

Got it. I do not think they are quite equivalent.

Unrelated String

how the fuck would it be pronounced with a /kj/

like what's the full transcription there

Queue, e.g.

I love this

The default interactive shell is now zsh.
To update your account to use zsh, please run `chsh -s /bin/zsh`.

Unrelated String

3:38 AM

the /j/ is the e
or i guess it's more like the /ju/ is the eu

@Pyautogui "kwih-nee"

Got it.

Thanks for the answers! That makes sense.

4:34 AM

Does the name of a code file count in the byte count?

thejonymyster

@Emanyalpsid only if you use it as part of the program

5:44 AM

→ 12 gallons of liquid nitrogen moved to The Nineteenth Bakery

Ice-cream radioactive cookie cake

Mmm tasty

At least it's not cinnamon mayonnaise, which was my other idea.

6:30 AM

CMC: Roll n d-sided dice given n and d.

(I have a quite neat 5 bytes in Vyxal)

output the sum, or a list of numbers?

The sum

6:45 AM

@emanresuA why liquid nitrogen, hmm

>>> class NobodyNeedsWhileLoops:
...  def __next__(self): return "Nobody Needs While Loops"
...  def __iter__(self): return self
...
>>> for i in NobodyNeedsWhileLoops(): print(i)
...

7:21 AM

@emanresuA flagless or with flag?

flagless

What do you have?

Probably the same as you

5 byte without, 4 with

Nope, mine can't be improved with a flag, although it's quite elegant IMO

Actually, 4 flagless, 3 with

Oh well :P

7:23 AM

What's your 5?

Oh nvm didn't see the link

Try it Online! for 4

s flag for 3

@emanresuA I had the same first two bytes but then vectorised choice followed by sum

Then I realised that vectorised choice means the range isn't needed,so I just removed that

7:42 AM

→ 2 millilitres of vanilla essence moved to The Nineteenth Bakery

@lyxal Smort.

7:57 AM

month rank #2 hooray

pass 1000 rep hooray, and one month ago i said i wasn't getting past 50...

8:45 AM

txtpad.cn/program

someone tell me what's the problem with this code thanks

password is CGCC

all print statements not in TEST are debugging purpose btw

thanks

9:01 AM

Yo @pxeger why you cracking down on dog food man?

look, I gotta subsidise the fossil fuel industry somehow

By not allowing me to claim dog food as tax deductible?

Why you gotta make tax evasion harder bro?

6

9:19 AM

@WheatWizard I added hgl to ATO: staging.ato.pxeger.com/…

Wheat Wizard

9:37 AM

@pxeger Wow, thanks. It's still under pretty heavy development, how often if at all should I bother you about updating it? I probably release changes on average about once a week, but it's not evenly distributed. Whatever you say is ultimately fine, having even an outdated version is pretty great.

It's not a lot of effort to update, so don't worry too much

Whenever you find yourself annoyed that it's out of date, you can just let me know

Wheat Wizard

Alright, sounds good.

Maybe worth mentioning that because things are unstable, there isn't a guarentee of backwards compatibility, so updates may break older answers.

I am building it with cabal, entirely ignoring nix. Is that likely to end up being a problem?

Wheat Wizard

9:56 AM

@pxeger Not sure, if it builds rn it probably won't end up being a problem, since I don't think there are going to be changes to the build. However last I checked cabal alone just didn't work for me so idk.

10:47 AM

Just for fun....

The letters a,b, c stand for non-zero digits. The integer ‘abc ’ is a multiple of 3; the
integer ‘cbabc ’ is a multiple of 15; and the integer ‘abcba ’ is a multiple of 8.
What is the integer ‘abc ’?

675 I think

but why??

c must be 5 because it's nonzero and cbabc is divisible by 5. If abc is divisible by 3 and cbabc is too bc must be, so a must be. It muust also be even becauuse abcba is, so it's 6. b is one of 147, and only 7 makes (ab)cba divisible by 8.

why can't a = 3, b = 5, c = 2 ?

for the "c must be 5 because it's nonzero and cbabc is divisible by 5." part

clearly that doesn't work for abcba is a multiple of 8

My solution is the only valid one

10:59 AM

that is cool

but I am not sure about the reasoning

@emanresuA you are right of course :)

11:15 AM

@emanresuA Can confirm

@lyxal nice!

12:38 PM

what is the best way to join by a number in python?

actually what is the best way to just join by newlines to a n dimesional array until it becomes a simple vector

An n-dimensional array of what? Numbers?

yeah

1:00 PM

snaaans

holy cr4p is that a sandbags underthetable reference?

uh

you know, sands undertable

Feb 3 at 12:51, by lyxal

that?

Sans (Undertale)

Sans is a fictional character created by Toby Fox for the 2015 role-playing video game Undertale. Initially appearing as a friendly NPC, he later becomes the de facto final boss if the player chooses to complete the "genocide route" and destroy the game's race of monsters. His name is based on the Comic Sans font, which is used for most of his in-game dialogue. The character was praised by critics for his dialogue and boss fight, which is considered to be the most difficult in the game. His popularity with fans inspired several user-created mods and other types of projects. Sans appears as a brief...

1:03 PM

:| |:

@PyGamer0 Whython, 43 bytes: f=lambda a:"\n".join(f(x)?str(x)for x in a)

@Ginger the great thing about being an RO now is that I can see exactly what you said in that even though I was in a different tab scouring the world of golfing languages for more stuff to steal for version 3 of a certain golfing language

I'm surprised how useful Whython's ? actually is (for golfing)

(imagine the reply is to the deleted message)

@lyxal Please tell me you're not looking at people's deleted messages tho

They're deleted so that people can't see them

1:09 PM

@user only if it looks like it was supposed to be a funny joke that was deleted for off-topic purposes

>:(

I wonder how many cringe messages that I thought were safely deleted were viewed by ROs :(

oh none lol it's only directly after sending

I'm sorry you can do what

Ah ok

I don't go transcript diving or anything lol

1:11 PM

So you only really see it if you were there when it was undeleted

exactly

@user Whython, 37 bytes: f=lambda a:"\n".join(map(f,a))?str(a)

I'm stuck at the part where ROs can view deleted messages

of any room

that they own

It could be useful when it comes to serious stuff

1:12 PM

oh shit you CAN

wow

well, TIL

anyhow back to stealing modifiers from brachylog

Final feedback GUIs?

1:35 PM

→ 27 paradoxes moved to Off-Topic TNB (still unlike radvylf btw)

@user honestly python should have that

why is it so gosh dang hard to find modifiers/quicks/adverbs to steal from other languages that aren't Jelly or Husk?

have you tried looking at APL, BQN, J, K?

yes

I've already raided those too

what more do you want then?

1:41 PM

More stuff that can be useful for golf

perhaps there is no more stuff to add

does a "check if the input equals Hello, World!" challenge exist?

I don't think it will be more interesting than a plain Hello, World! challenge

it could be more interesting if the string were longer

could be interesting in many hard-to-use esolangs (even plain BF)
(but I'm not going to create it)

@Ginger I only have 339 core functions. I need at least 500 because type overloads

so there's more to add

1:45 PM

@user but i want it to join newlines on say: [[50,50],[50,50]] and no joining of newlines on a depth 1 list [50,50,50,50]

you should just leave it blank and wait until the needs make themselves obvious

Jelly has lots of unused commands and it still does well

sure, but having a list of things you know you're going to have makes it easier to map functions to meaningful characters, instead of thinking of stuff character by character

you can just leave the codepage blank as well

there's nothing wrong with that

or change the codepage later

I think my point is that I want to make sure I have all the modifiers one could ever need, and I don't know if I've achieved that goal yet

You can't know that until people start using it

I agree with pxeger, the codepage could just be left blank for now. You could make new elements digraphs and then promote them to monographs if they turn out to be useful

1:51 PM

I need a modifier which toggles the lights in my kitchen on and off every time the function is called. You should add that.

7

@lyxal bruh

You can't ever have "all the modifiers one could ever need", you can only get asymptotically close to that

@user it's already filled though

I mean, you could just remove stuff from it

I need a modifier that restarts the Internet from Arpanet

1:53 PM

You don't need a modifier for that, just a normal function

but does vyxal have it yet?

we'll schedule that for v69 :p

v69 will be epic

2:33 PM

soap?

you want me to go get you some?

no

not sure how you would considering that you're in Australia

well what about soap are you asking then.

that was a test

OLIMAR is back

for the time being

I plan on adding help and a language list at some point

3:14 PM

@RadvylfPrograms I've implemented this for Befunge!

gist.github.com/Radvylf/734993e664e7989839bfa60d1713fa96

E.g., for the Hello, World! program on Wikipedia:

compress_befunge(`>25*"!dlrow ,olleH":v
                 v:,_@
                 >  ^`);
'22 3 >25*"!dlrow ,olleH":v_@,:v>  ^'

Nov 16, 2021 at 13:45, by pxeger

@RedwolfPrograms here's an idea I once had: stack-based, but if a function call has fewer arguments on the stack than it needs, it takes them all and puts a partial function on the stack; the next time an item gets pushed on the stack, it will be consumed by the partial function. This would make 1 2 +, + 1 2, and 1 + 2 all the same

@pxeger have you implemented ^ yet?

4:08 PM

~-n is a popular trick for golfing python. What exactly does it do?

it looks like ~-n is the same as n-1 if n is positive

is that right?

yeah

thanks

it's n-1 even if n is negative

I am trying to parse lambda n:[*map(len,bytes(sample([0]*100+[9]*~-n,n+99)).split(b' '))]

~-n -> n-1

-~n -> n+1

4:15 PM

thanks

I am working on a non golfed version currently

would the non-golfed version be just this code with ~-n replaced by (n-1)?

@thedefault. no... the split and bytes parts need to go too

L = ([0]*100+[1]*(n-1))
random.shuffle(L)

that's the start

now I need " list off the lengths of the n runs of ones separated by zeroes."

hmm

not sure how to do that yet

there are only n-1 ones so I am actually not quite sure what I should be counting

I think I am actually counting runs of 0s, right?

the ones are the "bars" in en.wikipedia.org/wiki/Stars_and_bars_(combinatorics)

please correct me if I am wrong!

def random_start(n):
    L = ([0]*10+[1]*(n-1))
    random.shuffle(L)
    return [len(item) for item in ''.join(map(str, L)).split("1")]

still pretty obscure :)

any tips how to make it easier to read?

5:03 PM

this is CGSE, you don't need to worry about unreadable code here

@Ginger Obviously it's not for CGCC then

Even when the "code here doesn't need to be readable" joke isn't horribly overused, applying it where it's not relevant is just telling people "I don't care what you're trying to do just do it my way"

...never mind then

I was just trying to be funny, didn't intend to imply that

sorry

I think I need to study my chatiquette

@RadvylfPrograms then it seems like the question was asked because OP was looking for a solution

5:32 PM

May 28 at 13:08, by mathcat

Looks like I'm gonna learn CSS now

I. Fricking. Hate. CSS. And. HTML.

I had 20 bugs and no errors

errors? in CSS? funny joke

why?

Errors were the best invention since computers.

6:01 PM

Chat Event Refinement: Make You a Lang for Great Good

9

HAHAAA I DID IT

push notifications are now functional!

kinda

6:36 PM

what

where