The Nineteenth Byte

Destructible Lemon

00:02

@PhiNotPi yeah but then it is semi easy to get a win by just playing to the side upwards or something

trichoplax

You could start with all spaces fully visible and gradually obscure them over time

Helka Homba

@Mego Thanks

Anyone want to play contact chat word game? Game starting: chat.stackexchange.com/rooms/53490/contact

PhiNotPi

@DestructibleLemon yeah, that's kinda the extreme option

ais523

huh, I just went and outgolfed Dennis in Jelly again

I should keep a tally of this as counterexamples for when people claim it's impossible

ATaco

I'd like to see people outgolfing me in RProgN2.

It's particularly good with quines, for some ungodly reason.

ETHproductions

00:15

"ungodly" is right ;P

ais523

it's because it's concatenative

Underload and 7 are good for the same reason

ETHproductions

Or are you not talking about the 1\n1 answer?

ais523

err, I should say, concatenative and eval operates on strings rather than parse trees

ATaco

I'm not talking about 1\n1, which is now considered invalid.

ETHproductions

Oh good, do we have a new consensus then?

ais523

00:16

actually, GolfScript is also good at quines for the same reason again

Adám

is there such a thing as a regex which will only match itself?

ATaco

RProgN2's quine works almost like golfscripts.

ETHproductions

492

A: Regex that only matches itself

PCRE flavor, 261 289 210 184 127 109 71 53 51 44 40 bytes Yes, it is possible! <^<()(?R){2}>\z|\1\Q^<()(?R){2}>\z|\1\Q> Try it here. (But / is shown to be the delimiter on Regex101.) Please refrain from making unnecessary edits (updates) on the Regex101 page. If your edit doesn't actually in...

ATaco

But instead of eval'ing/Calling, it just straight up starts reading the program from just after the function definition.

{`{.}{ is the quine right now.

Which bests Golfscript's {`"_~"}_~ by a few bytes.

ais523

Underload has (aS(:^)S):^, which is the same as GolfScript's if you allow for the fact that GolfScript has implicit output and Underload doesn't

I'm not sure if I've ever written a quine with an explicit :^ equivalent in 7 (because it does it implicitly at the end of the program), I should do that

looks like it's 3713256371325: Try it online!

that's 5 bytes long on disk:

00000000: 7cb5 73e5 ab                             |.s..

there are a ton of different ways to write quines in 7, though, of various levels of cheatiness

ATaco

00:28

7's quines are short, but are a pain to expand.

ais523

e.g. 23723 is two parts which each output each other, 223372233 consists of one part that outputs itself twice and one part that doesn't do anything at all, and 3713256371325 consists of one part that outputs itself and then a copy of the other part that's stored in itself

and no, they're very easy to expand

as long as you know the language

it's hard to write a quine in 7 that isn't payload-capable; even 23723 is

e.g. 22323722323 is a quine that works on the same principle and outputs itself twice (with a 7 in between because that's 7's equivalent of whitespace)

I just added a 223 to each half, "output a copy of the TOS"

Mendeleev

Hey everyone I just saw the question:

18

Q: Print the Russian Cyrillic alphabet

Apparently, we really like printing the alphabet in different languages. How about the Russian Cyrillic alphabet? Print/output/return this exact text (trailing whitespace is permitted) АаБбВвГгДдЕеЁёЖжЗзИиЙйКкЛлМмНнОоПпРрСсТтУуФфХхЦцЧчШшЩщЪъЫыЬьЭэЮюЯя This is the Russian Cyrillic alphabet wi...

It's a dupe of this one that I posted:

Zachary T

What language is this?

Mendeleev

codegolf.stackexchange.com/questions/97469/…

HyperNeutrino

@HelkaHomba Did you make the 2048 Sandbox app?

Mendeleev

00:31

@ZacharyT 7

Zachary T

Is 7 a valid program within 7?

HyperNeutrino

@Mendeleev 404

Helka Homba

@HyperNeutrino I did

HyperNeutrino

It's really cool :D

Mendeleev

@HyperNeutrino YOu can't see it yet

Helka Homba

00:32

Thanks :)

Mendeleev

Not enough rep

HyperNeutrino

Oh yeah right.

ಠ_ಠ

Mendeleev

O_o

Zachary T

ಠ_ಠ

(I will STOP now)

ais523

@ZacharyT yes, but it doesn't do anything, it basically entirely consists of trailing whitespace

HyperNeutrino

00:32

I set it to always spawn a 4 in the top-left corner and set voids so that it made a snaking pathway :D

ais523

in fact trailing 7s aren't even representable in the compressed format

Zachary T

How ... does this language even work‽

Is this an actual user? @ಠ_ಠ

ais523

it basically contains a bunch of primitives (six of them) which are very simple and collectively Turing-complete (apart from 3, which is rather complex because it's responsible for I/O in a number of different formats)

Mendeleev

I should make my username that (/s I'm not changing it until 2018)

HyperNeutrino

Wait it actually works

I thought it would be an illegal char

Zachary T

00:35

If you change it, it should be @wat again.

Mendeleev

Perfect username

I wanted to make my username U+202e once

ಠ_ಠ

Illegal char unfortunately. I tried :P

Zachary T

ಠ isn't invalid probably because it's used in the Kannada language.

What is U+202e

ಠ_ಠ

RLO: causes characters to render right-to-left

Zachary T

ಠ_ಠಠ_ಠಠ_ಠಠ_ಠಠ_ಠಠ_ಠಠ_ಠ

(I just looked that up, hence the spamming of that face)

ಠ_ಠ

00:37

But spam is still bad

Zachary T

LOL. @ಠ_ಠ

ಠ_ಠ

lol

I wonder why chat can't update immediately

like, from main profile

Zachary T

Is the interrobang a valid name‽

ಠ_ಠ

I wouldn't doubt it.

Zachary T

@‽

Destructible Lemon

00:38

@ಠ_ಠ ಠ_ಠ

Zachary T

Too much

ATaco

I don't want to know what that's going to do to I am Typing.

Oh, apparently nothing.

ಠ_ಠ

please do not ping me just because of my username

it should be changed already c'mon SE stop being slow

Zachary T

@ಠ_ಠ, of course we will, your username is ಠ_ಠ

ATaco

@ಠ_ಠ you are rather pingable my dear friend.

Zachary T

00:39

How do you change your chat name‽

ಠ_ಠ

That shouldn't even be my username anymore.

I changed it about 3 minutes ago.

there we go?

Mendeleev

@Mego

HyperNeutrino

no

Mendeleev

pls

Zachary T

There, it changed.

HyperNeutrino

00:40

There we go.

Zachary T

How does one go about changing the name that appears‽

HyperNeutrino

Ok done messing with usernames. I now realize I should have gone to Sandbox for that :I

Mendeleev

you have to toggle your parent user from PPCG to somethign else, then back

HyperNeutrino

@ZacharyT change your parent site to a random site to force it to update

lol ninja'd

Zachary T

Crap, I don't think it works, it's not a letter.

Mendeleev

00:43

You can have non-letter characters

Zachary T

‽

Mendeleev

Numbers, dashes, underscores work

HyperNeutrino

Hey look I can close vote things. I wonder when that happened

Zachary T

(Adám should suffix ‽ to his name) [jk, obviously]

ais523

can I just put on record that I really hate the interrobang character and get annoyed when people use it

HyperNeutrino

00:44

ಠ_ಠ I've become a challenger over an answerer lol

@ais523 I agree; it looks very "messy" almost

ATaco

I really dislike that character.

‽

It just, looks so ugly.

Destructible Lemon

this is why we need to standardise it

so we can get a not crap version

ais523

it's also massively overused despite being hardly used at all

the situations where it applies are very rare

HyperNeutrino

Usually !? is sufficient.

Zachary T

@ais523 Why‽ "!? just looks ugly" :p

ais523

00:45

usually ? is sufficient

HyperNeutrino

True.

ais523

questions normally don't work as exclamations too

they have to be highly specific questions (rhetorical, and disbelieving of the current state of affairs)

Zachary T

(I adapted my typed usage from Adám, I've always used it when writing)

HyperNeutrino

Apparently [compose] ! ? works to get it

ais523

yep

HyperNeutrino

00:47

Why exactly does the interrobang have an upside down version⸘

ais523

for Spanish, probably

Zachary T

Yeah, that should work ⍴robably.

HyperNeutrino

Oh yeah.

Spanish questions go like ¿<phrase>? right

ais523

(incidentally, for those who don't know, most Linux keyboard layouts put compose on left shift-altgr by default, although I rebound it to caps lock)

Zachary T

Si<insert acute accent here>

HyperNeutrino

00:48

Sí

[compose] i '

Destructible Lemon

☭

HyperNeutrino

I remapped it to Alt Car because apparently I have that key and it makes sense to use it for alternate characters

Destructible Lemon

[compose] C C C P

HyperNeutrino

@DestructibleLemon that looks like the communist symbol

Destructible Lemon

@HyperNeutrino it is, I think

HyperNeutrino

00:49

oh

Zachary T

⍳t ⍳s.

HyperNeutrino

oh yeah it is (zooming to 500% lol)

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Find the distance between two points Your challenge is, given two points, find the distance between them. Inputs Inputs must be in the order of X1, Y1, X2, Y2. It can be given as four inputs (e.g. 0,0,3,4) or with each point as an array (e.g. [0,0], [3,4]). Output You must give one output, a...

HyperNeutrino

I'd be surprised if that's not a dupe.

Zachary T

U+1F595. (DON'T KICK!(

HyperNeutrino

00:50

What is that even?

Do I want to know?

Zachary T

Can't you guess‽

Destructible Lemon

@HyperNeutrino it's not that bad

@ZacharyT ok stop forcing the interrobang

Zachary T

I am forcing it‽

(LOL)

Well, a nonstandard punctuation mark has no correct usage, so ಠ_ಠ

Destructible Lemon

@ZacharyT u wot. maybe, but you're still forcing it

ATaco

A sentence ending with an interrobang asks a question in an excited manner, expresses excitement or disbelief in the form of a question, or asks a rhetorical question.

HyperNeutrino

00:52

Why are you using it so much!?!?!?

^ works too

Zachary T

I have no idea, must have rubbed off to a horrible level.

HyperNeutrino

Half the time I use Sandbox is just to figure out whether or not my challenge is a dupe :P

Because I'm bad at looking things up.

ais523

almost everyone has trouble finding dupes

Zachary T

I should get to the 05AB1E of this problem!

ais523

I'm getting better at it but it's still nearly impossible

HyperNeutrino

00:54

@ZacharyT please no

Zachary T

I'm officially out of letters/numbers to do puns with.

HyperNeutrino

@ais523 yeah. it's hard but with a bunch of people looking usually someone can find it

Zachary T

It's not like I have 7 more up my sleeve. (that's bad. even for me)

HyperNeutrino

you still have all of the other characters

Zachary T

Do I look like a dead FISH to you or something?

HyperNeutrino

00:55

yes

ais523

one underrated maintenance task that people should be doing is editing old challenges to make them possible to search for

often they manage to describe a problem without using any words that a typical person would use to describe the problem :-)

HyperNeutrino

that would be a good idea actually. meta post?

although if they were mass-edited that would cause the same problem as the sandbox rapid edit

Zachary T

Seriously though, I just can't Cm to run out of puns, can I?

HyperNeutrino

@ais523 such as instead of "exponentiate" use "repeatedly add repeatedly" :P

@ZacharyT Actually, you will eventually.

Zachary T

Wow, Nice one. (I think that's one)

ais523

00:57

@HyperNeutrino right, I don't mass-edit

normally only when a question is duped to that question, or when I answer the question

Zachary T

OCaml, wherefore art thou Caml?

HyperNeutrino

I don't think mass-editing is ever good

@ZacharyT these are starting to not make ¢ anymore

ais523

(in the former case, bumping it is probably a good thing, in the latter case, it's being bumped anyway)

oh, or when I bounty the question

but in that case I feel entitled to bump it :-)

HyperNeutrino

:P true

ais523

actually, does bountying a question naturally bump it?

Zachary T

00:58

I had to try. OK (It's so eC when language names are 1 character)

Adám

CMC: Given a cubic array, extract the major diagonal.

HyperNeutrino

@Adám lambda x:[x[i][i][i]for i in range(len(x))]

that's what you mean?

Adám

[[["A","B","C"],["D","E","F"],["G","H","I"]],[["J","K","L"],["M","N","O"],["P","‌Q","R"]],[["S","T","U"],["V","W","X"],["Y","Z","A"]]] → ["A","N","A"]

@HyperNeutrino Yeah, I guess.

HyperNeutrino

yay

Darn it, ŒD just applies to each element because it only works on 2d arrays :(

Adám

OK, maybe that was too easy. How about given an equal-sided N-dimensional array, extract the major diagonal.

ais523

01:02

I think a Jelly solution is something like (untested) ŒD€Ḣ€ŒDḢ

I might have put the dot on the H in the wrong place

HyperNeutrino

@ais523 i tried something like that

ais523

did you remember the €?

HyperNeutrino

oh wait lemme go try again

Adám

@ais523 We use £ :-)

ais523

oh, looks like I forgot an € too, fixed :-)

Zachary T

01:03

$. $RunningOutOfPuns = false;

ais523

@Adám but that means "run a function from elsewhere in the code with no arguments", which is much less useful than "for each element" is for this specific problem :-)

HyperNeutrino

@Adám I think that's )%ab!e

uh what did I just do

ais523

looks like you wrote "05AB1E" but held down shift for some reason

with caps lock on

HyperNeutrino

no, I pressed shift on the wrong things :P

ATaco

console.log(puns.length); // infinity

Zachary T

01:05

That was more of a PHP: PHP: PHP: PHP: PHP: PHP: ... pun.

HyperNeutrino

console.log(puns.quality); // -infinity

Zachary T

More like undefined.

HyperNeutrino

f=lambda x:1+f(x[0])if hasattr(x,'__iter__')else 0
lambda x:[eval('x'+'[i]'*f(x))for i in range(len(x))]

Zachary T

GNU APL everyday keeps the good dfns away

HyperNeutrino

@Adám ^^

looks horrible but ¯\_(ツ)_/¯

Adám

01:10

@HyperNeutrino Ugh. That look complicated.

HyperNeutrino

?

Adám

@ZacharyT Can we stop bashing?

Zachary T

Did you just make a pun?!

HyperNeutrino

@Adám It's not too complicated; the first just recursively finds the depth of the array and the second repeats [i] enough times to get down to a single element. :P

Destructible Lemon

@ZacharyT pyth-e stop using the inteo'bang

Zachary T

01:11

There, happy?

HyperNeutrino

no

ATaco

Interrobangs make my monitor sad.

Destructible Lemon

did anyone get my pun?

HyperNeutrino

I didn't. It just looked like a major typing fail to me.

Destructible Lemon

pronounce pyth as pith

HyperNeutrino

01:12

oh well that's not pun-ny enough for me to get it (btw that's <----- not a pun)

Adám

@ZacharyT The people behind GNU APL have a noble goal of providing a free modern APL. The fact that they have less experience in APL development than certain others does not make their effort less praiseworthy.

Destructible Lemon

@HyperNeutrino notice that interrobang is missing it's Rs

HyperNeutrino

ok i'm bored time to make a new programming language

@DestructibleLemon I noticed that :P

Destructible Lemon

@HyperNeutrino please link together more parts of the language graph

pyth-e = prithee

Zachary T

I know, I just like bashing things for the hashhell o vim

That's the ⍴○⍳n⊤ when APL becomes totally unreadable.

Adám

01:16

@HyperNeutrino Dyalog APL (second challenge): ×∘⍴⍉⊢

@ZacharyT ^ better?

APL eats multi-dimensional arrays for breakfast.

Destructible Lemon

@ZacharyT change vim to vitsy

Zachary T

I totally do love the fact that GNU is open-source. Eh, at least you can change ⎕IO though!

Adám

@ZacharyT As opposed to ngn, you mean?

Zachary T

Yeah, thought that was implied.

It's still open source and cool.

Adám

@HyperNeutrino CMC: major diagonal of multidimensional array where sides are not necessarily equal length (although it isn't ragged).

Zachary T

01:19

And APL chokes on arrays as elements of arrays, to be honest.

(Unless I just am bad at APL)

Adám

@ZacharyT What are you talking about? APL allows infinite nesting.

Zachary T

I know that. Just the way it handles them compared to multidimensional arrays is ... blech.

⊃f/

Adám

@ZacharyT Really? Are we rehashing that again? Sometimes it is useful.

Zachary T

It just ... seems unnecessarily long. It's probably due to the fact that I'm used to APL code golf.

Adám

@ZacharyT You mean 3 instead of two characters? In most cases the ⊃ isn't needed, and btw, APL is not a golfing language.

Dennis

01:24

@ais523 This keeps happening. ;)

Adám

CMC: Given any arbitrary array, replace every leaf with a copy of the original array.

Zachary T

By leaf, you mean non-array element, right>

HyperNeutrino

@Adám How does that work exactly?

Adám

@HyperNeutrino Cut short according to shortest dimension.

@ZacharyT Yes, if you do not consider a scalar an array.

HyperNeutrino

@Adám ah okay

@Adám Darn it, Jelly's ṁ doesn't work :(

@Adám Jelly, 2 bytes: Wṁ

Adám

01:29

@HyperNeutrino Whoah.

HyperNeutrino

yup

Adám

I'll stop making CMCs. They're all too easy apparently.

HyperNeutrino

No, ~~Jelly~~ Dennis is just too good

Destructible Lemon

Jelly dennis

brb making jelly dennis image

Zachary T

Jennis. LOL

HyperNeutrino

01:31

oplz what have I done

Dennis

hovers over ban button

HyperNeutrino

sorry Dennis I didn't want this to happen D:

please don't ban me D':

Zachary T

@Dennis Said to which person?

HyperNeutrino

not... like... this..... DD:

Ð:

Dennis

1 min ago, by Destructible Lemon

brb making jelly dennis image

HyperNeutrino

01:32

oh so i'm safe for now

Adám

CMC: (n+1)×√((n-1)/(n+1))

Zachary T

Replace your o's with greek omicrons. 100% guaranteed to get you kicked eventually.

Adám

^^ Does Jelly have a built-in for that?

HyperNeutrino

@Adám No, but Mathematica would.

Zachary T

I'm gonna guess Jelly's solution is 6 ~ 8 bytes long.

HyperNeutrino

01:34

brb trying to mathematically simplify/modify

Adám

@HyperNeutrino Good luck. I don't believe it can be.

HyperNeutrino

It's just sqrt((n-1)(n+1))

which is sqrt(n**2-1)

Zachary T

Lol...

Adám

@HyperNeutrino How in the world do you get that from my formula?

HyperNeutrino

move the (n+1) under the square root

then multiply

shoot i messed up

Destructible Lemon

01:36

@Dennis why is this so bad

does it hurt your feelings when I edit your image?

HyperNeutrino

@Adám Jelly, 3 bytes: ²’½

Zachary T

New CMC: Calculate (n+1)*sqrt((n-1)/(n+2))

Dennis

@DestructibleLemon That pretty much depends on the nature of the edit.

Destructible Lemon

@Dennis ok, but so far I haven't made any mean edits

HyperNeutrino

@Adam

Zachary T

01:39

Pinged the wrong guy.

Dennis

@DestructibleLemon As a general rule, I'd refrain from editing any user's avatar into something they could consider not nice. My mod status isn't special here.

HyperNeutrino

@ZacharyT no I didn't. well actually I did; I forgot the accent :P

Destructible Lemon

I think the meanest edit I made was that one where I made jan dvorak on the yelling at illusion

Adám

@Dennis Your avatar is already a menace.

Zachary T

You still screwed up in the math. Where in the world does ] come from?

HyperNeutrino

01:40

@ZacharyT me being bad

Destructible Lemon

@Adám destroyed

HyperNeutrino

Adám: 1 Dennis: 0

Round 2!

Dennis

@Adám All the more reason not to mess with it.

Zachary T

= Round 2

(factorial)

HyperNeutrino

Round 3! = Round 6

@ZacharyT darn you (kinda?) ninja'd me (?)

Zachary T

01:42

I seriously ninja'D you?

HyperNeutrino

the D didn't make sense there, it was just a letter ffs

Zachary T

D goes with everything. D is one awesome language.

HyperNeutrino

Wolfram|Alpha agrees with me

This, however, can't be simplified

Adám

@HyperNeutrino OK. OK. You're right. But that means there is something wrong with the text where I found the formula. Why didn't they use a simpler form?

Zachary T

To make people Think

Dennis

01:45

@HyperNeutrino Wolfram|Alpha is wrong. Who can spot the error?

HyperNeutrino

@Adám to confuse you

oh wait

I assumed n>0

didn't i

Dennis: 1 HyperNeutrino: 0

ATaco

ATaco: undefined

Zachary T

W|A isn't wrong. (Assuming n is positive)

Dennis

It lists sqrt(n-1)sqrt(n+1) as an alternate form, without restrictions.

HyperNeutrino

complex numbers?

Adám

01:48

@HyperNeutrino Ah, yes, I thought we'd missed something. So, back to my CMC!

Dennis

Equalities between multivalued functions are hairy.

Zachary T

Oh, n = -1

^As well.

HyperNeutrino

That's just sqrt(2)i

sorry no it's 0

which isn't wrong

because the original relation also gives 0

Zachary T

The original relation gives division by 0!

Dennis

The original is undefined for -1.

New Sandboxed Posts

01:49

0

A: Sandbox for Proposed Challenges

Golf a program whose behavior is independent from Peano arithmetic Your task is to write a program that doesn't take any input, and whose termination cannot be proved or disproved inside Peano arithmetic. This is code-golf, so the shortest byte count wins! Rules Standard loopholes apply Short...

HyperNeutrino

oh...

riiiight...

Adám

CMC: (n+1)×√((n-1)/(n+1))

HyperNeutrino

sigh back to the drawing board

I did the math on the whiteboard behind me lol

Wheat Wizard

@Adám What is the CMC?

Zachary T

Just do it easily using increment, decrement, multiply, divide, and square root in Jelly.

I can't do it right now, as I'm out of here.

Wheat Wizard

01:50

Implement that in code?

Adám

@WheatWizard Given n, evaluate that formula.

Wheat Wizard

Write code to do that?

Adám

@WheatWizard Yup, but golfed, if you can.

Zachary T

cя

Adám

@ZacharyT Really?

HyperNeutrino

01:51

what does the flip('R') mean?

Adám

@ZacharyT See you. Clever.

@HyperNeutrino "Ya"

HyperNeutrino

oh

@Adám ’÷‘×‘

ooh, I held down the ^ key for a random amount of time and landed right on the CMC

Adám

@HyperNeutrino even for complex n?

HyperNeutrino

appears to work?

Adám

@HyperNeutrino What? What is the ^ key, and what does it do?

HyperNeutrino

01:53

(3+4j) gives (2+4j)

@Adám caret reply; it allows you to type n carets in from of a message and it replies to the message n above it

Adám

^ does that work in chat by default?

Apparently not.

HyperNeutrino

no, it's a tacoscript

Adám

@HyperNeutrino I get 2.94+4.08i for 3+4i.

HyperNeutrino

lemme do a simple implementation in python3 first

Adám

@HyperNeutrino https://www.wolframalpha.com/input/?i=((3%2B4i)%2B1)*sqrt(((3%2B4i)-1)%2F((3%2B‌4i)%2B1))

HyperNeutrino

01:58

Using closest Wolfram|Alpha interpretation: ((3+4i)+1)

Adám

@HyperNeutrino URL isn't working: Enter ((3+4i)+1)*sqrt(((3+4i)-1)/((3+4i)+1))

Transcript for