The Nineteenth Byte

@H.PWiz because it's not smart (the deverbosifier doesn't try to build it with commands at all). you can just slice the alphabet and that would take a lot fewer bytes

What Wizard

@ASCII-only No I mistyped.

ASCII-only

@WhatWizard I wouldn't be surprised if it was 1. ASCII picture of cat or 2. Brain-flak encoding

What Wizard

You might be surprised then.

ASCII-only

1:12 AM

:| sadly it's not a valid TIO link when converted to base64, nor is it base64/some kind of base32/binary/7-bit ascii

@WhatWizard :O it must be this

What Wizard

Uh, what. I don't know how you got that.

ASCII-only

@WhatWizard see the third screenshot

Zacharý

Have you tried base36 ?

What Wizard

Hm ok. Not the intention

quartata

My first thought was the swapping cases were some sort of delimiter

If it's unary however there's too many 1s

[1, 3, 11, 2, 9, 1, 1, 1, 6, 1, 2, 2, 9, 1, 1, 1, 6, 2, 5, 1, 1, 1, 1, 1, 1, 1, 1, 2, 7, 1, 2, 1, 6, 1, 1, 1, 2, 1, 2, 12, 7, 3, 2, 1, 5, 2, 1, 1, 1, 2, 7, 1, 2, 1, 6, 1, 1, 1, 2, 1, 6, 12, 3, 3, 7, 2, 5, 3, 7, 2, 8, 1, 10, 1, 4, 2, 4, 1, 6, 1, 2, 1, 10, 1, 2, 1, 6, 1, 1, 1, 1, 2, 8, 1, 9, 3, 8, 1, 3, 1, 2, 1, 11, 2, 8, 1, 2, 1, 6, 1, 2, 1, 4, 1, 7, 1, 1, 1, 1, 2, 8, 1, 9, 3, 7, 2, 3, 1, 2, 3, 7, 16, 8, 3, 5, 1, 9, 1, 6, 1, 2, 1, 10, 1, 2, 1, 6, 1, 1, 1, 1, 2, 8, 1, 9, 3, 8, 1, 3, 1, 2, 1, 11, 2, 4, 1, 3, 1, 17, 1, 6, 1, 2, 1, 10, 1, 2, 8, 6, 1, 1, 1, 1, 2, 8, 1, 9, 3, 8, 1, 3, 1, 2, 1, 11,…

ASCII-only

1:27 AM

@Zacharý oh yeah. forgot to say: someone else is using Charcoal :D

Zacharý

What does GetVariable do? I'm confused

ASCII-only

@Zacharý gets variables with ASCII names

New Main Posts

1

Q: Sign-Swapping Sums

Given a nonempty list of positive integers [X, Y, Z, ...], your job is to determine the number of unique values of ± X ± Y ± Z ... For example, consider the list [1, 2, 2]. There are eight possible ways to create sums: + 1 + 2 + 2 → 5 + 1 + 2 - 2 → 1 + 1 - 2 + 2 → 1 + 1 - 2 - 2 → -3 - 1 +...

ASCII-only

@Zacharý example

Zacharý

I found out what I needed, I still don't get Charcoal though

ASCII-only

1:40 AM

@Zacharý :| why not

it's basically like a C-derived language with ASCII-art functions. and functions instead of operators

Zacharý

I get the general idea, just some specific things just throw me off entierely

ASCII-only

@Zacharý :| like what

Zacharý

Question: does charcoal sometimes just do nothing as a bug-like error

ASCII-only

@Zacharý yes. like e.g. Actually, it sometimes silently fails

it's useful for polyglots, although it might not be the best idea

also, protip: the AST is really helpful

Zacharý

"Test case 1 failed" ... yeah, it's helpful alright /s

ASCII-only

1:46 AM

@Zacharý :| that's not AST

@Zacharý you're probably getting your command line flags mixed up

Zacharý

ASCII-only

@Zacharý --, not -

-a also works if you want to execute it too

Zacharý

How are ~50 backslashes helpful? I feel as if I'm missing something

ASCII-only

@Zacharý :| you do have -v as an argument right

Zacharý

Sorry, I is noob. (I accidentally deleted it when I put in the ASTify.

ASCII-only

1:53 AM

@Zacharý :||||||

Zacharý

ASCII-only

@Zacharý protip: just use an existing answer as boilerplate for something that works :P then just change the code

@Zacharý yeah TIO args don't work like that. you need them in two separate boxes

Zacharý

CRAP, I put two -'s before the v.

ASCII-only

or -va

Zacharý

I'm used to APL. Not much command lines there. Just a top line of <QUAD>IO<STORE>0

How does while work? It's silently failing there...

OH MY GOD

I FORGOT A BRACE

Still doesn't work, but I missed that

ASCII-only

1:58 AM

@Zacharý link pls

@Zacharý :| I should probably add syntax errors for verbose mode

Zacharý

Does Move(Cast(b),Direction(d)) even work, assuming string b and number d?

I am a total n00b at Charcoal

ASCII-only

@Zacharý protip: you can Direction a string (and direction actually) too (although it's super useless)

@Zacharý yes

Zacharý

While(Not(Equals(Cast(b),0))) {
	Assign(Peek, b)
	Assign(
		Add(d, 2),
		d
	)
	Move(
		Cast(b),
		Direction(
			d
		)
	)
}

I have no idea. It seems that section isn't working to me.

Hey, it's down ~~sheep~~ goat

ASCII-only

@Zacharý where's the rest of it

Zacharý

tio.run/…

It's for that arson's lullaby challenge

ASCII-only

2:13 AM

@Zacharý also can't condition just be while (Cast(b))

Zacharý

I was being pedantic since it didn't work.

quartata

@ASCII-only what would a Charcoal answer to this look like codegolf.stackexchange.com/questions/84050/print-a-lego-piece

Zacharý

Short

That's what it would look like. Charcoal's entire purpose is ASCII-art.

ASCII-only

@quartata The naive way (string operations) seems like it would be almost optimal

quartata

can I see it

or just a rough byte count

ASCII-only

2:16 AM

@quartata yeah hang on

@quartata something like this

quartata

hm 27 bytes

pretty short

ASCII-only

canvas is probably shorter lol

quartata

was trying to scope out the competition in case I ever start developing cinnamon gum again

I love its execution model so much but am less interested in doing the little things necessary to shave off bytes

this was its best attempt back when the challenge came out, it beat V: codegolf.stackexchange.com/a/86667/45151

ASCII-only

@quartata cinnamon gum's?

quartata

32

ASCII-only

2:22 AM

but yeah SOGL and Canvas often outgolf Charcoal

quartata

github.com/quartata/cinnamon-gum/blob/master/DOCS.md

It's self-modifying, in essence

It essentially just followed from extending Bubblegum to compress a lookup table

But I love the direction it was going

ASCII-only

:O nice

~~clearly a practical language where basically every operator is user-defined is nicer~~

quartata

if I can ask you for another sample: how would it do on this? codegolf.stackexchange.com/questions/59934/…

This is not really ASCII-art, a bit different

maybe not something it's suited for

ASCII-only

it'd probably do ok, brb trying

quartata

ignore the bonus

it's probably not going to make things shorter

Zacharý

2:27 AM

I'll just wait until Neil destroys the challenge

ASCII-only

@quartata :| generate mode. it sounds like it was made specifically for that challenge lol

quartata

It wasn't actually, it's come in handy in other challenges

The concept follows from the philosophy IMO. Modes generate strings from a template, in a sense

ASCII-only

@quartata first try: it does terribly (and terribly slowly too)

Zacharý

Is Actually or Seriously newer, I forgot already.

ASCII-only

@Zacharý Actually

@quartata 27 and less slow, but still terrible

quartata

2:38 AM

perhaps not its forte then

Since you have actual drawing primitives and what not the Lego piece challenge was a bit more of a fair fight :p

ASCII-only

@quartata :P

quartata

Perfection

ASCII-only

this might work too. idk

quartata

Oh right the bonus

ASCII-only

2:57 AM

@quartata wait actually that doesn't specialcase 1,1

:| so it'd actually be longer

oh well

ASCII-only

3:32 AM

@Zacharý ok so how does this not work exactly

ASCII-only

3:54 AM

@quartata i thought it would be shorter with the bonus because there would be no need to call StringReplace. apparently not though

Adám

8:27 AM

@LeakyNun Is it a problem that our student competition claims 孔子 was born in the year of the dog?

Leaky Nun

@Adám eh... which year was he born in instead?

Adám

@LeakyNun According to the (simplified) algorithm of the competition, it comes out like that, so our example says:

      your_solution ¯551 ⍝ Confucius b. 551 BC
Dog

Leaky Nun

so it's dog then?

you're confusing me lol

Neil

@ASCII-only Charcoal, 28 bytes: ¿⁼1×θＩηo«ＵＢo Ｆ_-«×⊗⊕θι↷²ιＩη↷ includes special case

Adám

@LeakyNun Well, afaict, the animal year system wasn't used back then. I'm just worried that the way it is stated would be an insult, as calling someone a dog in China is… questionable.

Leaky Nun

8:30 AM

@Adám I don't know if it is an insult

I don't have much emotional attachment to Confucius lol

ASCII-only

@Neil It seems like the first cast isn't needed

@Adám well zodiac year != calling someone a dog IMO. so as long as that's clear it should be fine

Neil

@ASCII-only fails for height 1

Adám

@ASCII-only Right, but the way it is stated could be read as: What is Confucius? — a dog

ASCII-only

:/ it does. oops

@Neil hmm. I wonder if this would be valid

Neil

I guess so... I keep forgetting about that

(also your ipv6 answers would print the last line twice)

ASCII-only

8:46 AM

>_> yes they would. oops

user202729

10:07 AM

Can moderators spam-hammer posts?

Adám

@user202729 I'm sure they can. They can do almost anything.

Anush

10:40 AM

hi all

I have an idea for a challenge which is a generalized version of gray codes

but maybe it has been asked

Leaky Nun

then ~~post~~ sandbox it?

user202729

@Anush What is it?

Anush

it is simply this... you are given an array n_0, n_1, ... n_k-1

you must iterate over all arrays S of length k where the values in S_i are in the range 0 to n_i

by only incrementing or decrementing values by one

so if all the n_i equal 1 that is exactly gray codes

user202729

szkopul.edu.pl/problemset/problem/-Wvu18KzrxHIevX0iW1QpsMM/site/…

Anush

intriguing

is that actually the same problem?

@LeakyNun codegolf.meta.stackexchange.com/a/16377/9207

user202729

10:56 AM

@Anush Yes it's identical (although not on this site)

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Generalized Gray codes Input: An array I of k positive integers. Output: Your code must output all possible arrays O of non-negative integers of length k with the restriction that 0 <= O_i <= I_i. To get from one array to the next you may add or subtract 1 to one value in the array. Your cod...

Anush

@user202729 it's intriguing.. is there some way to see what was submitted to that competition?

or if anyone managed to do it?

user202729

@Anush No you can't. But I have a solution.

Anush

@user202729 nice! I look forward to seeing it for my challenge :)

I did look at en.wikipedia.org/wiki/Gray_code but it seems not to be covered there

user202729

I didn't actually submit it, just have the idea.

Anush

10:59 AM

oh cool. Could you possibly do me a favor

I need examples for my question

if you have a quick way to make any

user202729

If it's code golf then nobody would do the smart way.

Anyway, my solution is -- (should I just say it here or move to somewhere else to avoid spoiler?)

@Anush Make it by hand :/

Anush

ah.. well can I make the challenge require a smart way?

@user202729 good point

@user202729 can we get a private chat room?

I don't know how this works

user202729

codegolf.stackexchange.com/questions/48880/… -

Anush

@user202729 that's not the same, right?

Erik the Outgolfer

12:18 PM

@user202729 that's how mods delete spam

unilateral spam flag

Dennis

The annoying thing is that it doesn't work on deleted posts.

Erik the Outgolfer

yeah, having to undelete the spam

Anush

12:49 PM

any comments on codegolf.meta.stackexchange.com/a/16377/9207 ?

I hope my examples are correct

AdmBorkBork

@Anush Does the output have to be how you have ordered it? Or, alternatively, does it have to be ordered at all?

ASCII-only

@AdmBorkBork it does not have to be ordered, but they do mention a restriction

Dennis

In fact, due to the restriction, it can't be ordered lexicographically.

Anush

@AdmBorkBork I think the output has to be ordered but not according my the order I gave

if it isn't ordered, it's rather simple

Erik the Outgolfer

I think the order you have given for the test cases is correct

Anush

1:04 PM

@EriktheOutgolfer thanks.. I was worried I might have missed out a case

Erik the Outgolfer

btw, I think the challenge is still simple with the ordering, but for fewer languages

that said, your challenge isn't bad

user202729

@EriktheOutgolfer How can ordering works? ...

Erik the Outgolfer

however, a few remarks to be made

user202729

9 mins ago, by Dennis

In fact, due to the restriction, it can't be ordered lexicographically.

Erik the Outgolfer

@user202729 it's described in the challenge

no, not lexicographical ordering

user202729

1:09 PM

@EriktheOutgolfer I don't see it described anywhere.

Erik the Outgolfer

@Anush 1) even though I kind of disagree with it, there's a policy that says you have to specify the winning criterion in the post (e.g. not just with a code-golf tag)

note: I don't think many people will VTC over that, but anyway

Anush

@EriktheOutgolfer aha! What is the standard wording for code-golf?

Erik the Outgolfer

well, you want a code golf challenge, right? there's no "standard wording", you can just say "This is a code-golf challenge, the submission with the source code with the shortest length wins." (you can even shorten it a bit, just be sure to include who wins)

Anush

do I need to say if it is in bytes or characters?

user202729

in bytes.

Erik the Outgolfer

1:13 PM

most common tie-breaker is earliest post, but you don't need to accept an answer (and are not recommended to anyway)

you don't need to say if in bytes or chars, the default is bytes

Anush

ok thanks

user202729

If you want, you can add the "Don't let the short answers in golfing languages discourage you from posting an answer in other languages. Try to come up with the shortest answer in any language.". A similar variant can be found on many of KevinCruijssen's challenges.

Anush

let me add those

Erik the Outgolfer

I also add it when I post a challenge

and I think Martin Ender also does, it's mostly something to encourage you to post your solution regardless if the language you use won't ever be competing compared to other submissions

Anush

@Dennis I actually do want to add a more dramatic time constraint

if we change the output format to be (index, +-) which tells you which index should be changed and if it is incremented or decremented, we can restrict the time complexity to be constant for each output

but is that too hard?

Erik the Outgolfer

1:23 PM

@Anush I'd say not time constraint, but restricted-complexity if you'd like to restrict it

Dennis

What do you mean by constant for each output? Complexity is asymptotic.

Anush

@Dennis O(1)

@Dennis so the time to give each output doesn't increase with k

@EriktheOutgolfer thanks

user202729

@Dennis So O(output size).

Erik the Outgolfer

that's O(1)...

user202729

@EriktheOutgolfer That's not O(1).

Anush

1:25 PM

O(1) per output is O(output size) over all the outputs

Erik the Outgolfer

@user202729 what do you mean then? if the output size is 6, as in the first case, the code would be O(6), therefore O(1)

Dennis

@Anush There can't be any O(1) or O(k) solutions.

Erik the Outgolfer

maybe you mean O(N)?

but that's not constant anymore

user202729

@Dennis "doesn't increase with k" means "independent of k", not "O(k)".

@EriktheOutgolfer How can that explanation make sense!? That's like saying "if the input is 5, O(n) is O(5), which is just O(1)"...

Dennis

@EriktheOutgolfer O(6) == O(1) And O(N) is meaningless unless you tell us what N is.

Erik the Outgolfer

1:29 PM

well, N here is actually the output's length, which makes it kind of confusing

Anush

@Dennis Sorry if I am being confusing. The time to give each new output (which will be of constant size) should be constant. That is it shouldn't increase when k increases. Recall that the input array is of length k

Dennis

So O(size of output)?

Anush

@Dennis Yes exactly. In this case each output is of constant size

so O(1)

user202729

Measuring in bit complexity, outputting each "index" takes O(log k)...

Anush

ah yes.. constant words

not bits

New Main Posts

1:31 PM

0

Q: How to design a tree from an emmental cheese?

TL;DR : code an algorithm that could transform an hierarchical display of a tree in an xlsx file into a tree data structure. The shortest code would win the challenge. If two code are as short the one with the other, the best answer would be the most self-explainable one. Plot : Once upon a tim...

data-structures

Anush

@Dennis but the problem is that I am not 100% sure it is possible :)

it clearly is for standard Gray codes

user202729

@Anush Ok, it is possible.

Anush

@user202729 cool!

so now I am sure it is possible :)

user202729

(note that I edited your challenge a bit, it is not completely equivalent to Gray code)

Anush

thanks.. I saw

Dennis

1:32 PM

To be honest, I'm not a big fan of restricted complexity challenges. Many of us (myself included) don't know how to compute the complexity of their own code, especially not with all the unreasonable assumptions (e.g., arithmetic takes constant time) that are commonplace in CS.

Anush

:(

Erik the Outgolfer

yeah, that, and restricted time is even worse

user202729

@Dennis Addition and subtraction takes logarithmic time in bit complexity.

Dennis

All arithmetic?

user202729

@EriktheOutgolfer Restricted time is measurable, but I don't like it. It involves all kiinds of "on which machine" problems.

Erik the Outgolfer

1:33 PM

with restricted time, you have to test all submissions on the same host, which can be tedious

and then what if the host's owner goes away from PPCG and then who does the test

Anush

I hope it won't be confusing. I just want the time per new output not to increase when k increases

nothing fancier than that

Erik the Outgolfer

the issue with restricted-complexity challenges is that, for example, what is one statement in a language can really be O(N) behind the scenes

user202729

After some more thoughts. If it's measured in bit complexity, the output size is O(number of arrays × log(k)), so it should not be a problem anyway.

@EriktheOutgolfer Check out the implementation.

Erik the Outgolfer

well, good luck bothering with that

Anush

forget bit complexity :) that's too annoying

dzaima

1:36 PM

@user202729 what if the languages only interpreters are online (aka no source access?)

user202729

@EriktheOutgolfer Empirically measure the runtime is an option.

Anush

@dzaima I don't know.. this is a good place to ask :)

chat room ^^^ ?

user202729

It's the answerer's work to prove the time complexity, not us.

Anush

I really don't think the time complexity should be a problem. I mean assume that all arithmetic on 64 bit numbers is constant time and then just measure the number of operations

user202729

@dzaima Closed source is a thing.

ASCII-only

1:38 PM

@dzaima well a lot of languages are closed source too. what about those

user202729

I remember a Mathematica solution being disputed on because no one can prove the time complexity. Too bad?

dzaima

@ASCII-only those at least have the assembly viewable

ASCII-only

@dzaima that usually is nowhere near helpful

Anush

@user202729 but in this case it has to be O(1)

ASCII-only

especially for e.g. 100MB executables

Anush

1:39 PM

if someone really comes up with an answer where we can't tell if it is O(1) or not, that's basically the same as an answer that we can't tell is correct to me

user202729

@Anush No, for a different challenge.

Anush

@user202729 got you

Dennis

@ASCII-only 100 millibits? That's one small executable.

dzaima

@ASCII-only I'm saying that it's theoretically possible, not that anyone should go trough the pain of actually proving that

ASCII-only

@dzaima it's theoretically possible to find the actual time complexity of an online interpreter by running it repeatedly (i think)

@Dennis yeah it's from the ultimate golflang

Erik the Outgolfer

1:42 PM

@Dennis ...which doesn't even need one whole bit to fit anywhere

user202729

@ASCII-only That's to estimate the time complexity empirically. That's not "proof".

ASCII-only

@user202729 that's true

dzaima

@Anush by O(1) I'm assuming you mean each subarrays outputting time? So doesn't just returning the array take O(k)?

Dennis

@ASCII-only How? Any interpreter will only allow you to test a finite (and thus, insignificant) portion of all possible inputs.

ASCII-only

well, you may be able to exploit server vulnerabilities to find the executable/source :P

Erik the Outgolfer

1:43 PM

uh, I'm not sure about it being repeatedly possible to find the time complexity by "repeatedly running", even empirically, there is other stuff going on between you and the server on e.g. TIO, and, if the interpreter is client-side, running it repeatedly could simply be so fast that the naked eye can't capture the difference

ASCII-only

@Dennis well, for a very small subset of challenges that 1. have a single integer input and 2. where the runtime doesn't vary by a lot

Erik the Outgolfer

and, if it's closed-source, you have to deal with the actual executable

Anush

@dzaima oh sorry there is a part you are missing. This is for a new version of the question where the output is (index, +-) which tells you which index to modify and whether to add or subtract

Dennis

@ASCII-only Any integer that can be represented on an actual computer basically rounds down to 0.

ASCII-only

well, not if the only supported inputs are integers within a certain range (but yeah i guess arbitrary precision is basically everywhere nowadays)

Dennis

1:47 PM

If the only supported inputs are integers within a certain range, all implementations are O(1).

New Main Posts

0

Q: Heuristic function in "cave puzzle"

I want to solve Cave puzzle (https://www.gmpuzzles.com/blog/cave-rules-and-info/) using A* algorithm in R programming language. I have already implemented Breadth-first algo. This is part of my code: v1 = c(2,0,0) v2 = c(0,0,3) v3 = c(0,4,0) v4 = c(1,0,0) v0 = c(0,0,0) arr <- array(c(v1,v2,v3,v...