The Nineteenth Byte

Am I the only person that thinks a hard ceiling of 256 characters in a programming language is a bad idea?

Once you assign the 256th character, you basically shut the door on further feature extensions

@Mego The herokuapp link in github.com/Mego/Seriously/blob/master/docs/index.rst#id1 is dead

Anonymous

@Pavel Blame @Sherlock9

@Mego btw, you participated in hacktoberfest this year, right? Have you getten an email or anything about it yet?

Having only 256 characters in a normal, non-golf programming language is fine, but in golf, every signle charater has meaning, so there's no way to fit a 257th command inside and make it fit

Anonymous

12:16 AM

@eaglgenes101 Not true. I've reserved one of the 256 characters as a way to switch command pages.

Okay... well

Anonymous

@Pavel Not yet

Ok. Me neither, and I was a tad concerned.

Anonymous

It's only been 2 days

Maltysen

@Pavel i did

12:18 AM

Who would have thought #hacktoberfest shirts would be collectors items? #TuesdayThoughts https://twitter.com/sheharyarn/status/925387773669818369

Tweeted by digitalocean on October 31, 2017 at 3:57 PM

It would seem that some people have already gotten their shirts

Maltysen

i got a tracking number

for my shipment

but it seems like they're doing them in abtcher

one for before oct 15

one when I finished it, which was like a wekk earlier

and i'm guessing for after oct is over like u

@Maltysen abtcher?

@Mego Probably not a gentle change, messing up some solutions, but I feel like - should push b-a instead of a-b. Same for /, %, and a few others.

Maltysen

@Pavel ._.

batches

o

Anonymous

12:23 AM

@Scrooble That doesn't make any sense in a stack-based language.

The purpose of a golfing language is to get as much practucal use out of each and every byte as possible

I'd suggest deprecating stuff like "Print hello world" and move them to an alternate codepage of "common tasks". A way to print hello world isn't going to help me much in most nontrivial golfing challenges, certainly not enough to hog one of the 256 bytes on the codepage.

And besides, purpose-built languages are always going to have you beat in tasks like those no matter what you do

The best you can hope for as a general golfing language is to approximate kologmorov complexity of human reasoning on code golf challenges

@Mr.Xcoder Try it Online! Not golfed, just wanted to do it in Funky.

You would probably like Jelly.

Anonymous

12:38 AM

@eaglgenes101 See, you're assuming that the goal of Seriously/Actually is to be the absolute best at golfing. That's not correct.

@DJMcMayhem @eaglgenes101 I still think the concept of Cinnamon Gum is more than viable but I don't really have the time or interest for adding all the little things it needs

Anonymous

It was an experiment in developing a golfing language

fair

Anonymous

If I was trying to make an optimal general-purpose golflang, I wouldn't have gone with stack-based

@flawr Your latest challenge reminded me of this a bit:

Polynomial kernel

In machine learning, the polynomial kernel is a kernel function commonly used with support vector machines (SVMs) and other kernelized models, that represents the similarity of vectors (training samples) in a feature space over polynomials of the original variables, allowing learning of non-linear models. Intuitively, the polynomial kernel looks not only at the given features of input samples to determine their similarity, but also combinations of these. In the context of regression analysis, such combinations are known as interaction features. The (implicit) feature space of a polynomial kernel...

The partition itself is still a regular plane, but in a new space created by applying the kernel (see the picture for how the plane captures the original points in an ellipse)

LegionMammal978

1:00 AM

Hmm...

If you were operating in 4/4 time, how would you notate a note which would take exactly one third of a measure?

...wait, duh, that's what triplets are for

@LegionMammal978 :P

RProgN2 was my first attempted at a golfing language, and it is still my best golfing language.

A: Sandbox for Proposed Challenges

Code bowling challenges are difficult

New Sandboxed Posts

0

Reverse a regex The .NET regex engine has the ability to process a regex in right-to-left order. There are some subtle differences between right-to-left and left-to-right order, so for the purposes of this challenge, right-to-left order is equivalent to running a reversed regex on a reversed str...

More specifically, designing good ones are difficult, since almost every language has a means of arbitrarily lengthening a program

1:11 AM

Lenguage does pretty good at code bowling.

@ATaco No it doesn't

(Depending on the challenge spec)

The most practical I can think of is to make it a robber challenge, and create a sister cop challenge to enforce the requirement that answers be nontrivial

Or is it the other way around?

Q: The Bowlers-Golfers Fraction War

Cops must bowl, robbers must golf the cop's answers.

Dennis

17

Introduction Let's start by arranging all fractions from 0 to 1 in order of lowest denominator, then lowest numerator: 1/2, 1/3, 2/3, 1/4, 3/4, 1/5, 2/5, 3/5, 4/5, 1/6, 5/6, 1/7... Note that duplicates aren't counted, so I haven't listed 2/4, 2/6, 3/6, or 4/6, since they already appeared in th...

1:15 AM

10^10^20's a pretty good score I think.

i found it hilarious that someone managed to hold an absolutely giant score for a moment just because no one could grok his code

Funky has been so much fun to write.

table.find = (t,val)=>{
    var iter = pairs(t)
    var found = false;
    var cur;
    while((!found) and nil != cur=iter()){
        if(cur[1] == val)
            found = cur[0];
    }
    found;
}
table.uniques = t=>{
    var out = {}
    for val in values(t){
        if((out::find(val)+"")=="false"){
            out::push(val)
        }
    }
    out;
}

table.xor = (a,b)=>{
    a = a::clone();
    b = b::clone();
    while(#b){
        var tmp = b::pop();
        var loc = a::find(tmp);
        if((loc+"")!="false")

This code actually looks C-Like :D

@ATaco *JS

(Well, I suppose this looks mostly like JS)

Ninja'd..?

The important thing is that it doesn't look Esoteric.

Perhaps the code bowling challenge could be increased by the fuzz criterion

For every character whose deletion doesn

't cause syntactic failure, subtract 2 characters.

Or something

NieDzejkob

1:24 AM

that would inappropriately destroy Actually

Yeah, could use more refinement, because I could trivially bypass it by stuffing my code with parentheses pairs

Funky would also die.

Funky doesn't throw syntax errors, it just ignores the code instead.

Any one parenthical removal would cause the compiler to crash and burn, but would be obvious filling mechanism

This is relevant if you remove the national flags

Trying to execute code with a syntax error in funky.

1:27 AM

@Pavel Wait is that location pulled from my location data????

Or perhaps use kologmorov complexity as the criterion for scoring; your score is the lowest number of characters someone can use to generate your code in pyth/cjam/golfscript/husk/jelly/whatever

@eaglgenes101 ...

that's not very easy to score then...

As I said, code bowling is easier if paired with a sister golfing challenge

@Christopher2EZ4RTZ The last two should be switched

@eaglgenes101 No... there have been several decent (and sane, unlike yours) bowling challenges

@ASCII-only yeah I would do that but I still won't get starred

1:31 AM

@Christopher2EZ4RTZ pls. no. starbait.

Yeah, I should probably put the bowling ball down until I get some lessons

@ASCII-only That was not intended as starbait.

> but I still won't get starred

At the very least, kologmorov scoring would at least kill off the algorithmically-expanded solutions quite quickly

@ASCII-only Wat, that was a fact

1:33 AM

@Pavel That means if ... end, for ... next etc. are safe though

@Christopher2EZ4RTZ well yeah? but you're saying you didn't do that because you still wouldn't get stars isn't it

@ASCII-only The meta levels of starbait is too high

@Calvin'sHobbies Probably too late: but for i in range(gradient > 1 ? x : y): fill line from f(x) to f(x+1) (or f^-1(y) to f^-1(y+1))

there'd be edge cases for lines passing through a corner though, maybe you'd need floor + ceil or something

Oh, I should make a Range function.

Calvin's Hobbies

@ASCII-only I figured it out by just using Bresenham's algorithm but adding an orthogonal block when a diagonal move is detected.

Oh, that would work too

1:43 AM

@Christopher2EZ4RTZ GeoIP

@ASCII-only if/end aren't that bad. I've never seen for/next.

@Pavel CASIO BASIC :P

TI and CASIO BASIC are horrible for other, more pressing reasons.

1-indexing...

All variables global...

Funky has 0 indexing, even though lua was a strong inspiration :D

@Pavel like what

1:53 AM

@ASCII-only The inability to put lowercase characters into strings.

@Pavel citation-needed

@Pavel you can put lowercase characters into strings...

In TI-BASIC, at least.

Dunno about CASIO

In CASIO BASIC too...

<- does not like TI calculators at all

I have come to realize in recent weeks that TI-BASIC is essentially a golfing language.

1:56 AM

@ASCII-only You can't enter lowercase letters

TI-BASIC was one of the first languages I learnt.

There is literally not buttons for that.

(I wasn't very good at it)

@Pavel Program A: AsmPrgmFDCB24DEC9

On the home screen: Asm(prgmA)

Have fun with lowercase!

This is for TI

@Pavel You can...

1:59 AM

@Scrooble I don't actually have a TI with me right now so I'm going to have to just ask what that does

@ASCII-only Not on my TI-84

6 mins ago, by Scrooble

In TI-BASIC, at least.

Enables lowercase. Alpha now cycles upper-lower-off

Probably not in older TI calculators of course

@Scrooble O_o this is hacky

Yup

@Pavel

whoah

3:26 AM

@Pavel yeah the lowercase letters are in the TI code page (they're two bytes though)

thats a pretty nifty little snippet for switching cases though

In other (sad) news about 20% of SE just got laid off, including a good chunk of the design team :(

@quartata :(

our design might be delayed until late November

3

also rip Pops

@quartata but it seems like they have a function to activate it considering it can be done in 5 bytes of asm

@quartata source?

@quartata don't you mean delayed by 6-8 weeks

3:32 AM

@ASCII-only yeah, it's like a little gameshark code. it's adorable

@Pavel check twitter or meta tavern transcript

@quartata meta tavern?

I'm not actually trolling

@Pavel chat.meta.stackexchange.com/rooms/89

fun fact I typed that URL from memory because I cannot be bothered to copy paste

Sad day

DJMcMayhem

4:01 AM

@quartata Delayed until November?

That implies we thought it would happen earlier than that

That would be an improvement

, as a pair operator, thoughts?

I think it's a Geobits meme

it was going to be yesterday

like, for real

Riiight

@ConorO'Brien back me up

4:06 AM

Yeah I'm going to need a source for that

also has anyone here used Backbone

Conor O'Brien

Hello who needs a back

@Pavel ye of little faith

November 1st is a meme

oh I just noticed the connection between those messages

back me up -> Backbone

that had to be subconscious

Conor O'Brien

There's a reason the meta meme page was fm burned

(Minus fm, blame mobile)

Q: Bijective function ℤ → ℤⁿ

4:13 AM

I see.

Or rather, I don't.

At all.

New Main Posts

4:36 AM

1

It is trivially possible to create a bijective function from Z (the set of all integers) to Z (e.g. the identity function). It is also possible to create a bijective function from Z to Z² (the set of all pairs of 2 integers; the cartesian product of Z and Z). For example, we could take the latti...

code-golf math number

Do the mods have access to tag burnination?

DJMcMayhem

6:08 AM

@quartata You could back yourself up with a tavern transcript link...

Anonymous

6:47 AM

in Tavern on the Meta on Meta Stack Exchange Chat, 15 hours ago, by Tim Post

Today was a rough day for us - you'll notice that some folks no longer have diamonds. We had to make some tough decisions and cuts in order to continue to advance toward some pretty critical business goals. While we're usually very open and public about our operations, please allow some privacy for folks that have been affected; the biggest thing for anyone to take away from realizing what happened is that we're taking exceptional care of the folks that had to leave us [1/2]

Anonymous

For once, this isn't quartata trolling

Anonymous

Shocking, I know :P

DJMcMayhem

@Mego Oh, i know there were layoffs

I'm skeptical of the November 1st thing

3 hours ago, by quartata

it was going to be yesterday

Anonymous

Oh yeah, that's 100% trolling, which means overall 50% trolling

Anonymous

7:01 AM

I think at this point, it's safe to just assume we'll never get our design :P

7:35 AM

SE designers are also golfing the amount of work they need to do for us to get a design; thus they only removed the "beta" next to PPCG

Anonymous

@Fatalize Even that took nagging iirc

Q: matrix 3*3 input , output how many numbers in the matrix are less than zero . i've been working on it since yasterday?

10:00 AM

Anyone from Philadelphia here?

flawr

10:47 AM

@quartata SVMs FTW:)

@AdmBorkBork what is owShel?

New Main Posts

11:40 AM

-2

import java.util.Scanner ; public class Assign1Q4 { public static void main(String[] args) { int c , d , e , f , g , h ,i , j , k ; Scanner in = new Scanner(System.in); int[] matrix = new int[9]; System.out.print( "Enter the elements of matrix : "); Scanner st = new Scanner(System....

matrix

New Sandboxed Posts

12:04 PM

0

A: Sandbox for Proposed Challenges

All ASCII Art Input You will be given a set of x, y, and A pairs, where x and y represent a coordinate on a 2-d plain, and A is some character. You may take these pairs in any convenient format, e.g. a list of lists, three separate lists, a list of pairs of lists, etc. Output You will output ...

mınxomaτ

12:16 PM

@Mego At least they are not ""pivoting"" (i hate this term) away from Talent & Jobs. That service is too expensive to suddenly stop existing.

AdmBorkBork

@flawr Har har

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Counting Fibonacci Orbits code-golfsequencefibonaccinumber-theorycombinatorics If we define a Fibonacci-like sequence as fk(n) = (fk(n-1) + fk(n-2)) % k, for some integer k (where % is the modulo operator), the sequence will necessarily be cyclic, because there only k2 different values for (fk(...

New Main Posts

1:19 PM

0

Q: How to code colors in hex

Hexagonal tessellations, or tilings of the plane, are three-colorable -- meaning that using only three colors, such as Red, Blue, and Green, every hexagon can be colored without any two hexagons that share a border sharing a color. A partial example showing such a pattern. Given a partial hexago...

code-golf ascii-art graphical-output

Stephen

2:10 PM

@Fatalize Not from Philadelphia proper but I live(d) somewhat nearby, though I'm in college now

@Stephen Do you recommend using SEPTA for transport?

(i.e. is it safe at all hours, is it reliable, etc.)

2:25 PM

Would someone be so kind as to verify if this answer of mine actually works? I can't test the colors right now because chromebooks :c

@HyperNeutrino doesn't seem to work

Darn. What does it look like? (pasteboard pls, imgur is blocked)

Stephen

@Fatalize didn't live close enough to the city, sorry :/

Oh wait I just realized that my color filling "algorithm" has a non-zero probability of breaking

Anonymous

for i in[-1,0,1]:q=2-abs(i);i+=k[0];r=C[k[0]][k[1]];C[i][k[1]+q]|=r;C[i][k[1]-q]|=r
IndexError: list index out of range

2:29 PM

oh? weird

yes

But TIO doesn't error. / for which input?

it IndexErrored for me too

first one actually, dunno how I missed that lol

Anonymous

Same input as on TIO, but as a string instead of a file (with appropriate modifications)

2:30 PM

It would have to be caused by there being no available colors left because it filled it incorrectly. My attempted golfing failed lol

Anonymous

Full code I used:

Anonymous

#!/usr/bin/env python3

def f(x):
	c=[[i,j]for i in range(len(x))for j in range(len(x[i]))if"*"==x[i][j]];l=lambda e:[[e]*len(r)for r in x];C=l(0);C[c[0][0]][c[0][1]]=1
	for k in c[1:]:C[k[0]][k[1]]=list({1,2,4}-{C[q[0]][q[1]]for q in c if 0<abs(q[0]-k[0])*2+abs(q[1]-k[1])<7})[0]
	for k in c:
		for i in[-1,0,1]:q=2-abs(i);i+=k[0];r=C[k[0]][k[1]];C[i][k[1]+q]|=r;C[i][k[1]-q]|=r
	return"\n".join("".join("\033[%dm%s"%(30+C[i][j],x[i][j])for j in range(len(x[i])))for i in range(len(x)))

inp="""     / \

so I actually need to breadth-first fill this? sigh

@HyperNeutrino No you don't need to.

@Mego the first two "lines" of inp are actually one line because of the \ lol

@user202729 oh rly? my current approach seems to bork though

2:31 PM

Just fill as if you are filling the whole range, but only fill those asterisks.

@Stephen ok no problem

Guess I'll take the taxi when I arrive

@HyperNeutrino I used sys.stdin.read() and still got an IndexError though

Anonymous

@HyperNeutrino Oh, right, duh. I added r and it looks like something

and to leave I'll ask the hotel staff whether it's fine or not

yay

huh. I think the IndexError is caused because the program is not filling colors correctly. It's filling in the first color of [1, 2, 4] that works, but that has the possibility of causing future ones to not work

Solution: brute force :D

there are only 3^(n-1) possibilities. Not a lot :D

2:33 PM

@HyperNeutrino So what is your approach? (I can't read Python, especially golfed Python)

Set the topmost (then leftmost if tied) to red, and then going through all of them in some order, find the colors of all colored neighbors, get a list of colors remaining, and take the first one

"some"...?

however, for an example like [[?, ?], [?, ?]], it might set [[1, 2], [3, ?]], which would give no remaining colors, while it could have done [[1, 2], [2, 3]]

@user202729 I think it goes right and then down but I'm not sure

anyway brb o/

At least it's not gtg.

Anyway color *'s by their x-coordinate mod 3 also works.

Which should be much simpler than BFS or something like. | It will not set the row 2, col 1 to 3 because it is not the first (smallest value) color remaining.

@user202729 oh... that's a really excellent observation +1. mind if I steal that trick? :D

2:39 PM

No problem.

yay thanks!

also I have an unnecessary lambda again :c

For the whole grid that is actually the only solution (up to color permutation), I only parametrize it.

Anonymous

@HyperNeutrino PNG of output with fixed input

what am I supposed to do with that :o

huh

that's an xxd

2:41 PM

Turn it into a png and view. Very inconvenient.

oh huh ok I'll see if that's possible on my chromebook lol

I think you can just do xxd -r hn.png.xxd hn.png and then open hn.png

> chromebook

Anonymous

@user202729 Not loading

2:44 PM

fixed and undel'd

@user202729 that image doesn't exist btw

@HyperNeutrino oh there's no xxd in there?

For the record I replace with regex .*: -> nothing, ` .*` -> nothing, and paste everything into a hex editor and save.

@EriktheOutgolfer nope

Anonymous

@user202729 Well when the most convenient option (imgur) isn't available, you go with a format that is easily transmitted over text and is easily recognizable and worked with (xxd hexdump)

You're suppose to click at the image. Pasteboard is weird.

What is xxd?

2:45 PM

ooh cool

Anonymous

@user202729 One of the most-used bash builtins

hexdump