The Nineteenth Byte - 2021-06-21 (page 1 of 3)

« first day (3796 days earlier) ← previous day next day → last day (1030 days later) »

12:50 AM

0

Q: Minkowski sum of two convex polygons

Background Minkowski addition is a binary operation on two sets of points (usually geometric objects) in the Euclidean space. The Minkowski sum of two sets \$A\$ and \$B\$ is formally defined as follows: $$ A+B = \{\mathbf{a}+\mathbf{b}\,|\,\mathbf{a} \in A,\ \mathbf{b} \in B\} $$ To say in plain...

code-golf math geometry

OldSandboxPosts

Sandbox posts last active a week ago: Recursive Stalin Sort, Is there a left-right connection?, Sandbox Notes

1:12 AM

do we have a challenge that's just take a set of points and output it in clockwise/anti-clockwise order?

not that i feel like posting one, just want to know if i can steal an approach for ordering points like that from anywhere

1:38 AM

@hyper-neutrino Do you mean ordering by atan2?

does that work though?

wait the polygons are convex

ok this'll work then lol

~~also i'd need to order by atan2 relative to some central point right~~ nvm yeah that's obvious

caird coinheringaahing

1:59 AM

@hyper-neutrino Interesting that Jelly allows you to do :,% with zero as both arguments, but not d: Try it online!

... interesting

Hello everyone

caird coinheringaahing

It calls Python's divmod rather than using Jelly's divide and modulo functions

Just submitted a BF solution to the half challenge

In spirit of code trolling

question: is the ability to exponentiate by a complex number ever going to be helpful

(not the base, but the power)

caird coinheringaahing

2:09 AM

As in (a+bi)^(c+di)?

yeah

since I'm not using Python i'm not sure if it's worth the effort

caird coinheringaahing

Probably. I think you can do x^i to do rotations in the complex plane or something like that

oh also the numbers may be arbitrary decimals

caird coinheringaahing

Check the complex-numbers tag and see how often it shows up

i think the best way to go about it is $exp(y log x)$ as suggested in the math chat

2:36 AM

Floating point errors are too bad

Can I directly convert result of a trigonometric function to fraction in python?

not really

just use sympy if you want precision

What python does is to first do the equation in float and after that convert to Fraction

Let me check if could use sympy

CMQ: Helpful ascii-art built-ins

ASCII line in any direction, mirror, vertical mirror, direct base conversion (like convert a big number in base-10 to base-3 with some defined letters) etc..

You might take a look at Charcoal, Canvas and SOGL

I'd add overlay to that list (replace spaces in image A with chars at that position in image B)

3:00 AM

\o/ my complex exp seems to be working

3:25 AM

@lyxal quad palindromize with/without overlap in required directions

Unrelated String

pad to rectangular is probably also good to have if you don't have a ton of ascii art oriented builtins

3:52 AM

@Wasif The result of any trigonometric function with an arbitrary number is very surely not rational (i.e. is not an exact fraction at all). You will likely get a better result if you encode your entire logic into sympy, instead of trying to get a fraction out of irrational numbers

4:34 AM

is there any way of preventing js from freezing the entire window when it's taking time executing

namely, i want to output infinite integers, but the window freezes and i can't see the output being produced as the code is running (and if i cap it at 128KB output, it blocks for a long time before display it all at once)

my "print" function just appends to output.value where output is a <textarea>

Redwolf Programs

Yes.

Just make it async, and add an await new promise(r => setTimeout(r, 0)) every 100-1000 iterations or so

Looks like I've misseda bunch of stuff :/

Vacations suck

er, wdym iteration lol

Redwolf Programs

Oh, for bigint stuff

Hmm, I don't think so

Because any sensible JS implementation is JIT and they're usually single-threaded, I think you'd have to find some way to break it up into smaller operations

hm :/ ok

is there any way to kill a function that's currently running?

Redwolf Programs

Not to my knowledge :/

4:43 AM

rip

Redwolf Programs

Maybe with some sort of web worker thing

But that'd require doing a bunch of work beforehand, it's not something you can do if it's already running

i guess i could set up some check somewhere and throw an error when a flag is raised but that'd only be able to break out of a loop and not if it gets stuck on a sub-operation

ok so i am not very smart

i tried adding a way to manually kill the program

but that doesn't exactly work when the program freezes the window until it's done

Redwolf Programs

The async trick I mentioned above could help

It gives the browser a chance to respond to things you do in the console

(How reliable and useful it is depends on how slow the sub-operations are)

would i need to add async to like all of my existing functions then?

Redwolf Programs

Only the ones upstream of where you add the await new Promise(...)

4:53 AM

hm. alright i'll think about it. thanks

most blockage will be caused by infinite or really long sequences and those are all lazily calculated so i might just be able to get away with restricting output like that

@hyper-neutrino Only if you use Web Worker (it supports .terminate())

Redwolf Programs

I wish instead of an await operator the JS people had just added a special .await() property to promises.

That avoids needing async all the way up

From what I learned from Rust, having a non-async .await() method would be highly nontrivial at least

To run a promise and block on it until it finishes, you need to create a new event loop dedicated to that promise

5:16 AM

this is why i like to stay away from threading related things lol

@hyper-neutrino This might be of interest. Basically put your language impl in the Worker and send input and receive output from the main thread

oh, thanks, i'll prob give this a shot later

5:42 AM

is there an easy way of seeing if a user in the user list is an RO?

when their user card is entering the room, their name will be italicized if they are an RO, but once they've entered the list, clicking their profile doesn't seem to show any indication (except that non-ROs say "add as RO" and ROs say "remove RO") unless i click to their user page itself and search for the room, or look at the room's RO list

6:32 AM

0

Q: Implement a cleave function

Think of cleave as the conceptual inverse of map. If map applies a function to each number in a list... map([1, 2, 3], x -> x * 5) -> [5, 10, 15] then cleave applies each function in a list to a number. cleave(5, [x -> x * 2, x -> x - 1, x -> x * x]) -> [10, 4, 25] More formally, given: \$n\$, a...

code-golf functional-programming

7:13 AM

@NewPosts friggin' ninjas

12 answers

Before I could even answer

lol

Well, it always happens on simple challenges

Unrelated String

^

there's not much to do in terms of either devising an approach to begin with or golfing its implementation

@Wasif where's the PowerShell answer

7:31 AM

That Java answer looks downright reasonable, but then you get to how it's actually called...

I think the lua answer is invalid, it assumes input as a table, but then outputs in StandardOut

Actually, the challenge specification just says that it can be any reasonable input or output. I suppose printing is reasonable.

I had an idea for a KoTH - Simon Says, on a collasping platform

Ooooh?

Basically, you can move on a platform, and when Simon says to do an action, you have to do it, like in the actual thing

but every ten turns, the platform shrinks by 1 unit

makes sense

or not?

what will be the criteria to decide who gets kicked off the platform

completely random?

7:38 AM

@Razetime would do soon

And what defines an action?

This sounds intriguing, but yeah I want to hear more!

@ATaco moving or staying still

@StackMeter ??? Couldn't understand

I intend to post it now

7:39 AM

@StackMeter without sandboxing?

@StackMeter So the action Simon would say might be move n blocks north?

And every turn you get to decide to move to an adjacent block or stay still?

And every 10 turns, the platform shrinks until there's one survivor?

Is there incentive not to stay in the middle?

@Wasif no, it's being sandboxed

Do we know where the middle is?

every turn, you can move up to half the length of the platform

Perhaps you'd start randomly placed?

7:42 AM

@AviFS yes

@AviFS also yes

So then why not stay there?

I have an idea - every other turn, you must move

And what about only one person per square?

Would that be interesting?

Wait - better idea, every ten turns, a random platform collapses

You can also have squares randomly be blocked off every turn, and forget the Simon Says piece

@StackMeter Jinx!

7:50 AM

lol I just scared myself by thinking I saw movement in the 'window', it was myself in the mirror

haha, nice

(If someone else can start saying "haha", I'd really appreciate it.)

(We're on the internet, aren't we?)

@AviFS I do a small amount

@AviFS that would probably end with unresolved ordering issues

Ah, good point.

I suppose there could be a rotating turn order

But that's probably unnecessarily complicated

@AviFS most people prefer to say "lol" or little of them "XD"

7:58 AM

@rak1507 Not good enough. I've filled up a page of "haha" chat search since June 12, and since then you haven't said it once

@Wasif Phew! Thanks so much. I was so worried, lmao

You're right. Just today and yesterday has entirely filled up a page of 'lol' chat search

I feel less unprofessional and childish now, haha

hyper-neutrino is the only person who has said XD, lol and haha recently AFAIk

pfft. professional? TNB is never professional

I feel like I still outdo any and all individual lollers...

caird coinheringaahing

@NewPosts @hyper-neutrino Is it allowed to take a tied function instead of a list of functions for Jelly? Especially when you can take a list of link indices which point to the functions

@hyper-neutrino Hey!

8:00 AM

@cairdcoinheringaahing not sure. The latter would probably be more correct

someone perform an "ok" search

Unrelated String

ok

but... i'm not sure how to map over get-by-index

ok has 372 pages, 50 per each

you can't exactly call the map quick on a quick

8:03 AM

HAH! I win on professionalism @hyper-neutrino

user image

Unrelated String

i want to say you could do something like ṛĿⱮ but knowing how jelly can be with adicity sometimes maybe not

user image

what professionalism?

That's the end of the first page of search terms. You filled up the page faster.

Of course, you're way more active...

Anyway, I still feel better, so thank you :p

caird coinheringaahing

@UnrelatedString Nope, Ŀ is monadic

Unrelated String

8:04 AM

as i feared

:|

Ŀ returns a link of arity max(1, index.arity) and calls a variadic link on the index function on x, y, gets the outmost_link at that index, and then calls that on x

so this might work

@UnrelatedString your exact code works

Unrelated String

never mind then, nice

do you want to post that

i only put my 6-byter in because I felt v<each> was too trivial

Unrelated String

yeah sure why not

i'll leave in my longer one cuz it's not the main sol anyway

caird coinheringaahing

8:14 AM

@hyper-neutrino Imagine using ÷2 :P

shhh

Unrelated String

it's not even easier to type than H

0

A: Sandbox for Proposed Challenges

I like my job, but sometimes it can get a bit tedious and boring. As motivation, I want to know how many hours of work are left before I finally get my two days of weekend and relaxation. I work every day from Monday to Friday, totalling 38.5 hours, starting at 9am and ending at 6pm. Every day ...

caird coinheringaahing

nuke as spam pls

ninja'd

caird coinheringaahing

8:18 AM

Ah :P

Well, I got a flag in :P

yeah, it was an answer not a question so since NPSP didn't post it I only noticed after smokey pinged me :p

caird coinheringaahing

Ngl it seemed kinda sus

Unrelated String

when i saw the account name i was definitely expecting a shitpost instead of a bot ad

ಠ_ಠ

caird coinheringaahing

> The ideas I gather from here cannot be paid with money

Is this supposed to be saying that "The answers here are so good I want to pay you as a reward, but I can't"?

8:26 AM

got a nice byte count to make people jealous at me

That's quite the formula!

gonna post a list of all my running challenge ideas here:
1. [here](https://codegolf.meta.stackexchange.com/a/23588/100887)
2. [here](https://codegolf.meta.stackexchange.com/a/23587/100887)
3. [here](https://codegolf.meta.stackexchange.com/a/23474/100887)

Feel free to leave any feedback

This part is giving me a lot of trouble:

> By that, I mean taking a single continuous slice of the set of panels, we can represent a number, and we can do that for 1 through n.

I'm still trying, but I think you'll lose a lot of people there

I followed everything until this paragraph:

> Your task is to, given a number 𝑛 and a base, 𝑏, find the fewest number of panels to represent 1 - n in base b. By that, I mean taking a single continuous slice of the set of panels, we can represent a number, and we can do that for 1 through n. Print a solution that you found next to the output such that it can be verified.

ok

let me explain that with an example then

Sure! But in the Sandbox as part of the official post, and I can tell you if what you write up there makes sense to me

Because even if you explain to me, I'm afraid you'll lose people there

Also, did you see my challenge also inspired from the no-repeating-digit-pairs?

You can credit the guy Dmitry Kamenetsky, because he has an account on code golf

He commented on the question I wrote based on that puzzle he posted

8:40 AM

@AviFS oh ok

Hello

sup

8:54 AM

Wooo finally back at 14k

Nice

Btw what am I doing wrong with this?

As soon as I try to join the sublists, it breaks.

Or is this a bug to do with tuples?

CMC: Given an integer raise it to the power of 2 and cube root it output it.

The negate's redundant.

` {(*∘÷∘3)(-⍵)*2}`

@Ausername oops

9:02 AM

@PyGamer0 Python 2: lambda x:x**(2/3)

I'm sure the golflangs can do that in 3 bytes

wait did I beat APL with Python

@StackMeter Wow, I was just going to say...

Lemme port

@AviFS 51AC8 (my lang) (3 bytes): ī±∛

*∘(2÷3)

or *∘÷∘1.5

9:02 AM

nope

fine...

whatever :p

I initially thought "raise it to the power of 2" means 2^x and was wondering why negate is redundant

51AC8 (my lang) (3 bytes): ī²∛

ignore the other one

BQN, 4: 3√×˜

@Bubbler ⎕*2÷3

9:19 AM

@Ausername does it break when using the j flag but not ⁋?

@PyGamer0 vyxal, 4 bytes, Try it Online!

ok someone break the record of 3 bytes

I think we need a language where 2/3 is a 1B built-in constant to get 2 bytes

or probably 1.5

random lang 2 bytes e⅔

implicit input

then raise it to the power of

2/3

Yeah, something like that

and there goes Python

9:27 AM

0

A: "Hello, World!"

All of these answers are pretty boring (one is a built-in and the string compression algorithm isn't even my own); consider upvoting some cool answers in tarpits / esolangs rather than this one. I'm just putting it here to have yuno officially be a language on CGCC, and because Hello, World! is ...

@lyxal Neither, it breaks when I try to join the indivual sublists.

@Ausername Try it Online!

Oh that's nice, thanks!

@Hello,World! upvotes anyway

@PyGamer0 yuno, 5 bytes: *┝2÷3, Try it online!

9:41 AM

Bye, world!

@hyper-neutrino Why do exponential literals in Yuno use ᴊ and not e.g. ᴇ?

Wait didn't you give up on Yuno?

@Adám similarity to jelly. but i suppose that would make more sense - i think i had something in mind for that symbol already though, but i'll think about it

@Ausername kind of. Hyper gave up on the stack based version

Unrelated String

maybe you're thinking of ash

9:42 AM

0

Q: ASCII-Plot the equation

You are given an equation, in the following format \$x = (c_0 * y^0) + (c_1 * y^1) + (c_2 * y^2) + ... + (c_n * y^n)\$ where \$c_n\$ stands for the coefficient of the \$n^{th}\$ power of \$y\$ You have to plot the equation on a \$10 * 10\$ ASCII matrix. The value must be floored to an integer bef...

code-golf ascii-art

@Ausername this is not the same thing, i archived the old project entirely so it more or less only exists for historical reasons

@Ausername Hyper then went on to make a tacit golfing language originally called mtrix or so

Why? Stack based is easy to implement and just as competitive as other methods...

so? I'm not used to it

tacit is easier for me to code in

I thought it was more that tacit was more fun than stack

9:43 AM

and honestly it's not that hard to implement at least via jelly tacit

@lyxal golfing in vyxal for LOTM made me realize i'm not good at stack-based golfing

@hyper-neutrino it can take a while to get used to stacks

Heck, I still struggle with stack mechanisms

like, the principle behind stack-based is far simpler than tacit in theory

golfing in Vyxal taught me that kolgormorov-complexity is free rep

just takes practice like anything else ¯\_(ツ)_/¯

@hyper-neutrino as well as the ability to somewhat visualise/manage value positions on a stack

9:45 AM

that too

Which, even though I've made 3 stack esolangs, I still don't do well

That's gotta be irony somehow

caird coinheringaahing

@PyGamer0 Jelly can get 3 bytes (ascii only) if it was cube then sqrt root :/

@StackMeter ascii-art is better.

*1. ?

caird coinheringaahing

Yep

9:47 AM

There's a reason I have ascii-art bronze

Like 90% of my Vyxals ('Vyxli'?) are those.

@Ausername true

@Ausername it would be Vyxales

Lyxal?

What is the plural of Vyxal;

Uhhh

@Ausername definitely Vyxales

if we're being fancy

why not Vyxles

9:49 AM

Vyxles is good

I like that

Vyxels

That's better

idea for a new golfing language by @lyxal Pyxal

2

mainly for ascii art

Laughs in experimental branch

0

A: "Hello, World!"

51AC8, 15 bytes ōHello, World!` Print Hello, World! or (15 bytes also) pHello, World!` push Hello, World! to the stack (implicit output)

caird coinheringaahing

@AviFS About the pop-con: I'd vote to close it as lacking an objective winning criteria, but only after someone else has. I'm not sure enough to close it first, especially after getting 150 rep (so far) from it, as it's borderline "Do this but creatively" which isn't objective. However, VTCs often cause "pile-ons" where people only VTC when other people have already done so

Unfortunately, as fun as "Do this but creatively" can be, they aren't on-topic here

'twas a long time ago...

9:57 AM

@PyGamer0 yuno, 3 bytes: *2ғ (Try it online!)

how is 2d matrices implemented in Vyxal @lyxal

numpy?

or normal python

« first day (3796 days earlier) ← previous day next day → last day (1030 days later) »

Transcript for

Jun '2121

The Nineteenth Byte

The Nineteenth Byte: General discussion for codegolf.stackexc...

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile