The Nineteenth Byte

I was gonna say since we're both going to be in the northern US we'd be pretty close to each other but then I remembered the US is very wide and I'm actually closer to Seattle now than I would be in the Northeast :p

caird coinheringaahing

Yeah, I also always forget just how huge the States is :P

@RadvylfPrograms That said, it's only roughly half the current distance between us :P

Radvylf Programs

12:21 AM

That's true yeah :p

caird coinheringaahing

Roughly, texas to the uk is 5000miles, and seattle to boston is 2500 miles :P

Radvylf Programs

Seattle's actually one of the few big cities I've been to

I don't really know how it compares to other cities, but it seemed nice

caird coinheringaahing

Oh cool :P How did you find it?

Radvylf Programs

My family ended up staying there for a few days before going on a cruise to Alaska, it was a while ago so I don't remember much

emanresu A

Next year I'll be at both uni and high school, which'll be weird and probably result in annoying scheduling conflicts

Radvylf Programs

12:27 AM

Some sort of dual credit or actual college classes?

(Idk how that works in NZ :p)

caird coinheringaahing

@RadvylfPrograms Fair enough. I have family friends there so have spent a decent amount of time there :P

emanresu A

@RadvylfPrograms Actual uni classes, from what I can tell

Radvylf Programs

Oh, gl with that!

Ginger

1:15 AM

@RadvylfPrograms now, I have something for you:

a screenshot of the Klein main editor! (kinda)

jezza_99

2:27 AM

@DLosc is there a logical xor in pip? I couldn

Couldn't seem to find one

emanresu A

Subtraction should work similarly if you have bools

Radvylf Programs

Same w/ !=, but also bool-only

Seggan

2:46 AM

streamable.com/j9hprn

im am ~~making a game~~ doing important research on electrodynamics

jezza_99

Ah sweet as, cheers guys

Seggan

now time to see what itll do with protons :P

Radvylf Programs

3 hours ago, by Radvylf Programs

Any feedback on KoTH: Don't Kill the Curve!?

@Seggan Seems cool

@Ginger Nice!

I tried making a text editor once

emanresu A

Same here

Radvylf Programs

Got about half an hour into trying to get CodeMirror Next set up and then decided against it

I'm perfectly content with T2xt lol

emanresu A

2:54 AM

I made a crappy vscode clone (in electron + codemirror) with no knowledge of fs APIs

Radvylf Programs

I've actually managed to create a weird hybrid thingy that uses Text's compiled HTML and extension manifest and stuff, but its uncompiled JS. It's really cool, I need to switch over to that as the extension source so I can start hacking it

I also need to add a dark theme

Seggan

@RadvylfPrograms lol same

Radvylf Programs

I think there's kind of a universal dissatisfaction with code editors

Once we've got a taste of customizability there's never enough :p

That's why I use Text on my chromebook

As soon as I use VSCode and can get it set up how I like it, I know I won't be able to go back :p

Have y'all ever thought about the fact that we don't actually really need to write code in monospace anymore

I wrote code in Atkinson Hyperlegible for a day, and aside from the fact that all the characters are narrower, it wasn't really that bad

Just a shower thought for y'all

Unrelated String

neat

Seggan

@RadvylfPrograms did we ever?

Unrelated String

3:03 AM

some formatting styles depend on being able to align things with more than just leading whitespace

Radvylf Programs

Yeah, there's some slight issues, but think of what we'd gain. It'd uh...it'd um...actually yeah idk

I would say it looks better except I really like monospace :p

Unrelated String

:p

Seggan

so for some reason my electrons are as afraid of protons as they are of electrons smh

Radvylf Programs

Maybe you've got a squared term

Seggan

nope i forgot an abs

now they can orbit protons yay

Radvylf Programs

3:10 AM

oh cool

I wanna see

Seggan

k will post rq

streamable.com/g705vd

electron-proton interaction ^

its quite interesting with 2 protons but i gtg bed

emanresu A

@RadvylfPrograms Do you know if there's a way to, in JS, return different stuff based on whether a function is called as f() or f()()?

Radvylf Programs

Ooh umm...the return value of f() would have to be a function, possibly Proxy'd. So you can't make it a primitive, but you can simulate any other object with it.

I don't think there's any way to know ahead of time if the output will be called again

In fact I know there isn't because of halting problem

emanresu A

Okay, thanks (trying to make JS into scala, sorta)

Radvylf Programs

Worth noting even if you can't make it a primitive, you can make it behave pretty close to one with valueOf and stuff

emanresu A

3:18 AM

I guess I can just special-case stuff like length and integers... when's valueOf called again?

Radvylf Programs

For stuff like +

I don't know exactly when

emanresu A

Oh well, special-casing I guess

Radvylf Programs

@emanresuA You could use a Proxy to make every .xyz or [xyz] behave exactly the same

Including length or an array's indices

Just map every proxy target to call the same target on the return value of f(), except call

emanresu A

I don't entirely understand what you're saying, can you show me some pseudocode?

Radvylf Programs

(also unrelated fun fact: .length on a function gives you its arity in JS, which is one of the most random overloads I've ever seen)

emanresu A

3:21 AM

ikr

Radvylf Programs

emanresu A

Best variable names. thanks :)

Radvylf Programs

Yeah I wrote "call_once" and "call_twice" and was like...wait a second :p

I'm actually kinda hyped about this KotH it's been a while since I've done one

And even longer since inspiration for a challenge has just hit me from out of nowhere

(and longer still since I've had the motivation to finish writing it out)

Radvylf Programs

4:12 AM

I'm thinking I'm gonna make one of those way overengineered controllers

Where it'll have a cool UI that lets you see really specific info about how bots are performing and stuff

If I hadn't had to temporarily murder the RTO server 'cause it was making weird noises, I'd attempt to do a multi-language KotH

At some point I need to boot it up again but I'm honestly kinda scared to :p

Anyway, good night o/

Promise me that if I start making weird noises you'll SSH before temporarily murdering me

UndoneStudios

11:09 AM

howdy 7 hours after

UndoneStudios

11:20 AM

how does this sound for a better primality tester: def isprime(n):return (n>1 and n<4) or(not str(n**.5).endswith('.0')and n%6 in[5,1]and 0 not in[n%11,n%5,n%7])

mousetail

it counts many non-primes as prime

UndoneStudios

like

mousetail

no

UndoneStudios

true

I need to fix cubes and squares

mousetail

ato.pxeger.com/…

2197

UndoneStudios

11:27 AM

any ideas for detecting perfect cubes?

mousetail

If you fix cubes I'll do 5th powers

And if you fix those I'll take 7th powers

this method can never work, unless you fix every prime but if you knew every prime you wouldn't need this method at all

UndoneStudios

{899, 901, 391, 779, 527, 403, 533, 793, 667, 923, 799, 551, 299, 559, 943, 689, 817, 437, 949, 697, 703, 323, 713, 589, 851, 731, 221, 989, 481, 611, 871, 493, 629, 247, 377, 893, 767}

mistakes

how do I fix those

mousetail

Any power greater than 3 of any prime greater than 13 will be a false positive

like I just said your method fundamentally can't work

UndoneStudios

but these mistakes aren't squares or cubes or anything

mousetail

Their factors are bigger than 11

UndoneStudios

11:31 AM

hold on a sec

these numbers look familiar

mousetail

Anything that's not a power with every factor bigger than 11 will also be false positive

UndoneStudios

@mousetail of course it won't work with you around

it's just a start

mousetail

No you don't understand

UndoneStudios

and I'm fixing it as I go on

mousetail

checking a finite number of primes can never work

UndoneStudios

11:33 AM

am I witnessing the same thing that made you quit the conway thing on rgb to hex?

if so I resist it

mousetail

You still haven't finished that btw

Please do

Also prove P = NP while you are at it

And create a algorithm that can do 90% lossless compression of random data

UndoneStudios

what

mousetail

Prove P = NP and create a algorithm that can losslessly compress random data

UndoneStudios

@mousetail I'm not working on that rn, I've got more important stuff, I'll probably do it in january

@mousetail why?

@mousetail of course P != NP

mousetail

Prove it

UndoneStudios

11:36 AM

it depends on what kind of question you want to ask

mousetail

Not to me

UndoneStudios

to science?

mousetail

Send a letter to the millenium prize comitee

You can win a million dollars

UndoneStudios

I don't need money

mousetail

Then give it to me

UndoneStudios

11:37 AM

why should I?

mousetail

Because I inspired you to solve it

UndoneStudios

@mousetail I probably could do the second though

@mousetail Since when did I start?

mousetail

Then do it

UndoneStudios

No way

I'm too young

mousetail

Please take a moment to undersand why problems are hard before trying to solve them

You'll understand why your solutions don't work

UndoneStudios

11:39 AM

yeah I'm just trying

I don't care if it becomes pseudomathematics

i'll just try

that's a little like saying public-key cryptography doesn't exist because finding a mathematical function will be too hard

besides I just realised what these numbers are

mousetail

Exactly what I just told you they where?

UndoneStudios

they're the exact same numbers that are used in RSA!

seriously screw you adleman

why did you do this to me

mousetail

Btw state of the art primaly testing algorithms are en.wikipedia.org/wiki/Miller%E2%80%93Rabin_primality_test and en.wikipedia.org/wiki/Solovay%E2%80%93Strassen_primality_test These are probabalistic so sometimes give false positives but at least are a lot better than yours

UndoneStudios

@mousetail this is my first attempt

seriously you have a talent for undermining triers

mousetail

TY

UndoneStudios

11:43 AM

I meant it negatively

mousetail

I've been convinced I found a solution to P=NP too many times to count

ok gtg

@UndoneStudios TY

@mousetail wp

@mousetail seriously

i wonder how you get things done

mousetail

11:58 AM

@UndoneStudios /s It's not my goal to discourage you to work in these famous very hard problems, in fact I'd like to encourage you. All I was saying is no variant of your algorithm could work. But very different ones may. I'd encourage you to continue searching for answers but just be aware these are hard problems and many very smart people have been trying to solve them for decades. Maybe you will be lucky and find a new solution.

Maybe not. In ether case you might learn something new, and that should be your goal.

UndoneStudios

@mousetail Why not? It's a variant of the 6k +-1 optimization

lyxal

12:30 PM

→ 49 goofy-ahh messages moved to Off-Topic TNB

UndoneStudios

12:56 PM

hi

@mousetail gimme some random data first

lyxal

I think the point was more that a "90% lossless compression of random data" is the same difficulty as solving "P = NP"

UndoneStudios

@lyxal gimme some random data first

then I'll see

mousetail

No, losslessly compressing random data is actually proven impossible

UndoneStudios

just gimme some random data

mousetail

217
52
12
107
49
14
211
193
38
6
115
247
110
100
124
94
42
76
139
249
214
92
142
15
34
168
33
103
41
10
156
235
179
18
64
158
93
201
119
245
187
126
149
50
32
112
227
145
150
253
215
11
212
71
157
255
8
167
184
188
185
216
135
174
134
239
16
159
113
241
236
28
136
240
125
9
206
238
95
44
225
197
118
106
73
189
43
66
62
133
166
243
198
85
82
121
250
170
79
224
99
98
207
30
220
140
122
117
138
77
89
203
210
81
237
53
102
191
199
59
177
20
83
26
208
35
90
226
232
27
194
67
63
91
120
37
205
104
24
234

UndoneStudios

12:59 PM

thank you

mousetail

These are byte values

UndoneStudios

now I'll use this for another purpose

creating obfuscated code

just like the fae

mousetail

I got it from here: random.org/sequences/…

Use it for whatever you like

lyxal

@mousetail oh wow, I didn't know that. Where's something I can read about it?

mousetail

I can't find a article right now but all compression works by making some patterns shorter at the cost of making others longer. So if the data is truly random and has no patterns any compression algorithm will, on average, make it slighly longer. This is average though, of course you can shorten some random data if it happens to contain some pattern

UndoneStudios

1:03 PM

@mousetail that's what you call shortening? I use indexing in my shorteners

this seems to use repitition only

lyxal

@mousetail dang, that seems obvious/intuitive when reading it, yet for some reason, that wasn't obvious when thinking about it

thanks lol

UndoneStudios

1:29 PM

I've finally written an obfuscated Hello World program

#5s[gAd\<xQKL?<&p@,-=PI#_>7%e|B}XRK9iMC~>Ybd#"c3[q"
#/<^UgZwx1c+9[1nJA@'BRu_fm?SvR"%{`VKwSShuB{`p% ~M#9
#4;#Ies2^]/V,sH%N~L7kSOFHNw A;],L2ZTfd*@ro_(18Jwk%)
#cL#lyn8uK3_A8fwKo'T15$>?d:gt"3<Gl/047?FhqThS^''\<*
#xF2Dd~_(*0DB-<av6xbZbF%ID;ov^^<5fXhg,6$q#{1nzK?h&"
##^wG[cHV`HLv36NlfC:^h,vo"6PqwwT3^9b^_w]w4gt~[HzTy&
#|)X.@KgpUQ(?[U{ItH%].{.JZkM|Ay^SV9d$~c*#85i@8Nz"[j
#z']pU"DGF60U_k0|j;CX:vCPKVpE48o#w:E!HD[j.h]hLxZktZ
#`AlCOW/!U2-cHkqEvms,J.>JR#ssA+sB3[(4yFzVq+Ty)CQQd`
#V"="{.Q5}KfzU=e'ohWh;Z^~F#e5eGso@jBCbhs$Ib21~/,[l]

mousetail

1:48 PM

All the lines are just commented out

that's not really obfuscation

UndoneStudios

What's interesting is that the obfuscation is important

if you remove it, it fails

mousetail

You are just indexing the file

but it's super obvious how it's done

UndoneStudios

yup

and it's my first attempt

I'll see if I can improve it later

like, say, compressing t

mousetail

If you did that except no comments just lines that happened to do nothing I would be impressed

UndoneStudios

I have an idea

what if the comments had the characters necessary to index some other characters that will print Hello World?

mousetail

1:59 PM

Still if you have unobufscated code at the end I think it's kinda boring

UndoneStudios

yeah

but python is clean, so not my fault

mousetail

2:23 PM

It is not

There's plenty of obfucated python

mousetail

2:46 PM

@UndoneStudios ato.pxeger.com/…

p=7-7 or("abnkir" \
and float and  (
"alph"and  chr. __self__. list ("l")and open)
) ;e= abw=IOError and repr.__le__ and eval("7"and "2" and
repr( "eval" )[1:   # 1 + 1 + 1 + 1 + 1 + 1 +
-1]) or 1 + 1 + 1 + 1; n = e( (u:="_"*2) +
"file" + u); d = type(p(n)).read(p(n));
eval(d[::7].replace(chr(10),d[:0]))

Oops the last eval can just be e

Seggan

3:03 PM

one hour left on my bounty!

@RadvylfPrograms heres a version with a more "normal" orbit

UndoneStudios

@mousetail yeah you know that's just whitespace and random stuff that don't do anything

and I can see it's reading the file

very uncreative

Sandbox Posts

3:20 PM

0

A: Sandbox for Proposed Challenges

Convert a string to RGB using HTML rules code-golf color Background This challenge is inspired by this Stack Overflow question. Given any string, HTML can convert it to a colour, using the below rules (from the answer to the linked question): Replace all nonvalid hexadecimal characters with 0’...

Seggan

thats a dupe for sure

cant find it tho

The Thonnu

3:43 PM

@Seggan I found it myself: codegolf.stackexchange.com/q/250535/114446

I've deleted the sanbox post

mousetail

4:26 PM

At least there are very few comments and strings, mostly it's executable code

that happens to do nothing

Radvylf Programs

TIL about Ctrl+\

Seems to be a sort of...super Ctrl+C

(sends SIGQUIT which is less likely to be blocked/used for something else)

Seggan

4:46 PM

anyone know of a font that covers a good range of unicode?

i need superscripts, greek letters, and combining diacritical marks

Adám

DejaVu Sans

Noto

Radvylf Programs

5:05 PM

I think most fonts will have greek and most diacritics covered

Or at least most I've used

Superscripts is gonna be harder to find so yeah, probably DVSM or Noto

Obligatory shoutout to google-webfonts-helper which will show which charsetsa font has that you have to opt into:

(That's for Roboto Mono, for example)

Seggan

5:22 PM

@Adám tried

actually wait

noto sans has all i need

ah wait i tried noto sans math thats why

ty adam

DialFrost

5:44 PM

Mornin

New Posts

9:24 PM

0

Q: CGAC2022 Day 10: Reduce, Reuse, Regift

Traditionally presents are kept secret in boxes wrapped in paper. Since the ice caps are melting Santa Claus has begun to investigate some ways they might make the whole gift wrapping operation a little more eco-friendly. The first thing they've decided to do is to put all presents in perfectly c...

Wheat Wizard

@NewPosts Posted early, undelete at UTC midnight.

Aiden Chow

10:52 PM

@WheatWizard wasnt day 10 already posted?

10

Q: CGAC2022 Day 10: Help Santa sort presents!

Part of Code Golf Advent Calendar 2022 event. See the linked meta post for details. Santa likes to sort his presents in a special way. He keeps "uninterleaving" the pile of presents into smaller sub-piles until each sub-pile is either full of toys or full of coal. Your job is to, given a pile of...

code-golf string binary sorting

Sorry for the reschedule, but can you post this on Dec 11? — Bubbler Dec 1 at 0:36

^ i think it should be day 11

Transcript for