The Nineteenth Byte

Seggan

00:07

@lyxal as a matter of fact I be getting my associates degree in a semester :P

I don't have a uni selected yet, but got a few candidates

Unrelated String

Whoa

@Seggan Where did you get in?

Seggan

A local community college

Unrelated String

Ah, so no applications yet

Seggan

Nope

Unrelated String

Well, enjoy that associate's degree!

Sounds cool to get at least something that soon

Seggan

00:11

Will start applying within a week or two

Unrelated String

Good luck

Seggan

@UnrelatedString I technically am getting it this semester but I apparently can't graduate high school and get a degree at the same time lol

Unrelated String

lmao

Seggan

So I have to wait till the end of next semester

Unrelated String

Are you even going to be attending the community college for the next semester, or are you just going to technically be a student while you wait for the next graduation ceremony

Seggan

00:13

Yeah I'm just gonna be technically a student

Unrelated String

Huge

l4m2

01:00

What languages use 4-bit charset?

Unrelated String

Nibbles, HBL

lyxal

01:59

I noticed an interesting problem with HBCS compared to SBCS the other day, and I don't think it's something that can be solved. Say you have a SBCS language (call it X) and a HBCS language (call it Y) solving the same problem. X and Y might solve it in the same amount of bytes using the same algorithm. But say that the algorithm can be shortened by removing a single operator/step. And that implementing this golf is equivalent to removing a single character

Removing the character from X saves an entire byte while removing the character from Y only saves a half byte

HBCS languages are really only good for the initial golf it seems

Ginger

lyxal's enigma

lyxal

Because once you get into harder to squeeze out golfs, the value of shaved characters is less in Y than it is in X

emanresu A

@lyxal What amounts to one operation in a SBCS lang could very well amount to two in a HBCS lang, considering their minimalism

lyxal

@emanresuA the example I saw the other day was length

In nibbles and vyxal it's both one token

It was able to be removed from both, with a full byte save for vyxal but only a half byte save for nibbles

emanresu A

Fair :p

Also my school wifi will occasionally randomly fail to load specific sites for some reason and I can't figure out why or how. It's currently refusing to access chat

lyxal

02:10

And then when you consider the fact your operation set isn't going to be as usefully large as sbcs (without getting into digraphs and trigraphs), meaning more complex operations will most likely end up equating to one sbcs character, I don't see the benefit of HBCS outside of novelty

emanresu A

¯\_(ツ)_/¯ Plenty of golflang design ideas don't pay off. Some do

And nibbles is actually fairly competitive - the author claims it beats Jelly 75% of the time, and while CGCC doesn't quite match that it's still decent

Seggan

Jelly IIRC is the golfiest language around, so id doubt that claim :P

mayyybe vyxal 3 + vyncode competes

emanresu A

Vyxal 2 (without vyncode) is in general slightly golfier than Jelly from experience

(also nekomata/thunno2 are doing surprisingly well)

Unrelated String

Jelly's remarkably competitive for its age, but it's old, and it was far enough ahead of the pack for most of the time it was being actively developed that it doesn't have a lot of the modern seasoned insights the newer gen golflangs take for granted

@emanresuA I'm honestly not surprised Nekomata's doing great--nondet is very powerful, Brachylog's just shackled in golfiness by some idiosyncratic/aesthetic factors (being made to actually feel uniquely like Prolog-the-golflang instead of a golflang with some Prolog inspiration)

Surely I'll get back to Perhaps one day 🤔

lyxal

@emanresuA although that could be due to lack of answers for both languages

Rydwolf Programs

02:24

@UnrelatedString And Risky, which kinda sucks most of the time but can highball occasionally

Unrelated String

Oh yeah lmao

lyxal

@emanresuA And a range of answers that isn't as diverse (e.g languages only being used where it's going to be shorter)

@Seggan hasn't been on top of the elo rating for at least a year now

Rydwolf Programs

03:45

I think my cat's a genius

The door to the room with her sandbox in it got shut, unbeknownst to me. But of all the objects she chose to pee on, she chose the only waterproof thing in the room

(my motorcycle jacket)

lyxal

04:26

mfw my headphones default to immersion (motion) rather than immersion (still) when I turn on immersion mode using the action button

can't believe I had to use the touch strip to select still

lyxal

04:59

> be me
> looking into how to use express.js and react in the same app
> find tutorial
> seems simple enough :D
> section titled "What the heck!! CORS ?"
> D:

as if i haven't suffered enough cors this year

New Posts

05:46

The meta FAQ bloat

10

emanresu A

@NewPosts Ayyyy I triggered the CAPTCHA while posting this

I'll pin this for now

Dannyu NDos

I just noticed that the usual binary representation of integers is justifiable as the inverse limit of $\mathbb{Z}_2 \leftarrow \mathbb{Z}_4 \leftarrow \mathbb{Z}_8 \leftarrow \mathbb{Z}_{16} \leftarrow \cdots$.

emanresu A

@NewPosts Also, it's interesting just how many things are covered by standard I/O - I guess it's not too surprising considering it's edgecase central

Dannyu NDos

*Cough* I'm installing Firefox on Windows just for MathJax.

Dannyu NDos

06:09

And on the second look, that doesn't identify just integers, but the entire 2-adic integers.

Sandbox Posts

06:37

0

A: Sandbox for Proposed Challenges

Factoriadic Fraction Addition Objective Given two rational numbers represented in fractional factoriadic as defined below, add them, and output the result in balanced factoriadic. Fractional factoriadic Fractional factoriadic is a numeral system that represents every rational number using \$1/2\$...

Dannyu NDos

Any dupes?

emanresu A

You mean Q[0, 1] right

Dannyu NDos

Eh?

No.

emanresu A

I don't understand how this is supposed to represent a) negative numbers b) numbers ≥ 1

Dannyu NDos

Added some more examples for you.

emanresu A

06:42

Ah, I missed "The digit at 1/2's place may be any integer."

CMC: Golf this algorithm from my computability theory class. Registers are nonnegative integers, and we have an increment operation, a decrement-unless-zero operation, and while loops / if(else?) statements that can check if a register's nonzero or zero. Goal is to create a macro that executes code block Q if r2 > r3, and code block Q' otherwise. I have nine lines:

while r3 ≠ 0:
 r2 = r2 - 1
 r3 = r3 - 1
 if r2 = 0:
  do Q'
  while r3 ≠ 0:
   r3 = r3 - 1
if r2 ≠ 0:
 do Q

(no, this isn't a homework question, I've done said homework (unless there's a "this is code golf, least lines wins" hidden in the fine print somewhere))

att

07:05

@emanresuA there's a hanging "tips questions don't need to have objective winning criteria. Neither of these" (end of paragraph)

emanresu A

07:18

@att (@Adám) fixed

Sometimes I start typing things and forget to finish them, in this case I forgot what I was typing in the first place

lyxal

@emanresuA gonna go and sneak this into the homework

l4m2

07:33

@UnrelatedString someone should try giving javascript 52 builtins :)

lyxal

while r3 != 0:
  r2--
  r3--
if r2 == 0:
  Q'
if r2 != 0:
  Q'

For 7 lines, no?

Subtract r3 from r2 and if there's still value in r2, it's bigger

emanresu A

Smart

(we have if-else statements)

lyxal

Doesn't change the line count, only changes the number of comparisons :p

and produces slightly shorter byte code if you were to compile it to some set of byte code

emanresu A

Yeah. These are theoretical register machines in a theoretical computer science / mathematics course, so I doubt it somehow

lyxal

@emanresuA also, it only works because of the non negative restriction

emanresu A

07:37

So does mine

lyxal

Ah didn't notice

@emanresuA theory of computing one of the best courses I ever did

emanresu A

Yeah it's fun

lyxal

Only behind compiler design ofc :p

emanresu A

I might look into some non-Java comsci courses at some point, not entirely because Michael Homer teaches some of them

(mostly postdoc stuff though)

Dannyu NDos

I just noticed that a std::set is just a free commutative idempotent monoid.

Sandbox Posts

07:52

0

A: Sandbox for Proposed Challenges

Reduce a string up to idempotency Objective Given a string consisting of printable ASCII characters (!0x21 ― ~0x7E), treat it as an element in the free idempotent monoid (as defined below) over the printable ASCII characters, and output the same element represented in the fewest characters as pos...

lyxal

@emanresuA have you looked at regular v context free yet?

Dannyu NDos

Suggested testcases much appreciated.

emanresu A

Don't think so unless we've done it under another name, we're only two weeks in

lyxal

Sounds like you haven't

Surprised you're starting with register machines then

The theory of computation course I did dove straight into finite state machines and regular expressions and the pumping lemma

emanresu A

¯\_(ツ)_/¯ this is specifically a third-year mathematics course

lyxal

07:55

Yeah no so was mine

emanresu A

Huh okay, I'll look into it

It's a bit weird because due to budget cuts it's a hybrid computability theory / mathematical logic course

lyxal

@emanresuA good luck with that btw

Basically all compsci uses java

emanresu A

yep :\

lyxal

Come to the software engineering side where we have python, react, and sometimes even whatever language you want

Mostly

emanresu A

Also network engineering has C (although from what I've heard it's not much actual programming)

lyxal

08:01

I don't think that's a degree type

At least not at the uni I go to

Networking is considered a SENG (software engineering) course

emanresu A

Yeah, it's not a separate degree type (there's comsci or software engineering afaik) but there are NWEN courses for network engineering

lyxal

I'll be doing one this semester :p

The course has an AI generated banner on canvas and everything

Rydwolf Programs

08:21

Finally got around to watching oceans 11, shows its age a bit but came for a good heist and got a good heist

New Posts

08:43

0

Q: Simulate Text Cursor

You are given a string [A-Za-z|]+ that contains exactly one occurrence of | denoting the current position of the cursor and a sequence of moves [><#]+ where: >: moves the cursor one position to the right <: moves the cursor one position to the left #: moves the cursor one position to the left, d...

att

09:38

@lyxal is it worth the risk of getting python or react

Themoonisacheese

10:06

@emanresuA in France we have Computer science, and Networking and Telecoms as degrees

New Posts

10:17

0

Q: Substring of Fizzbuzz

Decide if a string is substring of Fizzbuzz without separation 12Fizz4BuzzFizz78FizzBuzz11Fizz1314FizzBuzz.... Notice that it doesn't necessary cover whole tokens, e.g. zz78Fi is a substring. Input will contain only letters and numbers, aka. match RegEx /^[a-zA-Z0-9]+$/. True cases zz78Fi FizzBuz...

code-golf fizzbuzz

Unrelated String

11:39

@lyxal yep, same here

New Posts

0

Q: Golfscript whole-stack rotation

There isn't a builtin in golfscript which lets you rotate a bigger portion of the stack than the three top-most elements. I came up with this ]-1%(1$1$1$1$<@@)>+@@=\~]-1%~ But this is GolfScript, it can surely be done with fewer characters

golfscript

mousetail

13:23

CMC Given a range of angles (in counterclockwise order) and an angle, calculate if the angle is in the range. Must properly account for wrapping.

Or actually CMC Given a) a starting angle b) a span angle, either positive or negative, c) an angle. Determine if C is in the range given by a and b. Should probably handle wrapping, eg. 2PI is in the range -1/4PI and span 1/3PI

Adám

Isn't that just an "in-range" check with a modulus?

noodle person

^

mousetail

It's not that easy when you are intersecting 0

Or maybe it is but I'm not seeing it

The arc has a certain direction, if you do a modulo it might invert the range if you intersect 0 naively

Adám

13:39

Not if your modulo built-in is designed properly.

New Posts

14:02

0

Q: Double dequer sort

In this challenge Bubbler describes sorting a list by inserting its elements left to right into a double ended queue or "deque". Take a number from the front of the array, and push it into one of the two ends of the deque. Repeat this until the array is empty. That challenge asks if it is possi...

code-golf decision-problem sorting data-structures

mousetail

I came up with something like this:

fn point_in_angle_range(point: Vec2, start_angle: f32, range: f32) -> bool {
    let angle = point.to_angle();
    if range > 0.0 {
        (angle - start_angle).rem_euclid(TAU) < range
    } else {
        (angle - start_angle - range).rem_euclid(TAU) < -range
    }
}

@Adám is this what you where thinking of?

RubenVerg

15:43

@mousetail 1=⍸∘(360∘|) (in degrees just because it's one less set of parens)

@mousetail idk what this means

i think sol includes start endpoint but not end enpoint, not sure how to fix that

mousetail

Basically the angle can be either clockwise or anticlockwise dependingon the sign of b

RubenVerg

is b a difference from a or another angle?

mousetail

Yea but the sign matters. So for the same other angle you could have two different spans

RubenVerg

okay so you want to know if the angle is in the range (min(a, a+b), max(a, a+b))?

mousetail

Depends again on if you are going clockwise or anticlockwise

You can't really define "min" on angles

RubenVerg

15:48

i'm really confused

i'm either not understanding what you want or you're making it much more complex than it is

mousetail

Suppose you have one of those shitty coin operated binoculars. They start looking at a). They can turn up to b) degrees counterclockwise if b is positive or clickwise if it's negative. Can you see the mountain at c?

RubenVerg

does bqn 1=(360|·∧+`)⊸⍋⟜(360⊸|) do what you want?

mousetail

It might, if I spoke BQN lol

Adám

@RubenVerg Maybe add a try-it link?

RubenVerg

idk how to make one

anyways paste it into bqnpad and pass range on the left as a tie b (type tie by pressing backslash then space) and angle on the right, the function needs to be wrapped in parens

like this: 45‿¯30 ( 1=(360|·∧+`)⊸⍋⟜(360⊸|) ) 42

bqnpad

anyways this does "is the right argument modulo 360 between the range specified by the modulo 360 of the sort of the sum scan of the left argument?"

oh wait it's wrong

should be 1=(∧360|+`)⊸⍋⟜(360⊸|)

Adám

15:57

@RubenVerg mlochbaum.github.io/BQN/…

RubenVerg

"[...] specified by the sort of the modulo 360 of the sum scan of the left argument"

att

does this look right

RubenVerg

fixed code link

att

since the initial angle doesn't matter

RubenVerg

radian version

mousetail

16:08

Nice thanks

Adám

@RubenVerg ≠´(360|+`)⊸≤⟜(360⊸|)?

About to go golfing…

There we go: ≠´+`⊸(≤○(360|⊢))

Which will hopefully work in Dyalog 20.0 as ≠/+\⍛(≤○(360|⊢))

RubenVerg

@Adám oh, ≠/≤ is a cool pattern for "is in range"

is it in aplcart? looked for "in range" but had too many results

Adám

No, APLcart uses 1=⍸ because it is more general (TAO).

RubenVerg

i guess i could've figured it out myself if i wasn't so fixated on using Bins

@Adám yeah, but the former has the advantage of working with reversed ranges

Adám

Yes.

RubenVerg

16:19

i think it deserves inclusion

Adám

It is also scalar (and so, numeric only)

Should be ≠⌿≤ I guess.

caird coinheringaahin g

18:46

@Bbrk24 Not to brag, but I've been confused about aspects of my queerness for the past 7 years :P

Ginger

you win :p

att

~~only 7 pfft~~

caird coinheringaahin g

Wait, no, caird time drugs. The first queer identity I felt comfortable using was describing myself as gay when I was 14, and I'm currently 22 and questioning whether to switch to using bi or not :P

emanresu A