The Nineteenth Byte

PyGamer0

15:07

i got the yearling badge!

@Seggan i think its a feed so its slow

pxeger

it only updates every ½ hour or so

PyGamer0

what is necro bumping?

pxeger

have you tried googling it?

DLosc

@des54321 Looks like the @ actually makes it not work, so yeah, definitely remove that. ;) Also, it's the same length but you don't need the collectors. Alternate configuration.

@emanresuA ... Huh. I wonder if that's a bug in the program or the interpreter.

PyGamer0

@pxeger oh

Sandbox Posts

15:30

0

A: Sandbox for Proposed Challenges

Draw the Archimedean Spiral code-golf Overview I was playing with Desmos and discovered a really cool spiral: Doing some research, I discovered that this spiral was the Archimedian spiral. Every point on the spiral is exactly \$\theta\$ radians/degrees away from the origin, where \$\theta\$ is t...

Seggan

random poll from my school

🍪

Radvylf Programs

About 0.3% of the earths' mass in cookies

DLosc

In that case, I'm thinking I'll pick the water molecules. Don't want to bury myself thousands of feet deep in cookies. :P

Seggan

I think it will be all molten chocolate

DLosc

@Seggan Hmm, new site design

Seggan

15:42

lol i just noticed

Bgil Midol

yes

PyGamer0

i noticed that a couple of days ago

Radvylf Programs

RSNBATWPL is ready to be released

PyGamer0

?

Seggan

ping me when you do

@PyGamer0 thats his (practical) language

PyGamer0

15:58

oh

^^^

@Ginger you should probably add simple docs for etch...

Radvylf Programs

16:15

0

A: 1, 2, Fizz, 4, Buzz

rSNBATWPL, 57 bytes for{i=1;i<101}$print{(("Fizz"*1>i%3)+"Buzz"*1>i%5)or++$i} RSNBATWPL, or Radvylf Should Not Be Allowed To Write Programming Languages, is a "practical" programming language I've designed over the last week or so. It aims to be a sort of "super-JS/PHP/C++": a language that see...

github.com/Radvylf/rSNBATWPL

Seggan

👏

pxeger

> I to N are case insensitive; takes inspiration from FORTRAN

Radvylf Programs

Trying to read the source code already, I see

pxeger

> Enable bleeding edge optimizations, may cause some instability

eval(random string)

Radvylf Programs

There is a nonzero chance it can generate another eval(random string) with a longer length which generates another random string which creates a genetic optimizer which improves the performance of the program

DLosc

16:21

@RadvylfPrograms ... Is it okay that my first reaction is, roughly, "Ugh"? (I only say that because I think that might be the reaction you were going for.)

Radvylf Programs

Yes, the attractiveness of the syntax was very far from the top of my list :p

pxeger

my first response is a defeated sigh

Radvylf Programs

Fun oddity: You can't have parens or braces in strings

E.g., to access the (inputs) variable, you need to do var{'("inputs"')}

DLosc

@RadvylfPrograms My main 0_o moment is the right-to-left expression parsing. For some reason, it's fine when APL does it, but anything that looks C-like definitely shouldn't.

Radvylf Programs

The reason I did is so that = behaves reasonably

a=2+3 makes far more sense as a=(2+3) than (a=2)+3

Same with ~

You can use anything as an infix operator, even something that's not an identifier. E.g., 2 (x ~ y ~ y - x) 7 is 5

DLosc

16:26

Can you do a{1}=2?

if a is an array

Radvylf Programs

No, of course not

That would be totally unreasonable

DLosc

What about (a{1})=2? :P

Radvylf Programs

Although you're free to redefine , to return a function which acts like an array in every way except that a{1} returns an id which can be set

@DLosc That's parsed identically :p

DLosc

Also, whatever happened to using : for assignment? Or was that for a different project?

Radvylf Programs

I decided = was more reasonable

DLosc

16:29

:(

Radvylf Programs

uh oh I found an issue

Fixed

cast.float was returning something with a type that doesn't exist

Ginger

17:13

I DID IT

I GOT PUSH MESSAGING WORKING

:D

mathcat

> any way to negate a full list in desmos?
> I DID IT

Ginger

whoopsie

mathcat

context go

Ginger

og txetnoc

wait what do I do with the public key

._.

Radvylf Programs

Oh you keep that one. Put the private key somewhere very visible on your website /jk

Ginger

17:18

...I genuinely cannot tell what the public key is supposed to be used for

lemme try deleting it

It still works???

I'm going to leave it in case it's needed in the future

but it's public so I can post it here right?

Radvylf Programs

In theory yes

Ginger

oKAY THEN

-----BEGIN PUBLIC KEY-----
MFkwEwYHKoZIzj0CAQYIKoZIzj0DAQcDQgAELqxyRrjqbK6ZMOIbfWr7guvjW6qb
L6Wq81FQ8GPV+vvUEBDaAWNbSBpLE5YCUtPqmag6eZ350OgBMuwapaBaGg==
-----END PUBLIC KEY-----

DLosc

Seems kinda short?

Ginger

there it is ladies and gentlemen

ah yes:

> MuwapaBaG

it's a bag that you put muwapas in

well, now I just have to implement everything else

yey

Romanp

Now, the real question. What's a muwapa?

Ginger

17:28

Good question.

Radvylf Programs

Well as an adjective it would clearly mean "public"

Ginger

new katlani word!

Radvylf Programs

in TNB Conlang (katlani), 4 mins ago, by Radvylf Programs

WR: murapa (adj) Public

Ginger

oh

Radvylf Programs

How's that for a ninja :p

DLosc

17:43

@RadvylfPrograms You're on fire today. You also just ninja'd my edit to that Politics.SE question. :)

des54321

first weird quirk of rSN I've discovered: assignment operations always evaluate to the old value of the variable

so if foo is unassigned, foo = 15 evaluates to the reasonable Null

but if you then do foo = 7, that evaluates to 15 even though foo is now 7

Seggan

thats cool

des54321

okay i clearly dont fully understand this syntax, how do you apply the sort function to an array

Radvylf Programs

array sort function

Seggan

arr$sort?

Radvylf Programs

17:55

E.g., array sort cmp is the "default" sort

des54321

what

oooooh its an infix

thats cursed

Radvylf Programs

Any function can be infix in rSN

des54321

so you can also do sort{array}{cmp}

Radvylf Programs

Yeah

des54321

and you just pass it an arbitrary function to compare them

thats beautiful and stupid at the same time

Radvylf Programs

17:56

Yep. sort < and sort > are golfy, but less efficient, ways to do min and max sort

DLosc

@RadvylfPrograms If cmp is the second argument, does that mean you can't easily use the currying to make a "sort ascending" one-argument function that takes an array?

Radvylf Programs

Nope. You'd need x ~ x sort cmp

DLosc

Too bad

Seggan

next LYAL: rasin batwaffle

Radvylf Programs

I am going to add it to the LYAL voting at some point :p

Ginger

17:59

how nice that #01abda is a nice shade of blue

Radvylf Programs

labda?

Ginger

remember campfire? I'm making a version that's closer to SE Chat and calling it lambda.chat

and I'm going to use #01abda as the main color

des54321

fun fact about raisin batwaffle: it doesnt seem to have a maximum recursion depth

Radvylf Programs

It does, it's JS's max recursion depth

E.g., {} will instantly cause bad things™

des54321

huh, i just threw a recursive function into the online interpreter and it hasnt errored or anything yet

Radvylf Programs

18:04

It just freezes for me lol

des54321

oh hold on browser console is full of "out of memory" errors

Radvylf Programs

Yeah I just used a whole gigabyte of memory in a couple of seconds, with a two byte program

Seggan

whats the diff between cond and if

des54321

Ginger

._.

Seggan

18:08

i wish i had that much memory lol

des54321

im pretty sure thats only half the memory on this machine, i think i have 16 GB of ram

Radvylf Programs

@Seggan cond is more like an if/else

des54321

although im not convinced my computer knows that, because task manager was showing i was using like 80% of my ram there despite nothing else i have open using more than 200MB

Seggan

theres no else?

Radvylf Programs

No, with the way the parsing works it'd be close to impossible

Steffan

18:10

@Seggan you mean in ruby?

Radvylf Programs

No, in RSNBATWPL

Steffan

oh lol

Seggan

@RadvylfPrograms what about else if

Steffan

des54321

hold on i actually love the fact that you can use while as an infix operator, but rather than doing the readable code while condition you have to do condition while code

Radvylf Programs

18:11

@Seggan You have to nest conds, no else if equivalent

Seggan

oof

des54321

imagine using google and not duckduckgo

Steffan

lol i googled rasin batwaffle and it autocorrected to "raisin bar waffle" and is now showing be waffle recipes

@des54321 imagine using duckduckgo and not google

duckduckgo does not find things

lol

Seggan

^ ^^

Steffan

rasin batwaffle returns no results

des54321

18:12

duckduckgo aint all that bad, and at least google isnt tracking me

Steffan

but it doesn't find what you want

i tried it once and i ended up always going to google in the end

des54321

duckduck go finds rSNBATWPL

Seggan

oh wow that was fast indexing

Radvylf Programs

4

Steffan

wait that actually looks like a good lyal, rsnbatwpl

we should nominate it

Seggan

18:15

^

des54321

yeah this would actually be a good LYAL, break the cycle of semi-tarpits we've had the last few langs

Radvylf Programs

I'll nominate it

Seggan

ok ima use if{non't} forever now 🤣

Steffan

what about J

that would also make a good lang of the month imo

(I mean J, not rsnbatwpl)

Romanp

Is don'tn't still in the spec?

des54321

18:17

yes

Seggan

@Steffan agree

Steffan

imma nominate j for both

Radvylf Programs

0

A: Language nominations for the "Learn You a Lang for Great Good" chat event

RSNBATWPL A.k.a., rSN, raisin-batwaffle, or Radvylf Should Not Be Allowed To Write Programming Languages. Rsnbatwpl is a "practical" language with a lot of oddities. The parsing has a lot of quirks which make golfing interesting, and also mean it would be easier to learn with LYAL's format than w...

Steffan

oh my what on earth

github.com/Radvylf/rSNBATWPL/blob/main/rsnbatwpl.js#L124-L127

~~Radvylf should definitely not be allowed to write programming languages~~

des54321

may cause some instability

Steffan

18:22

lol

Ginger

logo!

Radvylf Programs

Whoa that's big

Ginger

It's an svg

Steffan

reminds me of haskell for some reason

Ginger

will be scaled down

Radvylf Programs

18:23

I mean in chat

Ginger

I just exported it at 500px png

Radvylf Programs

It takes up like an eighth of my screen

Steffan

about a fourth for me

Ginger

but do you like it?

Radvylf Programs

@Ginger Can you please edit an m into this, is what I'm tryna say

Ginger

18:24

outa time

sorry

Radvylf Programs

Nw

Ginger

but do you like it?

Radvylf Programs

I mean, it's a lambda on a blue background. It's good enough I guess

Ginger

nice

well, now I just need a contrasting header bar color...

Radvylf Programs

Dark blue

Ginger

18:26

hmm

Romanp

https://codegolf.meta.stackexchange.com/questions/2140/sandbox-for-proposed-challenges/24748#24748

Any feedback?

Ginger

@RadvylfPrograms what shade?

Radvylf Programs

No clue

Ginger

:/

des54321

@Romanp seems like an interesting challenge, i do wonder if theres a better solution than just "brute force every possible selection, discard the ones that spend too much money, and return the one that spends the most" though

Romanp

18:33

Huh, that's not my challenge. Sorry, let me check the link.

Mine's "make the apartment building," that other one's a lot better.

@RadvylfPrograms #3300CC is a nice deep blue

Steffan

0

A: Language nominations for the "Learn You a Lang for Great Good" chat event

J J is a high-level, general-purpose programming language that is particularly suited to the mathematical, statistical, and logical analysis of data. It is a powerful tool for developing algorithms and exploring problems that are not already well understood. (copied from the J site because I'm la...

Radvylf Programs

18:48

I did it I did it I made an ellipse with a polar equation!

Seggan

what is it

des54321

huh rSN lets you redefine builtin operators

Radvylf Programs

Yeah

You can even do stuff like id{"do"} = 2 to set the variable do, although do's a keyword so it won't be accessible

Seggan

wait lyal is still running

Radvylf Programs

Unless you do var{"do"}

Oh huh, just realized sort> is exactly as efficient as sort cmp performance-wise

Probably since it uses mergesort

Seggan

19:02

are you trying to fastest-code this lol

Radvylf Programs

No, I was just curious :p

des54321

just made a very fun to watch bitcycle dupneg bomb

0vvv
>~~~<
>~~~<
>~~~<
0^^^

Romanp

19:28

Even more cursed than the last set.

des54321

i dont even want to know how you abused a programming language into doing that

graffe

CMQ If n is an integer, what function tells me how many terms of sum 2^i for i = 0.. you need to get a number bigger than n? I.e. if n = 10 it is 1+2+4+8 so 4 is the answer

that is the smallest x so that sum_i^x 2^i >= n

Radvylf Programs

@graffe ceil(log_2), right?

graffe

@RadvylfPrograms oh thanks!

Radvylf Programs

It might be something like floor(log_2 + 1), you'll need to watch out for off-by-ones

Romanp

19:38

@graffe Working on this problem? codegolf.stackexchange.com/questions/247028/…

graffe

right if n = 1 you get 0 from ceil(log_2)

@Romanp sadly no

Romanp

Ok

graffe

@RadvylfPrograms floor(log(10,2)+1) looks good

Seggan

casually uses cmps for outsourcing challenges

seggan not admitting that he did that once too

Radvylf Programs

19:58

@Seggan sqrt(1/cos^2(theta)/h^2+sin^2(theta)/k^2)

Seggan

oh yikes

Radvylf Programs

It's surprisingly simple

emanresu A

> // THIS CODE IS MIT LICENSED TO HOBBYISTS. FOR-PROFIT CORPORATIONS MUST USE ANY 2 CLAUSES OF A 4-CLAUSE BSD LICENSE.

... what

Ginger

._.

graffe

given a power of two n, and a query integer n <= i < 2n, how can I find the smallest number of terms in n+n/2+n/4+... needed to sum to a number at least as a big as i?

is there a formula for this?

Radvylf Programs

20:08

i needs to be >= n and < 2n, btw

graffe

@RadvylfPrograms that's not for me is it?

Radvylf Programs

It is

graffe

oh I see what you mean

Radvylf Programs

But I'd guess it's something like log2(2n - i) - log2(n)

graffe

let me try that

Radvylf Programs

20:11

There are probably multiple off-by-ones in that, just a warning

graffe

is there a floor somewhere?

or a ceil?

Radvylf Programs

Yes, but I'm not going to attempt to guess where

graffe

n = 2**8
log2(2*n - 1) - log2(n)

Radvylf Programs

Likely one wrapping the whole thing

Unrelated String

@Steffan probably because of the lambda

graffe

20:12

that gives a number less than one

Radvylf Programs

Well because i is not between n and 2n

graffe

good point

Radvylf Programs

Yeah, so ceiling that should work

graffe

n = 2**8
log2(2*n - (n+10)) - log2(n)

that's legitimate right?

with i = n+10

Radvylf Programs

Yeah, as long as n > 10

graffe

20:15

2**8 > 10

the answer should be 2

in this case

Radvylf Programs

Ah, so it's ceil(...) + 1

graffe

this formula gives you -0.05748549466076014

Radvylf Programs

Wait I got it backwards

It's log2(n) - log2(2n - i)

graffe

aha!

Radvylf Programs

Yep, looks like it's working perfectly now

Math.ceil(Math.log2(256) - Math.log2(512 - 385)) + 1 // 3

graffe

20:19

what is the //3 part?

Radvylf Programs

A comment with the result

graffe

phew :) In Python that's integer division by 3 :)

thanks so much!

Radvylf Programs

Yeah, realized that right before you asked

graffe

what language is your code in?

Radvylf Programs

JS

Although it'd work in most C-style languages

(Assuming the math library was called Math in them all)

graffe

20:24

except for C I suppose

where nothing works

I should probably learn JS one day. People here seem to like it

Ginger

uuuuuuuuse pythooooooon

Radvylf Programs

@graffe I wouldn't say that necessarily :p

graffe

that is what I uuuuuuuse!

@RadvylfPrograms :)

the only think I know about JS is the website wtfjs

Ginger

link?

Radvylf Programs

There's a lot of stuff wrong with JS, but it's just convenient for a lot of reasons

Ginger

20:25

I'm actually using it right now!

graffe

github.com/denysdovhan/wtfjs

I would like a language where I could do simple animations easily. I mean it was so easy in the BASIC!

python has pygame but it is barely suppported

Radvylf Programs

JS canvases aren't overly difficult to use

graffe

you may get stuck on the wtfjs page :)

@RadvylfPrograms that's the impression I get.. hence maybe I will have to learn it?

Radvylf Programs

If you've used Python for a little while, using JS is really just writing the same code with different syntax

Ginger

^

Radvylf Programs

20:29

And the console built into basically every web browser is a really convenient way to get familiar with it

graffe

@RadvylfPrograms I guess there is no python equivalent of canvas though?

Radvylf Programs

No, since there's no Python equivalent of HTML :p

graffe

:(

github.com/adicu/…. :)

Radvylf Programs

Ooh nice

@RadvylfPrograms (Namely, I can write code in JS, and very easily move it between the browser and back-end/desktop)

Romanp

20:44

I think it finally gave up...

Every expression now returns errors like then when ran.

user

You've broken Mathematica forever, you monster

Romanp

Not yet...

des54321

requesting authorization to have Romanp and their copy of Mathematica thrown into the ocean

Romanp

That's next up after finding how to do ACE

I overwrote CompoundExpression which seems to have broken things

But hey, Print["hi"] still runs, it just lags like crazy

Radvylf Programs

Oh my...I just found an extremely dumb website

It does input validation on the back-end, for every keystroke

And it's meant for long-form typed content

user

20:58

That's, uh, very safe, I guess?

Radvylf Programs

So after every keystroke you have to wait a split second, and it sends thousands of requests as you type

user

Better than doing client-side input validation

Radvylf Programs

It's not input validation in that sense though because it's never sent anywhere

user

@RadvylfPrograms lmao, do they basically make you DoS them?

Radvylf Programs

It's like a spellcheck sort of thing

user

20:59

What

Why would anyone handle spellcheck like that?

Romanp

My error messages now create error messages when you click on them.

Unrelated String

21:17

hell yeah

emanresu A

22:13

Squares in bitcycle

It's really annoying how open collectors can output bits they've recieved while open

It breaks so much stuff

BCMC: Move a signed integer towards 0

forest

What does BCMC stand for?

emanresu A

BitCycle CMC

Look at the pinned message

user

@emanresuA Decrement it if positive, increment if negative, do nothing if zero?

emanresu A

yep

forest

22:29

Huh it doesn't show any pinned messages to me.

Oh I just had to reload the page. Chat was borked.

lyxal

23:19

My favourite boolean value is non't

And you can't change my mind on that

emanresu A

non'tn't

des54321

@lyxal personally im partial to willn't

@emanresuA ive got 12 bytes

emanresu A

Hm nice

10 based of that

des54321

alternate 10 byter:

dloscutoff.github.io/Esolangs/BitCycle/…

@emanresuA actually im pretty sure that is 11 bytes

emanresu A

Oops I can't count

des54321

23:30

the only multiplication BCMC we've had is multiplying unsigned ints, right?

emanresu A

Alternate alternate 10 byter

Jo King

surely just this works

emanresu A

@des54321 Yeah, signed woould be a pain

@JoKing Ooh nice

des54321

oh good one

ATaco

CMQ: calling with named arguments in functions, is foo(bar=x) or foo("bar"=x) preferable, considering it’s possible to do a variable assignment inside the arguments as well, which would require parentheses in the first example

des54321

23:31

@JoKing actually no, fails for input of 1

i knew there was a reason you had to actually detect the sign bit

Jo King

oh darn

des54321

@ATaco I feel slightly partial to the first one, but i think thats python prejudice

because of course in python, if youre doing variable assignment in that context you have to use := instead of =

ATaco

I prefer the first one, but I’m not sure I like overriding syntax like this.

emanresu A

The second one is overriding the syntax

ATaco

In the context of the language, assigning to a non-variable is invalid. So using string literals avoids any sort of collision with regular assignment

des54321

23:40

BCMC: given two signed ints, a specific one of which is guaranteed to be either 1 or -1, multiply them

Unrelated String

could use an entirely different syntax then; foo(bar:x) or some such

ATaco

That being said, I already use the first form for function definition

Unrelated String

if you have map/dictionary literals then perhaps use whatever those use

Radvylf Programs

@ATaco IMO named arguments should use a different syntax

ATaco

I have yet to implement them, but I will likely use the same thing as the calling

Maybe I use the walrus operator..?

emanresu A

23:43

Is this for funky3?

user

@ATaco I would prefer the first, but if it's ambiguous, I agree with Radvylf, maybe foo(bar:x)?

Unrelated String

my firefox and discord were just crashed by my windows insider build telling me it'll expire in like october

ATaco

23:59

bar:x would conflict with the Lua-style curried index operator, foo:bar(baz), but foo(bar:=x) is completely free for use

Transcript for