The Nineteenth Byte

True, that's what I always tell people. But Google just says "father" and it wasn't worth explaining to the fact checkers. Though that strategy backfired, haha

I am a person

what level of reputation is good

I say 1000+

hyper-neutrino

depends on your standards

AviFS

Although now I'm finding more reliable sources which do say "my father," so I will say that next time!

I am a person

6:29 PM

@hyper-neutrino Yes, good point.

Adám

@Iamaperson Don't worry about it.

AviFS

@Adám I'm certainly not my father either, though...

I am a person

@Adám :)

AviFS

But I am my own grandpa!

math

@Iamaperson 2000+

I am a person

6:30 PM

@math sorta agree but disagree

math

@Iamaperson You see, i'm already 1000+ and i'm challenging myself

I am a person

2000 is just too annoying to work for

Adám

@Iamaperson Here on code golf, 50rep is enough for most users.

I am a person

@Adám yes lol

math

@Adám because it takes less bytes?

Wezl

6:31 PM

1,402 is the perfect amount of reputations

5

Adám

@math No, because it is enough to chat and comment everywhere.

I am a person

2000 is too hard to work for, for me

But you got people on stack overflow hitting 1 mil+

:|

Adám

@Wezl So close.

Wezl

@Adám >:|

Adám

@Iamaperson Eventually, it becomes rather meaningless, other than something to spend on bounties.

hyper-neutrino

6:34 PM

It's a bit of an ivory tower stance for me to take, but working towards reputation rather than helping people out and enjoying your time on SE and letting reputation be a product of that experience IMO degrades both your experience and sometimes quality.

Wezl

@Adám if someone does that again I get to downvote 10 of their answers

math

@Wezl if they don't have any answer they're lucky

Wezl

it would be reversed anyway

I am a person

@Wezl So close

Didn't hit 1400 :|

Leaky Nun

CMC: Output the following mnemonic for multiplying by 9:

(in words: the tens digits are increasing 012345678 and the units digits are decreasing 987654321)

user41805

6:38 PM

leaky you're back

Adám

@LeakyNun KC?

Leaky Nun

yeah

Adám

So we have to have that 0? Hm.

Leaky Nun

yep

I am a person

So close to 1k!!!

(Reputation)

Adám

6:41 PM

@LeakyNun APL, 34: 9,'*',⍤1⊢i,'=',' '@1⍤1⍕⍪100+9×i←⍳9 Try it online!

hyper-neutrino

Python, 57 bytes: for i in range(1,10):print(9,'*',i,'=',str(9*i).zfill(2)) - Try it online!

Leaky Nun

@hyper-neutrino I have a 48-byte solution

Adám

@LeakyNun Extended Dyalog APL, 29: 9,'*',i,'=',' ','0'⍢⊃⍕⍪9×i←⍳9 Try it online!

Wezl

@LeakyNun can we output 9 * 7 = 6 3 for example because it looks better

Leaky Nun

@Wezl no, sorry

Wezl

6:44 PM

(for that I have ngn/apl, 20bytes)

Leaky Nun

url encoding fail? :P

I am a person

To anyone with over 100k rep, mad respect.

caird coinheringaahing

@AviFS It did?

Leaky Nun

@Iamaperson stackexchange.com/leagues/200/alltime/codegolf

I am a person

@LeakyNun only 5 people sadge

user41805

6:46 PM

@LeakyNun what do you think of this cmc chat.stackexchange.com/transcript/240?m=58542341#58542341 ?

Adám

@LeakyNun vanilla Dyalog APL, 29: {∊9'*'⍵'= ',1↓⍕100+9×⍵}⍤0⍳9 Try it online!

hyper-neutrino

@LeakyNun python 3?

Leaky Nun

@hyper-neutrino yeah

Wezl

@Adám ngn/apl, 25 bytes

hyper-neutrino

hm. was on my phone earlier but i'll try to get to 48 too now that i am on my laptop

@LeakyNun is this what you had? (python 3, 48 bytes)

user41805

6:48 PM

@Adám {∊9'*'⍵'= ',1↓⍕109×⍵}⍤0⍳9

Leaky Nun

@hyper-neutrino yeah

Wezl

I hate stranding and the default APL printing format, but at least they're abusable

Adám

@Wezl I would Looks wrong to me.

Wezl

whoops

Adám

@user41805 Yes, of course! Nice.

@Wezl fwiw, Roger Hui has started signing his emails with standing delenda est instead of ⎕io delenda est.

I am a person

6:52 PM

several questions indeed

Leaky Nun

@user41805 I think you're venturing into the halting problem area

hyper-neutrino

Jelly, 22 bytes: 9µ9⁶”*⁶⁸⁶”=⁶⁸’⁸9_”¶“”) (this is probably the ugliest Jelly code I've ever written)

there's probably a shorter way involving G but smashprinting like 13 nilads next to each other works xD

caird coinheringaahing

...is that just a chain of mostly nilads?

@hyper-neutrino G won't be that useful, but K€Y might be

hyper-neutrino

@cairdcoinheringaahing yep, it's great xD

Leaky Nun

@hyper-neutrino first line ends in 08 in the output of your code

hyper-neutrino

6:56 PM

oh wasn't paying attention oops

9µ9⁶”*⁶⁸⁶”=⁶⁸’⁸⁵_”¶“”) - Try it online! (still 22)

caird coinheringaahing

7:11 PM

What are some useful constants to have digraphs for?

hyper-neutrino

do you have a list of ones you already currently have?

caird coinheringaahing

All current Ø ones in Jelly, plus a couple more

Might go look through MATL's list, it always seems to have a weirdly specific but useful constant on hand

user

@cairdcoinheringaahing An infinite list of all natural numbers?

caird coinheringaahing

I was considering that, but I'd have to rework Jelly to add infinite lists

Adám

Uppercase hex digits?

user

7:18 PM

:(

caird coinheringaahing

I'm planning on adding them to Jam tho

@Adám Got 2 different ones for that, '0123456789ABCDEF' and '123456789ABCDEF0'

Or do you just mean 'ABCDEF'?

Unrelated String

some of matl's weirdly specific constants are for convolution and i have no idea if jelly's convolution builtin does the same thing

DLosc

@LeakyNun Pip -s, 20 bytes: Fi,9P[9'*i+1'=i.9-i]

user

What about [0, 1]? I often find that useful

hyper-neutrino

that's already a thing

if you just make a sequence class and implement __getitem__, you might be able to hack infinite lists in to some things, though it'll break a lot of other things

Unrelated String

7:20 PM

1,0 could be good though

hyper-neutrino

agreed

Adám

Ah, yes, the Boolean 3×3 matrices for the van Neumann and Moore neighbourhoods with and without self.

Unrelated String

i vaguely recall having needed it recently

caird coinheringaahing

@UnrelatedString Added that as Ø¹

Unrelated String

ayy

caird coinheringaahing

7:21 PM

⁵D sometimes works, but often requires ¤

@Adám Oh yes, I've been meaning to add those, thanks!

@cairdcoinheringaahing Not exactly a constant, but a nilad: Current time [hh, mm, ss] and current datetime [yyyy, MM, dd, hh, mm, ss]

Adám

Maybe the Boolean matrices [[1,0],[0,1]] and [[0,1],[1,0]]?

caird coinheringaahing

I have the second one as Øx, cause apparently that is used significantly more than the first (Jelly corpus)

Adám

@cairdcoinheringaahing Do you have complex numbers?

caird coinheringaahing

@Adám Yeah, I'm just extending Jelly's builtin set

Adám

@cairdcoinheringaahing Do you have the four directions?

caird coinheringaahing

7:25 PM

[1,-1,i,-i]?

Adám

Yes, but also in clockwise and anticlockwise order.

hyper-neutrino

not a digraph nilad, but do you have N => Nth identity matrix? not sure if it'd be useful ¯\_(ツ)_/¯

Adám

You might also want the four diagonal directions, and all 8/9.

caird coinheringaahing

What's the 9th?

hyper-neutrino

0

Unrelated String

7:26 PM

yeah getting the diagonals is much harder than just the main 4

Adám

8/9 are the same thing

Unrelated String

in terms of gaussian integers at the least

hyper-neutrino

rotate coordinate counterclockwise/clockwise might be a useful monad

caird coinheringaahing

Got both of those

Adám

That's the same thing as multiplication/division by i.

hyper-neutrino

7:28 PM

well, for coordinate pairs i meant

caird coinheringaahing

Not for matrices

Adám

How about [0,90,180,270] and [90,180,270,360]?

caird coinheringaahing

Got to love Jelly's greedy matching: Øı is a valid digraph :P

Unrelated String

as opposed to

caird coinheringaahing

Trying to parse it as a complex number

Or, splitting the chain at Ø and setting the argument to ı

Adám

7:29 PM

Pretty obvious that digraph prefixes are stronger.

hyper-neutrino

i mean it'd require some weirdness to match the ı first wouldn't it? cuz if you scan left to right you'll see the Ø

wait no jelly uses a giant regex to parse everything

caird coinheringaahing

I think that ØØ is also valid, but ØØ1 isn't (the first Ø currently breaks the chain, then Ø1 is yielded) :P

hyper-neutrino

lol wut

Unrelated String

oh

hyper-neutrino

cursed regex moment

caird coinheringaahing

7:32 PM

Or rather, ØØ could be valid, if you add it to atoms

Unrelated String

...so the digraph prefixes aren't actually coded as digraph prefixes, they're just the characters all defined digraphs start with

L|

:|

caird coinheringaahing

Yeah, it's why Œœ is a valid atom

Adám

I suppose that's like ⎕ in APL.

Unrelated String

you could definitely tokenize with the digraph prefixes being defined independent of the actual digraphs

caird coinheringaahing

@Adám von Neumann is orthagonal and diagonal, and Moore is just orthagonal, right?

@UnrelatedString Based on my interpretation of the regex, you could define AB as a valid digraph

Unrelated String

7:36 PM

i feared as much

caird coinheringaahing

There's nothing in the regices that references Æ etc.

Adám

@cairdcoinheringaahing No, opposite.

caird coinheringaahing

@Adám Is ⎕ by itself used for anything?

Adám

@cairdcoinheringaahing Yes, so when adjacent to something it can combine with, it needs separation.

caird coinheringaahing

Hmm. I suppose that you could assign quick/atoms to ØÆæŒœÐ, but I wonder how often you'd need to separate them

Adám

7:40 PM

Try it online!

hyper-neutrino

you can remember it as "moore" neighborhood has "mo(o)re" neighbors (just made that one up)

3

Adám

Maybe add nilads for the next level Moore and van Neumann neighbourhoods too?

hyper-neutrino

@RedwolfPrograms not a bug, but interesting behavior with your userscript - pressing ESC to cancel editing will stop editing but it won't clear the field anymore, meaning the message will just stick around in the box, lol.

@Adám like, in three dimensions?

Adám

@hyper-neutrino No, [[2,2,2,2,2],[2,1,1,1,2],[2,1,0,1,2],[2,1,1,1,2],[2,2,2,2,2]] and [[0,0,2,0,0],[0,2,1,2,0],[2,1,0,1,2],[0,2,1,2,0],[0,0,2,0,0]]

hyper-neutrino

ah, i see

caird coinheringaahing

7:43 PM

So, von Neumann neighbourhood would be [[0, 1, 0], [1, 0, 1], [0, 1, 0]]?

Adám

Yes.

hyper-neutrino

that makes more sense. also could be worth a monad

caird coinheringaahing

Considering adding a "given a matrix and md index, output the von Neumann neighbourhood" dyad

Wezl

CMC: Given a space separated list of words, uppercase the second to last one and place the last one in a random place strictly inside of it.

Adám

@Wezl Can we assume the words are lowercased?

Wezl

7:55 PM

example: it's barely below average => it's BAREaverageLY below or it's BARaverageELY below but not it's averageBARELY below

@Adám yes, but there may be non-alphabetic characters

Adám

Define word.

Wezl

@Adám [a-z'_0-9]+

Adám

What other chars can the string contain?

Wezl

space, and you may choose to accept a newline

Adám

Ah, so word means any sequence of non-spaces.

Wezl

7:58 PM

yes

bend it how you like

and you can output all solutions instead of a random one

Adám

@Wezl Can the second to last word be a single letter?

Wezl

no, because the last word couldn't be inside it, and there must be a solution

whoops, I meant uppercase the third-to-last one

it was a bad challenge anyway

and there's no CMC sandbox

Adám

Huh, ok.

A username

Nice PFP Wezl

Wezl

Thanks. It's animated but the animation doesn't work (bc &g=1).

A username

8:05 PM

:(

Nice anmiatttion

Wezl

it's for javascript awareness month :P

A username

bbyte

I can't believe I clicked that

Adám

@Wezl Extended Dyalog APL, 60: '\S+ \S+ \S+$'⎕R{((?≢1↓x)(↑,y,↓)⌈x),' ',⊃⌽x y z←≠⍛⊆⍨⍵.Match} Try it online!

DLosc

@Wezl Pip -s, 28 bytes: YUCg@-3yRARR#y-1.:DQgg@-2:yg

Sort of abusing the input format--the words are taken as command-line args, which makes them "space-separated"

Nope, that's bugged. Hang on...

27 bytes: YUCg@-3y@RR#y-1.:DQgg@-2:yg

31 bytes with input as a single argument, as it probably should be: a^:sYUCa@-3y@RR#y-1.:DQaa@-2:ya

caird coinheringaahing

8:20 PM

@Ausername status-completed

That's probably the reason New Posts has broken

hyper-neutrino

hm

N3buchadnezzar

8:43 PM

pastebin.com/QYfvu6rw

It might not be pretty, but it is working like a charm xD

caird coinheringaahing

9:11 PM

You know what's fun? Upgrading a lang to use symbolic integers in all it's math functions, then forgetting to change the builtin constants from floats :P

N3buchadnezzar

Changing / to // right? I had some similar issues with some python code

I kept getting round off errors when using floats, but switching to ints fixed it

caird coinheringaahing

No, more "why is e^0.5 outputting 1.6487212707001282 instead of sqrt(E)?"

N3buchadnezzar

ahhhhh yes

hyper-neutrino

why tf does µ,µ/ work

shouldn't / error on monads? (we've probably been over this :P)

caird coinheringaahing

why wouldn't it?

hyper-neutrino

9:16 PM

it's reducing over a monad, isn't it?

caird coinheringaahing

/ just assumes that the link its given is dyadic. It only errors on builtin monads because they have lambda z: ...

Sandbox Posts

0

A: Sandbox for Proposed Challenges

KEYBOARD FOR MONKE! (WIP) code-golf Flavortext So...this is awkward. It seems I accidentally returned to monke last night after eating one too many banana sundaes. This has made many things inconvenient, especially typing. You see, monkeys only need the following characters: uppercase letters (A-...

hyper-neutrino

so do monadic chains do like lambda x, y = None or smth like that

@SandboxPosts lol

caird coinheringaahing

For chains, because it just calls dyadic_chain(chain, args) each time, it doesn't actually matter

hyper-neutrino

*sigh* jelly is truly ~~cursed~~ blessed

user

9:17 PM

blursed

caird coinheringaahing

@hyper-neutrino No, it just pops the chain off the stack and only ever passes it to dyadic_chain :P

hyper-neutrino

aaaaaaaaaaaaaaaaaaaaaaaaaaaah

caird coinheringaahing

def reduce_simple(array, link, *init):
	array = iterable(array)
	if len(array) == 0:
		return list(init)
	return functools.reduce(lambda x, y: dyadic_link(link, (x, y)), array, *init)

hyper-neutrino

lol

caird coinheringaahing

tbf, it makes sense :P

hyper-neutrino

9:18 PM

it does but still :p

okay is there an example of a quick that requires a monad or dyad and will automatically coerce a variadic chain into a specific arity?

caird coinheringaahing

... /?

And \"þ{ etc. Anything that only works on a fixed arity will coerce the chain it's given into the required arity

hyper-neutrino

i'm trying to find an example where something µ something works but µ something µ something errors

or will that never error just cuz combined chain arities are a scam

caird coinheringaahing

I don't think that will ever error, assuming the only change is prepending µ

hyper-neutrino

ah. okay. well, it's not too important, just wanted an example for the tutorial i'm writing

N3buchadnezzar

mu seems complicated :p

caird coinheringaahing

9:30 PM

µ is the least complicated of the 5 :P

N3buchadnezzar

uh oh

caird coinheringaahing

Unless you have ð or ɓ in the link, µ basically means "reset the left and right argument to what the current value is"

Leaky Nun

9:44 PM

CMC: Take a string of lowercase letters (a-z); if it is "amogus", output "sus"; otherwise output nothing

caird coinheringaahing

@LeakyNun I have 14 bytes in Jelly

Leaky Nun

@cairdcoinheringaahing same

caird coinheringaahing

@LeakyNun Mine

user

@LeakyNun Scala, 30: s=>if(s=="amogus")"sus"else "". Nothing interesting

Leaky Nun

@cairdcoinheringaahing Mine

caird coinheringaahing

9:53 PM

@LeakyNun Nice

N3buchadnezzar

35 =(

Leaky Nun

@user you can save one by removing the space

@N3buchadnezzar you need to output "sus" not "gus"

user

@LeakyNun Oh nice, thanks

Usually doing foo"bar" doesn't work, never realized keywords are special

N3buchadnezzar

@LeakyNun 34 but less interesting

Leaky Nun

@N3buchadnezzar I have a 28-byte one

user

9:57 PM

lambda s:s=="amogus"and"sus"?

Leaky Nun

that returns False instead of nothing

user

Actually, that gives False

Yeah

N3buchadnezzar

lambda s:["","sus"]["amogus"==s]

This one is 32

user

As is lambda s:"sus"[(s!="amogus")*3:]

@LeakyNun Any hints?

rak1507

lambda s:(s=="amogus")*"sus"

user

10:02 PM

@rak1507 Ah, that's smart!

Leaky Nun

yeah mine is the two multiplicands switched

N3buchadnezzar

@rak1507 Ah since False = 0 and True = 1, clever!

CMC Take in a string "ABCD" + and a direction "l" or "r" and cycle the string to the left "BCDA" or right "DABC".

Leaky Nun

can I take other two values for the direction?

caird coinheringaahing

If so, Jelly 1 byte

Leaky Nun

@N3buchadnezzar

caird coinheringaahing

10:16 PM

Otherwise, 7 bytes

hyper-neutrino

I have this for 7 bytes assuming strict I/O is required

caird coinheringaahing

~~I have 6 bytes for strict I/O :P~~ No I don't

Leaky Nun

I have this for 7 bytes

N3buchadnezzar

I picked l or r because of Jelly :p

caird coinheringaahing

I had this for 7

hyper-neutrino

10:23 PM

oh, o- is a lot nicer than my double double-decrement :p

caird coinheringaahing

Jelly poll: for programs in the form ...µ<quick> where ... is all one chain (no separators), what're the 3 most common <quick>s?

hyper-neutrino

to quote Bubbler, "Trailing ɓ chain trick is indeed strong :P" :P

N3buchadnezzar

corpus?

caird coinheringaahing

Yeah, this is based on the corpus

Leaky Nun

idk, filter?

each?

hyper-neutrino

10:24 PM

i'm gonna guess "each" (before ) existed), "max for", and "filter over"

or maybe just filter over and filter against and each

caird coinheringaahing

Filter is surprisingly low. € is #1 (51 times), then # (28 times) and Þ/¡ (both 22 times)

Of the three filter quicks, they are used a total of 7 times

Leaky Nun

@N3buchadnezzar Python, 43 bytes: lambda s,d:[s[-1]+s[:-1],s[1:]+s[0]][d<"r"]

N3buchadnezzar

Kinda cool to see 3 different 7 byte solutions in the same language

@LeakyNun This one is a bit iffy as it allows other inputs than l and r

Aaron Miller

@N3buchadnezzar can we assume input is one of those?

N3buchadnezzar

@AaronMiller input will always be l or r yes

caird coinheringaahing

10:30 PM

Q: Which of qṇẉỵẓėṅ makes the most sense for a cmp(x, y) dyad?

Leaky Nun

@N3buchadnezzar then how is my program problematic?

N3buchadnezzar

A bit iffy makes my nose wrinkle, I did not say it was wrong :P

Leaky Nun

@N3buchadnezzar Python, 42 bytes: lambda s,d:s[2*(d<"r")-1:]+s[:2*(d<"r")-1]

Neil

@LeakyNun Retina, 28 bytes

rak1507

@cairdcoinheringaahing q for query

Neil

10:42 PM

(-2 bytes if trailing newline is allowed)

Leaky Nun

@Neil cool

caird coinheringaahing

@hyper-neutrino I'm modifying the behaviour of <monad>/ in my fork: <monad>/ is equivalent to ,<monad>¥/, given that µ/ can always be replaced with ð/

Neil

@LeakyNun Charcoal, 19 bytes: Ｅ⁹⪫⟦9 * ⊕ι = ι⁻⁹ι⟧ω

caird coinheringaahing

(Plus all of the other reduce quicks)

Aaron Miller

@N3buchadnezzar ₍Ǔǔ⁰\r=i should work in Vyxal, but there seems to be a glitch involving strings but also parallel apply?

I’m not really sure

Neil

10:48 PM

actually Charcoal, 18 bytes: Ｅ⁹⪫⟦9 *⊕ι=⁺Ｉι⁻⁹ι⟧

hyper-neutrino

@cairdcoinheringaahing nice, i think that makes sense :p

Anush

11:19 PM

How can you split a text file 0.9/0.1 into two files from the command line?

lyxal

11:31 PM

@AaronMiller fixed

Try it Online!

caird coinheringaahing

11:52 PM

Latest update to my Jelly fork is officially out. Adds a whole bunch of new builtins and functionality. Some highlights include h being aliased as lcm, some irritating bugs being fixed and proper symbolic math support

Feel free to drown me in pings because of various things I've done wrong ;P

N3buchadnezzar

Did you remember that Python 3.9 has lcm built into math? eg from math import lcm or math.lcm

caird coinheringaahing

I'm using 3.7

N3buchadnezzar

Eeeek, a caveman! :P

caird coinheringaahing

Besides, it already has a 2 byte lcm builtin, so I just aliased it (see the last few lines atoms ['h'] = atoms ['æl']) :P