The Nineteenth Byte

ATaco

05:21

My mission statement for ARBLE is to have someone complain that my answer is a snippet.

Sandbox Posts

05:57

0

A: Sandbox for Proposed Challenges

Repeat your program to print Fibonacci numbers code-challengesource-layout Write a program fragment so that, when repeated N times it prints the Nth Fibonacci number. For example, if your program is print(x) then: print(x) should print 1 print(x)print(x) should print 1 print(x)print(x)print(x) s...

Simd

06:10

CMC Given a list of positive floats, evenly spread the floats that occur between consecutive integers.

ATaco

Example?

Simd

0.1 0.2 1.4 goes to 0.333.. 0.6666 1.5

Unrelated String

ooh

Simd

:)

Unrelated String

i assume inputs are always given in ascending order?

Simd

06:21

yes

Unrelated String

and never integers

Simd

right

ATaco

I need a group construct in ARBLE...

Simd

:)

Extra gratitude points for a python answer

Unrelated String

can probably be shorter but ḞJ÷L‘$+ƲƙF` for 11 in jelly

er

`` ḞJ÷L‘$+Ʋƙ`F ``

fuck it

ḞJ÷L‘$+Ʋƙ`F

Simd

06:28

@UnrelatedString which language is that in?

Unrelated String

jelly

Simd

see what line I replied to :)

Unrelated String

:)

Simd

is it suitable for main?

I am tempted to post it there

although I would be very grateful for a Python answer :)

Unrelated String

floor, map over groups {divide [1 .. len] by len+1 and add the floors back}, flatten

Simd

06:30

:)

Adám

@Simd APL, 18: ∊⌊(⊂+∘(⍳÷1∘+)∘≢)⌸⊢ Try it online! doing exactly that ↑↑

Simd

@Adám nice!

Adám

@Simd Well, Py'n'APL exists…

Simd

@Adám it's a great name too!

New Posts

0

Q: Evenly spread values

Given a sorted list of positive floats (none of which are integers) such as: [0.1, 0.2, 1.4, 2.3, 2.4, 2.5, 3.2] Evenly spread out the values that fall between two consecutive integers. So in this case you would get: [0.333, 0.666, 1.5, 2.25, 2.5, 2.75, 3.5] (I have truncated the values 1/3 and...

code-golf

Simd

06:33

because your answers were so great :)

Adám

Answered 59 secs after posting :-)

Simd

:)

@Adám that's what I am talking about!

Adám

my code would work with repeats too

Simd

@Adám interesting

@UnrelatedString thanks for the answer!

does anyone here code in Python?

not that I am begging :)

Adám

@Simd I've added a full explanation. Should e possible to translate to Python. Maybe ChatGPT can even do it…

Simd

06:48

groupby(x,int) is new to me!

@Adám thank you

oh I see what it is doing (the python code that is)

everyone has gone for grouping

mousetail

@Simd There is a ton of useful stuff in itertoos: docs.python.org/3/library/itertools.html#itertools.groupby

Highly recommend looking at it

Simd

@mousetail I know it is there but have never spent enough time going through it

thank you

thanks for the python answer too

The Thonnu

@Simd you'll probably get a Python answer from xnor soon

Simd

@TheThonnu I look forward to it

although mousetail got there first :)

what is k+-~j for? I mean why +-

mousetail

I wouldn't be suprised if an answer with half my length is possible

@Simd ~ is equivolent to -x-1

So -~x is x+1

Simd

06:56

and k-~j doesn't work?

mousetail

It's k-(~j/-~len(i))

Order of operations is different

Simd

I think you mean -~ trick (in your answer)

@mousetail thanks

mousetail

@Simd Actually I can sort of do something like that

Simd

cool

mousetail

Since both operands of / are negative I can invert them both and save 2 bytes

Simd

06:59

for k,[*i] in ... is tricksy too!

@mousetail nice

I am never 100% confident when I see * in python

mousetail

Honestly same, there are a lot of odd edge cases

Be happy python doesn't have object desctructuring like in Javascript though, ton more edge cases for ... in JS than for * in py

Simd

If I have a function def foo(*x) I understand that if I call foo(2,3,4,5) then x becomes (2,3,4,5) but I am not really sure how to explain that

mousetail

All the remaining arguments get put into a list then assigned to x

Simd

@mousetail apparently a tuple

mousetail

Also ** does the same for keyword arguments

Simd

07:03

I have had bugs where I assumed it would be a list

mousetail

It used to be a list

Simd

ah... until python 3?

mousetail

Probably

Tuples are generally superior to lists

Simd

because they are hashable?

mousetail

And immutable

Simd

07:04

why does immutable make them superior?

mousetail

You can't accidentally modify them, which would modify all copies across your code

btw this is an excellent python resource: github.com/satwikkansal/wtfpython that explains these types of things well

Simd

ah ok. I am starting to get annoying by python (or maybe emacs). I really want to know if my indentation is broken or I have misspelled a variable before I run the code

@mousetail thanks!

inspired by wtfjs :)

mousetail

07:18

@Simd Can't you get a flake8 plugin for emacs?

Simd

@mousetail maybe :) I am trying to install eglot currently. It's not easy

New Posts

10:46

0

Q: Find separating sets

Two points pand q in a topological space can be separated if there are open sets U and V such that: p is an element of U and q is an element of V U and V do not have any elements in common Your task is given a topology (a set of open sets) and two points p and q to output sets U and V satisfyin...

code-golf math topology

mousetail

12:12

@TheEmptyStringPhotographer Great to see someone trying CellTail

Hope you try some non-trivial challenges as well

The Empty String Photographer

13:12

@mousetail I have a suggestion: remove integer overflow

mousetail

14:01

Remove?

Like just crash?

The Empty String Photographer

@mousetail it's annoying because if i have a number more than 9 digits i have to split it up.

mousetail

That's just how numbers on a computer work

It has 64 bit numbers

The Empty String Photographer

@mousetail signed or unsigned?

mousetail

Signed

Actually the wasm version might be less than the offline version...

Bbrk24

wasm is 32-bit

@mousetail 64-bit can handle 10-digit numbers easily

So "if i have a number more than 9 digits i have to split it up" and "It has 64 bit numbers" can't both be true

But wasm is 32-bit, because 64-bit integers cannot losslessly be converted to JS numbers. You either have to use the "i53abi" (in which long long is actually 53 bits) or use bigints instead of numbers at the JS/wasm border.

mousetail

14:43

It's probably 32 bits then

Still more than enough for any non-trivial program

Bbrk24

15:41

@Adám If you are concerned with efficiency, swift-algorithms has randomSample(count:). swift-algorithms is one of a few packages that are in a weird spot: they're first-party packages, but they aren't available by default, so you need to list them in your package manifest.

Being external packages is annoying since environments like Godbolt typically don't have them

Other such packages include swift-collections, swift-async-algorithms, and swift-numerics

mousetail

17:08

CMC Given a string, output the index of every : and , that is not surrounded by parenthesis

Ginger

@Bbrk24 that's how I plan to distribute most of Rabbit's pseudo-stdlib

as separate modules with their own release cycles, installed independently by the module manager

the few modules that don't follow this pattern will be for things like low-level networking and OS interfaces, and will be directly built into the interpreter and updated in lockstep with it

Adám

@mousetail You mean immediately adjacent, or even further out? E.g. does (a:b) give an empty result or not?

mousetail

Further out, with any level of nesting. So :((ab:cd):) should return only 0

Adám

17:28

@mousetail QuadR ≡, 25 23: Try it online! (1-indexed)

⍸⍵∊':,'
\([^()]*\)
⊣\⍵M

mousetail

Very nice

Bbrk24

@Ginger Here's the thing: without a package manifest, you can't have any external dependencies besides the stdlib and anything else built into the compiler, so simple scripts can't use them. By doing this, you massively increase the amount of boilerplate needed for some simple tasks.

Not that Swift is exactly a scripting language, but

Ginger

Rabbit (normal Rabbit, that is) isn't usable as a scripting language at all

all code has to be formatted as a module with a manifest

Bbrk24

Swift is trying to make Swift scripts a thing and this is the biggest obstacle

Ginger

it will be possible to use the Rabbit Script Host and schema modules to make "scripts", but I'm not sure how that'll work just yet

~~of course, all of this does assume I resume working on Rabbit~~

which I probably will, I like the concept too much to just abandon it

Sandbox Posts

17:39

0

A: Sandbox for Proposed Challenges

2D Percolation Model Per Wolfram MathWorld: "Percolation theory deals with fluid flow (or any other similar process) in random media." The model is a 2 dimensional lattice whose edges are either "open" or "closed" with probability p in [0,1]. At a percolation probability P, each edge will be eval...

mousetail

18:30

My CellTail sorting algorithm finally actually works: github.com/mousetail/CellTail/blob/main/examples/sort.ct

Simd

19:09

TIL that you can't really use a tuple of tuples of floats as a key in a python dict

mousetail

You can?

Simd

hmm... must be my mistake

I though I had tried it and it didn't find the key but I think I got that wrong

mousetail

Probably the floats are not actually equal

floats can be off by a tiny bit and have a different hash

Simd

ah yes

mousetail

Even if, when you print them, they look the same

Simd

19:13

good point

Simd

19:24

How can I make a hash of a tuple of tuples in python?

Md5 or anything similar will do

Bbrk24

There's the __hash__() method

>>> (1, 2, 3).__hash__()
529344067295497451
>>> (1, 2, 4).__hash__()
-4363729961677198915

mousetail

Or just hash((1,2,4)) will do the same thing

It works on any python type

For strings hash will randomize when you restart python though

Simd

@Bbrk24 what is the range it hashes to?

I am only doing it save space

@mousetail thanks

Bbrk24

~~let me google that for you~~

> hash() truncates the return value based on the bit width of the host machine.

Simd

@Bbrk24 I did try to look out up :)

Bbrk24

19:36

docs.python.org/3.10/library/functions.html#hash

Simd

@Bbrk24 ah!

Thanks

So it seems it is 64 bit signed

Transcript for