The Nineteenth Byte

« first day (5122 days earlier) ← previous day next day → last day (16 days later) »

Redz

00:53

@rydwolf 9/10, maybe you could make contrast between buttons and background a bit clearer

l4m2

Do a flagged answer automatically protect question?

emanresu A

several deleted answers, usually

5 hours later…

Sandbox Posts

06:07

A: Sandbox for Proposed Challenges

PizzaTian is a popular restaurant specializing in pizza delivery. Customers can place orders by phone, and each order comprises two components: Pizza Prices: The cost of each individual pizza. Delivery Fee: A fixed fee per order, regardless of the number of pizzas. Recently, PizzaTian has intro...

4 hours later…

Sandbox Posts

10:15

A: Sandbox for Proposed Challenges

Draw U+30EDE Biangbiang noodles are a type of Chinese noodle originating from Shaanxi cuisine. The noodles, touted as one of the "eight curiosities" of Shaanxi (陕西八大怪), are described as being like a belt, owing to their thickness and length. Biangbiang noodles are renowned for being written usin...

Themoonisacheese

cooking today

emanresu A

Mukundan314

emanresu A

Themoonisacheese

am i literally the only one that doesnt have it

i blame this on microsoft

any feedback though?

emanresu A

10:27

you should probably have a stronger spec on what's allowed as "simplification"

Themoonisacheese

yeah probably. i feel like "each stroke should be individually recognizable" conveys the intent, but i don't know how to make that intent explicit

lyxal

10:40

Steamed Hams but it's by Dr Seuss

►

Weird Glyphs

@Themoonisacheese It's also displayed like that for my Windows and Ubuntu

Themoonisacheese

maybe a chrome/firefox thing then

3 hours later…

mousetail 'he-him'

13:18

Suppose I have a cyclic number-like data type, such that if I keep increasing it it eventually comes back to the start. Then I have a function ">" that returns true if a number is in the next "half" of the cycle from the previous. Is it possible to write an algorithm to sort a set of these numbers?

Themoonisacheese

13:52

@mousetail'he-him' if i understand this correctly, no

suppose the max value is 255, and i give you the set (1,2). no amount of 1>2? checks will ever differentiate them

mousetail 'he-him'

Both orders are sorted though

It's a cycle so the starting point of the sort doesn't matter

Themoonisacheese

ok then assume a cycle of size 12, and you get given the set (0,6)

which one is first?

or do you not care and it only matters that, say, (0, 3, 6) would be valid as (6,0,3)?

mousetail 'he-him'

Yea then it's impossible. I think that means the > operation has to be antisymetric which implies the cycle size must be odd

@Themoonisacheese Yes, both would be valid

Themoonisacheese

i feel like it should be possible now but i can't prove it

mousetail 'he-him'

I write this impl but I'm not sure if it's sound: ato.pxeger.com/…

Themoonisacheese

14:04

i fear this is beyond my maths abilities

Mukundan314

are all the numbers unique?

if not what is the result when the numbers are equal?

*what

mousetail 'he-him'

Lets assume either the numbers are unique or a second function to check for equality exists

Actually, I think if you have at least 3 numbers, you can deduce which ones are equal and which are exactly opposite even if the cycle size is even. So this restriction may not be neccecairy

Mukundan314

then I believe the following algorithm should work:
- choose some number as your "start" number
- take all numbers which are within the next of the cycle to this number
- sort above numbers normally since they are within half of the cycle `>` can be used as a regular sort operator
- continue the process from the greatest number according to above sort
- if at some point you are unable to add more numbers and there is more than 1 number left choose a different start point and try again

I feel like there should always be atleast one valid "start" number, haven't proven this yet tho

mousetail 'he-him'

Interesting, I think that should work. I think you can guarentee one of the "left over" numbers should be a valid start number for the next ammept.

Themoonisacheese

this would be a fun question to ask in puzzling.SE or mathematics.SE i feel

Seggan

14:13

@SandboxPosts .

Themoonisacheese

Mukundan314

@mousetail'he-him' intuition for a start number always existing is that if current -> other is greater than half then other -> current must be less than half

4 hours later…

Neil

17:57

it seems to me that quicksort will always "work"

1 hour later…

Unrelated String

19:11

@Seggan I almost asked "is this biangbiang noodles" before I realized I can just. look at the post

TIL that's in Unicode

(and yeah it is also in my font apparently)

2 hours later…

ATaco

21:21

@mousetail'he-him' Theoertically you can sort it, but there are n valid solutions for list size

(Because any rotation of a sorted list is still sorted)

Actually, I guess that's not always the case

Given an arbitrarily high cycle count, [4,3] will always sort as [3,4].

Sandbox Posts

21:33

A: Sandbox for Proposed Challenges

Find the boolean logic to check if a number is prime! fastest-code primes binary integers Your task is to find a boolean expression that uses AND (\$\land\$), OR (\$\lor\$), XOR (\$\oplus\$), and NOT (\$\lnot\$) operators and binary digit positions (\$d_1,d_2,d_3,d_4,d_5,d_6,d_7,d_8,d_9,d_{10},\c...

New Posts

Q: Unap***peel***ing permutations

Given the height, \$h\$, and width, \$w\$, describing a rectangle of the first \$hw\$ natural numbers in row-major order produce the numbers in the order they are encountered by repeatedly removing a strip of the outermost numbers, each time starting with the top-rightmost and ending with the bot...

Seggan

22:08

what on earth

is it displaying like this to anyone else

ATaco

22:21

Sure is, probably an overzealous sanitation script

emanresu A

just grace period I think

DLosc

23:03

My knee-jerk thought was that there was something being censored by a userscript, but yeah, I'm sure the original title attempted to do bold/italics using markdown asterisks and was quickly edited when that didn't work.

att

@mousetail'he-him' can't you just quicksort this

att

23:23

ngn.codeberg.page/k/…

oh neil already mentioned that

ATaco

Hosting a separate instance of my interpreter for every version is probably a bad idea, but it's easier than ensuring each version is backwards compatible :)

Ginger

hey, it works for vyxal :p

lyxal

23:59

I found myself wanting to refer to the Jalapeño online interpreter as theseus (and the many misspellings like theosussy) online to realise hey wait a minute that name only applies to a single online interpreter not all of them :p

« first day (5122 days earlier) ← previous day next day → last day (16 days later) »

all rooms

Transcript for

Feb5

Feb6

Feb7

The Nineteenth Byte: General discussion for codegolf.stackexc...

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC