The Nineteenth Byte

Lyxal

08:51

Wow so apparently this site is almost a decade old

TIL

ASCII-only

@HyperNeutrino no

Adám

@Bubbler Unfortunately, I'm scheduled to run out of rep:

ASCII-only

@HyperNeutrino E=1, W=-1, NW=i, NE=1+i, SW=-1-i, SW=-i smh

Bubbler

@Adám Well, that doesn't mean I can't solve problems (and actually I'm willing to give out the APL bounty in case you're out of rep)

John Dvorak

@ASCII-only eh, I prefer the former. Either works though.

ASCII-only

09:02

NE=1.73+i :p

3

John Dvorak

too bad arbitrary floats don't form a group under addition

ASCII-only

i cant math

John Dvorak

too bad sometimes a + b - b != a in floating point arithmetic

ASCII-only

1.75 and call it a day

@JohnDvorak neither do integral floats so its probs a moot point

Razetime

@Adám Gonna be solving these anyway

John Dvorak

09:08

integer floats work just fine below 2**50

ASCII-only

09:26

@JohnDvorak i mean. same with 1.75 to 2**48 (not 1.73 ofc)

New Main Posts

0

Q: End of Year performance review

We here at "Behavior Analytics Inc." have been tasked to help a very important client under take an end of year performance review of his catchment area, which at the moment covers.... the whole planet. He currently holds data on all everyone within the catchment area, supplied to him by our sist...

New Sandboxed Posts

10:14

0

A: Sandbox for Proposed Challenges

Explain a Code Golf Answer code-golf string Background When writing Code Golf answers, it is often a good idea to add an explanation of the code so the reader understands what's going on. For example, this this answer by @Makonede (abridged): θ # last element of Σ # the input, s...

Neil

11:19

@ATaco JavaScript, 15 bytes: x=>y=>~-x%~-y+1

ATaco

if(!#undo.stack)return
var u = undo.stack:pop()

Spot my huge mistake

Returning a Variable declaration? Sure, why not.

Unrelated String

11:51

i'm taking another crack at day 20 and it's going even worse somehow

i think it's sleep time

Unrelated String

12:03

yeah on the example it's putting 3079 and 2473 on either side of 2311, then giving up

ASCII-only

giving up???

Unrelated String

i'm doing some funky pathfindingesque shit where it dynamically grows the grid outward, keeping track of what tiles it hasn't grown from yet

and it's just running out of those tiles

i don't have any idea how the borders are matching this incorrectly

ASCII-only

day 20 oh

@UnrelatedString you do reflect when needed, yes?

caird coinheringaahing

14:07

@WheatWizard Do we really need the featured on the Sandbox? With the infobox on main, it's kinda superfluous

Wheat Wizard

14:42

@cairdcoinheringaahing It wasn't showing up on main for me so I featured it again.

Redwolf Programs

17:36

Taking another stab at part 20.

Anush

hello all

Redwolf Programs

Hi!

Anush

how things?

HyperNeutrino

17:56

@ASCII-only oh :c oops

ASCII-only

:p

HyperNeutrino

pain

Redwolf Programs

For day 20, rotate-then-flip is the same as flip-then-rotate, right? Just want to make sure I'm not going crazy.

HyperNeutrino

18:12

@RedwolfPrograms no

Redwolf Programs

Oh. Guess I need to rewrite a bunch of stuff :p

Actually I don't, it seems like rotating the other direction then flipping gives the same result, as does flipping in the other direction, so it's like a XOR thing. Switching either one transforms it in the same way, but switching both keeps it the same. or something like that.

I have no idea anymore but I think what I wrote won't cause my laptop to catch on fire so it's fine for now.

HyperNeutrino

If you alternate transform/zip and flip, you will generate all 8 orientations

Redwolf Programs

Okay good, there are still only eight?

HyperNeutrino

Yes. The top edge has four positions and then the left edge must be beside it so that doubles the number of cases, and everything else may be determined by the placement of those edges.

Redwolf Programs

Okay good, everything I wrote up until now is still correct. Thanks!

HyperNeutrino

18:20

pog :D

Redwolf Programs

18:35

All of the input is now in a usable format, and I can determine which are corners and edges!

I can even make some assumptions about how tricky the input is(n't) by looking at this

Razetime

18:51

Cool

Anush

anyone here able to code in Julia? I am trying to translate some Julia code into python ~(and eventually cython).

I would love some help please

pxeger

20:25

Anyone have any opinions on my sandbox post? Specifically do you think it's a duplicate of this challenge?

Redwolf Programs

I'd say just related, not a dupe.

Anush

could someone help me out please. How can I compute x + y?

Redwolf Programs

Even though the title/purpose are very similar, the actual task is very different

Anush

the triangle is equilateral and there is a square inside it

I feel this is meant to be for 11 year olds :)

Adám

@pxeger Yours is range parsing + formatting. The other is sort order parsing + formatting, with the formatting part being identical. It is a good idea to isolate the core of your challenge, i.e. the range parsing, but then it ends up being a duplicate of this. So in conclusion, it isn't a duplicate of any one challenge, but it is a duplicate of the two together.

Redwolf Programs

20:30

@Anush You may want to ask in a math chat room. I know there's an easy way to solve that, I just can't remember it.

Anush

@RedwolfPrograms but everyone in here is lovely and helpful :)

Redwolf Programs

Without any angles for the big triangle I don't think you can solve for x and y.

Oh, it's equilateral?

Adám

yeah, 60

Redwolf Programs

In that case you know one angle on each of those is 60°, and that the other two sum to 90°

Adám

@RedwolfPrograms What do you mean by "the other two"?

Redwolf Programs

20:34

The other two in the triangle formed by the sides of the square, x/y, and the corner of the outer triangle

Wait that explanation makes no sense

Wait yes it does nvm

Anush

@RedwolfPrograms the big triangle is equilateral

Redwolf Programs

If we label the angle by x that isn't 60° as w, and the one by y as z, we get:
x+w=120
y+z=120
w+z=90
x+y=150

Anush

@RedwolfPrograms yes I can see the sum of the two angles to the and right of the bottom of the square is 90

I think I got it. Let's call the angles to the left and right of the middle bottom point a and b

we know a+x+60=180 and similarly we know b+y+60=180. We also know a+b=90

so we have a+b+x+y+60 + 60 = 360

therefore 90 + x+y+60=360

therefore x+y = 210

does that look right?

Redwolf Programs

No

Anush

grr

how come?

(fixed arithemetic error)

caird coinheringaahing

20:47

@HyperNeutrino What happened to the AOC challenges you were planning to do?

Redwolf Programs

Wait, so x+a=120 and y+b=120 and a+b=90, so a=-b+90, so x-b+90=120, so x=b+30, and y=-b+120, and x=b+30 can be written as -b=-x+30, so y=-x+150, so y+x=150

But your above one with x+y=210 also looks right to me...

Wait no, you left out a 60 between your steps.

a+b+x+y+60+60 should become 90+x+y+60+60, not 90+x+y+60

I don't think we have the information we need to find x and y individually, though.

@Anush ^

ASCII-only

@Anush this is ezpz x+y=150

Redwolf Programs

@ASCII-only Why? That doesn't match either of our answers.

ASCII-only

wait no

im high

@Anush this =300 not 360. @RedwolfPrograms is right

but yea we don't have enough info to know them invidually. you aren't given any info about the rotation of the square

New Sandboxed Posts

21:38

0

A: Sandbox for Proposed Challenges

Based Palindromes A palindromic number, as a refresher, is any number which reads the same forward as backwards. However, what about palindromes in other bases? Input Any integer where b > 1. Output All integer base 10 numbers below 1000 that are palindromes in base b. The output can either be a ...

HyperNeutrino

@cairdcoinheringaahing by the time i remembered that it was too late to finish them up and i was busy

finals

Anush

22:05

Thanks very much!

Redwolf Programs

No problem!

caird coinheringaahing

22:32

@HyperNeutrino Next year maybe? :P

Oh, hey @Mr.Xcoder! Haven't seen you here in a while :D

Redwolf Programs

I can't figure out why my day 20 p2 isn't working :/

Mr. Xcoder

@cairdcoinheringaahing Hi there, and merry Christmas! :-)

caird coinheringaahing

@Mr.Xcoder Merry christmas to you too :)

ASCII-only

@RedwolfPrograms ur a smol brain is why

caird coinheringaahing

Oh, and also:

Merry Christmas everyone!

(depending on your timezone)

ASCII-only

22:36

merry christmas everyone

~~anyone not on christmas day right now dont deserve the merry christmas~~

Adám

I should go to bed, I have an intense work day tomorrow…

caird coinheringaahing

@Adám My mind is trying to choose between "joke", "actually has to work" or "has young children" :P

Adám

I actually have work to do. I'm preparing a workshop with a colleague, and it has to be done by mid-January. I guess he doesn't keep xmas. I certainly don't. The reason it'll be intense is because it is both a winter Friday (so I have to rush to get ready for the Sabbath) and because it is a fast day (no food or drink at all).

Redwolf Programs

23:03

Can anyone help me figure out what's wrong with this? It uses a lot of custom functions/methods but they're all easy to explain. (It's my current partial solution to day 20 part 2 and it's doing some really weird stuff).

John Dvorak

/me silently scribbles over "javascript" in CV

4

Redwolf Programs

To be fair that is using 105 sloc of custom predefined array and string methods :p

Lyxal

@RedwolfPrograms I want to talk about how this message isn't a topic on your chat bio even though it is.

Redwolf Programs

Because this statement isn't true

Lyxal

Is this sentence true?

Redwolf Programs

23:15

But seriously I think my code is just messing with me at this point

It's not doing what I told it to.

Basically it's supposed to:
1. Get a list `blocks` of the tiles in the input
2. Make a list `forms` with every possible transform of each block, represented as `[transformed source, index of original tile, list where each edge is represented by the number of original tiles which share an edge with it]]`
3. Create a list `form_sides` which represents each item in `forms` as `[list of edges clockwise from top, index of original tile]`
4. Pick the first valid item in `forms` for the top left
5. Find all valid items in `forms` for right of and below the top left item, which isn't working

Redwolf Programs

23:34

i give up. I have no idea why this isn't working.

Guess I'm not finishing AoC this year.

Five hours and I still haven't made any progress at all.

Lyxal

Yes

story of day 19 part 1

for me at least

Redwolf Programs

It would have been a better use of my time to just print out pieces of paper with each tile and do it by hand

If it was just flipped or just rotated it still would've been hard. There's no way I can do this.

Lyxal

@RedwolfPrograms yes

It's sad that I can't overwrite the pop method of lists in python

Like if you try listobj.pop = some_function, it says no, you can't

Redwolf Programs

Can somebody else explain their approach to day 20 part 2? I don't think mine's going to work.

Unrelated String

why would you need to do that

Lyxal

23:41

@UnrelatedString because I'm making an esolang. Why else would anyone need to abuse python?

Redwolf Programs

Hang on wait

Why do I even need to piece them together, can't I just count the number of #s and subtract the number that are part of the two sea monsters?

Does anyone know if this would work?

Unrelated String

you probably don't have only two

Redwolf Programs

Oh, it's not two in the one given?

Unrelated String

you have to find how many by piecing it together

it is two in the example

Redwolf Programs

I guess since it's a 96x96 result there's way too many possible numbers of sea monsters to just brute force

Adám

23:47

CMC: Given a Unicode character property, answer the characters with that property (according to any version of Unicode). E.g. given Lt, answer ǅǈǋǲᾈᾉᾊᾋᾌᾍᾎᾏᾘᾙᾚᾛᾜᾝᾞᾟᾨᾩᾪᾫᾬᾭᾮᾯᾼῌῼ and given Pi, answer «‘‛“‟‹⸂⸄⸉⸌⸜⸠.

Redwolf Programs

Can someone please just help me by telling me if my approach to fitting the tiles together would work?

Here's a link to it

FrownyFrog

what do you mean first valid for the top left

Redwolf Programs

The first tile which is a corner

transformed so that it is a top left corner

FrownyFrog

yeah should work

Redwolf Programs

I guess my implementation is just wrong

FrownyFrog

23:52

there's actually never any contenders for any spot

if one edge fits, that's the one

Redwolf Programs

Wait, really?

FrownyFrog

totally

Redwolf Programs

I...I just realized what my mistake was and I feel so stupid

Five hours of debugging

It's because I removed items from a list when I was done with them and it messed up the indexes

FrownyFrog

I made a dictionary "top edge" : "list of tiles"

the list is either 1 long or 2 long

Razetime

@RedwolfPrograms soooo

FrownyFrog

23:59

so there is usually a contender, it's the tile you already placed

Razetime

You got day 20?

FrownyFrog

not yet

Redwolf Programs

Okay, it's finally giving me actual tiles instead of empty arrays!

Transcript for