The Nineteenth Byte

12:36 AM

@ASCII-only Apparantly DjMcMayhem

@Skidsdev No you do since it references itself

Skidsdev

@Pavel oh you did recursion?

Well, I didn't

DJMcMayhem

1:01 AM

@Pavel Yup

Veskah

1:34 AM

You ever make a Contact clue so good, it persists across several kill commands and a new game? I have

Lens is kinda crap

Skidsdev

yeah for real

3:55 AM

calculus is hard

send help

@Pavel help has arrived

@Skidsdev no i did

@Pavel ~~ffao's bots are kinda crap~~

@Pavel well imo a userscript is way, waaaay better for this

This question is asking for me to "Find the 3rd degree Maclaurin polynomial" of a function. I'm a bit confused, from what I can tell you have to specify a specific point to center the polynomial around?

They likely meant Taylor polynomial

Dennis

Some texts refer to the Taylor polynomial around 0 as the Maclaurin polynomial.

Ah did I get it backwards then

4:07 AM

Ok that sounds right

Thanks!

Dennis

Some texts refer to the Maclaurin polynomial around 0 as the Taylor polynomial.

All of these are conventions.

@Pavel well apart from that

Question: How many ways can you write the nth triangle number as the sum of n distinct positive integers?

First couple of terms of the sequence: 1, 1

Could be fibonacci ...

@Skidsdev oh yeah did you see the 85 byte one

"First couple of terms of the sequence: 1, 1"

4:11 AM

literally a couple of terms :P

primo

6 = 1, 2, 3
10 = 1,2,3,4
15 = 1,2,3,4,5

pretty sure the function you're looking for is f(n)=1

yeah.

@primo ~~who knows maybe someone will find a counterexample~~

This will be usable to my advantage.

definitely a good lower bound though

primo

for great victory

4:14 AM

@PostLeftGarfHunter lower bound for what

@ASCII-only 1 is a lower bound for the number of ways you can write the nth triangle number as the sum of n distinct positive integers

and also an upper bound? :P

That has to still be proven.

ok done

proof: think of triangular numbers as "lowest number that is the sum of n distinct positive integers" maybe

That is a good start but I don't think it is quite enough on its own. Unless you are seeing something I am not.

4:18 AM

hmm?

lowest n distinct positive integers = inclusive_range(1, n)

sum(inclusive_range(n)) = triangle(n). qed

Yes it is true and useful but I don't see how it proves the statement on its own.

Οurous

It's an upper bound because any other combination of n positive integers would be larger than the first n positive integers.

You have not shown that it is unique

we cannot assume that just because some sequence has the smallest possible sum that there is not another sequence with an equal sum.

:|

they have to all sum to the same number right. then let seq_2 = inclusive_range(1, n) as a, where a_x = a_x - m and a_y = a_y + m

a_x - m must not be between 1 and a_x - 1 since they're all taken. hence it must be 0

Yeah that proves it.

But that is an additional step lacking in the original

4:40 AM

wait

SE has hotdog flair :|

yes

5:48 AM

@cairdcoinheringaahing very late, but Germanic languages are pretty different from Romantic languages

6:16 AM

@ASCII-only usually it's really easy, but I had a bout with some ancient software that generated reports in CSV, except that multiple values in the tables were strings that had commas in them. With 200k lines across 50 files, it turned ugly, though it could have been worse

6:44 AM

@Neil the issues should be fixed once Dennis pulls

7:06 AM

now to implement actual operators. let's hope this works

8:25 AM

@ASCII-only I should probably get that done some day, but with it being a side project, and the compilation of all files and running them in a separate process to ensure that the testing environment isn't tampered with by the unit test code, things got too bothersome

@maxb tampered with?

:| what kind of black magic does your language need

@ASCII-only I just took my quick reference sheet and turned it into an explanation generator, it was like 10 minutes of work. But I just got lucky that I had structured my reference sheet that way to begin with.

:|

oh

it's stack based

of course the explanation is easy to generate

@maxb just making sure, sbcs to unicode translation (and vice versa) works right? :P

@ASCII-only I have had unexpected behavior when external scripts have altered the code page for my language, breaking everything. I don't really have a good structure as it is my first (and only) language, but I've had really fun with it. You can check out the compilation function to get an idea of why coverage would be harder than usual to get working

...

0/10 you don't call a function to run program

also protip: make the preprocessor a function

it's like 5 lines anyway

8:35 AM

@ASCII-only since I have a compilation step in between, as long as the input can be parsed by python, I can get a file of bytes containing the correct bytes.

@ASCII-only Yeah you're probably right... Looking through this I see a lot of sloppy coding. One of these days (before the heat-death of the universe) I hope to get around to refactoring MathGolf

@maxb :| that deadline is a while away

also. i've tried to refactor Charcoal several times :/

gave up every one of those times

and Charcoal has a lot (I mean a lot) of random/abandoned features

When I started with MathGolf, it was basically one class with 256 if-elif-cases. It felt like I was programming an advanced AI

:|

dzaima

@ASCII-only and I should get around to remaking Canvas. it has way too little random features :/

advanced AIs clearly have 256 if-elif-cases. clearly

8:42 AM

Then I actually got everything working quite nicely, with separation of operators by how many arguments they consumed from the stack, and turned the 256 cases to 3 cases and polymorphism (kind of)

@dzaima good luck trying to beat implementing part of the wolfram language :/

just look at the number of lines :/

dzaima

@ASCII-only yeah why did you do that

@dzaima so Charcoal could blow away all the competition for Bake a Slice of Pi

that's literally the only reason

also regex. very important

it hasn't been used afaik but that doesn't matter ~~;_;~~

@maxb oh god. it's still a massive amount of elifs. use a lambda dictionary

note to self: i need an explanation generator too

more random features incoming, probably

@ASCII-only It's a work in progress, but yes, dictionaries of functions is the approach I've used for refactoring so far, and it's probably the way I'll keep going.

another useless feature (part of wolfram. it doesn't expand it though, oops)

@maxb nice

and then you see charcoal's interpreter

you literally have no idea what's happening because different parts are in different files :/. plus, most of the logic is in the Canvas class (not to be confused with dzaima's Canvas)

@dzaima ok. real reason was mostly i was bored at the time :/

@maxb i'm not much better, i've used the same (self-written) parser for every single interpreter i've written (well, ones that need an interpreter)

on the plus side, that means the same ASTs. and this AST format is pretty nice

9:04 AM

@ASCII-only Ah yes, the Anamorphic strech transform, of course

@maxb :|

9:22 AM

@ASCII-only hmm...

I think you copied the code for Minus and ended up making all those function subtract arrays if you pass two arrays

L appears to be listed twice in multidirectionals

@Neil oops :/

@Neil intentional, one is floor (doesn't make sense to add char for ceil without adding the one for floor too)

oh, it looked like an L

github.com/somebody1234/Charcoal/blob/…

@Neil oops. what would be the most expected result? vectorize to length of shorter array?

cartesian product? hmm

@Neil also, yeah i did copy the code from Subtract. oops :/

@ASCII-only depends on the operator, for instance, should Equals compare two arrays rather than returning an array of their elementwise comparisons?

but I guess vectorising to length of shorter array would be a good default behaviour at least

true

hmm. i was thinking about set union/intersection for & and |, but i guess you can do that easily by making one of the arguments a python set

@Neil should be fixed now, I think?