The Nineteenth Byte

DLosc

00:00

Did you meet yourself? :P

Don't we always meet ourselves? Maybe not for the first time, but we definitely meet ourselves fairly regularly

But also, yes, I stared into the face of God earlier, and all I saw was affine subsets and quotient spaces staring back at me, and I had an epiphany:

I suck at linear algebra

emanresu A

00:19

Anyone know what the javascript bigint maximum is?

Rydwolf Programs

Whatever your memory can hold

emanresu A

... I shouldn't have tried to test that

Seggan

01:11

LDQ: should constructors be called new, constructor, or WhateverTheNameOfTheClassIs

or init

o

i just had a great idea

i was trying to figure out how to reconcile methods with ufcs

so in rol there will be no methods

everything is a function + ufcs makes for pesudo-methody syntax

so functions are resolved as specific as possible

so if i make a function that returns the same object, but as a type higher up in the heirarchy, the resolver will resolve the "super" function instead

example:

class A {}

class B : A {}

fun doStuff(self: A) {
    println("a")
}

fun doStuff(self: B) {
    self.super()  // the type of self.super() is A
        .doStuff() // since doStuff() is being called on A im effectively calling the "super method"
    println("b")
}

var a = new A()
a.doStuff() // prints "a"
var b = new B()
b.doStuff() // prints "a\nb"

emanresu A

01:45

new B::constructor()

Ginger

who is this Colonel Headers guy and why does linux need him to install my drivers

@lyxal I'm so proud :p

@emanresuA glorious

user

@Seggan you might want to check out nim

It kinda does what you’re thinking of

Also lets you choose between static or dynamic dispatch

The dynamic dispatch might not be as efficient or whatever as the way Java does it because it’s not a method associated with the object, but it’s very cool. It can sorta be overloaded for any of the arguments, not just the first one

@emanresuA the new and constructor together seem redundant

Unless you were joking

@Seggan new or init imo. Definitely not the class’s name, that’s a bit annoying in java. constructor is a bit long

@Seggan Rather than just self.super(), could you also allow treating it as any one of its supertypes? So if A extends F and there were a doStuff for F, then you could call that on b

Or just do what Nim does ig

Static dispatch

Seggan

02:00

@user how would you specify to recurse instead of calling super?

user

Just self.doStuff() right?

Seggan

yes, but not with your suggestion

emanresu A

@user How dare you accuse me of being Jo King. I'm totally not a sock.

user

@Seggan wdym

Seggan

@user ehhh i dont really like it

methods are not really separate functions then

@user how would you recusrse? how would you call super?

user

02:04

This specific ufcs is very similar to nim’s though

Seggan

in mine, for example, itd be self.doStuff(), and self.super().doStuff()

user

@Seggan recurse by calling self.doStuff(), super by calling self.super.doStuff() or self.super[A].doStuff()

Yeah same

I don’t see how my suggestion changes anything important

emanresu A

Require taking the this object in method, like python, but require it be called this

user

Anyway it’s not that important

I know Scala has that feature but I’ve never used it

I’ve only seen it used like twice

@emanresuA btw nim can overload based on any of the arguments, not just the first one. That can be useful if you’re going to separate methods from classes anyway

Seggan

@emanresuA no

emanresu A

02:09

How about either this.self or self.this, but not both

Seggan

why not just name it yourself

im trying to keep functions as functiony as possibly

methods do not exist, not even temporarily in the compiler

emanresu A

__self__::_getThis_(new ThisGetter::constructor())::get_tHis(__self__)

Ginger

AAAAAAAAAAAAAAAAAA

Seggan

are you messing with me

emanresu A

Nov 4, 2022 at 17:18, by mousetail

You have to understand when you ask me "What is the best X" I have a contractual obligation to suggest the worst possible option

Seggan

02:16

mmm

lyxal

LDQ: Thoughts on how this class syntax looks?

It's a translation of the example Seggan provided into a syntax style I'm experimenting with

Another example

Seggan

does gets my approval

lyxal

btw single inheritance only and whitespace is insignificant

Seggan

exactly

lyxal

Gosh dang

user

02:29

@lyxal not a huge fan of is. Maybe extends or a symbol like <?

Also, I think the does can be dropped

If it can’t, i like it. Shorter than where too

Rydwolf Programs

@Seggan This() or Self() (whichever you use) could be an option

Rust makes Self a shorthand to the name of the class

So there's precedent

user

@lyxal mimics implies it’s a proxy pr a mock object imo

lyxal

@user so what word should be used instead?

Rydwolf Programs

Also if you're making a classic OOP language, please for the love of your mom's grave, do not have a new keyword. Zero ambiguity in just using ClassName(...) as the constructor, instead of new ClassName()

user

@lyxal honestly, idk. I love implements but it’s long

I’d just use the same thing for both the parent class and any interfaces

Rydwolf Programs

02:35

new ...() has massively awkward precedence, especially if you allow new Thing without parens, and it's just a hugely outdated thing carried over from when it was necessary to state that you're dynamically allocating memory (and nowadays it's not like most constructors will actually do that anyway if the compiler's worth half a mom's grave)

user

On second thought is isn’t that bad

Rydwolf Programs

It's also short

user

@RydwolfPrograms why the obsession with mom’s graves?

Rydwolf Programs

And you can add an isn't keyword, separating certain classes from the rest, until tired of being oppressed, they rise against the bourgeoisie

user

Nonono

Oh wait it’s a joke

Didn’t read past isn’t

Rydwolf Programs

02:37

@user No I am serious

lyxal

@RydwolfPrograms no, you're Rydwolf

Rydwolf Programs

Class reductionism: the forbidden eighth principle of OOP

user

@lyxal instead of foo =bar which looks really confusing to me, how about foo -> bar?

If you really want to flip assignments

I really like has and the end markers

Itd be nice if you were allowed to do end Classname for long classes

Is this a stack based language?

emanresu A

I feel like most modern languages are missing an assignment assignment operator: ==. a == b should become a = (a = b), and a === b should become a = (a = (a = b)).

user

Wait is this vyxal?

lyxal

02:40

@user yes

Rydwolf Programs

@emanresuA Or a = (a == b)

user

@lyxal I’m so dumb lol

Okay the =foo stuff makes sense now

Rydwolf Programs

@lyxal I better not sit down for a job interview in twenty years and be asked to write Vyxal FizzBuzz on a whiteboard :p

Seggan

@RydwolfPrograms possible function call?

user

@RydwolfPrograms that’s the top secret ==== operator

Rydwolf Programs

02:42

@Seggan Functions with uppercase names?

Seggan

true

Rydwolf Programs

@user I thought that checked if the left hand side equaled the == function

user

Oct 23, 2022 at 16:18, by Razetime

what on earth does ==== do in scala?

Rydwolf Programs

x == (==)

Seggan

what do you think of my polymorphism thing?

user

02:42

We never found out what it did

lyxal

@user so symbolically, it'll just be |s, but literate mode has a chance to describe what's happening in each branch

hence why i'm giving examples in literate

user

@Seggan it’s very cool and I think nim’s way is even cooler and itd be really cool if rol borrowed from that

Seggan

nims is impure in my opinion

i dont wanna mix weird semantics

(btw i was talking to @RydwolfPrograms)

wait since when did you change your name

Rydwolf Programs

Wdym my name has always been redwolf

user

y

emanresu A

02:48

Are you going to change your website yet again?

Rydwolf Programs

Maybe

I'm undecided on my github

Breaking every link posted twice might be pushing my luck

Seggan

@RydwolfPrograms but youre rydwolf now

Rydwolf Programs

Yeah redwolf

Same as walways

emanresu A

huh?

user

Are you trying to gaslight multiple people at once?

lyxal

02:50

gaslight? what do y'all mean? the username has been the same as it's always been

emanresu A

03:01

Actually, it's called "gaslamping"

Rydwolf Programs

Yeah, you're probably just imagining things

It's always been "gaslamping"

Rydwolf Programs

03:25

Fun thing I did: Using ssh-agent, I set up a thing that'll prompt me for my SSH key password the first time I use it, then won't require it again until I log out

Did it by 1. making ssh_agent persist across different terminals, and 2. aliasing ssh to a shell script that runs it if I haven't provided it my key password yet

(I also have the alias check if I'm in school wifi and automatically proxy the SSH connection if I am)

emanresu A

03:42

For my peace of mind, I wrote a userscript to disable chat flags. I can't pass judgement on stuff like this, handle angry math people, or speak Spanish.

emanresu A

03:55

If anyone wants it

l4m2

04:05

Some golfing questions have bounty for e.g. solving in faster way to push solved range. What if I don't know many languages and obvious optimal solutions are all submitted? Still post an exactly same code and start solving bounty?

lyxal

hey so y'all know that github stats thing some of us have on our profiles? Well the link it uses is a little broken right now

meaning it won't display properly

but there's a fix: s/https://github-readme-stats.vercel.app/api?/https://github-readme-stats-git-masterrstaa-rickstaa.vercel.app/api?

Rydwolf Programs

@emanresuA 90% of the time they're just "a mod should freeze this discussion but nothing worth a 30 minute ban has been said yet", so probably a good call

emanresu A

I ignore them 90% of the time

The chat flag system... needs work

lyxal

and the earth orbits the sun :p

Rydwolf Programs

Eh, I don't think it could be much better

It just doesn't do much or need to do much

lyxal

04:17

~~The~~ chat ~~flag system...~~ needs work

UndoneStudios

06:38

I just realized today is a Friday the 13th

Adám

07:04

@UndoneStudios Indeed. My daughter's birthday.

mousetail

Congrats

Adám

Thanks. She's turning 0.

mousetail

Oh that's a very special birthday

hope it goes well

UndoneStudios

@Adám So all this while she's -1?

Adám

No, she's 0 now. She was -1 right before.

UndoneStudios

07:18

Basically, she's born today?

Adám

Yes, 3.5h ago.

Unrelated String

congrats

UndoneStudios

congrats

Adám

Thanks, all. A new potential code golfer. She's certainly short.

17

lyxal

adams_family ←adams_family , ⎕NEW daughter ⍝ gg man, hope she does well at apl :p

I don't care if that type errors, it is what it is :p

Adám

07:30

That should work fine, but you can golf it to adams_family,←⎕NEW daughter

lyxal

So, are we going to have to golf you baby supplies like we did for Dennis that one time? :p

Codidact Posts

08:31

New Posts - Challenges - Code Golf: K-Partition Sort

posted on January 13, 2023 by yjj‭

We all know and love the quick sort function, no? However, the usual partition functions (such as dutch national flag, five-way) are limited to a number of partitioning divisions. Your t...

Sandbox Posts

08:44

0

A: Sandbox for Proposed Challenges

Toggle some bits and get an actual square Tags: code-golfbinary-matrixnumbergrid Inspired by the title of the Toggle some bits and get a square challenge. In that challenge you output how many bits should be toggled, in order for the base-10 representation of the binary to become a square number....

PyGamer0

@Adám congrats

Adám

thx

PyGamer0

@lyxal tell me more :P

emanresu A

3

A: Polyglot Quiz (Cops' thread)

??? and ???, 54 bytes String: 8356830650 Works in A and B (hex): 00000000: 300b 3366 b87e 0b00 247e e239 2a83 2a3b 0.3f.~..$~.9*.*; 00000010: 0fd6 30 ..0 Works in A (hex): 00000000: a4ac b3bb 762b 3a74 d6 ....v+:t. Works in B (hex): 00000000: ...

I don't remember what this was, hopefully i have the files somewhere

mousetail

09:00

lol

Sandbox Posts

09:44

0

A: Sandbox for Proposed Challenges

Transpose binary numbers code-golf array binary The input numbers [34, 7, 109] can be represented in binary as 0100010 0000111 1101101 (one number on each line; and padding with zeroes on the left so that all the rows are the same length) Now transpose this array: 001 101 000 001 011 110 011 Th...

pxeger

10:09

The Low Quality Posts queue is truly useless

codegolf.stackexchange.com/review/low-quality-posts/66963 should be deleted, but none of the options when you choose to delete are really correct

and it doesn't even let you flag it from in the queue, I had to click and open it in another tab to flag it as not an answer

mousetail

It's a commentary on another post right?

lyxal

looks to be

pxeger

yeah

but the remarks in the first sentence are not "critiquing or requesting clarification from the author"

but it should still be converted to a comment

lyxal

"Thank you for an interesting challenge! Little later I'd like post here full answer on Mathematica, may be not for real challenge, just for survey."

It's not a comment

it's an half answer

mousetail

> So the question arose: should tangent circles be considered an overlap?

That seems like a request for clarification

lyxal

10:13

That's not a clarification. That looks like monologuing

"the question"

pxeger

yeah

mousetail

The question is: Should the output include the point (6,-6) and the rest of the post is noise

lyxal

lesobrod is a user who's interested in doing things for academic research purposes, and this reads like one such research thing

@mousetail could just be a commentary on their observations on what other answers do

mousetail

Possible, but to me (lacking context on the question) it seems like a request for clarification. The code demonstrates a edge case. Then they ask how the edge case should be handled

lyxal

If it was a clarification, why have a h1 header with the language name?

pxeger

10:22

it's that user's thoughts so far towards an answer

lyxal

plus there could be a little bit of a language barrier here. What lesobrod might have intended as a general observation answer might come across as a clarification request. I've noticed that english may not be their first language (which is okay, just an observation from prior interactions)

@pxeger exactly

mousetail

Or what you view as a general observation may be a clarification request. We know 100% sure there is a clarification request since it was said explicitly. There may also be a attempt at an answer

lyxal

I'm more saying from my experience with this particular user, it's most likely observations.

mousetail

But the code doesn't seem do do anything except point out the edge case, it's clearly related to the clarification request

lyxal

If it was a clarification request, I think it'd be left in the comments. This user isn't unfamiliar with the concept of comments.

honestly, it feels like there's a paragraph or two missing at the end

feels like there should be a "but here's what I concluded from my experiments with mathematica" kind of sentence

mousetail

10:25

If they knew comments they would have posted as a comment, or they wanted a bit more space to ask their question

pxeger

it really doesn't matter

just flag it as not an answer (or VTD if you have 25k rep)

lyxal

there is a distinct lack of delete button here

mousetail

Anyways, on another topic: **LDQ** So my language is based on polish notation. That means it has a "end". I'm wondering how to treat code that comes after the end. I see 3 options:
1. Discard it, good for polyglotting I guess but otherwise a waste
2. Treat it as a separate program, with implicit output it would simply print both results in a row. Could be helpful if a challenge asks for 2 or more unrelated things
3. Pipe them. Automatically create a variable for the result of the first program that the second can use. Saves bytes on variable declarations

pxeger

oh you can only delete answers with a score of < 0

lyxal

@mousetail pipe!

mousetail

10:32

Oh and option 4: Treat it like a string literal, to allow a convenient way to embed a single string without any escaping

Unrelated String

oh yeah pipe seems interesting

pxeger

presumably this is a golfing language not a practical one

mousetail

Yes

BTW there is already a builtin diad that does the piping thing at the cost of 1 byte

Seems like people mostly prefer piping so I'll probably do that

mousetail

11:44

Follow up question: If a program is shorter than it needs to be, I intend to insert implicit input in the first argument spot of the incomplete operator. However, if there are multiple piped programs and the last is incomplete, do I insert implicit input there or implicit reference to the output of the previous?

lyxal

so you want to have the last pipe implicitly use input and last pipe value?

or just one of the two?

mousetail

I can only add implicit things of any kind to the last chain, or it could increase the number of chains. So I can't do implicit input "left-first" as I'd like to. The question is what implicit thing to add if there are holes in the last chain

lyxal

first chain gets input, every chain thereafter gets previous chain's value

because realistically, if you're using more than one chain, you're doing it on purpose.

because if just discarding these chains was a viable option, it's clear that they weren't originally intended to be primary features

mousetail

Yea it's a golfing language, I'm assuming everything was done intentionally

lyxal

of course, I was just giving a bit of justification to my idea

mousetail

11:57

Ok yea, for every chain but the last input needs to be explicit. So in the case there is one chain I'll implicitly insert inputs. Otherwise I'll insert variable references

lyxal

sgtm

Ginger

13:38

I wish more languages had Python-like context managers

they're so helpful!

mousetail

In which language are you missing them?

lyxal

@Ginger scala. It has something similar to those but actually cooler

Ginger

@lyxal wat

lyxal

They're called implicits

And they can vary from automatic type conversions to just straight up filling in parameters

Ginger

wow

lyxal

13:41

And keeping track of something you want to pass between everything

New Posts

0

Q: By using BFS solve the following question in any programming language

We have an NxM grid, grid have one element named Bob. Bob can travel diagonally blocks only. The grid has some blocked blocks on which Bob can not travel. Write a function that returns on how many possible positions Bob can move. Solve this problem using BFS and submit the executable code in any ...

lyxal

docs.scala-lang.org/tour/implicit-parameters.html

@NewPosts no

Ginger

kachow

lyxal

@NewPosts that is not a question that is enjoyable here

mousetail

That's a cool feature but not really very much like a context manager

Ginger

13:44

^

lyxal

Wdym? I use it to manage context objects all the time! :p

mousetail

JS has a with statement but it's the absolute worst

New Posts

13:57

0

Q: A generated proof for g^(-1)kg ∈ K

Problem. Given two subgroups H, K of G so that for all g ∈ G, k ∈ K: g^(-1)kg ∈ K. I want to prove h_1 k_1 k_2^(-1) h_2^(-1) ∈ { hk | h ∈ H, k ∈ K } (no matter if we replace the instead kh). My wish is to do the same thing (a generated proof) to help with the problem for $$g^{-1}kg\in K\Rightarro...

code-golf math atomic-code-golf abstract-algebra proof-golf

New Posts

14:17

0

Q: Implement a bag without replacement

Intro The Tetris Guidelines specify what RNG is needed for the piece selection to be called a Tetris game, called the Random Generator. Yes, that's the actual name ("Random Generator"). In essence, it acts like a bag without replacement: You can draw pieces out of the bag, but you cannot draw the...

code-golf random data-structures

Ginger

lotta new questions today!

mousetail

Not really, 3 is normal, a lot of questions in 1 hour though

Ginger

that's what I meant lol

l4m2

14:40

"
lotta new questions tohour!

Ginger

huh?

mousetail

Not today but tohour

Rydwolf Programs

Updated UI for KotH controller:

mousetail

Are you using a CSS library?

Rydwolf Programs

no

mousetail

14:48

Ok looks quite slick

Ginger

wow

Rydwolf Programs

Got rid of the ugly asymmetry above the button

(the gap above it is slightly bigger)

mousetail

What language is the KOTH going to be in?

Rydwolf Programs

JS, so it can be run online

No RTO KotH tools yet :p

mousetail

Going to run it automatically?

Rydwolf Programs

14:53

Maybe, wasn't planning on it, but I could I guess

Rydwolf Programs

15:04

wha--rto.run's DNS records indicate I'm hosting it from my home IP address, but both servers there have been down for a month...

and it's been working during that time

Huh, apparently it's running from rust_wolf, my older server. Guess I have ghosts.

Ginger

spoopy

Seggan

so ive got a problem

Ginger

don't we all

Seggan

i realized that the scheme i described yesterday for rol is statically dispatched

so ig im adding full blown methods now

@mousetail piping is very similar to the way fig handles multiple lines

user

@lyxal Ackchually, that's not what implicits are generally used for

Seggan

15:17

@Adám congrats

user

I think the only reason Python has a special syntax for context managers is because it doesn't have multiline lambdas. In Scala, which does have those, you'd use something like Using

Seggan

@Ginger kotlin go brr (the with function)

user

Ackchually, that's also a different (and much cooler) feature. Kotlin, like Scala and other languages with multiline lambdas, can do context managers without with or other special syntax

Seggan

are you talking about lambda receivers?

user

No

Just normal lambdas are enough

Seggan

15:20

example?

user

kotlinlang.org/api/latest/jvm/stdlib/kotlin.io/use.html

Seggan

ofc i forget that one

user

In Java parlance, I think what Ginger was talking about is just try-with-resources

Seggan

ooooh

user

Which doesn't require anything special in languages that aren't Python (even Java doesn't need a special construct for auto-closing, even though it does)

Seggan

15:22

yes, kotlin lambdas are the best.

user

Gotta agree

Lambda receivers are really nice

Seggan

^

i love how you dont need parens around lambdas so it looks like cool new syntax

user

I hope contracts get good someday

Seggan

tbh i dont see their value

user

@Seggan And if the last argument's a function, you can just do {}

Seggan

15:24

yeahs thats what i meant

@user unless youll get smth like contract { param > 0 } then id see their point

user

Ah I thought you meant parens around the parameters

@Seggan Well also nulls

Seggan

thats handled by the type system

user

I think there's a limit to how much the current type system can do

Also, one cool thing is ensuring how many times a lambda will be called

Seggan

i dont really see the point of that

user

I think you might then be able to use lambdas for initializing final stuff

Actually, maybe not

Seggan

15:27

@user ah i see what u mean

returns() implies (param != null)

user

Yeah you can augment flow typing

Not just param != null but also param is List<String> or something

Seggan

true

user

If contracts got really good, this question would have a nicer answer

Seggan

o just read up on that, smart casting with contracts is real nice

user

My answer actually used a bug in contracts to get the desired behavior lmao

Hmm, it's been two years, it might be a little closer to doable now

Seggan

15:32

looks interesting

i still prefer prop = value and runtime exceptions tho

user

Yeah, you just can't put contracts into getters and setters :(

Seggan

hmm seems it works with the IR compiler

(i think its K2?)

seems the issue is on the back burner tho

contracts on getters/setts would be nice

user

As long as they also allow contracts on the getters and setters inside delegated properties

But that might be considered too complex

Seggan

oooh

user

@Seggan So the Kotlin 1.3 release notes have this example:

fun synchronize(lock: Any?, block: () -> Unit) {
    // It tells the compiler:
    // "This function will invoke 'block' here and now, and exactly one time"
    contract { callsInPlace(block, EXACTLY_ONCE) }
}

fun foo() {
    val x: Int
    synchronize(lock) {
        x = 42 // Compiler knows that lambda passed to 'synchronize' is called
               // exactly once, so no reassignment is reported
    }
    println(x) // Compiler knows that lambda will be definitely called, performing
               // initialization, so 'x' is considered to be initialized here

Seggan

15:40

hmm interesting

Rydwolf Programs

18:34

Okay, UI/UX question for my KotH

There will be a way to include multiple copies of the same bot

1. Should these all be in their own rows, as if they were entirely individual?
2. or should these be grouped together, and the list can be expanded out to show the individuals?

And then, what should the buttons look like to add/remove bots? Right now I have #1 with a ×2 button (duplicates the bot) and a - button (deletes it)

Ginger

18:59

I have discovered the power of Python generator expressions

mathcat

They are very OP indeed.

Rydwolf Programs

Hmm, interesting readability issue

What would y'all do if you had a list of indices you needed to iterate through, including their index in the list?

mathcat

enumerate I guess?

Rydwolf Programs

Yeah but for the variable names

i and index...which index is which? :p

mathcat

ah lol

I'd do ind, item then.

Although, that's even more confusing.

Ginger

19:18

@RydwolfPrograms index and indexIndex :p

user

@RydwolfPrograms Leave a comment explaining what each of those indices is

Alternatively, try a completely different approach so I don't have to deal with indices

Might be an X/Y problem. What are you trying to do?

Rydwolf Programs

@user Merge some iterators into a single iterator containing a tuple with every possible combination of their outputs

user

Ooh

Like a cartesian product?

Rydwolf Programs

Yeah

But with possibly infinite iterators

user

19:34

I looked at how Vyxal does it but it looks like we never bothered handling infinite lists, Vyxal just uses itertools.product :|

Ginger

lol

Rydwolf Programs

The way I'm currently doing it is recursive cantor pairing of the indices

That heavily biases earlier items tho

user

That's okay, though, right?

Rydwolf Programs

Later on we'll be able to use a weight representing how far into the list the answer is expected to be to construct the optimal recursive structure to reach a solution fastest

user

Oh, is this for Complement?

Rydwolf Programs

19:37

@user In theory, but in practice you don't want to be checking ninety thousand values of x to get to values of y and z that're in the low hundreds

@user Yeah

Simd

19:51

My challenge is still resisting the first C answer!

Seggan

which one?

Simd

@Seggan. codegolf.stackexchange.com/questions/256441/…

Seggan

ty

Simd

5 bytes of Jelly equals too hard in C :)

As does 52 bytes in python

user

@Simd There just aren't as many C golfers here as there are Python and Jelly golfers. Nothing wrong with not getting C answers

Seggan

19:58

im thinking of reduction by subtraction of the char codes and reverse sorting

(not saying ima do it)

Simd

@user this is sad. Well its not sad that there aren't as many but it's sad there aren't a lot of both

@Seggan do it! :)

Seggan

doing school rn

ive got appointments in the evening too

Simd

@Seggan which subject?

Seggan

american history

Simd

Are you a student or teacher?

Or both :)

Seggan

20:09

student

@Simd its been like that since copilot

Simd

20:36

@Seggan really??

Ginger

21:20

ffs I spent half an hour trying to figure out why my code was broken only to realize I forgor 💀 an underscore in a method name

Rydwolf Programs

21:33

Sounds like a failure on the language's behalf

Ginger

you misunderstand: I made both of the methods in question

I was using downloadUrlToPath when I should've used _downloadUrlToPath

more accurately, the bug involved downloadUrlToPath calling itself instead of _downloadUrlToPath

but now I've got an exciting new problem:

Downloading packages...
Downloaded typing_extensions-4.4.0-py3-none-any.whl
Downloaded ujson-5.7.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Downloaded virtualenv-20.17.1-py3-none-any.whl

Downloaded yapf-0.32.0-py2.py3-none-any.whl
Downloaded yapf-0.32.0-py2.py3-none-any.whl
Downloaded yapf-0.32.0-py2.py3-none-any.whl
Downloaded yapf-0.32.0-py2.py3-none-any.whl
Downloaded yapf-0.32.0-py2.py3-none-any.whl
Downloaded yapf-0.32.0-py2.py3-none-any.whl
Downloaded yapf-0.32.0-py2.py3-none-any.whl

see the issue here?

it's not supposed to download yapf 37 times

Transcript for