that gets automatically raised by my WS library if recv() doesn't get any data from the server after, in this case, a 2-second delay

see this

lol

@DLosc FunStack 0.0.1, builtin (and the motivation for the CMC): ToBase -1

Without the builtin, 47 bytes: From0 2 TakeCycle Neg Less? 0 Dec Abs if Double

Why is futures::select! so complicated

I just want to do Promises.any in Rust

Why does it need that complicated macro syntax and how does it even work

Okay figured it out

_ = thing => () seems a bit of a waste of space but whatever

@RadvylfPrograms Got to appreciate those programming languages that allow dummy variable names like the underscore. Looking at you, C and C++, among others. Not like those are disallowed, but technically implementation-reserved. Nice.

Still going to regard C and C++ in bad light and throw them the evil eye. Legend has it that Dennis Ritchie is secretly the one who had actually not only invented, but full-blown patented the buffer overrun error. Which means that authors of any program that happened to suffer from this bug and was abused by bad actors to steal sensitive data not only had to manage the aftermath damage, but also had to pay D. Ritchie a hefty fee for featuring his patented invention in their software.

@RadvylfPrograms Imports are a paiin, and standards are outdated

And toString doesn't always work

@mathcat FunStack 0.0.1, 56 bytes: Flatten Concat table over foldl Wrap "" Repeat flip Wrap

Or Flatten Concat table over foldl1 Repeat flip Wrap if you don't mind slightly odd output if the number is 1

wow my Rol's error messages are already reporting actual errors (not test errors): Type mismatch error at row 12, column 4, statement 'return luaRead("*l")': Expected String, got dyn?

fortunately all i have to do is stick a ! on the end. thatll be the non-null assertion dyn? -> dyn, and then dyn gets autocasted to String

anyone know how to convert a string to regex in pip, i try X but its escaping special chars

and i dont want it to do that

@AidenChow Concatenate it with an empty regex: ``."a*" -> `a*`

ah ok

@AidenChow I started to leave a comment on your Triangular polkadot numbers answer, but I think it'll be better here:

1) 0.5 can be /2

oh nice!

didnt realize unary / is reciprocal

I think it's a very nice feature for a language to have. Not sure why more languages don't.

2) Often, you can use D to subtract 1 and U to add 1 (if they're applied to a variable, they modify it, but if they're applied to an expression like DBa they just affect the value)

dlosc u should really continue making ur tutorial lol, im having some trouble with learning regexes

like regexes in pip, not regexes in general lol

Yeah, I know. :P What in particular?

i just need to see some example programs using regex so i can see how they are used, kinda confused lol

Well, here's a list for starters. Most of those should have some kind of explanation, but let me see if there are some good simple ones.

also could u check my court transcriber answer while u r at it, wanna see if im using regexes in a golfy way lol: codegolf.stackexchange.com/questions/252927/…

@DLosc u have to put ur own id there lol, cuz it put my user id when i opened the link

@AidenChow Whoops, lol. Fixed

Got that edit in right at the deadline

Here is a basic example of regex replacement with a callback function

Here is an example of the first-match operator ~

Here is an example of the all-matches operator @

@AidenChow You can just join on `.*`: most string operations, if you give them a pattern (regex), cast the result to a pattern

In this case, it probably doesn't matter because it's words not numbers, but in general you want to use N instead of ~ to test existence/non-existence of a match. That's because ~ will return the match itself, which could be falsey if it's entirely composed of 0s. N will return the number of matches.

@DLosc Oh, shoot. Actually, that won't work so well here because J has lower precedence than ~. :(

Still, (aJ`.*`) is one byte shorter than ``.aJ".*"

@DLosc ah okok

... Now I'm honestly puzzled how your original code worked. Lol. Maybe I'm missing something here.

@DLosc :\\\\

When something weird is happening, the -d flag (for "debug") gives you some extra information that can be helpful.

KXX is a byte shorter but doesn't fit in

Okay, this is a really weird one. Here's what's happening:

You're filtering on _~``.aJ".*"

Due to operator precedence, that gets parsed as (_~(``.a))J".*"

How does that work lol

Also maybe we need a "regex join by .*" operator?

Which means 1) convert every string in a to a regex, 2) find the first match of each of them in the function argument, and 3) join that list of matches on ".*"

Thing is, if any of the matches are nil, the whole thing is nil. So what this amounts to is "filter for words that contain all of the strings in a." What it doesn't do is check that they're in the right order or that they don't overlap.

For example

The way I debugged this was to remove @ (so as to get the full list) and then change FI to M so I could see exactly what the function was returning. Then that still didn't explain much, so I took out the J".*" too, and all became clear.

Bruh so basically my code doesn’t work

Yeah. But the good news is, you can make it work and golf it by a byte!

Lol

@DLosc I'm sorry, make working code? I object to that on moral grounds :P

@emanresuA genuinely i feel like it comes up so much lol

Tbf I would’ve never found that error on my own

like... i can even imagine it being useful in non cg code

@cairdcoinheringaahing Ah, sorry. Here, have some non-working code: #^&*&)(&%$$

Ah, perfect :P

Like I just ran the code for like 2 cases that I made up and it printed what I expected, so I’m like yep it looks good

Tho, I think that does work in Jelly :P

@AidenChow Yeah, the question does need more test cases. cc: @Ginger

@cairdcoinheringaahing You expect me to come up with a polyglot that doesn't work in every language? smh

@DLosc Obviously :P

(Could be an interesting challenge, but I'm not sure exactly how to score it)

CMC: Create a piece of code that throws an error in as many TIO languages as you can (according to the versions currently on TIO). Highest percentage (erroring langs)/(total TIO langs) wins. An error means any abnormal information being output to STDERR for the purpose of the CMC

@cairdcoinheringaahing ooh sounds interesting

+]

Errors in a lot of stuff...

don’t wanna test some code for like all the langs tho…

Anything infix will barf on the +, most other things will barf on the ]

Works in Jelly :P

It doesn't barf on the unmatched ]? oh well

@cairdcoinheringaahing Does an infinite loop count as an error? It does output something abnormal to stderr, but only because of the 60-second time limit.

don't know why i didn't expect unbalanced [] to be like any other unparseable tokens :P

I'd say no

@emanresuA looks like it's the same as q

i.e. it does barf, silently

+]# errors, and hoe=pefully doesn't fix other stuff

@RadvylfPrograms because a) less errors than a custom parser, b) less code than normal parser, c) Scala seems well suited to it, as it's a core library, d) I don't know much about Scala so the simpler the better and e) it's way easier to document - ebnf rules compared to having to explain a whole parser

@cairdcoinheringaahing Huh, actually it doesn't output anything abnormal to stderr. So I'd agree.

How about (+"'

Promising

Vyxal's almost-everything-proof but not on TIO

+] is a valid train in J btw

Everything suggested so far works in Befunge-98 (and probably other versions)

okay, time to add nonascii into the mix

Also in Del|m|t

+] works in bitcycle too :p

Also in Brian and Chuck

Bruh r u just going down the list and running it or what

Just cherry-picking. But the list is gonna be pretty long, so I'll stop. :P

Found a language where +] errors

Try it online!

As I've said in the past, I'm really good at language design

Nothing can error in e.g. ///

Oh bruh actually

What type of lang is that anyway

Substitution-based tarpit

@DLosc trying to use N but it’s not working? dso.surge.sh/…

Or am I using it wrong

Idk what's up with CLI args

@cairdcoinheringaahing simple just ACE TIO

@thejonymyster yep just ace TIO real quick, no problem whatsoever, rather simple, really

iirc you can do some funny stuff with bash actually but idr to what extent

Obviously TIO is ACE, pretty sure online interpreters don't experience sexual attraction, but okay

@thejonymyster Where's Aaron when you need him :p

get up for golf youre gonna be late

@DLosc You rang?

@lyxal dK?

dK

@emanresuA huh that actually works by the look of it

The issue is it's uniquify

he's the leader of the bunch

You know him well

Unless that's meant to happen?

he's finally back

to kick some tail

his coconut gun

i like this version better

if he shoots ya

@thejonymyster lmaooooo

it's gonna hurt

*grunt*

@Ginger add more test cases to ur court transcriber challenge

so you want me to make it bigger! faster! and stronger too!

make it harder better faster stronger

that is not the line

it is a thematically adjacent line

but it is not the first member of the DK crew

Huh!

there seem to be at least three remixes of harder better faster stronger with the dk rap

you can build the lego dk shack

someone do the next line, I can't do two consecutive ones

man yall can't sing to save your lives :b