Vyxal

Aaroneous Miller

01:36

@lyxal btw, did I get the intended solution?

lyxal

No

I'm still working on the next one

Which is taking a bit longer than I thought because I'm reworking the solution I had planned

lyxal

12:11

@AaroneousMiller @emanresuA codegolf.stackexchange.com/a/236107/78850

@AaroneousMiller I don't like how accurate this is

like seriously that's how I managed to remove r from my intended solution, by working on it really late and while feeling almost ready to sleep

Aaroneous Miller

12:39

@lyxal challenge accepted

pVC aecidiospore adduced

13:05

@AaroneousMiller sponsored by hot wheels - © mattel

Aaroneous Miller

lol

I don't understand v<nilad> at all. :/

!!/run 12345 v0,,,

Vyxal Bot

[@AaroneousMiller: 59315237]
⟨0⟩
⟨0⟩
12345

Aaroneous Miller

!!/run 12345 v0 ,,,

Vyxal Bot

@AaroneousMiller (output was empty)

Aaroneous Miller

!!/run 12345 v0W,,,

Vyxal Bot

13:15

[@AaroneousMiller: 59315249]
⟨0⟩
12345
0

Aaroneous Miller

!!/run 12345 vuW,,,

Vyxal Bot

[@AaroneousMiller: 59315257]
⟨-1⟩
0
0

Aaroneous Miller

!!/run 12345 v.

Vyxal Bot

@AaroneousMiller 0.5

Aaroneous Miller

!!/run 12345 v.,

Vyxal Bot

13:17

@AaroneousMiller ⟨0|`.`|5⟩

Aaroneous Miller

!!/run 12345 v.,,,

Vyxal Bot

[@AaroneousMiller: 59315273]
⟨0|`.`|5⟩
⟨0|`.`|5⟩
12345

Aaroneous Miller

._. wut

pVC aecidiospore adduced

so vyxal has failed the intuitive part
:P

Aaroneous Miller

always and forever

pVC aecidiospore adduced

13:20

¿

Aaroneous Miller

yep

Aaroneous Miller

13:50

I managed to get i by itself, so now I can index into stuff

If I can just get arbitrary numbers, I can grab C from one of the available builtins, and I'll have everything I need to crack it

Aaroneous Miller

14:12

lol I managed to get arbitrary negative numbers, which lets me index into kP with the aformentioned solitary i, which will let me grab basic arithmetic commands, which will allow me to increment, although it will be ridiculously expensive to do so

that moment when it takes 214 bytes to add 1 to a number

don't worry, I was able to golf it down to 208 bytes

wait no. it's actually 212 bytes

sorry, 210.

The incrementing is really expensive, but this time I don't have to increment literally a million times just to get a binary representation, so printing ! only takes 7235 bytes

all i have left to do is write up a program to calculate the total length and make the post

Aaroneous Miller

15:10

nice. only took me 2 hours:

0

A: Print X without X (robber's thread)

Vyxal, 29401143 bytes, lyxal get cracked on The backbone of this solution is the generation of arbitrary negative numbers, allowing for indexing to get all the commands we need. How negative numbers? u # Push -1 .o # Remove `0.5` Does nothing except convert to string Ė # Execute We c...

Aaroneous Miller

15:50

I'm still waiting for Vyxal DO, 255 bytes

Aaroneous Miller

17:08

I did the math, and if I were to increment with u- instead of 1+, the full crack would be only 19,651,753 bytes instead of 29,401,143 bytes.

emanresu A

19:27

@AaroneousMiller Your (ab)use of Ė is really nice.

Aaroneous Miller

Yeah, I found it kind of useful here and there :p

Aaroneous Miller

19:51

CMQ: String interpolation? e.g. JS's `${String} Interpolation`

2

Personally, I think from the perspective of taking elements from praclangs, it would make sense, and could be useful, though I don't think it would be too useful from a golflang perspective.

Do we still need the Scala CMP pinned, or have we come to a decision on that?

emanresu A

@AaroneousMiller osabie has it

@AaroneousMiller %

And I think there was a proposal for a superformat that pops items from the stack and replaces % until all have been replaced, somewhere in #160 although I can't find it

Aaroneous Miller

I know that there's the % command, but I'm thinking more like how you can insert a expression into a template literal directly like that JS example.

@emanresuA because of course

6/9/420

emanresu A

Lol

Aaroneous Miller

20:09

Also, can we special-case number conversion to base 1? At the moment it just errors, but it could be useful for making 0-filled lists of specified lengths. Right now, the golfiest way to do that is ƛ0;, but we would be able to do 1τ instead.

emanresu A

v0 in the rewrite?

That's what's intended at least

Aaroneous Miller

Yeah, that would do pretty much the same thing, and makes a lot of sense.

The base 1 thing could still be useful e.g. for base conversion challenges or w/e so we don't have to special-case it in a submission.

I guess I could just add those to #160 and then we can think about that when the time comes.

It looks like there's already a similar idea with doing Þ0, Þ1 ... Þ9, but having it be v instead would make more sense, plus then we could use other nilads like u or ×.

emanresu A

Ye

I didn't think of that when I wrote that request.

Aaroneous Miller

oh I didn't realize that was you lol

emanresu A

Why did you change to "Aaroneous"?

Not complaining, it's cool

Aaroneous Miller

20:19

in The Nineteenth Byte, yesterday, by lyxal

@Aaroneous Miller don't you know that the floor is lava

emanresu A

Gpt-3 rules the world.

Aaroneous Miller

I couldn't have GPT be wrong, now could I? ;P

lyxal

22:37

@AaroneousMiller @emanresuA codegolf.stackexchange.com/a/236125/78850

lyxal

22:50

in The Nineteenth Byte, 56 secs ago, by Catija

@Adám I do think that it does generally make sense when it actually gets explained... but for some reason we're just too tight lipped about some of this stuff... probably because it would make us seem lower value to potential investors (?) but I wonder whether that can change now that Prosus bought us.

> Prosus

That's a bit sussy

!!/sus

Vyxal Bot

@lyxal ඞ

user

@AaroneousMiller Would be good and help shorten code

But expressions like that don't work too well in a stack-based language, do they?

Actually, that could work, nvm

hyper-neutrino

the only thing i have against it is that it might not really save all that much because in normal languages concatenating or inserting things into strings can take a lot of overhead but in vyxal it's probably like 1 byte to just merge strings anyway

but due to stack structure it might take a bunch of bytes to manipulate item ordering so

i feel like it could work

as long as it doesn't interfere with strings that don't want interpolation (please don't make a flag that disables interpolation) it should not hurt byte count

lyxal

@hyper-neutrino meaning it'd have to be something guaranteed to not be used very often

user

I think a flag to enable interpolation would border on cheaty but wouldn't be too bad

lyxal

23:01

Maybe `%{}`

hyper-neutrino

@lyxal yeah

that would definitely not interfere much but it's 3 bytes instead of 2 which i'm not sure about ¯\_(ツ)_/¯

i don't suppose we could use non-ascii characters, since the dictionary is full?

lyxal

@hyper-neutrino let me just check if there's the possibility to do that

@hyper-neutrino currently, there's capacity for 26082 words if my calculations are correct

And there are 23102 words in the dictionary

Using a non-ascii char would make the dictionary capacity 25760 words

So it could work

hyper-neutrino

oh

Transcript for