« first day (3811 days earlier)      last day (1048 days later) » 

1:00 PM
@pxeger Have you thought anything about bckwards compability for languages? Up to language maintainers? (If they are wrong?
@N3buchadnezzar yes; each language has a (currently hidden) property called version which I'm intending to be the guarantee of which version a certain language will always be
That makes sense, so when a language updates the version updates 1.1 -> 1.2, but something like 1.9 to 2.0 (or some major updates), would split it
But even Python is not backward compatible
@cairdcoinheringaahing Monking
@N3buchadnezzar yeah, so like the current Zsh runner has :version: 5.8, so if/when zsh 5.9 comes out, that will be listed as a separate language
Just for the UI I think it makes sense to first select the language and then the version, but that is nitpicking (defaults to latest)
1:08 PM
hmm that's a good idea
@N3buchadnezzar Monking? We're not the 2nd monitor :P
Is your customary greeting instead to rick roll each other?
We discourage greetings here :P
They are wastes of valuable bytes :P
they say, wasting valuable bytes on :Ps
1:15 PM
:P is valuable, it tells y'all not to listen to what I say :P
we know that already
it's the default for anything you say
I want to disagree, but .... :P
@math it's nice that SE auto plays gifs, but does it work for mp4s too?
1:22 PM
looks good
@pxeger is that ato?
I've been considering using a more full-page selector like TIO has, where you can also see info about each language, but I'm not sure if it's best
@math yes
@pxeger I really like the ones you can search in
The dropdown
yeah, that's not so much of a problem with only 3 languages at the moment, but I think it would be a good idea
@pxeger I didn't clearly understand how to add a language
1:25 PM
it's not particularly simple
What language do you want?
Does not seem terribly complicated to make the dropwdown searchable: npmjs.com/package/react-searchable-dropdown. But yeah completely useless with 3 languages :p
@pxeger I dunno maybe some popular golfing languages
@pxeger Jelly :P
@cairdcoinheringaahing I put up a PR to implement it :p
C+ is not a language
and if you meant C++ then that's not a golfing language
1:28 PM
C+ is like 50% C and 50% C++
c++ is already like 50% C
@pxeger okay perl
that can be arranged
But I thought pxeger wanted to work on the infrastructure before adding more languages?
1:28 PM
@pxeger JAAAAPT!
@pxeger I actually just meant how to do it
@N3buchadnezzar well perl will be easy (for me) because it's generally available
Japt might actually be fairly easy - it's all in Javascript
@N3buchadnezzar it's added to the docker repo, but it needs a runner (on my todo list)
@cairdcoinheringaahing japt is written in japt? really?
I was joking; Japt can be way down your list. It hasn't been in active development for years so TIO and my interpreter do the job perfectly well.
1:29 PM
@pxeger Cant I just make another PR basically copying most of the contents of the zsh?
@pxeger JavaScript
@N3buchadnezzar yep
go and submit issues people, this is too much to concentrate on!
Issue: ATO is too cool
Do interpreters have to all be in one language?
1:31 PM
@pxeger ATO is made in python isn't it?
@pxeger Always had the same problem when I asked for feedback on my interpreter in the Japt chatroom. Never learned my lesson, though!
@cairdcoinheringaahing they need to be freely available, packageable into a Docker container, and interfaceable from the command-line, that's all
Jelly should be very easy then
@math yes, but that doesn't mean the languages have to use Python
1:32 PM
$ git clone -q github.com/DennisMitchell/jellylanguage.git
$ cd jellylanguage
$ pip3 install --upgrade --user .
@cairdcoinheringaahing indeed; @N3buchadnezzar is working on it already!
That installs jelly as a runnable command (e.g. jelly eun '“3ḅaė;œ»'), or you can pip install it and use import jelly; jelly.main
I don't want to add too many languages until I've finalised containerisation (which I basically have in my head, but I need to implement it :þ)
@pxeger What's your plan for updating languages?
because otherwise I'd have to go back and change them all
1:35 PM
is there a good way to split a list into a square in Jelly?
like sL½$$ but short
@hyper-neutrino Wait. That is short
@cairdcoinheringaahing look back at my conversation with N3buchadnezzar yesterday (I'm trying to find it and bookmark it)
5 bytes is quite long when the answer i'm trying to find is supposed to be 6 :p
@hyper-neutrino Gonna add that as a builtin in my fork :P
1:37 PM

how updating will work on ATO

yesterday, 3 hours 6 minutes total – 180 messages, 4 users, 0 stars

Bookmarked just now by pxeger

oh no don't worry
@hyper-neutrino That doesn't even work, you need sLƽ$$
@pxeger I learned a lot sorry
Pxeger i think I got everything ready
what is sbcs?
oops I edited the message instead of sending a new one
Single Byte CharacterS
so that's true for jelly
So that one should be true yeah
Single Byte Character Set :P
1:43 PM
same diff
@hyper-neutrino Should the builtin mimic sLƽ$$ for non-square-length lists, or should it format into a rectangular matrix?
e.g. what should 7RŒ½, 8RŒ½ and 9RŒ½ be?
1 2 3  1 2 3  1 2 3
4 5 6  4 5 6  4 5 6
7      7 8    7 8 9
I did not change the whole ATO/yargs thingy except adding the jelly part, not sure if zsh and take inputs differently
1:47 PM
@N3buchadnezzar ah that reminds the reason I didn't add a runner for Jelly yet: because it doesn't have options at all, so I kind of wanted to implement a way to hide the options field so people weren't confused by it
if the intent of the built-in is to format as a square I think what I listed makes the most sense
possibly add a dyadic version to squarify with filler
Agreed. What about for 12 tho?
I think a 3x4 still makes more sense than a 4x3
@pxeger Makes sense, but those who use Jelly are not your run of the mill Windows XP users
It would be nice, but imho not a burning necessity
@N3buchadnezzar well I wouldn't count on it :þ
1:49 PM
@hyper-neutrino So [1,2,3,4],[5,6,7,8],[9,10,11,12]?
I think with a trailing ,[] would make sense but IDK
@Bubbler Charcoal, 11 bytes: IEN⁻⊕ι⊗÷⊕ι³
@hyper-neutrino So would 5RŒ½ be [1,2,3],[4,5]?
IDK how I feel about either [1, 2, 3], [4, 5], [] or ^
1:52 PM
Empty seems a bit strange
I dislike having trailing [], especially when it doesn't make the most sense
I'll add a square with filler dyad as well
fair enough
@Bubbler Charcoal, 10 bytes: ⪫''⪫⪪S'¦''
Today I learned that zebras are actually piebald horses in the process of defragging themselves!
brand new sentence
1:59 PM
@Bubbler Charcoal, 9 bytes: IEθ№…θ⊕κι
@pxeger I am not a frontend expert, but this seems to do it stackoverflow.com/questions/33667773/…
@Bubbler Charcoal, 10 bytes: ⭆θ⎇§ηκ↥ι↧ι
But yeah, as you say. It is a bit more work, having to update the languages whether they take options or not :p
@N3buchadnezzar oh I know how to do it, I just haven't
2:08 PM
can i get independent confirmation that this guy is being a prick? codegolf.stackexchange.com/a/231018/76162
YEah, seems like its a bunch of stuff to update for a minuscule change
@JoKing Has he deleted all his comments but the last one?
@JoKing I think prick is a strong word
@cairdcoinheringaahing yes
And yes, it looks like he's being unnecessarily antagonistic
2:10 PM
but he is actually right
actually I think it depends on the interpretation of external
well, the question of whether or not he is being unnecessarily rude is still "yes", but I don't know anything about Go to judge the validity of the solution ¯\_(ツ)_/¯
because many existing answers are already using libraries, so I think the answer definitely follows the spirit of the rules
But what constitutes as an external library in Go?
oh, which ones?
it's just unfortunate that the letter of the rules (albeit accidentally) end up requiring different things for different languages because of the import system in Go
2:14 PM
I think the main issue is that it might use the internet, specifically, rather than an "external library"
@JoKing like... most of them? All of the short ones that don't encode the data themselves
A: Print the character names

David ArenburgR, 54 bytes 62 library(Unicode) cat(u_char_name(utf8ToInt(scan(,""))),sep="\n") Edit: per @flodels comment, I need to read it from connection first, so had to add scan. This is also probably non-competent solution according to all the rules. Usage > cat(u_char_name(utf8ToInt(scan(,""))),sep=...

they're all using standard libraries or functions as far as i can see
ah, ok, idk about that one
@JoKing Yes!
@Bubbler Retina, 31 bytes
Hey look, banning vaguely defined programming things has lead to confusion and anger. Who would've expected this???
2:16 PM
I think this is just a poorly worded rules issue
...and I maintain my position that invalid answers should not (necessarily) be deleted
which rings true here IMO
Blatantly invalid answers should be. Answers whose validity is potentially uncertain shouldn't be
it could be worded better, but i thought that "external libraries" basically refers to anything that's not in the standard installation
If you think this answer "requires the use of the internet", then so do all the other answers, because oit requires you to download the langauge interpreter/compiler and hence the libraries (which are apparently not external) from the internet
Does the package/library used come "pre-installed" with Go?
e.g. importing fmt is fine, but golang.org/x/whatever is not really
@cairdcoinheringaahing nope, at least not in my installation
2:19 PM
Same way that import math would be fine in Python, but not import numpy, as it's an "external" library
@JoKing anything on golang.org/x/ is in the standard installation in the same way that /usr/bin/find is "in the standard installation" of Linux
If I really wanted I would fork golang to include golang.org/x/whatever in the standard installation
and then it would be valid
ergo it is valid
okay regardless of the clarity of the rules or what everyone's opinion of processing invalid answers or potentially invalid answers is, surely this could've been solved if OP explained what pxeger did in their first comment
This seems like a standard loophole
This seems like an issue that could be avoided entirely by simply not banning things
Like forking go, is basically creating a new language specifically designed to solve the problem at hand
2:22 PM
rather than banning you could just use what the standard rules always are - when using external libraries like numpy, the language name becomes "Python + numpy", and people can just indicate that built-ins or non-standard things are being used. If it's boring, people can downvote. If it isn't, people can upvote. Nobody needs to argue the rules ¯\_(ツ)_/¯
I mean we all agree the question was poorly worded / unclear, and the user should have used the sandbox. But that is in the past. X have already happened, now the issue is Y :p
@N3buchadnezzar hmm good point
Regardless of the validity of the answer, neither of the user's comments have been particularly pleasant
and the removed ones weren't particularly exceptional either
BOD: they have a borderline case, and JoKing has been somewhat one-sided in their consideration of the post
I can, at least, understand why they're annoyed
2:27 PM
I mean, Jo King literally just started off by stating a rule, OP could've explained instead of being antagonistic from the start
(not that I expect you to know that they were - they have several removed comments that you can't see, of course)
¯\_(ツ)_/¯ one could argue either way. not particularly productive in any case
you also can't see any of the comments where all they say is that I have zero knowledge of Go and that the answer breaks no rules without explaining anything at all
The easiest would be just to outgolf him in Go :p
it looks pretty well-golfed to me though
2:31 PM
i can't really do that without using a non-standard library
Someone else can roll that back, I don't want to wade any further into this
i guess that's confirmation huh?
@pxeger thanks for the link, it seems Jo should excuse themself from further discussion, as they are not being objective (childish name calling from a moderator)? — Steven Penny 52 secs ago
I don't remember any "childish name calling"
"prick", I guess? that's not name-calling, that's literally just a descriptive noun ¯\_(ツ)_/¯
2:33 PM
ah ok
"prick" and the slightly more offensive original message
well, technically it was a slightly ruder word than prick before i edited it
23 mins ago, by pxeger
@JoKing I think prick is a strong word
Less drama, more golf
@hyper-neutrino ...I'm struggling to see how that doesn't constitute name-calling actually, tbh
2:34 PM
Is my guy missing an arm?
You need to escape the \
anyway, can a RO move this to a different room?
Why and where?
it might just be a difference in culture
as an australian, i have the right to call people cunts
@cairdcoinheringaahing a new room, I mean; and because it doesn't seem to be interesting the other ¾ of the room
@JoKing perhaps not as a SE moderator though, don't you think?
2:36 PM
yup, that's why i tone it down
Honestly, it seems like the most annoying part of being a mod is that you can't take any actions without being accused of being powerhungry and overzealous
Which of course, is mostly BS, but it still must be annoying
It comes with the domain I guess
looking at the Go standard library page I haven't found anything under either src/text or src/unicode that indicates runenames is in the stdlib
I may be missing it, so someone else should take a look
18 mins ago, by pxeger
@JoKing anything on golang.org/x/ is in the standard installation in the same way that /usr/bin/find is "in the standard installation" of Linux
it's not techincally part of the standard installation, but most things you want to install will require it
¯\_(ツ)_/¯ okay all I'm saying is that it isn't in the "stdlib" section
2:40 PM
Q: Roll up a D&D character, with options

user3934737I'd like to roll up the ability scores for my Dungeons and Dragons character. But I'd also like them to be balanced (and slightly better than average). Typically, a character's stats are created by rolling 4 6 sided dice, adding together the three highest results, and doing this 6 times. The 6 sc...

as my final word before i go to sleep, i'd just like to note that according to the answerer's logic, the only time they'd explain to me how it wasn't an external library were if I already knew why it wasn't...
@NewPosts Uhm, why cant I answer this question by always picking the lowest allowed values
Yeah, that should be clarified
2:47 PM
a=random.sample([h,t], k=6); a[0]=l
As it is cat;shuf -e 1 2 3 is a valid answer (output the three inputs followed by 1 2 3 in some order)
because the first input will always be <= the first input; the second will be >= the second, and the total (first+second+third+1+2+3) will always be >= third
(actually, seq 3|shuf is shorter)
Isn't the third test case wrong?
no, why?
There's no elements below l, or above h, and the sum is equal to, not greater than, the threshold
same goes for all the test cases
i'll edit non-strict inequality in
2:55 PM
@pxeger Is it not reasonable to assume that "6-sided dice" means standard dice? — Adám 1 min ago
@Adám maybe, but it isn't specified
Isn't "rolling 4 6 sided dice, adding together the three highest results" a non-observable requirement?
@Adám it would be, but I don't think the challenge itself says that's necessary
So simply using random numbers in the range [3,18]?
it doesn't specify anything beyond "randomly generated"
that's the problem with the challenge
@Adám If run multiple times it would be observable different from uniform [3,18] pick.
2:59 PM
@xash But problem doesn't state distribution.

« first day (3811 days earlier)      last day (1048 days later) »