« first day (3895 days earlier)      last day (25 days later) » 
00:00 - 20:0021:00 - 00:00

12:00 AM
Q: Why should I register my account?

ripper234I can use many of the site's features while unregistered. I am looking for a simple explanation of what benefits a user can have from registering. For more information, see "Why should I create an account?" in the Help Center. Return to FAQ Index

they do have an email address though
or at least, they've listed one. i guess maybe they didn't confirm it from their inbox or smth
@cairdcoinheringaahing you have my approval to "borrow" that line :D
see that falls under the category of things that probably aren't particularly professional or official-looking but it's some guy plagiarising i don't care :p
12:34 AM
hi ppl
I just watched 2 balloons fight for 20 minutes what's wrong with my life
remind me of bloon tower defense
12:52 AM
@cairdcoinheringaahing Pip lacks a factorial function too
There's only a handful of challenges where it'd be useful. Now, prime factorization & primality checking is something that I probably should add.
(However, I somewhat agree with your sentiments on GolfScript.)
yoooo i get vaccined
now i can go other country
Sandbox posts last active a week ago: Extremely small data compressor
can we close the SE suggestion of accepting answer?
it suggest me to accept or add bounty lol
Have you considered accepting an answer or starting a bounty for this question?
2 hours later…
2:54 AM
My hobby: coming up with horrible jokes to announce my monthly prime resub for my favourite twitch streamer
3:11 AM
@Niko Nice!
First or second? (or third?)
I just tried to write void LinkedList<T>::operator+=(const LStack<T>& other) inside the LStack class implementation.
Very cool
very nice
That just looks like normal C/C++/Java to me lol
That is to say, very cursed and scary
@RedwolfPrograms the part about it that is so horribly wrong is that I accidentally tried to declare the operator+= as a method of LinkedList<T> instead of LStack<T> inside a file which is 100% LStack
@RedwolfPrograms first
3:46 AM
@lyxal you are too used to a linkedlist
@pVCaecidiosporeadduced so it would seem
well in 20 mins; i have the vyxal (the first language that shakespear learnt) exam
Good luck
sounds cool
3:51 AM
lemme go get the printout of the vyxal elements :P
And remember, if you ever get stuck, use the program lyxal to do a little trolling to everyone in the room
i wish my school have some sort of codegolf competition
i have to use a camera for the exam
sooo dont hack pls :P
4:07 AM
ok this exam is in 2 parts;
objective and subjective, rn i have to do objective
i think i will go now
4:57 AM
1st part objective done
it was easy
had some vyxal diagraphs and commands
2 hrs later i would have to write actual vyxal programs :P
2 hours later…
6:44 AM
guess who has an assignment to write an algorithm within specific complexity constraints in a formal model of parallel computation where it's gotten hard enough to reason about that he has implemented a simulator for the formalism with far too many python generators and a stack for process-local data and is now implementing a binary search in it
That sounds almost like a graduate-level assignment... except that I expected Coq instead of Python
i think it is a graduate course actually
Then why the hell are they building formal model in Python
writing an interpreter isn't actually part of the assignment it just seems easier than actually working through the formalism by hand for making sure the expression i have for the right length to cascade at actually gives me the right work and span bounds
1 hour later…
8:09 AM
Random idea: when calculating the ELO rating, if a challenge is solved in language A but not in language B, count it as A winning over B
That'd reshuffle the leaderboard a lot.
JS and Python would become the top languages
Yeah :P
1 hour later…
9:11 AM
weight that much lower
it might work better
The intended purpose is to incentivize solving challenges with fewer answers, btw
though the resulting ELO wouldn't be the "pure golfiness" ranking anymore
It is a bit unfair how SOGL's in the lead because it wins most challenges it enters.
answer normal challenges in it and it will naturally go down
i used canvas instead of sogl because canvas has searchable docs, sogl would be close to that or worse, lilely.
Real problem is when only the language author knows how to use it
hehe yeah pretty much
9:20 AM
Yeah, then they know what to answer and what not to, and it's inevitably going to win.
this was sort of the case with husk as well
until the lotm
SOGL accept rate: 35 / 263. Vyxal accept rate: 6 / 851.
I know that's partially due to changes in accepting, but still...
It's no coincidence that most of the accepted answers in SOGL are either ascii-art or kolmo, and also most of them include a huge compressed string
and also no one knows how to compress strings for SOGL like that
Iit's getting late so 'ngiht
9:41 AM
Oh, Sledgehammer has finally come to the table, though with a mediocre elo (which is understandable because it still has very few solutions so far)
Wondering why Vyxal is still not on the table
10:22 AM
@Bubbler because SEDE
For some reason, it really doesn't want to include many 2021 results
CMP: How about modifying the SEDE script to return the most recent 50000 answers?
11:08 AM
All the drama on MSE about sponsored sites is making me wonder what site/service would a) want to sponsor us, and b) would be a good fit? :P
Q: Find the longest line

Wheat WizardGiven a string containing only the characters -, |, + and newline determine the longest straight line contained in it. A straight line is either an uninterupted run of -s and +s in a single row or an uninterupted run of |s and +s in a single column. So for example: | | ---- |...

@cairdcoinheringaahing obviously a sponsor would have to be fine with the stuff we go on with here
Sep 20 at 23:45, by hyper-neutrino
well that diverted the conversation, so let's instead talk about Code Golf and Coding Challenges Stack Exchange as we always do in this very on-topic and focused chat room
Having said that, I'd totally sponsor us if I had a corporation that had the capacity for sponsorship
11:25 AM
@cairdcoinheringaahing dyalog maybe?
@Bubbler yes
@UnrelatedString joe?
@user what's ligma?
Bit of a haskell wat. null [ id ] returns False but [ id ] == [] errors.
Brownie points to any non-haskeller who can guess why.
Is [ id ] a different type to [], so the equality comparison fails?
The type checker will infer the type of [] to be the same as [ id ] and has no problem doing this. So for example [ id ] ++ [] works just fine.
In short no, that's not the issue.
I have no idea what null does, but does it, by any chance, just ignore what it's argument is and return False, meaning that [ id ] is invalid, but just isn't evaled in the first one?
11:38 AM
null returns true on empty lists and false on non-empty lists.
it's not that confusing when you think about it
It is if the closest you've come to using Haskell is using Husk :P
To be more direct [ id ] is perfectly valid and Haskell does not error trying to evaluate it.
I have no idea then :P
The thing is that functions are not comparable, since checking equality would require solving the halting problem. And by extension lists of functions are not comparable because if they were then [a] == [b] would have to compare a and b. Comparing to the empty list is a degenerate case where we can always determine equality, so it trips people up.
I get this error all the time in my code.
12:09 PM
Yeah, a -> a is not Eq
On the other hand, if a function's domain is finite and its codomain is Eq then it's mathematically comparable, though it's practically (almost) meaningless I guess
It is comparable assuming there is no bottom.
Which to be fair is an assumption you pretty much have to make for any Eq.
For example cycle [1] == cycle [1] will not halt.
Of course
Time to add Eq1 ((->) Bool) and Eq1 ((->) ()) to hgl. :þ
12:42 PM
Don't forget Ordering :P
Also Int is technically finite :P
I think you can generalize it to (Enum a, Bounded a) => Eq1 ((->) a)
Int might also actually be useful, whereas I find it unlikely that the other ones would be, particularly () and Ordering
In a similar sense I can imagine a technically working Ord1 instance
Yeah, Int -> a is basically List a but with a really large fixed size.
... and then Enum, and even Bounded for functions with bounded codomain
Actually Bounded doesn't need any constraint in its domain I think
12:58 PM
@Bubbler I did an edit
@AncientSwordRage I get the purpose, though CGCC community is generally less concerned about accessibility than other sites
@Bubbler I think it does. Maybe could you elaborate?
@Bubbler .... maybe it should be?
Not entirely sure about the code block -> table, but I don't see anything wrong with that. Definitely +1 to approving the image description
1:13 PM
is re builtin in python?
What do you mean by that?
the re module
is it provivded by pythoñ
It's part of the standard library.
I think any installation of python will include it, but in order to actually use it, you always have to do import re
1:15 PM
builtin is the part of the question that is unclear.
@WheatWizard To my instinct, the natural minBound for Bounded b => Bounded (a -> b) is \_ -> minBound, and the same for maxBound, whatever a is
Python has "built-ins" and "standard libraries" clearly separated in its docs. The former are in Libraries Reference chapters 2-5, and the rest belong to the latter
re is a library, not a built-in
> pythoñ
yeah by builtin i mean its in the standard library
@AaronMiller The thing over the n is the python that everyone keeps talking about :P
@cairdcoinheringaahing thats the snake :P
1:22 PM
ah, of course
totally didnt long press n on my tablet
lol I was wondering how you managed to accidentally type that
@Bubbler This is interesting. Bounded doesn't seem to have any rules, but I would expect that there would at least be a partial ordering which the minBound is a terminal object, which would not hold for your given type.
Just realized it's easier to type Jelly chars on my phone than on my computer
so tmrw is my hindi exam which is hard so i must study bye o/
1:25 PM
Phones are nice, since you can pretty easily make your own keyboard.
@WheatWizard how?
@Bubbler Really? How?
I always have to copy-paste on mobile
That applies to some chars only :P
@emanresuA I'm planning to have per-tag rankings in my script
So you can see which ones are better for math, strings, ascii art, etc.
That won't entirely fix it, but it provides context for why certain languages are near the top
The default Samsung keyboard does have quite a few accented chars and some non-ascii special chars that happen to be also Jelly chars, but I don't see dotted letters
1:31 PM
@lyxal Why tf don't you combine multiple queries
Maybe someone should make a Hacker's Jelly Keyboard at this point
If the ELO script isn't including all of the answers it's practically RNG
and/or heavily biased toward languages from a specific time period
@Bubbler better idea: make a fully customizable android keyboard
so you can type any chars you like
maybe add multiple sections
@pVCaecidiosporeadduced Hacker's Keyboard supports custom layouts and switching between layouts
dzaima has made one for APL
@RedwolfPrograms I was wondering that too, and I have a relatively easy fix for that
@Bubbler isnt it outdated
like it uses deprecated ap8s
1:37 PM
Just strip the first line of all but one and cat them with that one coming first
@RedwolfPrograms because I'm bad at SEDE
It doesn't require SEDE knowledge
Well, you need to filter out a specific date or ID range, but that's not too hard
It does require basic SQL knowledge though
If I could figure it out anyone can :p
(I am not good at SQL because it wants me to die)
I checked a few hours ago and there are roughly 130k eligible answers in total
1:43 PM
I modded the IDs by four, to download them as four separate files, then just merged them as part of my ranking script
what does :£ look like?
A kitten who's a superspy and being sniped at but expertly evading the bullets
@pVCaecidiosporeadduced I was going to say duplicate and set register but then I realised you meant emoticon wise
@RedwolfPrograms so then how do I add new results?
And how do Iimit the years?
Use a where to filter out a specific date range
@lyxal in Vyxal?
1:46 PM
Do that 3-4 times to get all of the results
Then join the files after removing the extraneous headers
@RedwolfPrograms that means I have to download 4 csvs each month
Sure does
@pVCaecidiosporeadduced my life has gotten to the point where I just recognise it even when it isn't there
@RedwolfPrograms then I might as well just download all pre 2021, merge those, and just append 2021 each month
1:48 PM
Sounds reasonable enough
Just do that then
@RedwolfPrograms I do not like bullets anymore
Not after [hyperlink blocked]
@RedwolfPrograms tomorrow. Because it's sleep time
ok what about :₩
@lyxal o//))
@lyxal (before I go) some of us will soon understand the meaning of this message
@pVCaecidiosporeadduced About to kiss someone but you get hit in the face with a sword multiple times
1:50 PM
@cairdcoinheringaahing It looked like a table, smelt like a table and tasted like a table....
@RedwolfPrograms ok this :@×!
it also means I can put explanatory headings on the columns
@pVCaecidiosporeadduced Someone screaming with a weird goatee
ok this? :^&%÷%@/£@=?£2/*2
Someone with a big nose, a weird mouth, and a body which has gone through a paper shredder
2:50 PM
Llamas with wheels would be cool
I've got a scientific rendering
well that's an interesting thing to see first thing opening up tnb
The ironic thing is that the Inca didn't have wheels, so wheeled llamas would actually have been quite revolutionary
my computer assignment: 1 mark. Convert this to an equivalent java expression: (0.05x²+y²)÷10
me: (0.05 * x² + y*y) / 10
3:02 PM
my teacher: full marks
i wrote x² and got the mark
hmmmm excellent javaing
@rak1507 ty :)
me: sees x*4, also me: sees x<<2
@pVCaecidiosporeadduced -1 byte: ¥2↲
3:08 PM
CMP: x*8 or x<<3?? (for general use)
for optimization i think i would for bitshift
but in general multiplication
:D I just realized today is LYAL
which lang?
ooh, it looks like we're doing cascade. that's one that i've been wanting to check out
any decent compiler should compile x*8 to x<<3 anyway
@rak1507 unless you use an interpreter
3:12 PM
if you use an interpreter it will be slow either way, micro optimisations like that probably won't make much difference
@AaronMiller when does it start?
@RedwolfPrograms bad rendering; i see the legs
No you don't
> i just did
@pVCaecidiosporeadduced those are part of the wheels
Nope, you didn't see any legs in my high quality rendering.
3:14 PM
@AaronMiller oh right, the axles
Ooh nice, got a 113% on a calculus test
@pVCaecidiosporeadduced about 9 hours from now
@RedwolfPrograms whats the derivative of 113?
3:16 PM
@RedwolfPrograms no the value changes 1 then 1 then 3
@AaronMiller i think so
@AaronMiller wait, no. I forgot to carry the one. :facepalm:
@AaronMiller actually a recursion error
in Vyxal, 47 secs ago, by Vyxal Bot
[@pVCaecidiosporeadduced: 59251247]

maximum recursion depth exceeded while calling a Python object
Most recently popped arguments: [[], 0]
Final stack: []
makes sense
@RedwolfPrograms no
in Vyxal, 1 min ago, by Vyxal Bot
[@pVCaecidiosporeadduced: 59251261]

maximum recursion depth exceeded while calling a Python object
Most recently popped arguments: ['00', 0]
Final stack: []
again a recursion error
3:20 PM
yall like recursion so much :p
@RedwolfPrograms no
hmm. using e to get every zeroth letter of a string give one of two different errors depending on the order of the arguments, even though it performs the same function
getting every ath letter of b gives a recursion error, but getting every bth letter of a gives an error about the slice step being 0.
like i said yesterday
vyxal's pretty well designed and not as much so executed :P
also why is the exponent function called "exponate"
@AaronMiller this is because types == (str, Number) is coded as a separate if block but the rest are part of the type-map block, which doesn't have a case for (Number, str)
Conclusion: e needs rewritten
@hyper-neutrino I see, so it's not that 0 string e breaks, it's that number string e breaks. nice.
3:56 PM
I think my program can now correctly parse or reject the headers of basically every code golf answer on the site
There are a few situations where ambiguity can cause issues, but it's able to get something reasonable out of 98% of them
There are a few situations where ambiguity can cause ambiguity? Inconceivable!
E.g., Perl 6 could mean "Perl 6" without any byte count, or "Perl, 6 bytes" (my program treats it as the first, and rejects it)
@WheatWizard Comment markdown leaves a lot to be desired... I edited the test case to what I had intended.
One iteration of my Pip submission could have saved bytes by concatenating the list of lines together, which would turn that test case into |+--+|, giving an incorrect answer of 4.
(Pretend there's spaces on the outsides of that. Chat markdown leaves a lot to be desired... :P)
@RedwolfPrograms What's your opinion of Pip's -r flag? It changes the input source from command-line args to stdin, which basically just makes it easier to enter the input, especially on TIO.
4:17 PM
100% fine
Cool, then I won't add a 1-byte syntax for it.
Except for really low level languages on challenges with restrictive I/O, I think using flags to specify where or how input is taken are fine
As long as that isn't being used to actually process the I/O
I had just lumped it in with the other flags, until I wanted to use it in an answer today and went "Wait a minute..."
4:40 PM
@RedwolfPrograms For Vyxal, the b flag is ok, but not the T flag, right?
Aren't those both timing related?
lol yeah
I don't think there's any legitimate use for them in an answer, but they wouldn't be cheating if you just used them for their intended purpose of timing out after a little while
If you used that in a challenge requiring timing out after a certain amount of time, it'd be cheating
E.g., submitting "while 1" with T would be flag abuse for "make a program that runs for 60s then terminates'
I can name 2 answers that use timing flags
Show them to me so I can downvote them :p
4:48 PM
this and this, though it technically isn't needed for the bf one, it's just necessary for pretty much any substantial bf program
I don't think those flags belong in the headers, except maybe for the CnR (but that's different)
i agree
The BF interpreter should put the flag in the TIO link, but not the answer header
Like you did here
hmm, was that upvote you, by chance?
Yeah :p
4:51 PM
would you say that CnR is flag use, or abuse?
just curious
For a CnR, it's a bit vague I think
I'd say it's a bit boring, but probably not necessarily cheating
what about using +startinsert in Vim, like I did here?
Seems like a pretty clear flag abuse to me
Saving i/a by putting it in the header
yeah, that's one I'm not so sure about
that's the only place I've used it, though
I'm going to write up another flags.md detailing my three black boxes model
I think it's a good step in the direction of making flag usage more objective
5:11 PM
@RedwolfPrograms I'm going to do a flags writeup too, and I disagree with your three black boxes model ^^'
Out of interest, anything in particular about it you disagree with?
I definitely think that flag usage should be more objective if possible, and that flag abuse is bad, but I also feel a bit more lenient about what actually constitutes flag abuse.
@NathanMerrill Hi!
@RedwolfPrograms I think it sets up an unobservable requirement to say that an output of Hello, World! is valid if it's the string "Hello, World!" but invalid if it's the list ["Hello"; ", "; "world"; "!"] which the language outputs by concatenating all the values.
If I'm remembering your argument correctly.
It is still unobservable and/or ambiguous in a lot of cases, yeah
E.g., if the language spec says that that should be outputted as "Hello, World!", I personally think it'd be fine
5:36 PM
I have an idea for a golfy way to represent programs in 2d languages
Probably already exists
Basically, you follow the path the IP would take, writing down any non-duplicate instructions along the way
And for instructions that can branch multiple directions, you do it prefix-style
E.g., if there was a language where . was where the IP started (and it began by going right) and , ended the program, and slashes are mirrors and x sends the IP in a random direction, you'd write this program: as
5:52 PM
What about a branch that could go into multiple infinite loops? How would that look?
You'd never write the same instruction twice, so if you end up in a loop, after the first iteration you can leave
  d b
Would be .#a\b/c\d/
Where # skips the next instruction
So you can't do infinite loops, then?
What do you mean, that's how you'd do an infinite loop
oh I thought you meant that it would leave after the first iteration
If you had a condition branch before that, once you finish with the infinitely looping branch you jump out of it and start on the other branch
Oh cool we get to do something with HCl in class today...I should put a drop of it in my eyes and see what happens
Wait what
I halfway want one and halfway really don't want one
@RedwolfPrograms I kind of want to stick my finger in the HCl to see what happens
Just a little bit
I should accidentally trip and hit my face on the thing of it so I'd have a giant acid burn on my face and look like an evil supervillain
(And everyone knows I can't actually get hurt since I'm wearing pants as requested)
6:23 PM
@RedwolfPrograms Did you and lyxal switch accounts again? You're shitposting more than normal :P
dude this keyboard is absolutely not worth it but i really want one lmfao
@cairdcoinheringaahing I'm just bored :p
Also apparently the HCl wasn't very pure D:
@hyper-neutrino Honestly, one of the funniest things SE has done in a while :P
See they say it's a joke right now, but I think the April Fools thing was a leaked early prototype
Is it LYAL or BMG today?
6:43 PM
What language do people want for LYAL?
7:02 PM
what will we call the Cascade CMCs?
Hopefuly there's no user called MC...
posted on September 28, 2021 by celtschk

Given a non-negative integer up to $999\,999\,999$, write it in English. The input number can be in any form other than English (though you'll typically want to use the native integer ty...

7:15 PM
I suppose if all else fails we could just use LMC as a general LYAL MC, but I like doing something different for each one
We could default to the second letter in this case and call it an AMC
Wait can you ping someone in bold or italics?
did it ping?
@AaronMiller Yep that pinged me
7:23 PM
I'm pretty sure it works regardless of formatting: @RedwolfPrograms
Huh, even in code?
Let's try a code block: @AaronMiller
wow TIL
7:24 PM
Neat, pinged me still
Chat is weird
The neat thing is, that's a Vyxal function
It didn't highlight green though
def RedwolfPrograms(f):
    """decorator to fix a function according to Redwolf's rules"""
    def inner(*args, **kwargs):
        return "".join(c for c in f(*args, **kwargs) if c.lower() not in {*"jlmhe"})
    return inner

def greet(name):
    return f"Hello, {name}!"
@RedwolfPrograms yeah that pinged me, i just wasn't paying attention
7:27 PM
@AaronMiller we can tell it pings you because it highlights as a reply to your last message
@pxeger What are redwolf's reules?
What about in a
multiline message
Wow, usually multiline messages break things

redwolf letters

Sep 16 at 18:57, 13 minutes total – 46 messages, 4 users, 3 stars

Bookmarked 7 secs ago by pxeger

@RedwolfPrograms when you pinged me in code, it only showed up green in the transcript, how about you?
Same. I think it's my chat userscript, though.
Unless you're not using it
Oh wait huh
7:30 PM
@pxeger that makes it sound like Redwolf enjoys is "letter"ing, whatever the verb "letter" is supposed to mean
It's not green even without the script
I turned off my userscripts and its the same
it's strange that it's different in the transcript
7:40 PM
> An inconsistency exists in chat
Have you ever used chat? :P
00:00 - 20:0021:00 - 00:00

« first day (3895 days earlier)      last day (25 days later) »