« first day (2606 days earlier)      last day (2256 days later) » 
00:00 - 19:0019:00 - 00:00

12:00 AM
goshdarn parents, forcing you to play video games
@Downgoat -1 still no TIO link. :P
does puts guarantee a flush?
@ETHproductions he is wimping out and coding some macros
imma be casting and invoking every spell manually
@Downgoat I was wary about searching for you on twitter bc the name you go by isn't super unique, but I relaxed when the avatar of the first result was a goat
@Downgoat No.
Also, [Try it online](lib/vsl/parser/parser.js) is not a TIO link. :P
12:07 AM
i did what
On the bright side, it wasn't your password.
how do you know? ;P
Fair enough.
@ASCII-only y is ternary like:
   -> Assign _ "?" _ Ternary _ ":" _ Assign {%
why isn't the last part another Ternary
:| i did a stupid is why
12:17 AM
btw how would I implement bitcast
because syntax example is: Pointer<UInt8>::string.rawValue
where LHS is a Type instead of Expressoin
so not sure how we'd do this
@ASCII-only oh no we did a javascript:
vsl::asts> 1..2
Syntax Error: Unexpected token `.2` (1:2)

    > 1 | 1..2
        |   ^^
maybe require floats to have at least one digit after decimal point then
@ASCII-only btw what does this mean:
                data[0][0].isClosure ?
                    (data[0][0].isClosure = false) || true :
                    data[4][0].isClosure ?
                        (data[4][0].isClosure = false) || true :
                        data[0][0] instanceof t.Whatever || data[4][0] instanceof t.Whatever,
oh that handles whatevers
like 3 + ? * 5 will become {3 + ? * 5}
not 3 + {? * 5}
@ASCII-only btw i have quetion on lambda semantisc
if you have like:
// inside func f
let a = 1
let b: () -> Void = { print(a) }
a = 2
What should this be
I was thinking 1
because integer would be capture by-value
it would be good for that oisc challenge
12:53 AM
Idea for a language: brainflak but it has function definitions
1:15 AM
@DestructibleLemon so basically lispflak
not exactly
lisp functions can take more than 1 argument
and also support functions as a type
and don't have a global stack and side effects as the default
speaking of which
brainflak but without side effects as much as possible
tbh i'm not sure if that idea is even coherent
1:31 AM
has CrazyPython &| Tux been around SE chat recently?
you mean crazypythou? idk
2:06 AM
@Neil Really? No, if all elements are odd, it should be Math.floor(a.length/3).
@DestructibleLemon So, "In BF, overflow errors out immediately. Your generated program must not overflow in any case."?
2:22 AM
Can any close-voters of this explain? (cairdcoinheringaahing, totallyhuman, Mr.Xcoder, NieDzejkob)
@user202729 no, i mean, there's no way i can provide an example where bf overflows, because the exact compilation of bf is entirely up to the persons program
I'm getting the answerers to compiler to bf, and if they wanted to not have any bf overflow for some reason, that would be weird, but they're entitled to do that, except that it would probably be ungolfy and overcomplicated
@DestructibleLemon I've played around with this idea a lot
@DestructibleLemon But still... having a test case where the naive implementation gets it wrong can be useful.
@user202729 the spec means overflow wraps around, for the challenge
I corrected that. it was an oversight due to forgetting to save edits and such
One language I sort of designed was called Flap-Jak, which had no loops and instead of stacks of integers it had stacks of pieces of code
2:29 AM
either way, if overflow was undefined, they would have to not have any overflow at all, but it is defined
I never finished the interpreter because it made my head hurt
I designed a language to be 2d /// but then i didn't finish the main command
Anyone know how can I type less-than-or-equal on my keyboard? (Arch Linux - English (intl., with AltGr dead keys))
anyway, for example, in the most recent revision: "- : decrement the cell under the pointer, unless the cell is 0, in which case set cell to 255"
@user202729 i have a compose key for that, which you might get maybe? idk
@user202729 ≤
2:31 AM
haha i have more typing power
(Which key?)
@user202729 i followed some tutorial to get the compose key
also can i share with you a garfield remix i might submit?
I did alt+,
... I only have AltGr dead keys. Although it has "keyboard layout", it's not really clear.
Nope, Alt+, gets ç for me.
@user56656 ≤ oh yeah i can do that as well
i probably should have used that while writing sad-flak programs i guess
2:37 AM
My keyboard layout : i.snag.gy/PN3CQR.jpg
CMP: In a language where function declaration inside another function is not possible and code start execution in a main function (C++ or Java), what would you do if you need a recursive function to process the data?
(1) use global variables,
(2) use lambda,
(3) other.
You can use a recursive function without function definitions inside of other functions..?
just recurse?
int fib(int n){
        return n;
    return fib(n-1)+fib(n-2);
please i need feedback on my sandbox challenge i'm not kidding: codegolf.meta.stackexchange.com/a/14968/55896
apparently cheese is technically illegal in Florida
Tyramine ( TY-rə-meen) (also spelled tyramin), also known by several other names, is a naturally occurring trace amine derived from the amino acid tyrosine. Tyramine acts as a catecholamine releasing agent. Notably, it is unable to cross the blood-brain barrier, resulting in only non-psychoactive peripheral sympathomimetic effects following ingestion. A hypertensive crisis can result, however, from ingestion of tyramine-rich foods in conjunction with monoamine oxidase inhibitors (MAOIs). == Occurrence == Tyramine occurs widely in plants and animals, and is metabolized by various enzymes, including...
2:57 AM
@DestructibleLemon You say the tape extends infinitely to the right. What about the left?
@Dennis i think i forgot to save edits possibly... clarifying
std::vector<std::vector<int>> adjlist; // <-- global variable, do you use this?
void dfs(int node) { /* something */ }
int main(){
    /* process adjlist */
3:43 AM
@DestructibleLemon See the starboard >>>
There are ≃5 room owners in the room right now...
If your TNB is glitching too, refresh.
@Dennis well, it's not exactly glitching
@user56656 Can't, mobile.
Also, Dennis beat me to it
@user56656 Typo...
4:06 AM
@DJMcMayhem ofc
@user202729 nah, it's in Thanish
4:18 AM
@user202729 who said
Wheat Whizard!! user56656!
also for reference:
Apr 19 '17 at 17:20, by betseg
And "dansk" is "Danish" in Danish.
@ASCII-only Whoops...
Wait a minute... That is a word?
Or is it not?
No, it isn't.
I believe the term is "burninate" rather than "delete" when speaking of tags.
4:30 AM
@anyone know of any golflangs not here yet?
Obviously it refers to electronics worshippers.
Q: Make a BCDto7SEG chip

l4m2Source Given a decimal digit in BCD format in cell a[0][3] to a[0][0] (so 8 will be represented as an 1 in a[0][3], and 2 be one in a[0][1]), you are required to output its 7-segment expression in a[10][0] to a[10][6]. Converting Table: a[0][3] a[0][1] a[10][6] a[10][4] a[10][2] a[10][0...

Any idea why this is unclear?
BTW... this is 1 vote from being reopened.
@user202729 well for one there's no spec at all?
@ASCII-only Which specification? ...
there's no output format
which means every answer will have a different format
which will be terrible
4:43 AM
@ASCII-only The format is not very important. This is circuit-golf, not code-golf.
@user202729 of course. but if it's inconsistent it takes more effort to read
You can expect people will use reasonable format, or there is only one answer.
oh yay logisim
@ASCII-only No idea if it's real, but it's definitely unverifiable.
2 hours later…
7:27 AM
A: Sandbox for Proposed Challenges

BubblerFind out if this a valid Stack Cats program, in Stack Cats style! code-challenge restricted-source Background Stack Cats is a reversible esoteric language made by Martin Ender. Each command in Stack Cats is either the inverse of itself (represented as a symmetric character, such as -_:T|), or ...

2 hours later…
9:54 AM
Q: Blue duck, red duck, grey duck

AJFaradaySo, last week I posted a challenge to play Duck, Duck, Goose. This lead to a number of Minnesotans commenting about their regional 'Grey duck' variation. So here's the rules: Given an array of colours: Red Orange Yellow Green Blue Indigo Violet Grey Write a program to follow these rules: S...

10:23 AM
@user202729 you're right, I confused myself about what needed to be odd. I guess it needs to be a=>Math.min(o=a.filter(e=>e%2).length,(e=a.length-o)+(o-e)/3|0)
10:34 AM
@MartinEnder not to break it on you, but Brachylog beats Retina on a string challenge :)
10:50 AM
I'm not surprised :P
11:09 AM
We've had a week or two of tape, scissors and ropes. I feel we're heading into a week of ducks and geese...
@StewieGriffin *ducks*.
markdown next to period == good luck
oh wait that was escaped
@Adám :| this is two years old
@ASCII-only So? We did one year of ducks, but we have many many more at HQ and more are being added frequently.
@Adám It's just weird seeing '16 referred to in the future tense I guess?
11:19 AM
@ASCII-only I guess, but all old news will have that. I suppose (I don't instagoogletweetface) that people's old posts have things like "I'm so excited for tomorrow".
@Adám oh yes they do, if you were to edit every one old post then your life would be over
@Adám Hmm. I guess it's just a bit misleading in that it looks more like a (non-time-specific) info page than an update then?
@ASCII-only I'll ask that it be corrected then.
Anyone know the C# equivalent of Python's __import__ function?
@ASCII-only Why?
11:33 AM
@user202729 Because I need reflection black magic interop
I think you can just use fully qualified name. TIO.
(if that's what you meant by import)
@user202729 No. Assume you're given the string name of the namespace or module you want to import
So... no, I don't know much about reflection.
Even better, is there a way to get the object referred to by any string (no matter whether it's a namespace or module), without having to climb up the class tree
12:19 PM
(I will make the controller after I get some final feedback; basically, feedback -> controller -> post)
@HyperNeutrino pls don't disallow posting too many submissions...
ok... I don't think the limitation is too bad but I'll remove it for now
I guess I just don't want someone to create a ton of submissions and cross-team them all but I suppose that would go against the standard competitiveness rules
12:36 PM
A: Sandbox for Proposed Challenges

Kevin CruijssenQuine Time Bomb with Reset Functionality code-golfquinetime Challenge: Create a program or function that takes an integer input, which outputs a new program/function as specified below. Input: Integer n: Time in seconds before the Time Bomb explodes. Output: The original program which had ...

12:51 PM
morning o/
@HyperNeutrino Well you can just run one bot from each player max?
what's the scoring for execution speed challanges again?
i assume just the speed, no byte count
@moonheart08 yes
by editing your thing i got pinged twice. Huh.
1:07 PM
@moonheart08 what did you think happens when you edit >_>
true. :p
Landed in Athens 22 hours late. 0/10 never going to Greece again -.-
Cant wait for the xray project (from the Atom team) to be usable. Atom's slowness is finally starting to get to me
@moonheart08 atom's slowness got to me a long, long time ago
@ASCII-only i have a really beefy laptop, so i notice less :p
when you have a good/overthetop system, it's hard to notice as much, and atom turns out to be a really nice editor
1:30 PM
A: Sandbox for Proposed Challenges

Buffer Over ReadOctacontakaipentagon code-golf Given a polygon with the number of sides n, the a-th n-gonal number is: Given an input of two numbers a and n, output the a-th n-gonal number. Standard loopholes apply. Shortest code in bytes wins.

Here the answerer indeed misunderstood the challenge, correct?
they did
@BufferOverRead Dupe.
@user202729 :(
It's not a dupe
There's a twist at the end, did you spot it?
@BufferOverRead it will usually not be a whole number then, right?
1:37 PM
......... So that's just a "multiply by 3/2".
@ASCII-only I misread that as "plus"...
Okay, I get it, I get it, dumb challenge.
don't worry, that's why the sandbox is there
I've been working on a research paper with something similar to that, so that inspired the challenge
Basically, the research is about data compression using polygon-like structures
1:40 PM
@ASCII-only I guess, but I want to allow people to have multiple submissions running together but not completely dominate the whole thing
(also this thing about a fifth of the problems doesn't allow anyone to submit until there are 4 submissions :P)
@BufferOverRead Gets 139% improvement over 7zip in best case.
@BufferOverRead Use :<message number> <content>.
@cairdcoinheringaahing Just saw your answer-chaining post in the sandbox. I've been wanting to do in almost the exact same form for months but never found the time to build the base interpreter. But since I've been thinking about this a lot, here are some ideas/suggestions:
@HyperNeutrino well IMO I'd take more solutions over fewer, KotHs tend to get fewer solutions so any extra is a plus
@MartinEnder link?
Which language are you going to write the base interpreter in? Python might be good?
1:41 PM
@user202729 Ah I see
I guess that's true. yeah I'll keep the restriction out then
You can avoid "new built-in must not be specifically for one challenge" by requiring that each answer shortens at least 3 (or so) of the problems
You could also consider letting each answer add two new built-ins, in case anyone wants to implement matched brackets or control flow or something.
@user202729 All of the interesting challenges have mostly been already made. And esoteric math with quaternions will just be ignored
Also, will the initial interpreter be Turing-complete, so that there is a base score for each problem already?
A: Sandbox for Proposed Challenges

caird coinheringaahingBuild a golfing language with me Tags: interpreter answer-chaining multiple-holes (maybe) Our community has found it necessary to create a set of languages designed specifically for golfing, "golfing languages" as we call them. Such languages have evolved from the once brilliant, now clunky Gol...

@cairdcoinheringaahing there's also the question of the memory model, which I think is something the base interpreter already needs to provide.
"If you really wish, you may add a code page (or equivalent) as well as a command in one turn." I have no clue what this means.
Also, get rid of the multiple holes tag. The multiple challenges are more like test cases in this context.
You'll also want to allow people other than the last answerer to golf the current solutions further (and then require the answerer to edit them in and lower their score) so that people can't do a suboptimal golf in order to make a future solution easier. In fact, you might even want to require that the updated solutions wouldn't have worked in the previous version of the interpreter.
(instead of the current "They do not have to use the new command", which seems counterproductive to me)
@MartinEnder clearly Charcoal is the ideal language for this since it is 36 bytes shorter than the next shortest in Bake a Slice of Pi :P
1:48 PM
I am, for an intristic, going to calculate the most commonly used pairs of letters (like, say, ca or is). What'd be a good resource to use to actually calculate this?
Now that I think about, Bake a Slice of Pi was the reason I added Wolfram to Charcoal in the first place :/
aka, i just need a datapoint to feed into a program that can do it for me. (I'm writing the program)
oh neat
@cairdcoinheringaahing oh, I just saw "Regarding the new interpreter, you may use any language you wish." That seems like it'll get messy. It means every answer has to reimplement the entire language if they switch languages, which is bound to introduce errors and is a lot more effort than necessary. It also makes it hard to judge what "a command" is.
I think you should start from a more rigid framework, where the base interpreter is fixed (and written in a popular language), as is the character set, and each answer may add a function implementing one more command.
Then you could also limit the size of each implemented command to something like 256 or 512 bytes (which should be plenty).
1:52 PM
About character set... that's possible, but will make commands unintuitive.
Sure, you could also use bytes and let the user pick a Unicode glyph to go with that byte.
(I guess that's what caird meant with "you may add a code page".)
For task 7, I'd definitely prefer output the sign.
@MartinEnder I like the idea very much, but I want an opinion from a mod: Might it end up closed for being multi-part?
not from me
Neither from me
this isn't a multi-part challenge. the challenges are really just a test battery here.
1:58 PM
Oh, then caird shall consider adding that tag
@cairdcoinheringaahing I'm also not a huge fan of having the 20 problems as links to other challenges. a) it requires people to read a lot of challenge specs, b) what if those challenges get changed (or worse, deleted)? you can reuse the problems, but I'd try including each problem as a very concise spec in the challenge itself. especially with a common header/footer for admissible I/O formats.
having randomness in the test cases means that the base interpreter already needs a built-in for RNG. not sure whether you want that.
Can users with <100 rep edit their own CW answer?
users can always edit their own answers.
someone said "add C# interop"
now I'm sitting here staring at the abomination I'm in the process of writing
2:15 PM
@ASCII-only add interop for anything that compiles to a object file
Bam, even worse mess
@moonheart08 eh, that's for later wait can you give an example of a non-c# object file
Uh, a object file compiled from C/C++/Rust/D/etc
in this case, just your standard shared library
I'm probably going to have a much more abominable abomination once i'm done reworking QUARK to be a compiled language.
At least the compiler is somewhat streamlined (actual compilation is put in it's own rust module so i can replace it later if needed)
@moonheart08 yeah that won't work if the interpreter doesn't know the signatures for literally everything
@ASCII-only actually, it can.
there's a reason libraries have symbol tables
and said symbol tables often contain info about the signature
(other times it's a guessing game, and it's up to the supplied program to actually provide the signature)
@moonheart08 exactly?
2:21 PM
i just talked myself back into your problem
for .NET, unless you want to basically copy the entire PInvoke, you basically can't guarantee you can get the signature
QUARK is going to cheat, and is going to require LLVM bitcode files for anything you want to use alongside the base language :p
...i just realised that the stance on using libraries for codegolf is kinda a grey spot
The rules loosened up a lot recently, so may as well just assume
2:39 PM
@moonheart08 It's not really a grey spot IIRC?
I just don't remember any rules on it. I'll doublecheck
It's just that if you use a library, the library becomes part of the submission language essentially
3:02 PM
I'm trying to parse Jelly like code. Turns out that jelly like code is absolutely horrifying to parse
@Dennis help i'm lost (sorry for the ping) but how the heck do you parse jelly code?
@moonheart08 "Postfix parser".
(you can look at the source code for more details)
oh ok. Silly me.
Sorry dennis
@moonheart08 IDK what you're doing but you might be able to just use Dennis' Jelly interpreter as a library and not reimplement parsing.
Cant. Using rust :p
@Pavel Using a whole library just for parsing postfix code is... somewhat overkill.
I'm just the kind of person who's written a miniature bootloader, yet has never parsed postfix code.
It depends on what you're actually trying to achieve
Like I'm really intrested in why you're parsing Jelly in the first place
I'm not parsing jelly. Working on my own thing
Jelly is just what came to mind for the syntax
> Please note that Jelly is currently under development; features will change or disappear without prior notice.
Given that any breaking change might break a TIO permalink, I find this extremely unlikely
3:23 PM
@Pavel Actually the code would be still there, just the output is different. Only if someone re-run it (as TIO caches).
@user202729 TIO caches only the last ~1000 runs, after that they're gone
They don't tend to last very long
@MartinEnder So originally I was thinking Python or JS as they're used quite commonly, but wasn't sure whether it should be available for all languages.
@MartinEnder Adding the builtin seems to be the biggest issue with the challenge at the moment, especially as control flow doesn't really count as a builtin. I'd (ideally) like to have an original idea regarding the memory model, but if that fails, a stack is a solid fallback. The requirement that updated solutions mustn't work in previous interpreters is a good idea, thanks!
Q: Workplace Puzzles I

SouthpawThe Challenge: Write something that takes a string as input, rot13 the string, reverse it, output the transformed string. A program was considered valid if it could handle all the characters here: abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ 1234567890!"£$%&*()-=_+/?.,;:'@~#<>^|`¬{}[] ...

@NewMainPosts That's a dupe of at least two challenges :/
@MartinEnder That's another thing I'm still undecided about. Ideally there'd be one language which everyone can use, but CS doesn't work like that, plus I mainly use Python, so I think Python might be a good base. Thoughts?
3:40 PM
@cairdcoinheringaahing I would provide a bunch of memory storage options and let people choose what to do with them. E.g. a deque, an infinite tape, 5 registers and a finite indeed array.
@MartinEnder How do you mean? As in during the chain or while in the Sandbox?
As part of the base interpreter.
I'm on mobile right now, I'll explain it in more detail when I get back to a PC.
4:11 PM
A: Sandbox for Proposed Challenges

SokSelf Numbers A self number (wiki link) is a natural number that cannot be expressed in the form n + <digit sum of n>, for any natural number n. For example, 21 is not a self number, as n = 15 results in 15 + 1 + 5 = 21. 20, on the other hand, is a self number, as no n can be found which satisfie...

A: Sandbox for Proposed Challenges

Dom HastingsQuine Suite code-challenge quine restricted-source We've had a lot of quine challenges, but a lot of the quine formats are similar, lets create some variance in our quines. Your task is to create a selection of programs, all of which output their own source code, however, none of these pro...

4:35 PM
Q: Does anyone know how to remove a massive folder from github without using git? I tried with git, but really don't understand it
@cairdcoinheringaahing git rm -r path/to/folder
Then add it to .gitignore
+1 for using git. Github isn't a very good UI
meh, I don't like using the command line much, especially Windows's cmd >_>
@EriktheOutgolfer Installing git for windows installs mingw
git is useless for me. It's perfectly fine cloning the jelly repo, but not one of mine ಠ_ಠ
4:43 PM
@cairdcoinheringaahing How? What doesn't work?
ಠ_ಠ Never mind, I mistyped my username facefloor
Good job
5:42 PM
@cairdcoinheringaahing happens way more often than I'd like to admit.
@EriktheOutgolfer yay for PowerShell (although that has its own problems)
powershell isn't at all convenient for such tasks
such tasks? using git?
inb4 "Powershell is great and does all things."
well, basically calling external utilities
it's just so slow at least for me
@cairdcoinheringaahing so what I meant is that you could provide a base interpreter that already has all the basics in place (reading and parsing the source code, providing some functions to retrieve input and print output, as well as various memory stores and some functions to work with them). when you have that, each answer only needs to define a function that implements one command, in terms of those framework functions you provide.
5:47 PM
So it already has, say 8, commands, each of which operates on a different memory model?
Or something similar?
@EriktheOutgolfer Aww, man...
TFW Python tells you a comment is invalid syntax ಠ_ಠ
there's such a thing called a magic comment
@cairdcoinheringaahing it doesn't actually have to have those commands as part of the language yet. but the interpreter already has functions to access those memory models which can be used to implement new commands.
5:59 PM
it can be assigned to, so it can be considered a variable
ಠ_ಠ TIL
And I say I know Python...
I always assumed that for implied a declaration, and couldn't be used for a variable that was already defined.
That actually makes it a lot easier to create a language with variables, because you can simply store them in a dictionary :P
6:44 PM
Isn't that literaly just Python with globals()?
Q: Add Rainbow Pairs

Dan9erThe rules are as follows: Take in a standard input of integers and add the first and last numbers, the second and second-last numbers, and so on. The program must output the numbers separated by spaces all on one line. If an odd amount of integers are inputted, just print the lone integer at th...

00:00 - 19:0019:00 - 00:00

« first day (2606 days earlier)      last day (2256 days later) »