« first day (2438 days earlier)      last day (803 days later) » 
00:00 - 16:0016:00 - 00:00

12:04 AM
Oh, that looks kind of fun :P
12:19 AM
@Downgoat SO seems to say you need a 303 for that
12:47 AM
Playing /bin/bash as u8 raw audio actually sounds pretty interesting
@KennyLau tiling of the plane with heptagonal convex polygons
it can be done! :D
wait what
does it extend to infinity though
12:50 AM
they just get stretched very, very thin
@orlp trippy
@orlp Not really regular or even semiregular
@LegionMammal978 did I claim such things?
Huh, a file called ~/Downloads/huge-prime.html sounds like a buzzing insect
Has some weird faint overtones
12:54 AM
@LegionMammal978 huh?
@orlp Been piping random binary files into aplay (which treats them as raw u8 audio)
@orlp did you see my comment?
@KennyLau an older version required !! to get unique values, but this version doesn't anymore
@LegionMammal978 this is awesome
most things just sound like static cuz they're effectively random tho
but sill awesome
12:59 AM
@Maltysen Mostly compressed resources which do that
A small example, using the /usr/bin/aplay binary itself:
@ATaco wuuut
@ATaco Nice
A little laggy right now, if you shrink your window it'll run a bit faster.
@Doorknob do you know how to use sasl w/ irssi?
@user2284570 it's not that common really
@Christopher2EZ4RTZ why? there's nothing obvious wrong with it
1:07 AM
Dang it
Just tried to do oldNewPipe=newPipe;newPipe=function(){oldNewPipe();oldNewPipe()}
Just made my browser freeze up
This might be assigning newPipe before oldNewPipe is called, because Javascript
»    a = b; var b = function()end; console.log(a);
«    function b()
ReferenceError: b is not defined [Learn More]
function b(foo,bar){/*foo the bar*/};a=b;b=function(){};console.log(a,a.length);
function b() 2
So it's definitely setting a to the initial b
Yep, my testing confirms.
And b is being set to the new b
1:38 AM
Coke has release "Coke No Sugar Plus Coffee" in Australia.
It's also in the UK, apparently.
The taste is rather different.
coke without sugar? isn't that just water?
still has the flavoring, and color
Is Coke No Sugar not in America?
1:43 AM
@ATaco doesn't seem to be in uk
only japan and AUS rn
according to the googles
@ATaco diet coke is, not coke no sugar
Code No Sugar is gr8.
It takes like coke, but contains Phenylalanine, which is an antidepressant.
so antidepressant coke
Only mildly, but yes.
interesting idea
2:01 AM
So it's Diet Coke
Nope, different formula.
Phenylalanine is also in Diet Coke though
(Why tho)
fun fact: (normal) coke contains so much sugar that it would make you vomit if there were not the phosphoric (?) acid
Then again, it's probably almost identical.
2:04 AM
just another name to get more money
Besides the fact that the one I'm currently drinking literally contains coffee.
It's a part of aspartame
@xnor I'm planning on making a problem on anagolf. What is "deny exec"? I couldn't find any references
@ConorO'Brien it means that command line exec doesn't work (so Python's exec still works)
command line exec being...?
2:08 AM
Shell access
oh ok
I tried solving This Challenge in TacO, Managed to get all the primes less than or equal to n into a list, then remembered I have no way to evaluate the product of a list.
@ATaco a moment of silence for your loss
@ATaco meaning it's impossible? or you haven't devised one yet?
2:13 AM
I haven't devised one, and it might be impossible.
likely impossible
TacO is quite intentionally not TC.
Also, Arithmetic, not Arithmatic
@ATaco I mean if you have primes...
2:14 AM
Primes are easy to get, surprisingly.
so have you proved TacO's non-TC quality?
No, but there's to my knowledge no way to loop infinitely.
I don't think that's a requirement of TC-ness; I'm not sure though...
It's a strong indicator, at the very least.
let me take a crack at it >:D
2:15 AM
This code gets all primes less than or equal to n.
taco code looks un-taco-like sadly :( I was hoping to make code look like tacos xD
@ATaco 0/10 debugging
TacO is one of my older languages.
It's hard to wrap your head around, hard to write, and any particular problem may actually be impossible to solve, meaning writing it can be both frustrating and worthless.
huh sounds like proton
and positron
2:22 AM
TacO was actually TC for a whole 10 minutes before I revered it.
Why'd you revere it? Was it that good? And did the reverence make it return from deification?
On a more serious note, why did you reverse it?
@Sherlock9 It didn't fit the themes and ideals of the language.
2:23 AM
the ideal of the language is to not work? (jk xD)
POLL: For my new language, should I use identical link chaining to Jelly for familiarity's sake?
be original
hm ok
the thing is
my ideas never are
this language is kind of meant to be a way to shorten Jelly for a decently sized class of challenges
See Jalapeño for how not to do a tacit language.
example programs?
I've decided to use Jelly chaining, at least until I come up with a better idea
I want to be original but I always fail because all of my ideas have been used already xD
also, how is a niladic chain evaluated where there are dyads?
It's not the individual ideas that are original usually, but the combination of the ideas
And how the ideas are used together
2:32 AM
@ConorO'Brien you should keep "deny exec", it stops languages from calling the interpreters of golfier languages
for decisions problems, i suggest including many many test cases
@HyperNeutrino I'd supply some, but I don't remember how to write it.
@xnor how? there's only 3 blanks. (or am I supposed to ask people to filter over numbers from stdin?)
@ConorO'Brien submissions can hardcode the answers and some languages are really efficient at compression
@HyperNeutrino originality comes after being unoriginal many times. (I've made three-ish BF derivatives, 2 ><>/fungeoids, and an arguably worse version of pyth) - not saying my ideas are all that original
@ConorO'Brien yes, unfortunately you're only allowed 3 so usually you put one test case per line
which means there's some boilerplate in answers, but i don't know a better way
2:34 AM
hm. and of course the test cases are the only ways to judge an answer, right?
yes, exactly
How do we define "turing completeness" here?
@ConorO'Brien also, beware the 2 second time limit if you have a lots of test cases
oh, hm...
yeah, decision problems are hard to make work
2:36 AM
@xnor how do people take input in python anarchy golf usually
is it sys.stdin.readline()?
usually while 1:s=raw_input()...
would it be better to do a generation problem? e.g. print all n-type numbers below X?
@xnor so it doesn't have to terminate?
it terminates with error, which is fine
2:37 AM
see this for comparison of other input methods: noe.mearie.org/python_code_golfing_tips
i.e. if you need to take multiple lines at once
@ConorO'Brien yes, or maybe just fix a big X and have a fixed output
do fixed output submissions get more time?
no, always 2 seconds
hm ok
is your sequence hard to compute?
sometimes requiring efficiency is interesting
I don't think so. it's just involving prime factor trees
2:52 AM
@xnor do you suggest endless or with a deadline?
@ConorO'Brien i'd generally suggest a deadline of 1 or 2 weeks
with endless, people think they'll do it later, then never do
and it's nice to see answers after
3:04 AM
CMC: Make a stack without using a list datatype.
cod egolf?
what operations should be implemented?
stack of integers?
so many questions
Codegolf, and just push pop and a way to determine if it's empty.
Stack data can by anything, even bits, but there needs to be more than 1 unique setting.
can I assume the user provides allocated memory initially ?
No, Stacks should be dynamically sizable.
it will be, but can I assume the initial memory is provided?
e.g. you give me a pointer allocated with 4 bytes, and I'll dynamically resize it if necessary
3:07 AM
You can assume the memory for the stack itself, but it's data should be dynamic.
Although there's no particular restriction on stack size, but most methods that would abuse this would likely be list solutions, thus invalid.
define list datatype
If you're asking that question you're almost definitely going to be using a list datatype.
that's not true
does list == array? how about a region of memory? is that a list?
In this case, yes.
The expectation is to manually define a linked stack.
3:27 AM
@ATaco should we handle popping from an empty stack?
Try it online! probably suboptimal
3:41 AM
@ATaco Here you go:
public class StupidStack {
    const string StackFile = "stack.txt";
    public StupidStack() {
    public void Push(int i) {
        File.AppendAllText(StackFile, i.ToString() + Environment.NewLine);

    public bool IsEmpty => File.ReadAllText(StackFile).Trim().Length == 0;

    public int Pop() {
        if (IsEmpty)
            return 0;
        var text = File.ReadAllText(StackFile);
        return int.Parse(text.Split(Environment.NewLine).Last());
@ConorO'Brien Undefined behaviour.
4:02 AM
@xnor, wait the while 1: is making me put the body on a seperate line, is this normal?
4:33 AM
nvm, that was just my interpreter being weird
I tied u for digit tally tho :D
5:20 AM
@Maltysen hmm, it should work as usual as long as there's no control flow on that line
2 hours later…
7:12 AM
Anyone remember the challenge about reading ascii-art text? The letters were written in fixed width using #-signs...
A: Sandbox for Proposed Challenges

Stewie GriffinParse written integers Take a number in the range 0-100 as input in its written (text) form, such as one, eleven or twenty-three, and output the value as number, such as 1, 11 or 23. Builtin functions: Built in functions that can transform a number to text (1 - one) or transform text to numbe...

7:27 AM
Never mind, found it...
Is this one ready? Is it good? 2 upvotes and the comments are addressed.
A: Sandbox for Proposed Challenges

Stewie Griffin90 % fail this challenge! I guess I'm not the only one that have seen this kind of image on Facebook (and other sites). The picture above was posted 12 days ago, and has accumulated 31 k comments. Some answers: 0, 4, 8, 48, 88, 120, 124 and so on. Challenge: The mathematics in the question...

@StewieGriffin I'd also recommend allowing N as a fraction rather than as a percentage. The ×100 or ÷100 operation isn't interesting.
7:52 AM
@StewieGriffin Suspected copy-paste mistake in the second example (5 should be 6)
Q: rotate bit array

xakepp35Given an array of tightly packed bits. Also its length is known and is fixed, but is not guranteed to be multiply of 8. Task is to rotate array, shifting bits by a specified amount positions in either direction. Positive shifts go to the left and negative values go to the right. Bits should circu...

@Adám Edited. Thanks :)
@Emigna You're right. Thanks :)
8:14 AM
@orlp ... that suggests you know someone who found the rules of Jelly intuitive...
8:26 AM
Q: Chess Related Coding Problem

Sasha RSuppose a new piece namely the camel is introduced in a chess board.Camels can move from a position(x,y) to (x+1,y),(x,y+1),(x-1,y),(x,y-1).How many such camels can be placed in a NxN chessboard so that no two camels attack each other? For a 1x1 board,there can be only 1 such piece. For a 2x2 bo...

@StewieGriffin Is it OK that I "borrowed" your tagline from the bit rotating problem?
8:56 AM
@HyperNeutrino iirc if you can solve the halting problem for TacO, it's not TC
That is true
1 hour later…
@orlp how does it work?
@LeakyNun same algorithm as everyone
I highly doubt everyone uses the same algorithm
factor s=p-1
check pow(b,s/f,p) == 1 for any factor f
@LeakyNun however, for each p I only ever use two mods :)
@LeakyNun the trick is montgomery multiplication
it takes some precomputation
but once you've done that you can calculate a*b % m without any modulo, just multiplication (or rather, you change the modulo from mod p to mod R, where R is a power of two, so it's free)
10:10 AM
@LeakyNun can you run the code on your machine?
I want to know how much faster it is than your code
not now
@orlp right, I’m stupid
you only ever need two mods
@LeakyNun not without montgomery multiplication
why not?
I mean, if p-1 has three prime factors, you only need the first two
also that's not what I meant
nvm im wrong
10:15 AM
int modpow(long b,int e,long n){
	long r=1;
		if(e&1) r=r*b%n;
	return r;
do you see how often you call %n?
        for (int i = 0; e; ++i) {
            if (e & 1) r = montmul32(r, pows[i], p, pneginv);
            e >>= 1;
(I precompute pows, but that has nothing to do with it, you could do that too)
pows of what?
// Returns (ab)R mod n given aR mod n, bR mod n, n and -n^(-1) mod R, with R = 2^32
inline uint64_t montmul32(uint64_t a, uint64_t b, uint64_t n, uint64_t nneginv) {
    uint64_t T = a*b;
    uint32_t m = T*nneginv; // m = T*-n^(-1) (mod 2^32)
    uint64_t t = (T + m*n) >> 32;
    return t >= n ? t - n : t;
@LeakyNun b, b^2, b^4, b^8, b^16, etc
10:17 AM
but note how montmul32 calculates a*b mod n (although a and b are in a different form that requires precomputation) without any modulos
I don’t really know the algorithm
In modular arithmetic computation, Montgomery modular multiplication, more commonly referred to as Montgomery multiplication, is a method for performing fast modular multiplication. It was introduced in 1985 by the American mathematician Peter L. Montgomery. Given two integers a and b and modulus N, the classical modular multiplication algorithm computes ab mod N. Montgomery multiplication works by transforming a and b into a representation known as Montgomery form. For a modulus N, the Montgomery form of a is defined to be aR mod N for some constant R depending only on N and the underlying computer...
@Adám Yes. I deleted my answer though. I think the challenge should be deleted too, to be honest.
I’ll read it later
11:11 AM
Q: Shortest code to println the first byte of a function’s code

user2284570Rules The code must be a function which takes another function as parameter and will print the first cpu instruction byte of parameter such as : . func dummy() { } print_first_native_instruction_byte(dummy) would print : 0x90 which is a nop instruction on x86. You must don’t use any p...

Q: 90% fail this challenge?

Stewie GriffinI guess I'm not the only one who have seen this kind of image on Facebook (and other sites). The picture above was posted 16 days ago, and has accumulated 51 k comments. Some answers: 0, 4, 8, 48, 88, 120, 124 and so on. Challenge: The mathematics in the question doesn't make sense1, so we ...

11:53 AM
I think we should reopen this, as it has been edited and now it's a perfectly valid challenge.
This is a great case of "edit-it-to-help-new-users", we should do that more often.
@Mr.Xcoder Yup, nice team effort there.
@Adám Yeah, you (and others) really made it look better
What's wrong with this Jelly? ²‘HḞ
Links probably, let me check
@Adám You try to square, increment, halve and ceil right?
@Mr.Xcoder Floor, but yeah.
12:00 PM
Thanks for the positive feedback I've received on my challenges in the past. I had fun writing them. Unfortunately, I've lost the hang of it so I've decided to stop. I might give it a try some other time.
@StewieGriffin Uh, you're leaving PPCG?
@StewieGriffin So you are leaving?
I'll stick around and post some answers when I can, but I won't post any challenges for a while...
:( I'll miss your challenges, which are very fun to solve :D - In fact I am working on an answer rn
It's fun to make challenges when people like them. It's not that fun when people don't...
12:03 PM
@Adám Hahaha
You have put the input in the "Input" field
@Mr.Xcoder ?
@Mr.Xcoder Thanks , I believe I've written a few good ones in the past :) But it's been a while...
@Mr.Xcoder Oh :-(
@Mr.Xcoder Right, I understood.
12:04 PM
ok :-p
@Mr.Xcoder Thanks. At least my Jelly was alright. Dibs on it when the challenge re-opens!
Yay, my first non-trivial Jelly. And correct on the first try too.
Sure, I'll let you post it :) - I'll go with Pyth and Husk (and maybe 05AB1E)
Well, at least somewhat non-trivial
The userscript fails to detect the deletedness of posts :-(
Huh? - Am I the only one that fails to understand this?
@Mr.Xcoder Compiling makes human readable code become "stuff" in memory. Your task is to return the first byte there.
12:14 PM
Q: Shortest code to print the first byte of a function’s code on stdout

user2284570Rules The code must be a function which takes another function declared in the same program as parameter and will print the first cpu instruction byte of parameter from ram such as shown in the pseudo code below (for example) : . package main { func dummy() { } println_or_printf_first_...

@Adám 3 bytes for your Jelly answer: ²HĊ
@Mr.Xcoder Thanks.
It's double in Pyth >_>
.E**.5 --- Hmm, I can do better
12:27 PM
@Mr.Xcoder I have Jelly and J submissions ready for when it re-opens.
@Adám I got Pyth rn (and maybe Husk, not done yet)
And will try in Ohm, Gaia, M and Pyke too
Done in Pyth hahaha
@Adám It's open!!!
@Mr.Xcoder Thanks. Ugh, I have to wait 60 seconds.
@Adám Why do you have to wait?
I posted mine 3 seconds apart from each other
@Mr.Xcoder Anti-spam, I guess? It is my third answer.
When a man posts 3 answers on stack overflow it gets suspicious.
12:35 PM
We are not SO
@someone We are not men (we are code golfers)
ಠ_ಠ Husk uses for squaring.
I still have no idea how to ceiling on C#. Currently i use System.Math.Ceiling(), what is very c#. The Husk square operator is amazing.
@Mr.Xcoder That literally says "square". Can't deny that.
@someone that is how to ceiling
12:45 PM
I've came up with n=>n*n/2+n%2/2, and that works.
(that is an answer, not an ceiling function)
@Mr.Xcoder what is this for?
hi do any of you know a good way to temporarily kill tiredness for like 12 hours or so
(other than sleep)
Stare into a wall for 12 hours. After a while you'll fall asleep and wake up not tired at all.
> (other than sleep)
12:49 PM
@HyperNeutrino Grab an energizer
what is that
Although i do not recommend it, really >_>
@HyperNeutrino 24oz. of Red Bull
Energizing juice
you know I think I'm feeling less tired already as the sun came out in my class's window
because when I biked here this morning it was about 10m visibility of fog but that's all cleared now
12:52 PM
@HyperNeutrino Caffeine. Lots of it.
@HyperNeutrino (yeah, why would you?)
actually this is really weird i feel like 100 times less tired than a few minutes ago for no reason
@Adám because class xD otherwise really I would
(:o hyperneutrino sleeping? what is this)
looks like 3 is the limit
you mean three bytes for camels?
it could be shorter if there were a right-shift-by-one operator in any language...
@LeakyNun MaximumNumberOfCamels@ ?
(edit: I'm dumb, that's floor)
@Adám ಠ_ಠ
12:59 PM
@HyperNeutrino (not dumb; tired)
@LeakyNun I still think 6 is not the limit for Pyth :-)
No such thing as "being tired"; you're either full of caffeine or not.
@J.Salle There is no such thing as tiredness, only LBC (Low Blood Caffeine).
1:07 PM
I've actually never had decent amounts of caffeine at once before
by decent I mean like one cup of coffee
I never ever had any coffee
the only "coffee" i've had is starbucks (because my parents like starbucks so on occasion I can buy some from the bookstore when we go there to get books) but that's like .01% caffeine lol
Palindrome reputation!
12821 is an interesting number... I somehow missed the moment when my rep was roughly equal to Dennis' id >.<
1:11 PM
darn :P
Caffeine pills are way more powerful than even the strongest coffee.
Poll for my history essay, should I write it quickly and then edit it or spend more time writing it and make it better the first time
@HyperNeutrino Quick-write + edit.
@Adám Those can cause death from caffeine overdose lol
@Adám okay I'll try that (I've always just not gone back to revise it lol)
Quick write + edit
1:13 PM
@HyperNeutrino Yes, part of the excitement. (LD50 is about 40 pills a day over long time.)
> part of the excitement
On rats?
When it come to history, I do Quick write + edit
it takes 82 cups of coffee to kill an adult man by which time that much fluid may become slightly uncomfortable
don't ask why I know that number off the top of my head
You know 7 liters of water can kill a person, if drunk in less than 3 hours? (Btw excuse the grammar, I feel like it is incorrect rn)
1:18 PM
It would be "drunk" which may sound slightly awkward, which is why most people just go with the grammatically incorrect past tense form of the verb instead of the participle. But this is chat, grammar is not really a thing here :P
Fixed, thanks
is that a hi, a bye, or a something?
@HyperNeutrino 🙋
1:28 PM
i is confuz
brb switching classes
the annoying thing about switching classes at my school is that the school is divided into two parts separated by a single central staircase, and within half a minute of the bell going off that staircase becomes a flood of about 90% of the school population so it takes like 2 minutes to get through a single flight :P
That's the problem with polymorphism
If it's not set up right, it's a pain switching classes
it's because the second half was built kind of as an afterthought because the school population expanded too quickly
originally this high school was built in the middle of nowhere back in 1960 with hopes to draw more people into the region
it was too successful with that so it had to expand, and now it's beside a bunch of residential areas and near two universities
That's how it usually goes
1:38 PM
and now they need to demolish it because it's no longer safe by the new safety standards (so I could die today from falling through a staircase, who knows) and rebuild it, which I wish they did earlier because this school has air conditioning in about 5 rooms xD
2 hours later…
3:21 PM
someone please post a math challenge :(
no good new challenges :(
@Geobits CMC: Switch class. (Given the name of a class, create a new class with the same members)
Hmm. Well this sounds like it works >_> A extends B
Although B extends A sounds much better to me :p
@Geobits ?
3:37 PM
@Adám that's not as hard as you may think
I have a Haskell question: What's a golfy way to write a two-argument function call? I currently have a@b=div(length$a!b)(length$p a)
@Adám see here for an example (python 2)
Also, it turns out that I need to do float division and (length$a!b)/(length$p a) throwing a no instance for (Fractional Int) error at me
3:53 PM
@Sherlock9 If you need Fractional, you could define f=fromIntegral.length somwhere else. Although there is probably a golfier way.
Ah well, I also forgot that it was a fastest-code challenge :P This one
@Sherlock9 f x=sum[1|a<-x] Then a@b=(f$a!b)/(f$p a)
@Sherlock9 Oh nvm
Well, is sum faster than fromIntegral.length?
Probably not
Just tested, definitely not
Well, this is what I have so far
p a=[x|x<-[2..a],and[mod x y>0|y<-[2..x-1]]]
a!b=(f[q|q<-p a,take 1[i|i<-[1..q],b^i`mod`q==1]==[q-1]])/(f$p a)
Since I'm calling p a twice, I'm running through [2..a] twice. That's a lot of overhead
00:00 - 16:0016:00 - 00:00

« first day (2438 days earlier)      last day (803 days later) »