« first day (2375 days earlier)      last day (1636 days later) » 

@ConorO'Brien it would be longer, but also cleaner... this... is...
it's not talking about the code, it's talking about the memory layout. :P
@ConorO'Brien this is, uh,
that sounds like a terrifying criterion
I thought there was a standard loophole about cryptographic things, but I didn't see it on re-reading the post. In particular it is trivial to create an answer to this cops and robbers question that requires reversing a hash to crack
12:21 AM
@Liam The author banned it, I believe
@Liam He also started a meta question about it: codegolf.meta.stackexchange.com/questions/13421/…
I'm not too good at reading. Thanks
It's pretty implicit that crypto is banned from Cops & Robbers, even if it's not explicitly, it's just not very good sportsmanship.
yeah, that's why i thought it was a standard loophole
especially because it comes up in a decent number of cnr challenges
there's always that one person who just wants to win, unfortunately
It might be worth closing by default?
12:23 AM
I don't really see "Winning" as a major factor of CnR, as shown by the recent meta post.
@ATaco or at least, for cops, to identify a single winner if multiple are safe
Like writing challenges is it's own reward, Being a cop in CnR is rewarded by writing interesting challenges to crack.
Of course, that's my opinion and many people just want the green tick.
A: Sandbox for Proposed Challenges

GryphonFloor Planner! ascii-art code-golf kolmogorov-complexity Disclaimer: The story told within this question is entirely fictional, and invented solely for the purpose of providing an intro. I have a friend who is an architect, and, after explaining the concept of code-golf and this site to h...

fun fact: 7 can be encoded in base -1: 1010101010101
I'm not going to question base-1.
12:37 AM
Could anyone help advise how I should edit this question if it isn't acceptable codegolf.stackexchange.com/questions/136269/…
@William Is this a challenge..?
@ATaco that's why its codegolf? Shortest wins
You need to explain that A: Codegolf is the winning criteria, and B: How you want LISP to be converted to JSON in detail.
I have added that the shortest one wins. Plus I have the code gulf tag. What else should I do to clarify how to convert it from LISP to JSON? I have given an example how to escape the sequences that are necessary.
Assume that everyone has no idea what either LISP or JSON is.
Hi laser eye guy Thomas o/
@ATaco do I need to post spec type documentation
@William In future, try posting a question in The Sandbox
You simply need to clarify what constitutes the minimum for that challenge. People will generally not put any more effort in than that.
12:48 AM
Someone should continue the OEIS challenge. The next one is so easy it's almost not even funny xD
"0 Bytes, 0 Indexed"
Well, with 1-indexing it's 0 bytes in almost all golfing languages
I know it's obvious, I just feel it needed to be stead.
not in 05AB1E because it's weird
Write the challenge, but write it with PopCon.
12:50 AM
I'll see if I can do that next one in BF
StepHen is going to try Taxi-ing A000027 xD
As in, can I find the algorithms for input/output in decimal
I don't think you need to but since you can't do 1 byte anyway you might as well xD
anyway brb o/
Well the sequences have to work up to 1000 right?
Otherwise you could just do ,.
the next one has a formula yay
1:00 AM
@HyperNeutrino I did it fast enough good xD
yey xD
@HyperNeutrino Every number can be encoded in base -1
and in infinitely many ways right? xD
because appending/prepending 11 does nothing
so you can 11 it infinitely :D
1:08 AM
phinary is better
no base (i-1) is better
or was it (i+1)? o_O
you mean the quatimaginary thing?
something like that
someone ought to make a language where numbers are in phinary
@HyperNeutrino also 00
1:12 AM
1:26 AM
Why don't we have code.golf as a shortener
oh right it must be taken
screw it
nobody would buy that
@Nobody the real one is codegolf.se
@Nobody as cool a site we are we aren't going to get the name of the sport
oh yeah
Much better
Consider me blind
1:36 AM
I would have preferred s.tk/ppcg but beggers can't be choosers.
Lets wait for a golf.se to come and it gets s.tk/golf, and we get s.tk/ppcg
@ATaco people who constantly get on hnq so much that it's an issue*
A: Sandbox for Proposed Challenges

GryphonEfficient Robot Movement code-golf string My boss has gotten a new toy robot, and he wants me to help program it. He wants to be able to enter simple arrow instructions to get it to move. These instructions are: ^ (for move forward) < (for turn left), and > (for turn right). However, now that...

i dont understand this next oeis sequence
namely a000236
anyone pls i need a character for charcoal for all/every and any/some
@Mr.Xcoder Yeah Charcoal doesn't have any room in its codepage to spare for things like factor :P (although I should probably add more arithmetic builtins soon it's not very high priority)
1:54 AM
hi I have a maths question
lol just found this again
@ASCII-only for all symbol and exists symbol? Or needs to be ASCII?
@trichoplax well in charcoal ASCII characters are part of strings so :P
@trichoplax hmm those seem a bit specific (to math) though, i want a more visual representation
also maybe zip as well, there were suggestions a while ago but idk if they're good
hmmm I will help you
also gtg
1:58 AM
do you have filled circles and unfilled circles?
maybe try those
unfilled for any and filled for all
Hmm what would be the mnemonic though
How about Freehand circles :P
@ASCII-only oh, because filled shows all of them, and the circled shows that at least one of them is, but not all of them
If you can find a Unicode one, sure
@DestructibleLemon whaaat
it's hard to explain
I mean, the filled circle kind of shows that all are in the set
whereas the boundary shows there is a valid one within the boundary
● ○
actually... maybe these are preferable? (vertical offset is weird) ◯⬤
@ASCII-only does that make sense?
2:14 AM
@DestructibleLemon boundary shows there are none
I disagree... but whatever
@DestructibleLemon well to me unfilled means it's empty
maybe get a dotted circle
The boundary is just there so you can see the circle
how about that
2:18 AM
@DestructibleLemon this would work
bigger one: ⨀
we did it team!
size comparisons, with double dotted circle bonus, and right dotted circle bonus: ⚇⊙●⚆⨀⬤
oh suddenly that character does not render?
weird bug
anyway don't use that big one I guess
you could use a black dotted circle for not all, a white or crossed circle for not any too
Nah that's wasting too much of the codepage
my interpreter is working without me trying too hard and it's blowing my mind
@ASCII-only yeah ok
@StepHen O_o how
Also what interpreter
2:31 AM
@ASCII-only I don't really know, and it's for an esolang I brainstormed today
It's a crazy one too
@StepHen interpreter in one day how D:
@ASCII-only I don't know that either
yeah one time I thought an interpreter was working
although it's fairly simple
2:31 AM
I made an interpreter in 20 minutes once
@HyperNeutrino for BF?
Charcoal probably still has dozens of bugs haha
Wait I made this 2 weeks ago? o_O
I thought it was a lot longer ago
There should be an interpreter called ~brainfuck that naturally interprets as many variants of BF as possible, without you telling it what you're giving it
@StepHen not possible
A missed opportunity to call it (.*)fuck
Too many variants use similar command sets
@ATaco you implement it, you name it whatever you want :P
2:33 AM
OK this lang is going to be the absolute worst to code in, but whatever
I was like "I should keep up my record of not using the F word around here" but I've used it once already lol
@StepHen ???
@ASCII-only the one I'm finishing up the interpreter for
I don't even know how it would add yet, although I bet it could
I leave an exception when I'm referring to the name of a programming language.
2:36 AM
@StepHen not true, Malbolge is worse
@ASCII-only yeah it's designed to be bad though
this I just made
and now, it's going to feel like making a TIOQuine without any base-64 shenanigans
but I bet it's turing complete (I'll probably put a challenge to prove it one way or the other)
Let's make a BF variant named Holyfuck
And make it real good
@Nobody how can there be good/bad variants of BF 0.o
Although RProgN2 is really good at quines, I have no idea how to tackle the TIOQuine, because it has no support for zips.
@StepHen Oh
2:39 AM
@Nobody name change?
@ATaco brute force it?
@HyperNeutrino he's Odysseus
@HyperNeutrino Yes
@Nobody I recognized you by your profile description, actually :P I remembered the WIP embedded thing
Also hjroh
that embedded thing was great
CodeReview chat has a bot that chats comments 0.o
2:41 AM
Really want to post a language specific challenge but probably not a good idea
that's not all good tho
@ASCII-only what's the reason it's language specific?
I'm trying to disguise a Charcoal feature request question as a challenge to outgolf Charcoal by as high a ratio as possible
@ASCII-only you mean find a challenge and outgolf Charcoal in it?
2:45 AM
Maximizing Charcoal length / other language length
I think that might go over ok
@StepHen not really? That's not hard :P
Hey Ceres is now capable of primality checks! (builtin)
are they outgolfing a specific Charcoal program?
@HyperNeutrino how did you implement it?
In cQuents I tried to keep to a design pattern of creating sequences and then manipulating them
I may have cached all primes and composites up to 10000
2:47 AM
@HyperNeutrino I noticed
That, does help with some initial speed, I suppose.
but I mean client side :P
@ATaco yeah xD
how do you use it in Ceres? Is is just a function that returns truthy/falsey?
2:47 AM
Sadly Big O notation cares not for caching.
Not really? Just find a challenge where Charcoal would do badly I guess? Or just post two snippets that do anything (I'd probably need to restrict it in this case)
@HyperNeutrino it isn't going to make much of a difference at all
yeah probably not
@StepHen yes
@HyperNeutrino in cQuents I have it so that the only way to check membership in a sequence is to generate the sequence and then use query mode (?)
ah I see
in fact cQuents doesn't even have ifs
2:49 AM
grr for whatever reasons the functions themselves are getting pushed onto the stack o_O
anyway I gtg now. o/
OK, tomorrow night I'm going to release this wacko turing* tarpit I have made
:( I thought was ready to release
@HyperNeutrino you made a mistake
@ASCII-only it is but I don't have docs yet
3:06 AM
oh also @anyone pls suggest overloads for charcoal, add more of charcoal's operators if you want
3:22 AM
@ASCII-only I did?
oh you were talking about that
well yeah duh xD
anyway my wifi might cut out any time now don't know why I still have it but I'm not complaining xD
@ASCII-only how about subtract with a string to remove the last n characters?
@DestructibleLemon :| doesn't really make sense because it's not the opposite of add
plus that's what slice is for
:| ok I guess
how about prepending rather than appending concatenation?
@DestructibleLemon :| idk
3:29 AM
you can just change the order in + to do that
anybody else here watching rick and morty tonight?
well you should add that to the add functionality?
@DestructibleLemon ??????
idk it just looks like you've only got concatenation when the first arg is str
3:33 AM
I don't think so
I'm just too lazy to write the opposite order on the hackmd
4:07 AM
Hello all
Any interesting challenges come in?
so lang as you give me the opportunity... there's this
4:12 AM
Was "lang" a pun or a typo?
I don't know how I typoed that but I managed to anyhow
Wow. Considering that a is nowhere near l, o, or n :D
That's an impressive typo
I am proud lol
Oh, just realizing that I hit 7k about a month ago
1 hour later…
5:44 AM
Q: Find the first duplicated element

Thomas A. AndersonGiven an array a that contains only numbers in the range from 1 to a.length, find the first duplicate number for which the second occurrence has the minimal index. In other words, if there are more than 1 duplicated numbers, return the number for which the second occurrence has a smaller index th...

Is O(n) time and O(1) extra memory possible?
@ConorO'Brien maybe lang is long in German?
@LeakyNun I can think of one way you could, but then again that way I'd say leans towards O(n log n) but depends on what an n is
@Downgoat I seem to have found a solution online
oh my god it's so clever
@LeakyNun does it have to do with giant relatively prime numbers
@Downgoat no, it's extremely clever
one list stores two information
you mean like [1] -> [[1, otherData]]?
5:52 AM
It's not hard, you just follow cycles.
we store which items have appeared, somehow inside the original array
Storing extra data is cheating.
@feersum it uses a clever way
to store the extra data inside the original list
@LeakyNun uhh how is that O(1) memory?
@Downgoat because it uses the original array
5:52 AM
you're duplicating part of array?
that would be O(n)
@Downgoat it uses a clever method to store which numbers have appeared
ok please ping me when you have posted
Brute force is probably shorter, so there's no reason for it to be posted.
@feersum in the spirit of the challenge
@LeakyNun What?
5:55 AM
@feersum posting a solution with O(n) time and O(1) extra memory is in the spirit of the challenge
I would assume you can set an entry to 0
Although 0 is not in the input range.
It's definitely easy if you can do that.
@feersum how?
by cycle do you mean a[a[i]] cycles
I'm not sure how you would do it with cycles
5:58 AM
kinda sorta like how Arnauld was approaching it
though I'm hesitant to call it O(1) memory for larger values
A: Find the first duplicated element

Leaky NunPython 3, 88 bytes O(n) time and O(1) extra memory. def f(a): for i in range(len(a)): if a[a[i]-1]<0:return-~i a[a[i]-1]*=-1 return -1 Try it online! Source of the algorithm. Explanation The basic idea of the algorithm is to run through each element from left to right, keep track o...

@Downgoat here
:39125717 def f(a):
    a = [None] + a
    for i in range(1, len(a)):
        if a[i]==0: continue
        start = i
        while a[i]:
            i2 = a[i]
            a[i] = 0
            i = i2
        if i != start:
            return i
    return -1
Well that was interesting, the reply identifier got itself code-formatted.
@LeakyNun Your solution is cheating because it uses integers 1 bit larger than the input.
@feersum interesting.
you're right.
But I think that is the intended solution. Otherwise it would mean that the challenge is impossible.
What about the one I just posted here?
I'm still reading it
6:07 AM
@LeakyNun you could use 1-indexed lang
@Downgoat wrong reply?
I think it should not be too hard to remove the use of 0 as a sentinel either.
@feersum very clever. I'm still figuring out why it works.
I could use a[1] instead of 0 everywhere.
apparently you used the fact that if there are no duplicates, every cycle will loop back to itself
6:11 AM
It's just basic cycle detection.
@feersum I know how it works. I don't know why it works
@feersum close enough
@LeakyNun Close? Close to what?
@feersum it should return 3
I can't fix it.
Heh, I just looked for any old duplicated element.
It can be fixed.
Instead of returning right away, always run to the end.
I can't fix it because there's no way to guarantee that the elements you zeroed are the first occurrences
@feersum above
6:17 AM
And keep track of the minimum index duplicate seen overall.
Do you want me to fix it for you?
4 mins ago, by Leaky Nun
I can't fix it because there's no way to guarantee that the elements you zeroed are the first occurrences
There is an issue when the first time an element is seen is at its second occurrence.
So it's not that easy after all, with this added rule.
@feersum yep that's what I said
6:34 AM
If we could somehow run the algorithm twice, once starting at the beginning of the array and once from the end, maybe we could do it.
But it's non-reversible.
@feersum an element can appear thrice
2 hours later…
8:56 AM
Woo just hit 5k :)
Hi there
@TheLethalCoder Congrats!
9:44 AM
Q: Count up diagonally!

Mr. XcoderWe have lots of horizontal axis for numbers, but I honestly think they're kind of boring. Your task today is to build me a portion of a diagonal axis between two distinct non-negative integers given as input. How to build a diagonal axis? Let's take an example, with the input 0, 5. Our axis sh...

The solution to my challenge is surprisingly easy in Python
I now have 83 bytes
10:02 AM
@HyperNeutrino APL, 5 bytes: ⊢+⍳∘≢
One of those days today. 3 support calls in less than 2 hours -_-
@Mr.Xcoder APL, 3 bytes: ⊢+≢
@TheLethalCoder Were they interesting?
Perhaps some fun issues should be introduced into the product to make support call handlers' jobs more interesting! :-)
10:19 AM
No just boring routine ones pulling me away from programming.
I know that feeling mate
Q: How to compute a unique set of squares where each side is one of 3 colors

Helen.MafloyIn class we were given a problem to solve and implement in java, and I still cant wrap my head around it. There is a set of 24 squares, where each side of every square has a border of either blue, red or green. Every possible permutation is allowed. I've been trying to use for loops to outp...

One was a four way conversation of me constantly saying I need files x, y and z. Then them going okay heres file x. So I'm having to repeat with I need files y and z etc. etc.
@TheLethalCoder You couldn't have emailed them a batch file that grabbed and zipped them, could you?
No, I needed the files to get the information to be able to do that! And they're so strict they probably wouldn't have allowed it.
10:26 AM
@TheLethalCoder And you couldn't have connected them all to the same line so you didn't have to keep repeating the same information.
@Adám running it without stringify seems to work fine: Try it online!
That's why I try to bother phone support people as little as possible.
@Cowsquack Yes, my mistake, because I used ⎕R for development.
@wizzwizz4 No they just couldn't be bothered to do it properly, they never can be. What annoys me more is that they are IT support for their company and we've told them 100's of times what files we need when they report an issue.
> IT
well there you have it
10:30 AM
@TheLethalCoder They are... what?!
IT support that aren't competent sysadmins?
Okay tech support... They're their companies helpdesk
@TheLethalCoder Ah. Helpdesk.
Do you feel frustrated being a programmer dealing with people who are supposed to be "techy people"?
One of our customers has an IT guy who always brags about his Cisco certification. He couldn't even plug in a router with our instructions
It's not the techy part that annoys me, it's the lack of common sense. And the usual "HELP WE NEEDZ IT NOWWWW!!!!!!" but we'll take 3 days to give you all the information you need.
@TheLethalCoder Have you tried a web wizard that lets them submit a ticket if they upload the correct files (and does simple validation on those files to make sure they haven't uploaded the same one three times or something similar)?
10:39 AM
@wizzwizz4 Oh I have thought about that. Problem is we only have 10 people in the company with a team of 3 developers (me included) that do support, programming, lead etc. And we're always busy on new projects that we never have time for in house things like that.
@TheLethalCoder How do they currently submit things?
@wizzwizz4 Call, email, FTP files if they're too big. No proper process.
@TheLethalCoder File upload from a POST request to a PHP script from a basic HTML5 page shouldn't be too hard. I'll have a bash.
@TheLethalCoder we have around 14-16 employees, and we use ZenDesk, it's a lifesaver
@wizzwizz4 No I know it shouldn't be hard. Half of these companies are sensitive about how they send stuff. So when they upload to FTP it's actually their own one server that we connect too.
10:44 AM
Wooo 2999 month rep :)
Most of them are PCI compliant due to the nature of what they deal with
@Mr.Xcoder Go you
@Mayube We have our own in house helpdesk but nothing for customers to use.
@TheLethalCoder ... That's less secure if anything else is hosted on that FTP server.
@wizzwizz4 Well SFTP and I don't know, it's their server not ours.
@TheLethalCoder ^
We use Jira for development tracking, and ZenDesk for both internal and external supoprt. It's great because we just have a hypervisor monitoring our support email and automatically screening and converting relevant emails to ZenDesk tickets
if they call in with something that isn't "Our system is down, help" we tell them to make a ticket
10:46 AM
@TheLethalCoder Well, if they do something like emailing the credentials to you, everything on that server is potentially compromised.
@Mr.Xcoder Now we need to make one rep to please the OCD in myself. How can we do that?
@wizzwizz4 Oh they're not that daft
@TheLethalCoder +5 rep, -4
@TheLethalCoder Already did, I now have 3009 :)
@TheLethalCoder Get one question updoot and two downdoots.
@wizzwizz4 Please don't.
10:50 AM
@Mr.Xcoder 3009 ewww that's even worse
@TheLethalCoder You can downovote (joke, don't) 4 answers and then I'll downvote another answer and I'll be back to 3k
Too late :)
I reached 3014 :(
@Mr.Xcoder I won't.
@wizzwizz4 That was a joke :P
Someone is serially upvoting my challenges!
@Mr.Xcoder Don't worry. It'll be reverted soon.
You can @ping a mod if you really want it looked into.
There is a voting ring that is actually a ring shape!
Damn too far now we'll have to run a voting ring to get you to the next OCD number.
10:53 AM
(It's pretty.)
@wizzwizz4 No, I don't think it's one person. It's probably because HNQ, because there are 2 of my Qs that get upvoted.
I'd say challenge idea but rounding to the nearest 10 is boring
@Mr.Xcoder Ok. That's not serial upvoting, so it's not a problem.
I know. I was just surprised by the fact I got +15 in 3 mins
Yes, it is because HNQ.
They're both on the first page, so

« first day (2375 days earlier)      last day (1636 days later) »