« first day (2632 days earlier)      last day (676 days later) » 
00:00 - 17:0017:00 - 00:00

12:51 AM
@xnor going from 58 to 55 bytes on e for me was just using some python stuff I didn't know to golf the decimal point
I have also matched you on phi now, by changing from while to exec
1:49 AM
@cairdcoinheringaahing Error in next_index persists.
MCVE: { [ ] } 123 ]
1 hour later…
3:02 AM
@user56656 (What Wizard) Did you miss an e in your name?
@user202729 wait wat
@NewMainPosts re: this challenge, I've already found a loophole-abuse question in Jelly; it's unlikely to be reopened, so shall I post the answer CMC-style?
err, @cairdcoinheringaahing, see the previous comment
Loophole-abuse...? What did you mean?
3:07 AM
the question says "Aside from the above restriction, any other printable ASCII characters may appear in F"
this means that any answer that never produces printable ASCII in the output is a valid answer
Then it's possible to do this in most golfing languages...
or in other words the question is "given an input string, generate a program that prints that string and contains no prinable ASCII"
so yes, most golfing languages can manage it
@user56656 Submission to what?
Jelly, 10 bytes: Oḅ⁹”‘x“ṃ⁹Ọ Try it online!
3:09 AM
@user202729 The closed challenge
do X without Y
I was considering working on an Incident answer but it'd take days
@user56656 (but parentheses are printable)
@ais523 Can incident work with only one character
(and why didn't your chat name changed?)
@Downgoat what's this for
3:09 AM
(the hard part is not "write a program to not use these characters" – that's trivial in Incident – but "write an Incident program to print this string", which is a hard task even in practical languages)
@user56656 no, but it's invariant under replacing any character with any other character
learn Incident
@user202729 Yes but parentheses are required.
it's a puzzle language though (i.e. intentionally difficult to write in)
I have to say I like incident a lot. It's really fun
I'm also really happy it is in the polyglot.
@user202729 Takes some time to update. I can force an update but I don't really care.
I left it until "late" intentionally (I wasn't expecting the polyglot to get that big; hoping, but not expecting)
but I had to take 33 when I realised it was a string that's surprisingly non-painful to output
3:13 AM
@ais523 (and why is your PPCG account different from your network account? I think 101 association rep can write CW as well?)
Incident has progressed to the point where I am no longer sure where tokens come from. I used to be able to see what I was doing to create or destroy tokens by accident but now everything is so carefully stacked that it's behavior is unpredictable.
PPCG account is 11 rep intentionally
this account is from Puzzling and has enough rep to chat
that's the only reason I'm using it here
out of interest, has anyone here actually learned Incident control flow / semantics or the like? or are they just preserving my program literally and simply keeping the token sequence the same?
4 mins ago, by user202729
learn Incident
(the advantage of Incident though is that it basically doesn't exclude anything, the whole token thing was designed so that it'd always be possible to create or uncreate tokens regardless of fixed requirements)
I know how incident works (or I did, need some refresher), but I just adjust the tokens so that 0mo is at the center.
3:15 AM
all this said, I think people were expecting Hexagony to be worse, until we found the Hexagony capsule that's invariant on hexagon size
I don't know, writing linear Hexagony is easy.
It also has to be a polyglot too though.
"capsule" is the term we use in the polyglot for an independent piece of code that does what you want and gets moved to the correct place
I think that writing a Hexagony program for this should be easy.
For what?
3:17 AM
or any place where control flow naturally goes in the polyglot you have so far
@user56656 (if you're replying to me, "the linked challenge")
Q: Script that outputs a script that prints a given input

QuartzicWrite a piece of code that takes a string as input, and outputs a piece of code in the same language that, when run, will output the initial input string. It must be able to handle any combination of characters A-Z, a-z, and 0-9. Example in Python: import sys print "print('" + sys.argv[1] + "...

that one
maybe I'll try it in Incident some time
Oops looks like that I link to an answer instead...
I'd love to see that incident answer.
Oh wait the same language, oh boy
you have the duplicate challenges of a) writing a program in Incident and b) representing a string in Incident
it's like the Unlambda self-interpreter challenge
3:19 AM
@ais523 on the topic of puzzle languages have you tried Lost
no, I haven't seen that one
which is weird as I read basically everything submitted to Esolang, I must have missed it
It's pretty fun if I do say so myself.
Not as hard as Incident though.
ooh, I really like that
would be nice to prove it TC
although that language is a pain to get arbitrary memory storage in, you'd have to use the Mini-Flak construction
I got pretty close. You can build some program structures pretty easily but I never finished the proof.
@ais523 Huh? could you clarify.
your only data storage is one stack
3:22 AM
:O ais523 is back
so you can't access things near the bottom of the stack without forgetting things near the top
Lost has 2 stacks
oh yeah: i was wondering if you had some tool to help code in incident?
@ASCII-only (ais523 was on PPCG all the time, just not have an account with >11 rep ...)
*back in chat
3:23 AM
this means that you have to do something like use the top stack element as a temporary counter that holds the entire state of the program while you're moving up and down
that's actually a bit harder than the mini-flak construction, but should still be doable
mini-flak's second stack is compile time determined, lost's is not
well, mini-flak has one stack + the third stack (which is now the second)
since you can move freely through space and thus hit the same ( a bunch of times.
lost has one stack + the swap command
Lost has ( and )
3:25 AM
oh, I missed that there was a second stack called the "scope"
that makes things much easier
Which are not mentioned on esolangs...
it is mentioned, just easy to miss
Oh it is.
it should be TC even without, just a pain to write in
Lost reminds me of My Unreliable Past a lot
Yeah. The goal of lost was to make it easy to write programs despite the randomness. I hope to one day make a second version that has fewer commmands available.
3:27 AM
I'll cross-link them
(what is the exception used for?)
> There's also an exception that changes the meaning of an otherwise valid program:

If a command would trivially cause an infinite loop, it is skipped and has no effect. A command is defined as causing a trivial infinite loop if it pushes a value to a stack, it has pushed that value to that stack before, and no stack has been popped since then.
@ais523 What a language name ...
@user202729 do you want spoilers? there are two reasons
Ok, I have to go. Thanks for talking @ais523!
... it is a puzzle to figure out? Then no.
one of the reasons isn't a puzzle, it's basically to rescue polyglots that get stuck in an infinite loop early
the other one is
(that said, you could also break the infinite loop simply by screwing with the tokens in it…)
Incident was designed pretty slowly over a period of months so designs that seemed necessary early in its design sometimes turned out to be unnecessary later on
3:41 AM
@ais523 :| unimplemented
that's fairly common on Esolang for languages people aren't really expected to program in
it shouldn't be too hard to knock up an implementation, but it's almost 5am so I shouldn't do it right now
:| it's almost 5am? what are you doing lol
@ais523 I guess that's true since a lot of Esolang languages are more theoretical/exploratory I think?
many of them can't be implemented
which is annoying for languages like Tableaux which were intended as golfing languages
(it was an interesting idea, working out the most powerful primitives to use for a language with a very small number of commands; unfortunately they ended up too powerful and the resulting language can do things that Turing-complete languages can't)
@ais523 O_o
it only has four primitives: a) zero, b) add one, c) index into 2D memory, d) equals
but the memory works weirdly
3:47 AM
At least it (seems to, I didn't read carefully) formally described, unlike, e.g., English (which is a joke language).
yes, I try to have an unambiguous spec for my esolangs
there's nothing ambiguous or unclear about Tableaux's spec and it's possible to work out how programs behave in it mathematically
it's just impossible to implement, in the sense that for any Tableaux interpreter, there must mathematically be at least one program it doesn't handle correctly
Ok, let me read it...
(the same would be true of, e.g., a halting problem solver; those can be defined in a clear and rigorous way too)
3:50 AM
too many infinites
this sort of thing is why most language specifications ask for unbounded memory, rather than infinite memory :-)
1 hour later…
4:56 AM
CMC: given positive integers n and r, where n>r, return the sum of the products of every combination of r numbers in 1 ... n
i.e. if n is 4 and r is 2, then return 1*2+1*3+1*4+2*3+2*4+3*4
what is it with those spaces
@LeakyNun For readability.
Basically n f r = (n-1) f (r-1) × n + (n-1) f r with base cases n f 0 = 1, 0 f r = 0
5:43 AM
Which leads to this polynomial time algorithm: Jelly, 15 bytes
6:09 AM
Q: 99 Bottles of Beer in Pure Lambda Calculus

Llew VallisYou must write a program in your desired language to return the following output: 99 bottles of beer on the wall, 99 bottles of beer. Take one down and pass it around, 98 bottles of beer on the wall. 98 bottles of beer on the wall, 98 bottles of beer. Take one down and pass it around, 97 bottle...

2 hours later…
7:44 AM
hi @Neil
in response to your comment on my sed answer, I noticed that in the string "Word Icicle!", it removes the second c before the first c, when it should do it the other way around
re: new hello world submission, i just looked at the interpreter
i'll need to go wash my eyes now
8:36 AM
A: Sandbox for Proposed Challenges

flawrCurve Matching code-golfarray Given two lists a, b of the same length n find a third list x of indices such that a(i) = b(x(i)) for all indices i and x(i) <= x(i+1) for all applicable indices i and x(1) = 1 and x(n) = n. Details The list x is not necessarily unique (for instance when b has a...

Q: Multiply two numbers

MendeleevInput: Two decimal integers. These can be given to the code in standard input, as arguments to the program or function, or as a list. Output: Their product, as a decimal integer. For example, the input 5 16 would lead to the output 80. Restrictions: No standard loopholes please. This is code-go...

9:04 AM
hi algorithms people
if I give you lots of binary strings of the same length, is there a more efficient way to find the pair with the minimum hamming distance than just naively testing each pair?
Hamming distance? Number of different pairs of bits right?
The hamming distance between two strings of the same length is the number of indices where the bits differ
In information theory, the Hamming distance between two strings of equal length is the number of positions at which the corresponding symbols are different. In other words, it measures the minimum number of substitutions required to change one string into the other, or the minimum number of errors that could have transformed one string into the other. In a more general context, the Hamming distance is one of several string metrics for measuring the edit distance between two sequences. A major application is in coding theory, more specifically to block codes, in which the equal-length strings are...
(just look it up, no need for onebox)
9:33 AM
@Lembik More details please? How long are each string and how many strings are there?
(and is it a competitive programming problem?)
n strings of length m
it's not a competitive programming problem! I don't even know if there is anything cleverer than naive you can do
if there is, I may post it as a ppcg challenge :)
I think there is, but it's somewhat dependent on the input. For random strings it may perform well. (I didn't try implementing it...)
I would be very interested to learn more!
The first page in JScript documentation uses document.write function which doesn't exist in real JScript. Microsoft, as always.
'document' is undefined
Nope, doesn't work.
10:05 AM
Q: something I don't know what though?

Muhammad SalmanGiven an input (a string , array , integer) , Your job is to append one of the following to it. st , nd , rd , th . If you are unable to understand what this is. Here is an example : 1 ---> 1st 2 ---> 2nd 3 ---> 3rd 4 ---> 4th 11 ---> 11th 12 ---> 12th 13 ---> 13th 14 ---> 14th and so on and ...

10:18 AM
@Lembik It's similar to matrix multiplication, only xor'ing the cells instead of multiplying. I don't know if it helps computation though.
@Cowsquack ah, good catch, I hadn't spotted that
At what point did Jelly become JLI?
@ATaco At v2.
how the hell I get rid of the annoying "cannot open output file invalid argument ld returned 1 exit status" in Code::Blocks ?
@AlexKChen Open task manager and kill the process.
10:32 AM
@user202729 Well I know, but which process ?
(the program.exe that you compiled to)
Well I don't see any such thing
@AlexKChen Go to Details tab
(why is TNB online interactive help? :/)
(and besides, why don't you use Code Blocks (64 bit)?)
@user202729 Well I'm asking here because I'm frustrated restarting the computer everytime the error is showing. And idk the difference between 32 bit codeblocks and 64 bit codeblocks. Also which app in details you're talking about ? is it .cpp or .exe ?
@AlexKChen Of course an executable is .exe. Anyway what's the screenshot of your details tab?
(probably the 64 bit version would be faster? TBH idk what's the difference)
stackoverflow.com/q/4077885 Looks like the difference is not very large. I don't think any program needs 4GB anyway. Also I don't have 4GB of RAM...
10:39 AM
@AlexKChen that means the file isn't being closed properly
Basically if your cpp is abcd.cpp, then kill abcd.exe.
Well my code is named codes.cpp and the executable is named codes.exe but it's not in the list ?
Ok then...
try to delete the codes.exe?
Opened in another program. Can't be deleted.
10:40 AM
If not succeed, just rename it to another filename.
@AlexKChen yeah that's a problem with Code::Blocks
@user202729 ... opened in another program = can't be modified in any way
@ASCII-only I don't think C::B open the exe.
@ASCII-only I believe rename is possible.
Q: Make me some curry

BMOHaving a function f that takes arguments x1, x2, …, xn                                                        – ie.  f : X1 × X2 × … × Xn → Y – currying redefines f as a function taking a single argument a1 which maps to yet another function. This technique is useful for partial application, fo...

A: Sandbox for Proposed Challenges

Muhammad Salmanordinal string check Description : Given a string as input , check if it is a valid ordinal number or not. If it is valid return true otherwise return false along with the real ordinal number. Possible inputs : 21st ---> true 12nd ---> false 12th 1nd ---> false 1st .... Your output can eith...

@user202729 how does it run the exe if it doesn't open it...
@user202729 if it's open then it's not possible...
10:41 AM
@user202729 oh that works (by copying the text by opening the file in notepad and renaming it). Thanks a ton :)
@AlexKChen (why do you need to copy what?)
well because apparently i can't open codes.cpp from code::blocks. idk why. but after renaming it works fine.
@AlexKChen wait what O_o
Probably a problem with the virus scanner?
I mean just rename the codes.exe to something else.
Wait C::B can't open codes.cpp too? How? /:
@user202729 yeah you can't do that if it's still open
10:43 AM
no yeah that works. that's why I told you thanks.
@ASCII-only I believe you can...
but I'm using Linux now and I don't want to reboot just to try that.
@user202729 if you can't delete it then why would you be able to rename it...
btw what's the benifits of using linux instead of windows for competitive programming ?
@AlexKChen Good question.
@AlexKChen There isn't any real advantage tbh
10:44 AM
like some guys who went to my country's national oi camp suggested me to use linux.
Sometimes I find the ability to view stack trace on _GLIBCXX_DEBUG/assert failure with gdb useful.
I guess it's mainly that linux has a more sane API
For example on Linux when the program (debugging in GDB) crashes I can almost always view the backtrace.
On Windows I know a hack (namespace __gnu_debug{void _Error_formatter::_M_error()const{__builtin_trap();}}) but that's not as nice.
Of course that's only useful if I know how to use gdb.
3 mins ago, by ASCII-only
@AlexKChen There isn't any real advantage tbh
@Neil: I don't know JavaScript but if I understand correctly you can remove the ternary.. I changed the specs a few minutes after posting, st. you're only required to support input-functions with ≥2 arguments, maybe you didn't see that - sorry!
@BMO thanks for letting me know
but because of the recursive nature of my answer, I don't think it makes any difference
10:53 AM
Ah, now I see. I missed the f(...) part.. I was thinking there was some JavaScript magic going on -.-
In that case you need it, yeah.
11:09 AM
... I'm a bit confused on what to do with this suggested edit.
11:33 AM
A: Sandbox for Proposed Challenges

ngnFinnish cinema You're given the map of a cinema theatre as a boolean matrix: 0 represents a free seat, 1 - occupied. Each Finn who walks in chooses the seat farthest away (Euclidean distance) from the nearest occupied one, or if several such exist - the first among them in row-major order. Outpu...

11:58 AM
@cairdcoinheringaahing (what should I do with the bug (in Bugle) now?)
@user202729 Request the next answerer fixes it, or you post an answer that fixes that bug
Well, RIP the Curious badge :/
12:26 PM
@cairdcoinheringaahing I prefered the old rules, but I thought that they just had a few holes.
I think it was definitely patchable.
also fwiw I didn't downvote but if things were fixed I probably would have upvoted. Seemed like an interesting challenge.
> dovetailing
> virtual machine
Terribly hard.
I think it's difficulty was entirely dependent on your language choice.
I also don't think there is anything wrong with terribly hard.
1:13 PM
that's a happy birb
2:11 PM
I was just creating an example of OOP and almost named the classes Cat, Bird and ><>. Too much PPCG
Fortunately ><> is not a valid identifier name (in which language?)
(sorry, misread your message :v)
The language is mine, unreleased
2:34 PM
Q: Hexadecimal, octal, binary, oh my!

Redwolf ProgramsThis is a code golf puzzle (Shortest byte count wins). The program should to do the following (Standard loopholes forbidden): Input a hexadecimal number Convert to octal Create a number with exactly the same digits in binary (765 octal -> 765 decimal) Divide the number by two Convert to hexadec...

@NewMainPosts Not very hard, but it's no more trivial than the "up go the bits" challenge.
3:06 PM
Q: ordinal string check

Muhammad SalmanDescription : Given a string as input , check if it is a valid ordinal number or not. If it is valid return true otherwise return the corrected ordinal number. Possible inputs : 21st ---> true 12nd ---> 12th 1nd ---> 1st .... Your output can either be an array or a string (You can replace sp...

3:28 PM
Is anywhere a list of unicode characters that are compatible with most monospace fonts?
It's called printable ASCII.
But I don't want to limit them to ASCII only
It's the only common denominator though. Everyone can make their own font in like 5 minutes, so "most monospace fonts" simply isn't possible.
Well, I meant the most used ones
@Soaku Define objectively.
3:32 PM
You have to match the font to the characters you're displaying. You can't do it the other way around.
@user202729 Most of unicode characters don't work with monospace fonts (they don't show up with fixed length)
Otherwise just pick your favorite fonts and take the set-intersect.
Probably you can choose some subset of this; it works on TIO and SE code font, except for a few chars
On mobile, I can't even see on PPCG, because of the font order SE chose for monospace.
@Dennis What you said seems to work.
3:38 PM
@Bubbler I wish there was a single SE font. TIO embeds DejaVu Sans Mono, so it should work even on devices that do not have it.
That's the problem then :(
FileFormat seems to have a nice list. The link above is list of Consolas characters, because this seems to be the font that has the least characters of default monospace fonts in IDE's.
Didn't check for IntelliJ tho
@Angs thanks
Suppose you have an odd number of white balls and the same number of black. How many partitions have an odd number of white balls and an odd number of black balls in each subset?
3:56 PM
Closed form or calculate efficiently? Partition into 2 or more subsets?
either would be awesome
closed form, I presume
can't be 2, for obvious reasons @user202729
actually if you calculate it efficiently that's great too
@LeakyNun I figured it out... it's so obvious.
so for 3 white and 3 blacks the answer is 2
4:01 PM
Sounds very hard.
Obvious dynamic programming approach is O(n⁴).
code that computed it would be great
even that would be great!
for the moment
n⁴ is not great :/
maybe a fastest code challenge is coming on...
@Lembik how? (w w w b b b), (w)(w)(w)(b)(b)(b), (www)(bbb), (wb)(wb)(wb), ...
it gives me test cases
@Cowsquack (www)(bbb) isn't valid
0 is an even number
4:02 PM
oooh I see
Q: ordinal string check

Muhammad SalmanDescription : Given a string as input, check if it is a valid ordinal number or not. If it is valid return true otherwise return the corrected ordinal number. Possible inputs : 21st ---> true 12nd ---> 12th 1nd ---> 1st .... Your output can either be an array or a string (You can replace spa...

but (wb)(wb)(wb) ...
Oops. n⁴ is only the memory complexity.
that looks ok ...
(w)(w)(w)(b)(b)(b) isnt' valid
so the answer is 2
@user202729 ah ok.
hmm. .let's try to work it out for 6, 8 and 10 balls so I can ask a challenge :)
Sounds like something that generating function would be useful.
"ask a challenge"
4:05 PM
pose a challenge may be better :)
Consequences of PPCG being on the SE network.
@user202729 is that an answer?
I did wonder :)
what do you think the answer is for 8 bins?
(wwwwwbbbbb), (wb)(wb)(wb)(wb)(wb), (wwwbbb)(wb)(wb), (wwwb)(wbbb)(wb). Did I miss any?
4:10 PM
@WhatWizard That's for 10.
Yes, I hadn't meant to post it right after you asked the question.
I'm reading on partition function to see if it's useful for this.
I think doubles of odds might be pretty easy.
@WhatWizard I also get 4
@WhatWizard that's interesting
4:13 PM
@Lembik but 8 isn't valid, isn't it?
I am really hoping someone is writing some code to check for small numbers
@Cowsquack sorry!
(wwwwwwwbbbbbbb), (wb)(wb)(wb)(wb)(wb)(wb)(wb), (wwbb)(wb)(wb)(wb)(wb), (wwb)(wbb)(wb)(wb)(wb), (wwwbbb)(wb)(wb), (wwbb)(wwbb)(wb). For 14
forget that
@WhatWizard are you doing it by hand?
But I'm about to write some code.
so the numbers so far are 2, 4, 6
4:14 PM
@Lembik That's possible, but coming up with an efficient algorithm is more interesting.
someone should double check my 14
@user202729 yes absolutely.... at the moment I just want some answers we know are correct
then we can go awesome :)
@WhatWizard (wwbb) isn't valid
yeah, I made a mistak
4:16 PM
was my intention but I made a copy paste error
so in terms of a generating function it seems to be
sum_{k=1}^{\infty} (x + x^3 + x^5...)^k (y + y^3 + y^5 + ...)^k and you then look for the coefficient of x^a y^b when you have a white balls and b black balls
I think
if anyone can code that up
It's a polynomial
Doesn't look very correct...
@user202729 oh!
4:20 PM
@Cowsquack correction: add 0001 0111 000111
so 14 gives me 8 (?)
@user202729 is there an example where it gives the wrong answer?
I feel that this is more correct... wait a minute doublechecking
@Cowsquack Here is 10, might be more:
14 seems to be the first mysterious case
are we sure about 10?
I missed (wwwwwwwb)(wbbbbbbb)
4:27 PM
(wwwwwwwbbbbbbb)(wb) is incorrect
yeah, I miscounted.
all the ones above the last partition appear to be for 16 instead
I don't think I am allowed to pose a challenge without worked solutions I think
otherwise I would just pose it now
it looks like more fun than I realised :)
Of course you can but it's not very encouraged ...
(QFT (Quest For Tetris))
4:32 PM
Here is everything I have for 16:
For 6 white balls I have:
do you consider 00011111 0001 01 to be equivalent to 11100000 1110 10?
I missed (wwwb)(wwwbbb)(wbbb)(wb)
@Cowsquack I haven't been
I have a couple of duplicates, it seems
What are the existing results?
4:36 PM
@WhatWizard then (wwwwwwwbbb)(wbbbbb) = (wwwwwb)(wwwbbbbbbb)
@user202729 1,1,2,3,4
I haven't been
Q: An uncommon approach to common factors

Redwolf ProgramsThis is a code golf question. Shortest byte count wins, and standard loopholes are forbidden. Everyone has heard of factors, the list of numbers that can be multiplied to result in a number (24=1,2,3,4,6,8,12,24). Common factors are the factors that 2 or more numbers share. The Challenge: Buil...

@H.PWiz For n=2,3,4,5,6?
Seems to match this (main diagonal)
For 1,2,3,4,5,6
The first in the diagonal is n=0 -> 1
My 8 for 6 white balls is:
4:39 PM
@WhatWizard I meant to say that you included both of those in your results
And I meant to. I consider them different
oh nvm I misunderstood your statement
@betseg @Riker /r/expectedfactorial/
@WhatWizard I like that name a lot better than the last
user image
xd no
4:55 PM
what is the current best guess for 5 and 7 white balls?
5 -> 4
Also you're requiring unobservable behavior (not all languages have array, etc.) — user202729 17 mins ago
@user202729 Why does that matter? If your "programming language" doesn't have an array-like construct, you just don't use it for the challenge.
I've seen a lot of those comments on questions
@H.PWiz is 7 still open?
@flawr Flagged as offensive.
4:57 PM
@Lembik I'm looking at it now, I think it's 12 tho
@H.PWiz cool
I am writing a challenge :)
00:00 - 17:0017:00 - 00:00

« first day (2632 days earlier)      last day (676 days later) »