« first day (2515 days earlier)      last day (2343 days later) » 

12:25 AM
\o/ Reclaimed hat leadership :P (for PPCG)
12:50 AM
Q: Requiring non-builtin answers

caird coinheringaahingEver since PPCG was created, answers using builtins have been an issue for challenge writers. Sometimes, an interesting challenge can be posted, only for some languages (not necessarily golfing languages) to have a builtin function that does all the work. Some people don't mind these answers, oth...

@NewMetaPosts That was quick
Does anyone here have experience with Swift?
1:29 AM
Best open source project name: github.com/commonsense/conceptnet-numberbatch
Isn't that the name of that one actor?
1 hour later…
2:44 AM
I want to make a really Functional language using the Tokenizer I wrote for Funky, but I can't find the drive to start.
>>> False == False in [False]
@orlp that's (False == False) and (False in [False])
because of chaining
but you got me for a second, good job
3:09 AM
@LeakyNun that is the most awkward chaining I have every seen o_O
also i'm mad. i killed a zombie in minecraft and then 5 zombies appeared out of nowhere and killed me; 1 of them even had an iron sword like where the heck did THAT come from
oh and also it was hardcore mode so i'm sad
also hello leaky :D
oh well i put about 30 minutes of work into it so far so i don't even care that much. but it was a good world spawn and i'm still sad :( :P
I'm way too proud of \s=toInt o(%)s
3:51 AM
@Fatalize Do you still golf in Prolog? Does anyone golf Prolog?
@WheatWizard: Only a few hours ago, I downloaded a copy of "The Art of Prolog" because I'm interested in a new programming paradigm :)
You reckon it's worth it?
Logical programming is cool, prolog is kindof antiquated
What should I learn instead?
The Art of Prolog is pretty old, I haven't read it but I've heard its pretty outdated
@BruceForte Prolog is still the best, as far as i know, but that doesn't make it great
old ≠ bad
But if it's outdated, maybe I'll look for something else then
3:56 AM
That's true, but Prolog is both
It's still supposed to be good, so it might be a good starting point, I really don't know, I'm just parroting what I've heard.
4 hours later…
7:29 AM
@WheatWizard: How do lists work in Prolog :S
What do you mean?
I will admit they are not very easy to deal with
If I call l(123,X) it starts complaining because it expects a list & l([1,2,3],X) complains about Illegal numbers
(trying to test your solution)
Oh, it expects a string not a atom
try string_to_list(123,X),l(X,L).
Yeah, that works
Maybe you should add a TIO.
I didn't know enough prolog to make TIO work.
I plan on figuring that out soon, I'm still working on golfing my program back down.
7:36 AM
Yeah, np
Anyway, gtg. Thanks for raising my interest in Prolog :)
8:23 AM
Q: Print a flag in as little lines as possible!

james jakeJust for fun, I wanted to see how little lines I could solve this problem in, using python. I managed to condense my code into 9 lines. size = int(input("Size: ")) a = [x for x in range(size-2, -1, -1)] if size % 2 == 0 else [x for x in range(size-2, 0, -1)] b, counter = a[::-1], 0 print("\n"+"...

Just found a trick so that gold code-golf users can VTC as dupe without hammer. Remove the tag, VTC and rollback
8:40 AM
@Mr.Xcoder Its always been possible, but I think it is heavily discouraged
fun fact: the reverse does not work, you can't add a tag so you can hammer a quesiton.
8:52 AM
I've only gotten hats on Worldbuilding thus far, weirdly enough
But I have mainly been away for the past week
1 hour later…
10:14 AM
Q: Write Program in few lines

ArtierOutline of challenge Your quest of this code-golf post is to create a flag (as printed bellow) in as few bytes as possible. The flag has an X shape and a border of #'s. Given any integer n create an appropriate flag. Basic Rules No use of standard [loop holes][1]. Least number of bytes wins...

@NewMainPosts Why does Community ♦ behave like that?
(vote to close as duplicate)
Should this have a leaderboard?
11:16 AM
Will editing a CW post notify the OP? What about commenting?
11:41 AM
@WheatWizard not much, Brachylog has the same principles as Prolog so it would be kind of redundant to use both
2 hours later…
1:21 PM
@user202729 because it's OP who agreed with close votes
@user202729 Yes to editing; I forget for comments but I think so.
Q: A very long Terza Rima

LiefdeWenDescription Output the rhyme scheme for a very long Terza rima. Input none. Output ABA BCB CDC DED EFE FGF GHG HIH IJI JKJ KLK LML MNM NON OPO PQP QRQ RSR STS TUT UVU VWV WXW XYX YZY Rules You can pick between seperating stanza's with whitespace or newlines so either: ABA BCB... OR AB...

There are always some atoms I don't know about... My solution is 10 bytes.
Cool, can I see your approach? :D
This time I also forgot that "a niladic link start with a nilad is equal to the monadic link evaluate at that nilad.
(of course the µ is completely unnecessary)
1:35 PM
Ah okay. Interesting
Why aren't there any image recognition APIs that provide coordinates for the object in the image -_-
cutting edge research ought to be a bit more cutting edge and a bit more free :(
What about openCV?
1 hour later…
3:22 PM
:O I've got 160 rep today without posting anything
From answers I posted yesterday
i've gotten 85 today and i've posted 3 answers :p
I posted 6~7 answers yesterday, got about 130 rep, then came home from school and had a +160 in my inbox :P
lol :P
i'm 5 away from gold code-golf :D
i went from 992 to 980-something because a user got deleted so i was sad :( but then a challenge got posted and I could just spam-answer in all the languages and farm rep and score
3:30 PM
Q: Yo Dawg I Heard You Like Code Golf

Stephen LeppikWrite a program that outputs another program that outputs another program that… etc. for as many iterations as you want. The innermost program's output should complete the meme: So I put a code golf inside your code golf so that you can golf your code while golfing your code Rules At lea...

Any final feedback? Thinking about posting it later
Looks good.
ಠ_ಠ Two upvotes away from taco hat or 2 upvotes away from Sherlock hat
PPCG theorem: the harder a question is, the more upvotes/less answers it gets. (correct?)
Yeah, sounds about right
3:41 PM
the answers part is definitely true. probably the upvotes as well because most people see it and are like "nope too hard bye"
I don’t agree with the first part
Is there a way to search for sandbox challenge?
how to get upvote: 1) meme question 2) profit
@user202729 in the question part of the sandbox post there should be a search link put there by someone (I think me or the other guy)
The most votes question on this site is also (probably?) the hardest...
@user202729 yeah but it doesn’t apply for all challenges
:( I am into a Pyth outgolf war on mobile
3:48 PM
CMP: Who would be interested in answering a KoTH sometime in the near future? It's been a while since there's been a KoTH (I think) and I'm planning on posting one soon.
... Not sure. If there is something interesting I would answer.
BTW @HyperNeutrino this is O(n*(2^n)) unless you implement it extremely poorly. I have a (fairly obvious) n^2... Anyone want to solve this?
I'll try to make it have some elements that make it significantly different from other KoTHs (not just adding more defenses/attacks for bots) but I'm currently having trouble thinking of something creative enough but not outrageously hard to implement (both for controller or for a fighter bot)
I have two KoTHs in the sandbox, but I haven;t got round to making the controllers yet
3:51 PM
@user202729 It is, but the second most voted question is to print 2014.
@user202729 Hm okay. If you can justify optimality then the bounty will go to you :D
@cairdcoinheringaahing I can help implement controllers if you want because implementing controllers is (kind of) fun (until everything starts breaking and people yell at you in the comments)
You guys are all better at golfing Python than me, any suggestions?
@ Dennis : Obvious counterexample... popcon in the past can get a lot of votes.
@HyperNeutrino I'm going to have a bunch of spare time in the summer, and will probably finish them then, but thanks for the offer
> the summer
is it just me or is that in like 7 months
unless you live in the southern hemisphere?
but UK is in north... ¯\_(ツ)_/¯ idk i won't judge
3:54 PM
@HyperNeutrino Yes that's also what I think of...
I mean in the southern hemisphere it's already the summer :P
End of June is the end of my exams, so I have a bunch of free time from then until September :P
ah ok. so it is in like 7 months (closer to 6 but w/e) :p
anyway gtg o/
@El'endiaStarman In case you missed the notification, I commented about posting your sandbox challenge.
@user202729 That's why I dropped in here, actually. :)
Erik commented that my "shortest solutions" for levels 4 and 5 have issues. Gonna fix those then solicit final feedback in here.
3:58 PM
Being longer than necessary or cannot eat all carrot?
Not 100% sure.
Wait... if this is atomic-golf, why is the objective find the shortest solution?
4:13 PM
Because, then the objective is absolute... I suggest "program must terminate in 10 seconds. Output the shortest program possible you can" or something like that.
Q: Create a regex to match a given range of integers in a given radix

Marathon55Task Write a either a function for a compiled language or a command line script that, given three integers n,a,b prints a regular expression which matches all of the base-n integers from a to b (and no integers outside of that range). Your algorithm must be deterministic (i.e., produce the corr...

A: Sandbox for Proposed Challenges

El'endia StarmanGoogle's Hopping Bunny atomic-golf code-challenge On December 4, 2017, the Google Doodle was a graphical programming game featuring a bunny. The later levels were nicely non-trivial and they seemed like a great candidate for an atomic-golf challenge. Details Game There are four available mo...

Final feedback requested!
@user202729 Yep, I added that.
> Winning criterion: sum of number of moves of shortest solutions found for each board with a total runtime of 1 minute to analyze all boards.
1 minute process all boards? Also you may want to say "on your machine".
Yeah. I'm not totally sure that 1 minute is the right limit. It's always hard to find that line between impossible and super easy for stuff like this.
Alternatively, I could say one minute per board. Edit: Yeah, I like that better. Edited the meta post.
@El'endiaStarman looks like you fixed the issues with the solutions
however, something I'm unclear about is what you mean by "total runtime of 1 minute"
4:27 PM
I suggest "shortest solution your program can", because in 1 minute the program may not be able to find the globally optimal solution.
@EriktheOutgolfer Yep. I suspect there might be a way to solve level 4 with a solution that's the same as level 5, just with loop counts tweaked, but I'm not going to go to the effort of finding that out. If that is the case, I'll let solvers find out.
I earned so much rep from my answers on meta se
Is this better?
> Winning criterion: sum of number of moves of shortest solutions found within one minute for each board.
yes (nija'd so you ruined my joke :()
@El'endiaStarman what if your program doesn't find a shortest solution for a board in one minute?
4:32 PM
@EriktheOutgolfer That's a good question. Two options are 1) score of infinity (so programs MUST find a solution) or 2) score is worst solution found by another program (infinity if there isn't one).
Shortest you can.
anyone have a challenge i should answer?
@El'endiaStarman 1) is 100% not feasible at all
no matter how fast the program is, if you give it a board big enough, it will take over a minute
45 mins ago, by caird coinheringaahing
You guys are all better at golfing Python than me, any suggestions?
parents called me in the middle of school :P
4:38 PM
Just give extremely large score - 6 × number of carrot. (I think this is enough)
2) is very hassling
@cairdcoinheringaahing tabs not spaces
@El'endiaStarman doesn't look like a good criterion to me
@Christopher2EZ4RTZ It's Python 3, so you can't mix indents
Damn python 3 forcing sane coding practices
4:39 PM
@Mego i kinda hate that :P
It's a very good thing for writing clean code, but it hurts code golf
@Pavel i mean we are on PPCG so that is a pointless thing
Anyone open this? (is it ready for opening?)
Since Python is designed for writing good code, and not for golfing, it's perfectly fine
@Mego lol
4:40 PM
If you want to write bad code, use Python 2 :P
@Mego "good"
@cairdcoinheringaahing I wasn't going to go there :P
@Dennis thanks for making TIO. It saves me from being bored when away from my compilers :P
@user202729 looks ready for reopening, but extremely trivial
4:41 PM
@EriktheOutgolfer Well, of course. Dunno if any of the boards currently given are that big though.
not sure if I should reopen, hmm
@user202729 Aiming for the "I'll handle it" hat?
> Your code should be able to take any board as input, not just the ones given as examples below.
@user202729 Yeah, that could work. Score = 6 * number of carrots (unmarked squares).
@EriktheOutgolfer Should be able. That's there to discourage hardcoding.
@Mego except when using {*...} sets :p
4:43 PM
@EriktheOutgolfer Splat sets are a good addition. It's a happy accident that they're good for golfing, too.
Generalizing operators (where they make sense) is a good thing
or *var=, [*...], ...
generally unpacking stuff
It's a shame that you can't do l=lambda though to save bytes
and in the rare case you can't have quoted input, input vs. raw_input
@cairdcoinheringaahing Already got it.
4:45 PM
lambda is the biggest problem with Python golfing IMO
@cairdcoinheringaahing Eh. Keyword aliases isn't a good feature for writing clean code.
in a case you aren't allowed a trailing newline, you have the ability to do print(end=string) instead of import sys and then sys.stdout.write(string)
@cairdcoinheringaahing Yeah, I'd like a λ spelling for it
@EriktheOutgolfer @user202729 How about this? "Winning criterion: sum of number of moves of shortest solutions found within one minute for each board. If your program does not find a solution for any particular board, your score for that board is (5 * number of squares)."
@EriktheOutgolfer I hope that was atomic... Anyway because OP decided.
4:47 PM
@El'endiaStarman now looks fine
I suggest 6. Euler path is 2×, each move take at most 3× (2 turn, 1 forward).
"The over all goal in golf, is to play less golf." source
@user202729 I hope that was
@Mego How many bytes is λ though?
4:48 PM
@user202729 Yeah, I considered that. I think multiplying by 5 is good in that it makes such a program worse than even those that build the most trivial solutions.
\o/ Tying Riker at hats :P
That looks like the wrong way to spell tie-ing
;P I finally arrived at a solution which is nearly optimal after golfing my initial approach (which I wasn't satisfied with) by over 5 bytes...
@cairdcoinheringaahing Depends on the code page
well, Python uses utf-8, so for Python purposes it's 2 bytes
@EriktheOutgolfer You can declare the encoding as ISO-8859-7, though, and potentially save bytes (depending on how many lambdas you'd be replacing with λs)
4:52 PM
@Mego well, that would be extremely rare
#coding=iso-8859-7 is already very long
Lol two of the comments on one of my answer (I just realised I have typed nearly the exact same thing) – someone: Outgolfed you back by 1, now at 25.; me: @someone Outgolfed you back by 1, now at 24.
@DJMcMayhem I haven't used brainflak in so long i forgot the syntax :(
@EriktheOutgolfer 4 lambdas replaced by λs and it's shorter
@Christopher2EZ4RTZ The syntax is basically just ''.join(random.choice("(){}[]<>") for _ in range(100))
@DJMcMayhem yeah, i wish there was an output char
On an unrelated note, you're always more than welcome to ask for brain-flak tips and I'll help you out. :)
going to be a huge pain for this challenge i just started
@Mr.Xcoder because λ doesn't exist as of yet ;-p
@Christopher2EZ4RTZ Brain-flak classic has an output monad: [...]
Oh I am off-topic then :P
4:56 PM
@DJMcMayhem do you have the github specs?
also isn't that negative?
If you're wondering about syntax, the general syntax is just () for a nilad and (...) for a monad.
@Christopher2EZ4RTZ In regular brain-flak, yes. But not in classic. Also, in classic [] == -1
do you have the language specs still?
Yeah, just a minute
@mınxomaτ This is really cool. Any chance you could explain what word embeddings are and/or why vectors are involved? Basically, why does each line of data have a word and then 300 floating point numbers afterward?
4:59 PM
@EriktheOutgolfer #coding=greek is shorter.
a wild Dennis appears he used GOLF.... It is super effective!
@Christopher2EZ4RTZ I don't think BFC has an official spec, but there's an unofficial spec here:
Q: Write a brain-flak classic interpreter!

DJMcMayhemBrain-Flak (a cross between Brainf**k and Flak-Overstow) is a stack-based esoteric language. Since this challenge was posted, the language has evolved and updated, but this first revision of the language is known as "brain-flak classic". You must write a program or function that takes a string o...

@Christopher2EZ4RTZ you mean orlp's GOLF?
@EriktheOutgolfer what?
@DJMcMayhem dang idk if it will work. I need ascii output :P
Oh, if you use the modern interpreter, the -a flag should still work with classic. You can do -l classic or --language=classic or something like that, I don't remember
Hmm... interesting
5:05 PM
Haha, I just checked up on JHTBot; it has over 400 notifications:P
[...] will always output in decimal, regardless of the -c or -A flag: Try it online!
cc. @WheatWizard
IMO that's a bug
so I guess when i finish the code I will post but remove it until the bug is fixed
5:18 PM
@DJMcMayhem would my answer be invalid then?
Until the interpreter is fixed, I'd say yes.
Is it possible to output after the program is done executing, or does it have to be during execution?
You could totally do that in regular brain-flak
5:19 PM
but why
and you can but it would take up more bytes
At this point people are like "using python version bugfix 4daer421 at 1:25 PM during the summer eclipse of 2017 in north Antarctica to save 1 byte"
@Christopher2EZ4RTZ Just FYI, I'm not sure if it'll work in BFC because that outputs during and after execution
it works
(10 is newline)
I know :P
since you doubted my brain flak skills anything was possible
Oh, I see you're not pushing anything
5:24 PM
your avatars aren't confusing at all
@DJMcMayhem anything i could golf?
Yeah, [(()()()()()()()()()())]{} is redundant
You're pushing and then popping. Try just printing instead
Also, ()()()()()()()()()() could be (()()()()()){}
@Christopher2EZ4RTZ I have it almost working in regular, but I can't figure out where the newlines are supposed to go
@DJMcMayhem every 3 bytes
5:30 PM
No, in the code.
I'm not sure where in the code the push(10) part should be
one sec
that is just getting you to the ascii value you need to start with
and gets the 2nd stack to be the amount of times you have to print the 3 letter thingys
@DJMcMayhem Do non bracket characters get ignored in brainflak?
@Christopher2EZ4RTZ Try it online! is super close
@cairdcoinheringaahing yeah
@DJMcMayhem nicee
So a Lisp polyglot is quite possible :P
5:33 PM
@DJMcMayhem tio.run/…
@Christopher2EZ4RTZ BTW, after I help you get it working, I am going to outgolf you
@DJMcMayhem i fixed it
added {}
Still missing zyz
Just increment the loop times?
don't need it
right from the challenge spec
But the output is upside down
5:35 PM
sounds like a personal problem :P
Uhm, it's your answer?
well shit
You could always {({}<>)<>}<> at the end, but not super golfy
Oh don't worry, I've got some ideas
5:38 PM
that sentence should not have scared me as much as it did
Q: Foam Bath Letters

MatthewIf you have a small child in your house, you may have come across foam bath letters. These can be moistened and stuck to flat surfaces such as tiles and the side of the bath to make words and messages. The range of words and messages is somewhat limited if you only have one set though, since you...

@DJMcMayhem won't you have to understand brain-flak code first?
anyways gtg bye o/
Do you mean your code?
Cause I do understand my language :P
@DJMcMayhem Congrats! That's an achievement to be proud of!
I don't understand any of my languages so...
I understand V 2/3rds of the time
5:45 PM
V is black magic
that's about it
in Jelly, Apr 1 '16 at 18:42, by Dennis
I don't understand my language. ._.
@Christopher2EZ4RTZ Be scared...
A: A very long Terza Rima

DJMcMayhemBrain-Flak, 90 bytes ((((()()()){}){}){}()){(({})<((({}((((()()()()){}){}){}){})())[()])((()()()()()){})>[()])} Try it online!

@DJMcMayhem wow that's incredibly short
i'm already at 92
and i haven't even started printing anything
Thank you :)
5:47 PM
why does brain-flak have an SBCS on TIO lol..
@HyperNeutrino SBCS is to distinguish from multi-byte schemes like UTF-8
but brain-flak only has 8 applicable characters lol?
wait you could make each character 3 bits in brain-flak :D
@HyperNeutrino Yes, so it can be compressed, but it will also need padding to integer byte lengths
it can also be implemented to implicitly close brackets
5:53 PM
And to have only one close bracket type
@EriktheOutgolfer If brain-flak was designed to be golfy, there are a loooot of design choices that would have been implemented before bothering to implicitly close brackets :P
For example, making it not suck to use
I mean, brain-flak inherently isn't particularly designed for golfing :P
@H.PWiz no
no because then autoclosing wouldn't be as golfy
How about a niladic close type and a monadic close type?
5:55 PM
that would not be helpful at all :D
No, it would if the goal is a better encoding
(()[){} can become (()[]){} but ((_[_{} is ambiguous
Because then there are 6 characters instead of 8, and you could huffman code it so that the closing types (which will be used extremely often) are less bits
I mean, half of the characters are unnecessary. Just have 8 characters, 4 for nilads and 4 for monads
@DJMcMayhem like
5:56 PM
and then just make it a prefix language
@HyperNeutrino Well, monads have to be two characters so that you can tell what is an argument to it
not necessarily
Because (()) and (()()()()()()()()()()) are both equally valid
5:57 PM
So you don't know how much is in the scope of the monad
you're right :( i'm dumb
i forgot momentarily how brain-flak works i guess :P
Scope is pretty much the foundation of all of brain-flak
what if we make 12 chars instead
(using APL chars here)
A nilad that doesn't do anything but has a value of 1 isn't useful unless all the monads can tell the value of everything inside of it's scope
Which I suppose technically <...> can't, but it's still very useful
O -> ()
⎕ -> []
⌺ -> {}
⋄ -> <>
those are the nilads
the monads are as usual
5:59 PM
that could work
@EriktheOutgolfer If you're doing that, then the closing types are redundant. You could just use | instead of individual closing brackets
If the goal is less characters

« first day (2515 days earlier)      last day (2343 days later) »