The Nineteenth Byte

Geobits

19:00

Of course, asking the question usually doesn't hurt either way.

ais523

@El'endiaStarman fair enough, I was just explaining the meaning for people who didn't know; and it's something that benefits from an abbreviation because it's a) frequently useful and b) really long, so it's not surprising that one grew up

Rainbolt

After nearly two hours of fiddling, I am nearly able to elevate a PowerShell script, in the same working directory, passing a path that contains spaces as an argument, without losing the output from stderr.

ais523

in general it's not surprising that people try to use IRC conventions here out of habit, because the two systems work very similarly

redstarcoder

IRC conventions seem to bleed into everything

FlipTack

When you have an infinite loop that takes input, and eclipse is shouting at you for not closing the input scanner, but if you do close it after the loop it shouts at you for "unreachable code"

:(

ais523

19:06

@redstarcoder I find IRC a much more convenient system than this one (e.g. on IRC I don't have to move my hand to the mouse for basic operations like replying to a post)

Geobits

@FlipTack I think it's trying to tell you to make your loop exit at some point rather than making it infinite :P

ais523

but IRC's also designed to be a lot more transient than SE chat is; everything here is logged as a matter of course, so I guess we're supposed to ensure that everything we say makes sense in retrospect

rather than living in the moment like IRC does

TimmyD

@FlipTack Simple. Construct and destroy the input scanner each loop iteration. If the code runs slowly at that point, that's a hardware issue.

redstarcoder

@ais523 me too, though I do like the reply feature here. I wish I could adjust my ping noise.

ais523

on IRC I use a visual ping, rather than an audio one

also my speakers are off atm, so I get pinged only by the title bar

FlipTack

19:08

@TimmyD yeah, originally I had that, but I'm probably closing it wrong because it seemed to completely close input and error when another scanner is opened

redstarcoder

I used both, but my ping was a very soft noise.

ais523

(you can turn sound off in chat just above the title where it says "The Nineteenth Byte"; that would have the same effect without having to turn off your speakers)

if only I could turn off images and the like

redstarcoder

Userscript?

ais523

one of the huge disadvantages of SE chat is that everything takes up so much vertical space

well I use temporary profiles for SE because it's so JavaScript-based

FlipTack

Anyone, does anyone know if you're allowed to return a function which returns the result in PPCG? For example, for the add two numbers challenge, could you submit something which is called via f(4+5)() instead of just f(4,5)

Geobits

19:09

@FlipTack Why is the loop infinite? I assume it has to end at some point, so just ake it exit the loop, close the input, then do whatever (even if that's only exiting the program).

ais523

@FlipTack you can't require the person providing the input to do the calculation themselves

FlipTack

An example in python would be for finding the length of a string, returning f=lambda s:s.__len__, which than has to be called again like f("string")()?

ais523

however, f(4)(5) would be a legal calling convention; I suspect f(4,5)() wouldn't be though

FlipTack

I know that's longer than normal code, it's just an example

redstarcoder

@ais523 I think he just means it's ultimately returning the pointer to a function that hasn't been called

ais523

19:10

you're basically defining a curried function there that takes two arguments, but the second one doesn't matter

quartata

I'm pretty sure it's fine but I would have to find the meta post

FlipTack

what I meant is, for example in Java, returning the toString method rather than the actual string, and get the user to call the function and then call the result

ais523

even so I think that's failure to comply with the spec

I think I understand what you mean

redstarcoder

I would agree that seems to be against the spec ^^

Geobits

While I don't think it's appropriate, I'm failing to see why it's different in theory than using an anonymous function.

FlipTack

19:11

alright

Geobits

To use it, you have to assign/call it.

quartata

Or any statement that evaluates to a function. Just an extra layer of indirection.

FlipTack

I might bring it up on meta? unless it's been discussed before

redstarcoder

@Geobits An anonymous function takes the input as parameters

@FlipTack maybe that's best

FlipTack

alright, writing post

ais523

19:13

err, is there any way to edit a closed post without it going into the reopen queue?

quartata

@FlipTack 99.99% sure it has been brought up

so dont bother

Geobits

@ais523 I don't think so. Most edits to closed posts are done with the intent to make them better/reopenable.

quartata

Lemme search for it

mınxomaτ

The camera thing: github.com/turbo/c4

TimmyD

Fascinating hardware solution to a usability problem

ais523

19:15

@Geobits in this case the edit would be an attempt to let people know how to help the question be reopened (it's still receiving upvotes), and putting the post into the reopen queue early would defeat that

Geobits

Yeah. I just mean I don't think they built it with that in mind.

ais523

there needs to be a distinction between making a post better but not yet reopenable, and making a post reopenable in its own right

Kritixi Lithos

@mınxomaτ All of this is legal, right? :P

quartata

@mınxomaτ Well that's terrifying

30k? Seriously?

mınxomaτ

@KritixiLithos Viewing is. Controlling depends on the "hacker" laws in your country.

Geobits

19:16

There have been cam lists public for years now. This one's just a bit bigger :)

You can find plenty just by googling with intitle: iirc.

FlipTack

I've written most of the post

@quartata find anything?

mınxomaτ

@Geobits It's actually better to use Shodan and look for the magic header length that basically all these cameras share.

Geobits

Probably. I just know they aren't really all that hard to find. I like the compiled list, though :D

Kritixi Lithos

@mınxomaτ How did you get list of cameras in the first place?

mınxomaτ

I did my own scan.

Basically a banner fetch for IPv4. Then look for manufacturers and known header lengths in the headers.

FlipTack

19:20

0

Q: Returning a function, which returns the result?

Submission formats are always flexible here on PPCG. Named functions, anonymous functions or even full programs that perform IO, are all fine. Even currying is acceptable, for example, taking arguments as f(4)(5) rather than f(4, 5). But what about returning a function that returns the result? ...

discussion

New Sandboxed Posts

-1

A: Sandbox for Proposed Challenges

Solve all of these OEIS questions at once! code-golfsequenceinternet Task Your task is to write a program or function that takes the numeric name for any OEIS sequence (e.g. A000001), and outputs the first N elements of it in any way. Input Your program or function will take two inputs, ...

Geobits

@mınxomaτ FTR, I'm using xdg-open instead of open for this. I didn't have it configured, and it's just as easy.

mınxomaτ

Ew, more bytes.

Geobits

Only the first time. After that it's just an up-arrow away for either :P

redstarcoder

@mınxomaτ find anything cool? I found this spooky-looking 216.16.241.216/mjpg/video.mjpg?COUNTER

mınxomaτ

19:23

@redstarcoder Looks like a Dexter kill room in progress.

redstarcoder

It's so weird, it even shakes around sometimes

Geobits

Is it looking upward? I keep seeing things fall toward the camera.

quartata

I'm scared to click on it

mınxomaτ

@Geobits I think that's a shipping container and the "things" are larger insects.

Geobits

Open it, cast it to tv, invite company over. Don't say a word about it and see how long it takes for someone to mention it.

Ah, yeah it could be moths or something drawn to the camera's light.

quartata

19:26

oh I'm so happy I didn't click on it then I hate large insects

moths are the worst

Geobits

Not that large if I couldn't tell they were insects ;)

mınxomaτ

Hm, it's in Canada.

That might be snow.

{"ip":"216.16.241.216","country_code":"CA","country_name":"Canada","region_code":"ON","region_name":"Ontario","city":"Toronto","zip_code":"M4W","time_zone":"America/Toronto","latitude":43.6827,"longitude":-79.373,"metro_code":0}

Geobits

Probably more likely. The "things" are fairly consistent in direction, which I wouldn't expect from bugs.

Nathan Merrill

it doesn't look outside though

like, it looks like its inside of a warehouse or the back of a truck

sadakatsu

I haven't tried drafting any instructions yet, but would there be interest in a KotH + Cops and Robbers competition where the "Cops" compete to build the most difficult Sudoku puzzle generators and the "Robbers" compete to build the fastest Sudoku puzzle solvers?

FlipTack

19:29

could be coming through the roof?

Geobits

Yeah. Really weird.

quartata

It is snowing in Ontario though

FlipTack

-6 celsius

light snow

according to google

mınxomaτ

Meanwhile I've found the nail salon from Better Call Saul.

Nathan Merrill

@sadakatsu I've actually made a sandbox about this concept. I was going to do a different puzzle (NP--hard puzzles are beter)

Geobits

19:29

Keep watching the feed though. I expect someone to start dragging bodies in at any time.

Nathan Merrill

who is serving these videos?

is it the camera?

mınxomaτ

Yes.

Nathan Merrill

if so, wouldn't all of us watching be overloading it?

Geobits

Apparently not yet :P

mınxomaτ

Not really. It's low framerate and heavily cached.

New Meta Posts

19:30

0

Q: Returning a function, which returns the result?

Submission formats are always flexible here on PPCG. Named functions, anonymous functions or even full programs that perform IO, are all fine. Even currying is acceptable, for example, taking arguments as f(4)(5) rather than f(4, 5). But what about returning a function that returns the result? ...

discussion

Nathan Merrill

0

A: Sandbox for Proposed Challenges

NP Cops and Robbers This is an idea I've had for a while, and I really want to get it to work, but there are some large hurdles. Hurdle 1: I need to pick a puzzle, preferably one that is NP-Complete. I think lots of Nikoli puzzles are good candidates, and I am leaning towards Light Up Cops: ...

FlipTack

are they giving that woman's foot a massage?

Geobits

Probably. Might come as a package with the pedicure.

quartata

no pls I don't want to know

sadakatsu

@NathanMerrill I'm not sure NP-hard puzzles are necessarily better. Having a well-known puzzle could attract a different audience.

Nathan Merrill

19:33

if you don't use an NP-hard puzzle, you're going to have a hard time measuring submissions

PhiNotPi

@zyabin101 I confirm the submission

TimmyD

> Members of the "hacking" community known as PPCG have assisted authorities in seizing one of the largest single-day drug hauls in the history of the United States. Asked how they infiltrated the criminal gang's hideout to secure the location of the drugs, one of the PPCG residents said "I dunno, I just clicked on a random camera link." -- CNN

18

sadakatsu

@NathanMerrill I don't see why. Sudoku may be P (not quite sure it is), but it still takes a fair amount of search to brute-force an answer. I am unaware of algorithms to calculate a solution formulaically, but maybe there are some.

DJMcMayhem

@NewMetaPosts @FlipTack I know this is unrelated to your question, but you could just do len instead of f = lambda s: s.__len__

Nathan Merrill

solving sudoku is actually quite easy, IIRC

devRicher

19:35

@TimmyD Source? ;)

FlipTack

@DJMcMayhem I know. It's not a golf example

TimmyD

It's on the Internet, it must be true.

FlipTack

It's just a simple example to get the point across

Nathan Merrill

I mean, if you really want to go with sudoku, then make the puzzles huge, like 400x400

or something like that

that'll do more than adding more puzzles

Pavel

HAHAHAHA 24 HOUR BAN LIFTED SPAM ALL THE THINGS
(Not actually please don't hurt me Endia)

FlipTack

19:36

in languages like Python it wouldn't really help but in Java you'd be able to return a "function property" of an object without the parenthesis, saving bytes @DJMcMayhem

DJMcMayhem

@FlipTack I know. I just can't resist golfing things :)

FlipTack

same. I had to write (i+1) in the middle of some math expression the other day, not golf, and I really had to stop myself from changing it to -~i

DJMcMayhem

I'm not posting this as an answer, because I don't have arguments for or against, but that sounds perfectly reasonable to me

Allowing that is very similar to what I argued in this answer

8

A: What counts as distinct and consistent?

Outputs are consistent if they are equal in the sanest and most obvious way of comparing them. For example, if you have two char * strings in C or C++, and you try a == b you will get a falsy value even if the strings are equal. But every half-decent C programmer knows that you don't compare s...

Pavel

Anyone here know how to invoke Java lambda expression? I need to test this and I don't know how:

n->!new String(new char[n]).matches(".?|(..+?)\\1+");

FlipTack

You need a boiler plate for it

java.util.function has some builtin boilerplates but if it doesn't fit into those you write your own

Nathan Merrill

19:40

do like, a = n ->

FlipTack

@NathanMerrill you need to give it a type.

Nathan Merrill

then a.get()?

oh yeah

Pavel

boolean a = n -> ...?

FlipTack

@Pavel what are your input and output types?

no, not like that

sadakatsu

@NathanMerrill I still don't see why you're saying that, but maybe it's because we're looking at this differently. I am curious to see if there is a way to generate standard Sudoku puzzles that can reliably slow down Sudoku solvers. The only way to test that is to have programs generate many puzzles, and then see if different solvers consistently take longer for any generator's puzzles than other generators'.

Pavel

19:41

int to boolean

sadakatsu

@NathanMerrill Finding the solutions is likely to still be fast, but a puzzle that averages at 1s is more interesting than one that averages at 10ms, right?

FlipTack

give me a sec I'm writing it

so you do static interface Lambda { boolean run(int x); }

And then you assign it with Lambda f = ...

and call with f.run( your value )

Pavel

Is there a way I can get it to work in TIO?

Nathan Merrill

@sadakatsu for cops, that's somewhat interesting, but you also want it to be interesting for the robbers. If a cop is using some weird puzzle generation that slows down most solvers, I think it'd be fairly trivial to make a solver that solves it quicker.

Pavel

Thanks though. I'll just have to test later.

Nathan Merrill

19:43

if you make the puzzle bigger then that gives cops more room to come up with difficult variations

FlipTack

@Pavel I'm writing a working example on repl.it now

Pavel

(The regex matches strings of composite length)

Oh cool

Nathan Merrill

which means that robbers have to handle more and more difficult stuff

but once again, since sudoku is P, I don't think there's really much to be found

(it is P, right?)

FlipTack

@Pavel repl.it/FBit

you should only really use lambdas if that's your whole answer, otherwise you have to include the boilerplate to invoke it which costs bytes

but if the expression itself is the whole answer you don't need to include the boilerplate

Nathan Merrill

"The general problem of solving Sudoku puzzles on n^2 × n^2 boards of n × n blocks is known to be NP-complete.[1] For n=3 (classical Sudoku), however, this result is of little relevance: algorithms such as Dancing Links can solve puzzles in fractions of a second."

ah, nope its not

Mathematics of Sudoku

The class of Sudoku puzzles consists of a partially completed row-column grid of cells partitioned into N regions each of size N cells, to be filled in using a prescribed set of N distinct symbols (typically the numbers {1, ..., N}), so that each row, column and region contains exactly one of each element of the set. The puzzle can be investigated using mathematics. == Overview == The mathematical analysis of Sudoku falls into two main areas: analyzing the properties of a) completed grids and b) puzzles. Grid analysis has largely focused on counting (enumerating) possible solutions for different...

Pavel

19:46

@FlipTack Thanks a lot!

FlipTack

no problem

I only just got back into Java, I need the practice :D

sadakatsu

@NathanMerrill, Yeah I just found a different link stating that it Sudoku is NP-complete.

@NathanMerrill So, ironically, a P-complexity solver for Sudoku could solve the P=NP question : P

Nathan Merrill

so yeah, I think that sudoku is a good candidate. however, I'd definitely recommend doing something bigger than a 9x9 board

sadakatsu

I guess it depends upon how long individual solutions will take. If they're so short that factors like context switching and OS lack of fairness will determine the winners, then bigger boards are necessary. If not, then 9x9 is fine (and more interesting to me since I will want to take a crack at the winner's puzzles XD ).

TimmyD

Hexasudoku

Nathan Merrill

19:51

@sadakatsu right, but since you don't know, and you can't change it once you post, go on the safe side, and make them bigger

sadakatsu

@NathanMerrill No, I can put in a clause requiring that the puzzle generators be able to generate a puzzle given a size, and solvers need to be able to determine a puzzle's size based upon the passed input, then have them generate puzzles with 9, 16, 25, etc. puzzles when I run the competition.

Nathan Merrill

so, who wins? on size 9, 16, or 25?

if you aggregate their times, then it basically comes down to who does the 25 the fastest because it'll have a lot more weight than the rest

sadakatsu

I was thinking I'd use something like a Wilcoxon rank-sum to determine the winner. That would not care so much about the puzzle sizes or the actual times spent.

Nathan Merrill

I have no idea what that means, so you'd have to explain it in your post

and it seems like a lot of work to allow scoring for arbitrary puzzle sizes

sadakatsu

I had been planning that scoring system before I thought of different puzzle sizes.

Basically, all times for all puzzle solves are ordered from fastest to slowest. Each is assigned an index for where it is in the list. All the indices for the solvers and generators are summed together. The solver and the generator with the lowest sums win.

Nathan Merrill

19:59

that's not a bad system. It's interesting because it makes languages more competitive against each other than against themselves

like, I need to be hyper aware of the other submissions to identify where I should improve so that I can rank past more submissions

however, if 9x9 still comes down to microoptimizations like output speed and context switching, they'd still get a better score for improving on those factors

sadakatsu

By having each generator generate a lot of puzzles (I was thinking 100) and allowing each solver to solve each puzzle multiple times, I think it will be fair to both generators and solvers. Generators will not be faulted too much if they accidentally generate one problem that is too easy, and solvers will not be faulted too much if they perform poorly against just one problem.

New Main Posts

0

Q: Print all ASCII alphanumeric characters without using them

Challenge Print the following characters: abdcefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ1234567890 The catch is that you may not use any one of them in your code. You may print them in arbitrary order, with or without a leading or trailing newline, but you may not print any other char...

code-golf string restricted-source

Nathan Merrill

oh, generating lots of puzzles is a good idea. I just also think having them generate bigger puzzles as well

sadakatsu

Hm. I also want to make sure that I can run the competition in a reasonable amount of time. If I have COPS * 100 puzzles, and each solver solves each puzzle 100 times, then I have to run COPS * ROBBERS * 10000 tests.

Nathan Merrill

yep. Welcome to the world of fastest-code and king-of-the-hill

sadakatsu

20:05

But I am also not interested in a microoptimization puzzle, especially when the contestants do not have access to my computer.

Nathan Merrill

I'm very familiar with having my computer run for hours for a particular challenge

sadakatsu

And I'm not giving it to them, even if drug dealers seem to do that with their camera feeds ; P

Kritixi Lithos

I'm very happy with V's range operator right now :)

sadakatsu

I could theoretically thread the solvers to get through the results faster, but then they might compete for resources and hurt each other unfairly.

TimmyD

TIL PowerShell variables don't need to be alphanumeric provided you encapsulate the variable name in curly brackets -- i.e., ${#} is a valid variable name

Nathan Merrill

20:07

@sadakatsu wait, why are you generating different puzzles for each robber?

you definitely want the robbers to run the same puzzles as the cops

sadakatsu

Yes, I am.

TimmyD

Now to figure out how to cast as [char] without using char ...

Pavel

@TimmyD What are you trying to do?

sadakatsu

But if each cop generates 100 puzzles, and each robber has to solve each of those puzzles 100 times, that's COPS * ROBBERS * 100 * 100.

Nathan Merrill

no, you have the robbers solve each puzzle once

so COPS * 100 + ROBBERS*COPS*100

TimmyD

20:09

@Pavel wat's challenge that was just posted

Pavel

;-; RIP java

Eᴀsᴛᴇʀʟʏ Iʀᴋ

@devRicher what lang are you using?

sadakatsu

If I do that, then the robbers could be penalized or rewarded unfairly by luck of the draw.

Eᴀsᴛᴇʀʟʏ Iʀᴋ

I think so, it just depends on the lang and how else you can get input

Nathan Merrill

why?

Eᴀsᴛᴇʀʟʏ Iʀᴋ

20:11

for me or sadakatsu?

Pavel

I suppose possibly a lambda expression using _ as a variable, but I doubt it. (I don't actually know if _ is a legit Java var)

sadakatsu

@Pavel, I believe a Java identifier has to start with a letter.

Pavel

RIP java

betseg

@NewMainPosts there are two answers with less bytes than Dennis. woot

sadakatsu

(Looks it up) And an underscore is a "letter".

Pavel

20:13

Also, RIP most non=esolangs

Kritixi Lithos

@DJMcMayhem Why doesn't Ó×|_ work instead?

Eᴀsᴛᴇʀʟʏ Iʀᴋ

@betseg my challenge has no dennis answer, but slip (complete eso lang) is beating jelly by 5 bytes :D

Nathan Merrill

@EᴀsᴛᴇʀʟʏIʀᴋ that was for sadakatsu

Eᴀsᴛᴇʀʟʏ Iʀᴋ

ok figured

Geobits

Maybe it's just me, but having cops submit puzzles, with robbers writing a solution to all of them, sounds like it needs a new label. It's not exactly cops-and-robbers imo.

DJMcMayhem

20:14

@KritixiLithos because | isn't magic by default. You need \| or ü in V

Kritixi Lithos

Aah, thanks :)

Pavel

@Geobits It's like a weird kind of test-battery, almost.

DJMcMayhem

Anytime

sadakatsu

@NathanMerrill If someone writes a stochastic robber, they could get really lucky and try the right solution quickly, scoring really well for his timing and hurting the generator. Similarly, operating system performance variations could make a fast robber perform slowly on a puzzle in comparison to other solvers. However, multiple runs for each robber for each puzzle should smooth that error out.

Nathan Merrill

@sadakatsu simply having multiple puzzles solves that issue IMO

DJMcMayhem

20:16

Nice answer BTW

Nathan Merrill

it also drastically reduces the amount you have to run, and allows you to run more cops iterations

sadakatsu

So something like COPS * 10000 + COPS * ROBBERS * 10000...

Nathan Merrill

like, if you are running each robber submission 10000 times for each 1 cop submission, measuring the time the cop ran is kind of silly

sadakatsu

Oh, cop times would only be for disqualification for running too slowly or tie breakers.

Pavel

I was actually really surprised that so many people thought my prime-checking regex solution for this is so good. It isn't all that well golfed, and the regex has been floating around for a while.

sadakatsu

20:18

Robber times are used to generate cop scores.

devRicher

Oh, sorry, was doing something else meanwhile @EᴀsᴛᴇʀʟʏIʀᴋ

Eᴀsᴛᴇʀʟʏ Iʀᴋ

lol it's okay

what lang?

devRicher

Yeah I'm using RBX.Lua in Roblox studio

Nathan Merrill

@sadakatsu oh, that's problematic IMO. Like, generating an NP-compete puzzle (with one solution) is just as hard as finding the solution I think

Pavel

I'm actually only 28 rep away from my 1k, from having seven hundred something two days ago.

Nathan Merrill

20:19

maybe not, not really sure

Eᴀsᴛᴇʀʟʏ Iʀᴋ

@devRicher oic

sadakatsu

@NathanMerrill I figure it's a difficult problem to generate puzzles at all. However, the quality of the problem is dependent upon how difficult it is to solve, and that is measured by how robbers perform when solving it.

betseg

@EᴀsᴛᴇʀʟʏIʀᴋ in that challenge, can i pass the size of the maze? c doesnt know the sizes of arrays

Eᴀsᴛᴇʀʟʏ Iʀᴋ

?

in what way?

Nathan Merrill

right, but if you aren't time-scoring the cops, and simply setting a time limit, then cops have to spend as much time as possible generating a puzzle while hoping it doesn't go over the time limit on your computer

Eᴀsᴛᴇʀʟʏ Iʀᴋ

20:21

@devRicher yes, you can take input like that, since roblox doens't have really any other input method

devRicher

I mean, I don't expect to win with it anyways, but I like golfing in lua.

Eᴀsᴛᴇʀʟʏ Iʀᴋ

yeah, good luck!

@betseg how would the function be called?

betseg

@EᴀsᴛᴇʀʟʏIʀᴋ func(array, lines); (i can find the columns)

Eᴀsᴛᴇʀʟʏ Iʀᴋ

oh, I see

yeah, that's fine

sadakatsu

@NathanMerrill You don't think keeping cop times for tie-breakers is enough?

Nathan Merrill

20:24

I think you have to use it directly in the score. the chance of two submissions taking exactly the same amount of time is rather small

sadakatsu

@NathanMerrill Otherwise, someone will write a cop that removes just one cell to see if he can win on cop timings.

Nathan Merrill

right, you have to take both the cop times as well as the robber times

sadakatsu

Maybe a cop's score will be a value generated by 90% of his score from the robber times plus 10% of his score from the cop times?

Harder puzzles will still be better, but a reasonably hard generator that is fast might beat a super hard generator that is slow...

Oh, but cop scores from robber scores want to be high in a rank-sum score.

Pavel

I'm writing a 2d language similar to befunge or ><>. Does anyone know I can handle integers? I don't know how to differentiate push 4, push 8 and push 48

Nathan Merrill

what if you do copTime/robberTimeSum?

sadakatsu

20:28

I'd probably have to hack some simulators to see what those ratios would end up looking like.

Nathan Merrill

@PhiNotPi I'm interested in what you think (above conversation). What's a good way of scoring cops in a cops-n-robbers challenge where they generate/solve np-complete puzzles?

DJMcMayhem

@Pavel how about digits push to an incrementor, and a different command pushes the incrementor?

So, 0-9 means "multiply incrementor by 10 and add n" and <insert character here> means "push incrementor, set it to 0"

Pavel

That makes sense! Thanks.

Eᴀsᴛᴇʀʟʏ Iʀᴋ

@PythonMaster how in the world do you have 13k rep on meta.se but nothing even close on any other site? o_O

that's not bad, just suprising

Pavel

The language is called sushi. It's similar to ><> in the 2D language sense, but I can't actually understand ><>'s docs so I don't know how it works.

sadakatsu

20:35

@NathanMerrill (still waiting to hear what PhiNotPi thinks) If I think about the typical idea being Cops and Robbers, these are cryptographic puzzles in some way. Thus, a Cop that generates puzzles that take the most time to crack are better than those that take less, even if that Cop is slower than those other ones. I am worried that any kind of composite score will lose that.

Nathan Merrill

right, but its an arms race. you need to be able to generate puzzles faster than the robbers can solve them

while the robbers try to do the opposite

while a time limit would work, because you are running it on your computer, it's really hard to know whether you are under the time limit

betseg

@EᴀsᴛᴇʀʟʏIʀᴋ oh... i found that you can in fact find size of array of string

Nathan Merrill

because you really do want to use every last second

Eᴀsᴛᴇʀʟʏ Iʀᴋ

@betseg well than nope, sorry

Pavel

I'm so close to my 1k rep... it seems so close but there's nothing I could do to make it faster... It's been driving me nuts.

Eᴀsᴛᴇʀʟʏ Iʀᴋ

20:37

I don't C well, so input should be given as described in the question except when your language can't

Geobits

@Pavel You mean besides posting useful content to main?

Eᴀsᴛᴇʀʟʏ Iʀᴋ

@Pavel COMMENCES STRATEGIC DOWNVOTING

Pavel

There's nothing trivial enough that I can solve it easily.

wat

Hi, why was I kicked?

Pavel

I'm not actually good at this whole programming thing.

Eᴀsᴛᴇʀʟʏ Iʀᴋ

20:38

@wat prolly cause you were spamming chat

Pavel

@wat I was kicked for 24 hours not long ago, you're fine.

wat

@EᴀsᴛᴇʀʟʏIʀᴋ ...how was I spamming chat?

What did I even say?

Eᴀsᴛᴇʀʟʏ Iʀᴋ

i have no clue

but I assume so because that's the main reason people get kicked from chat ;p

Geobits

@wat Because you posted a giant picture of carrots directly following a conversation about chat noise.

4

TimmyD

Dangit, doesn't seem to be a way to cast to char without using alphanumeric

wat

20:40

sorry.

@Geobits oh, that was it.

Eᴀsᴛᴇʀʟʏ Iʀᴋ

there has been way too many flagged messages around here

that's not because of overzealous flaggers

no it really doesn't

why would it

Pavel

@wat For your challenge, can we write a function to return a string or char array?

wat

@Pavel reread the question, I just edited it

Pavel

Cool

TimmyD

@wat I suppose returning an [int] array of the corresponding ASCII values is too much of a stretch, eh?

wat

20:43

@TimmyD Yes.

TimmyD

Well, it was worth a try. :D

Pavel

8 rep to go...

betseg

ill uptove one of your posts and downvote one so you have exactly 1k

wat

@Pavel I'll upvoet a post

Geobits

-_-

wat

20:46

there did it

Pavel

:D

Eᴀsᴛᴇʀʟʏ Iʀᴋ

no

don't upvote posts like that

12

Pavel

^-^

Eᴀsᴛᴇʀʟʏ Iʀᴋ

seriously, not cool

Geobits

And they say voter fraud doesn't exist...

Pavel

20:47

I have 1032 now. You guys are awesome. AND I managed to not directly ask for upvotes.

Eᴀsᴛᴇʀʟʏ Iʀᴋ

that's not impressive at all

sadakatsu

@NathanMerrill An arms race was not what I had in mind, but that does seem to be more in keeping with the Cops and Robbers theme... hm.

wat

@Pavel I upvoted all your answers

betseg

-_-

Pavel

-_-

Geobits

20:48

@wat Congrats. Now let's wait for the script to reverse the serial voting.

wat

@Geobits wait it detects that?

Eᴀsᴛᴇʀʟʏ Iʀᴋ

of course it does

Pavel

I don't have enough answers

wat

I thought it only reversed serial voting from new/inactive users

Eᴀsᴛᴇʀʟʏ Iʀᴋ

somebody else probably did also

@wat no, anybody

wat

20:49

or just general bot like patterns

Geobits

That's the definition of serial voting

DJMcMayhem

47

A: I gave someone +200 rep and a Mortarboard Badge

I'm sure this is not really brightening anyone's day. You're not doing that person any favors. Someone might look at that and think this person is guilty of vote fraud. Please stop voting this way, it's more annoying than anything else.

You're not doing him a favor

Eᴀsᴛᴇʀʟʏ Iʀᴋ

it'll probably take a bit to reverse it, but it will

Geobits

@Pavel Nine answers is definitely enough to trigger it.

Pavel

That's fine. I was prolly going to get the 1k today anyway.

Eᴀsᴛᴇʀʟʏ Iʀᴋ

20:51

no it's really not fine

you begged for votes in chat, and somebody serial upvoted you

both of you are at fault

Pavel

Oh yeah, mortarboard badge definitely trips serial voting. So it's fine, the system will revert it, and nothing changes.

Eᴀsᴛᴇʀʟʏ Iʀᴋ

except for the fact you were begging for votes in chat and that it happened at all

wat

are snippets allowed?

Pavel

Usually not.

wat

wait, you got mortar board badge?

Pavel

20:55

I generated a bunch of upvotes organically today too.

wat

mk

Pavel

Snippets are only allowed if OP specifies.

Eᴀsᴛᴇʀʟʏ Iʀᴋ

@Pavel doesn't matter

@wat define snippet?

wat

@EᴀsᴛᴇʀʟʏIʀᴋ In this case, a string literal

Eᴀsᴛᴇʀʟʏ Iʀᴋ

no, sorry

TimmyD

20:56

(((((${#}=+!''+!'')*${#})*${#})*${#})*${#})*${#}+!'' <-- The number 65 in PowerShell without using alphanumerics TIO

wat

In ruby I got somewhere with $$

Pavel

That's uhh... really cool.

How does that work.

devRicher

A little bit of magic is required.

DJMcMayhem

It's a cool moment for me. Both of the two languages I've written were used to answer the same challenge within the first hour it was posted, and neither of them was written by me. :D

Pavel

Which languages have you written?

DJMcMayhem

20:58

V and brain-flak

Pavel

I though V has been a thing for a really long time, being built into Vim and all.

DJMcMayhem

Nope. It's an extension of vim, been around for a little bit less than a year

Pavel

Oh, cool.

Transcript for