The Nineteenth Byte

Alex A.

8:00 PM

@PeterTaylor He changed it to a code golf.

Beta Decay

Wooooo I am the best Beta of them all

Alex A.

I don't think it should be open in general, but the close reason at the time was no longer accurate.

Peter Taylor

@Sp3000 ??

Alex A.

Sp had a question for you a couple lines up from that

ETHproductions

I'm #402 overall

#36 this month

Peter Taylor

8:05 PM

Did he? I'm either not seeing it or not able to work out from the context what it is.

Alex A.

2 hours ago, by Sp3000

I feel like there should be a difference between if(!1) and if("!1") here... (confused)

Peter Taylor

Where?

Alex A.

Two lines up from what you replied to

Peter Taylor

No, there should be a difference between if(!1) and if("!1") where?

Alex A.

If you sift through the crap in that area, Sp and Mauris were talking about one of Mauris' solutions to Dennis' primes where he linked to something you said.

ETHproductions

8:11 PM

I'd gotten so used to my default blue avatar, I can't find any of my messages on here while skimming through :P

Dennis

This answer prints the string !1 as a falsy value:

1

A: Is this number a prime?

C, 63 bytes i,n;main(p){for(scanf("%d",&i),n=i;--i;p*=i*i);puts("!1"+p%n);} Prints !1 (a falsey value, by Peter Taylor's definition) if (n-1)!^2 == 0 (mod n), and 1 otherwise.

@PeterTaylor ^^

ETHproductions

There's someone who's earned -76 rep this year: stackexchange.com/leagues/88/year/codegolf/…

Peter Taylor

Ah, I see. Then I agree with Sp3000: I don't think it's a falsy value by my definition.

Alex A.

@Mauris ^^

Peter Taylor

I also don't think it's as golfed as it could be. Wouldn't putch(48+p%n) save a char?

(Or putch(49-p%n), whichever gives the correct output)

Mauris

8:13 PM

putch is nonstandard conio.h stuff

feersum

putch is Windows-only

Mauris

I'd need putchar, which is 1 longer

Geobits

I don't understand why it wouldn't count if printed as a string. The question says "output the string representation of a truthy/falsy value".

It's not printing the literal " marks.

feersum

!1 is an expression, not a value

Geobits

You could argue that 1 is an expression, too ><

feersum

8:15 PM

it is, and it's a value

Mauris

The definition as written says nothing about expressions or values

Geobits

It clearly fits the if(!1) test either way.

Mauris

ideone.com/43fb9s

^ All Peter's definition says is "If this construct doesn't compile or error, it's the final judge".

I'm okay with changing it to putchar, but then I feel like the definition should be expanded/clarified.

Dennis

The wording of my question seems a bit unfortunate. My intention was to allow truthy/falsy values as well as string representations thereof. So there's really no point in arguing, since putchar(p%n) is already shorter.

Mauris

That would print a raw 0x00 byte for false and 0x01 for true, which feels iffy

feersum

8:20 PM

@Dennis How can there be a difference between these two, when a full program is required?

Dennis

Not pretty, but I'd allow it in brainfuck.

Mauris

Sure, I'd agree, but in C I dunno...

I think I will change it to putchar(48+p%n)

Dennis

@feersum For numbers, no difference. But empty output is an empty string, as well as "".

At least that was the intention.

feersum

OK, string representations of strings aren't involved here though.

Geobits

I was originally hoping that I could finally use an exit code for output, but noooo, Dennis wants it printed :P

Mauris

8:22 PM

Hmm. What if I print 1 for true and nothing for false?

feersum

The empty string isn't falsy in C

Unless you want to say it's a string representation of a null pointer :P

Mauris

Hah: I could say puts(p%n*"1") printing either 1 (a true value) or (null) (or whatever your compiler's string representation of a null pointer is).

That is probably hella UB.

feersum

won't you need -fpermissive?

cjfaure

do i sense a doorknob

Doorknob

no

cjfaure

8:26 PM

doorknob senses u

Mauris

Ah, can't do int * char*.

Rainbolt

If a Russian tells a Russian meme joke, does it come out in the proper order?

cjfaure

@Rainbolt I doubt people who were alive for the whole Soviet deal are young enough to meme

Mauris

I could also say:

main(){puts("0");}

Rainbolt

@cjfaure Oh, is the meme specific to Soviet Russia?

Mauris

8:28 PM

For primes, it outputs the string representation of the string "0", a truthy value.

For non-primes, it outputs 0, a falsey value.

cjfaure

@Rainbolt I believe so

feersum

That doesn't print any quotation marks

Mauris

(I'm joking)

ETHproductions

@Rainbolt In Soviet Russia, joke incorrectly orders YOU!

feersum

The truthy char* "0" or a string representation of the int 0, I think you mean

ETHproductions

8:33 PM

"0"==false returns true in JS.

Oddly enough, !!"0" returns true...

Peter Taylor

@cjfaure Who are you calling old? Get off my lawn!

Doorknob

@ETHproductions And that is why you never use == in JS.

(For values of never meaning only if you have a really good reason.)

ETHproductions

Like golfing one byte?

feersum

s/ == in//

Doorknob

@feersum Yeah that works too

Mateon1

8:36 PM

JS is full of weirdness, I love doing things in JS

Dennis

@Mauris I wish I had noticed it earlier, but you cannot use Wilson's theorem in C because it won't work for all possible inputs.

Doorknob

oops, that's not the google search bar

5

Mateon1

Also, Python unary regex solution is way suboptimal for the prime golfing challenge

I can't get below 58 chars

Mauris

@Dennis Because of overflow?

feersum

Did you test it for 1-255?

cjfaure

8:38 PM

Hey look, you can star removed messages

Alex A.

No you can't

Mateon1

@Dennis Can't you insert the modulo into the factorial, to avoid overflow?

Mauris

Oh, right

cjfaure

aw damn, i figured the button being there meant that it worked

Mauris

Yes, I was thinking something like that might work @Mateon1

cjfaure

8:38 PM

as I did earlier cough @Geobits

Dennis

@Mateon1 That should work, yes.

Doorknob

@cjfaure The button isn't there.

Wait that's my own message.

I'm an idiot

14

Dennis

@Mauris Yes, the rules allow breakage because of overflow, as long as input 1 to 255 is supported.

feersum

Does it actually fail though? I thought you might be lucky and not have it go to 0 for any primes

even numbers tho

Mauris

i,n;main(p){for(scanf("%d",&i),n=i;--i;p=p*i*i%n);putchar(48+p);}

Mateon1

8:40 PM

Factorial of 255 is muuuch more than int can store. 255*254*253*252 is on the edge, and multiplying by 251 overflows (assuming 4 byte int)

Mauris

That might work. Do you know the inputs it failed on, @Dennis?

Alex A.

@Doorknob No you're not. You're one of the smartest people I've ever met on the Internet. <3

feersum

Why do you have to square i?

cjfaure

even smart people post their google searches sometimes

Mauris

I'm using Xnor's Corollary to Wilson's Theorem.

Alex A.

8:42 PM

You should be using Beyoncé's Theorem for Singleton Sets: If you like it then you should have defined a ring on it.

Mauris

int(isprime(p)) == fact(p-1)^2 mod p

Calvin's Hobbies

@MartinBüttner Does your language quiz challenge specifically say Dennis won in a comment I can link to or something?

Martin Büttner

no it didn't, I can edit the challenge though

Dennis

@Mauris It prints * for input 10.

Mauris

Oh. Hm.

Dennis

8:44 PM

The original, not the modified version.

Mauris

OK, my new one gives 0, here, at least

ETHproductions

Suggestion for Stuck v2: Identical to v1, but the blank program is a primality test.

Martin Büttner

@Calvin'sHobbies done. will edit again if there's a challenge to link to

Calvin's Hobbies

k

feersum

Wooo beating Java

even though Snails doesn't have numeric types or variables of any kind hehe

Nathan Merrill

8:51 PM

I'm working on a feature selection problem. How do you identify pairs of variables that are important? (Aka, Variable A and B are only important if both of them are high, otherwise they are both irrelevant)

Alex A.

High as in both exceed some threshold value?

yes

Are they correlated?

unknown

Random idea of the day: JSolang - Just like JS, but the syntax is as short as possible while still retaining all of its data.

Nathan Merrill

8:55 PM

I'm working on a game-playing AI that is able to (hopefully) play a human-designed game. I need to identify things that are important within the game

sometimes, those things come in pairs

cjfaure

ooh yes, AI

the field I pretend to know stuff in

Nathan Merrill

for example, if I'm playing poker

Peter Taylor

@feersum Actually, tying, once Geobits edits in my corrections.

ETHproductions

Can't wait to see it when you're done :)

Nathan Merrill

and there is a King on the board, and King in my hand

that's important

Dennis

8:55 PM

@AlexA. The ngn/apl online demo is the only interpreter I'm aware of that prints automatically in APL, and it doesn't work with ⎕. :(

feersum

:P

Nathan Merrill

but if there isn't a King on the board, then its much less important

Alex A.

@Dennis baruchel.hd.free.fr/apps/apl

feersum

guys what's a good tool for drawing squares

Nathan Merrill

excel

feersum

8:56 PM

maybe with soem arrows

Nathan Merrill

it does it automatically

ETHproductions

Random AI question of the day: In deterministic 2048, shouldn't it be possible to get to the 65536 tile?

feersum

don't think it does arrows though

Nathan Merrill

probably not :P

feersum

even getting h = w is a PITA for Excel

Nathan Merrill

8:58 PM

turtle graphics

in all seriousness, I'd probably use paint

ETHproductions

^ Me too

I actually got this:

define six as 1+5;
define nine as 8+1;
return six * nine;

To run as JS, using a custom regex before eval()

It returns 42, as expected ;)

cjfaure

well yeah, 1+5*8+1 == 42

ETHproductions

@cjfaure Here's where that came from: codegolf.stackexchange.com/a/21887/42545

cjfaure

@ETHproductions ah yes~

Alex A.

@Dennis I wrote and tested it in the desktop version of Dyalog APL (but TryAPL doesn't support input)

Dennis

9:09 PM

@AlexA. If it works in Dyalog, sweet. The GNU APL online interpreter is a REPL though.

Alex A.

Oh right

Dennis

And GNU APL is pretty much unusable for golfing otherwise. :P

Alex A.

Haha

Dennis

I'm serious. It refuses to interpret EOF as the end of the file. You have to write )OFF.

And I still couldn't figure out how to print.

Alex A.

Oof. I'm never used GNU APL on desktop. Seems unnecessary when I have the desktop version of Dyalog.

(Which you can get for free, btw. Just claim you're a student.)

Dennis

9:13 PM

You also have to choices: script mode, where everything goes to STDOUT (including error messages) and normal mode, where only the banner goes to STDOUT (the rest to STDERR).

Alex A.

Who... who thought of that

ETHproductions

Well, that's dumb

Mauris

Can someone link me to the What counts as a programming language? thingy?

Dennis

Saw !!SuperPrime?

Mauris

I can't find it on Meta (I suck at finding things.)

ETHproductions

9:14 PM

17

Q: What are programming languages?

Often, answers to questions asking for "programs" or talking about "programming languages" utilize things like sed, awk, … in order to get around having to write an actual shell script. Therefore, a question comes to my mind: What qualifies as a programming language? Sure, ultimately the OP ca...

discussion

I think that's it

Mauris

No, writing up something silly.

The sequel to Help, WarDoq!, you might say.

Dennis

Haha.

Alex A.

Help, WarDoq! 2: Help, WarDoq Harder

The WarDoqening

Mauris

Hmm, does Help, WarDoq! satisfy the first constraint, there?

ETHproductions

Well, it has a primality test and can add two numbers

I think that's what he was going with

Mauris

9:17 PM

It feels like it can add two numbers without supporting a representation of them.

ETHproductions

IMHO, HWD isn't a valid programming language for challenges on this site.

Dennis

It supports them in the input, which is something. :P

Doorknob

@ETHproductions Why not?

ETHproductions

Well, even if it's valid, it unuseful for all but 2, AFAIK.

Doorknob

Yes, but it's still valid.

Dennis

9:20 PM

valid != useful

ETHproductions

I know, but...

Doorknob

Something like "Java, except every program that fails to compile prints Hello World" is also valid.

ETHproductions

>
Support a representation of natural numbers and of tuples.

Does it support tuples?

(from the top answer on that meta post)

Mauris

I'm sort of siding with ETHproductions, but at the same time, I realize that there is nothing stopping a new version from fixing this issue in what is still a very useless way.

Doorknob

The easy solution is "don't use Hello, World!—be more creative."

Mauris

9:23 PM

Like, add in a stack, on which you can push, pop and print integers; and pack/unpack them into tuples.

Doorknob

@Mauris Or this:

3 mins ago, by Doorknob

Something like "Java, except every program that fails to compile prints Hello World" is also valid.

Dennis

The string "Hello, World!" is a tuple, isn't it?

Optimizer

That moment when your favorite anime starts playing traditional Indian music in the background o.O

Mauris

I dunno... There are no projections (X1, X2, ... Xn) → Xi

ETHproductions

I would add as a requirement "contains at least one memory cell that can be written to and outputted from" or something similar.

Mauris

9:26 PM

Arguing that "Hello, World!" is a tuple of characters feels a little like arguing that 12 is a tuple of bits

mbomb007

I have a Python question, if someone could help. I'm trying to trim a floating point number to a certain number of decimals for printing. When I try this method, it works with the small example, but not in my program, which prints letter frequencies by position in the word.

Mauris

If there's no way to access the individual elements, I mean

Alex A.

@Doorknob Rust question for you

Mauris

(From a category theory perspective, those projections are all that defines a tuple!)

Doorknob

@ETHproductions What's the point? You could still create "Java unless it fails to compile in which case interpret as Help, WarDoq."

mbomb007

9:27 PM

It's just printing zero, instead of the floating pt values.

Doorknob

(Cause, ya know, Java.)

Dennis

@Mauris I should have taken your original suggestion and implemented a interpret-source-code-as-CJam built-in.

Help, WarDoqs! is on its way. :P

Mauris

@mbomb007 You are probably running into Python's integer division thingy. Try adding float() around an expression on either side of /?

Dennis

(That's what I get for reading only the executive summary.)

mbomb007

@Mauris It doesn't work in Python 3 either, so I don't think that's it.

Doorknob

9:29 PM

@Dennis Or you could do "CJam, except one-character programs print variations of HW, FizzBuzz, 99 bottles of beer, etc." :P

Dennis

Or inflate, then interpret it as CJam. Bubblegum with steroids. I can already taste the downvotes.

mbomb007

@Mauris Python 3. It keeps the floating point decimals far out.

@Dennis To Beta Decay, downvotes taste like candy.

Doorknob

@mbomb007 I didn't know Beta Decay liked to eat Geobits...

3

@AlexA. I'm waiting... :P

Alex A.

@Doorknob Oh haha. How does read integer from STDIN? :/

Doorknob

@AlexA. Is the integer the only thing on its own line?

Alex A.

9:34 PM

Yes

mbomb007

@Doorknob Geobits sound like they could be some variety of cheesy snacks. That, or pet food.

Doorknob

And is this intended for golf, or actual code? :P

Alex A.

Haha golf

Doorknob

Oh. This might change the answer, then.

Hmmm. There's lots of ways to read from STDIN; I'm not sure which is shortest.

Alex A.

I don't know any ways, at least none that work in the Rust playground.

Doorknob

9:36 PM

The way that immediately comes to mind is with std::io::prelude, but that's lots and lots of chars.

mbomb007

@Mauris Example: print(float('0.500000001')), once I trim the value and wish to convert back into a decimal, it has inaccuracy due to precision.

So does Python, despite having arbitrary precision integers, still have limited precision floating point values?

The above prints 0.50000000099999997, btw.

Alex A.

@Doorknob Is that like use std::io::prelude::*; then BufRead or something?

Mauris

You can import Decimal or something.

Doorknob

@AlexA. BufRead? I've never heard of that. But it's the super long use statement, yeah. I'm trying to figure out the shortest way now...

Martin Büttner

@ETHproductions please don't approve suggested edit which golf an answer

10

Q: Suggested Edits: Reject Golfing?

Every now and then, I see an edit where somebody does some golfing (most recent). From what I can tell, our standard is to reject those edits as invalid and post a comment saying what was suggested, because we feel that golfing tips should be comments. And I have always done so (after I learned)....

discussion editing reviewing

Alex A.

9:39 PM

^

@Doorknob Maybe just Read? I was looking at the docs for prelude.

Doorknob

Ugh. I'm trying to test stuff, so I made a file called asdf.rs. Turns out the file called asdf already exists, so the compile failed. :/

Alex A.

Haha

Doorknob

Anyway. Baseline is this:

    use std::io;let mut s:String=String::new();std::io::stdin().read_line(&mut s);

(That avoids having to bring in prelude.)

Alex A.

There's no way to do a direct to i64 or something rather than starting with s: String?

New Meta Posts

1

Q: Disallow edits to answers that only golf code further

This answer had 2 suggested edits that only golf the code further, and the edits were both approved. I personally disagree with edits that only golf the code further. Here's why: It shouldn't be up to the reviewer to make sure a code edit works. That should be the job of the answerer. If an edi...

discussion editing

Doorknob

9:42 PM

Oh. Nope, I don't think so.

(I could be wrong though.)

Alex A.

But I could use s.parse() to get i64?

Doorknob

Uh. No, I'm not sure how that would work.

Alex A.

Hm, okay.

Doorknob

The normal way would be with from_str, not sure if there's anything shorter.

Where'd you find that parse method?

Alex A.

Internet

haha

ETHproductions

9:45 PM

@MartinBüttner Oh, sorry, I didn't realize that was a bad thing (although I guess I wouldn't do it, so I should have thought of that...)

Doorknob

Oh, is it this?

Martin Büttner

@ETHproductions no worries, not everyone can have read every meta post. that's why I'm telling you ;)

Doorknob

That's a generic method, which means you have to specify a type.

Calvin's Hobbies

If anyone is curious about the PPCG Minecraft Server, I made a video tour:

Minecraft - Tour of the PPCG Server

►

EvilJS

Hello. I was about to ask about code golf Question, but some similar was rejected. Is raytracer with some constraints (and the same task for all) proper?

ETHproductions

9:49 PM

@MartinBüttner Thanks! :)

Alex A.

@Calvin'sHobbies You have a very radio-friendly voice.

Calvin's Hobbies

@AlexA. That's good to hear :)

ETHproductions

@EvilJS Welcome to PPCG! If you post your challenge to the Challenge Sandbox, you can recieve feedback before you post it on the main site.

Alex A.

@Doorknob is.gd/SKVkTq

I get error

Doorknob

Huh? No error for me.

EvilJS

9:52 PM

@ETHproductions ok Thank you.

ETHproductions

@Alex Is is.gd a generic link shortener, or some other shortener designed for that website?

Alex A.

@Doorknob Or warning, rather: warning: unused result which must be used, #[warn(unused_must_use)] on by default

ETHproductions

Ah, nvm, I found it online

Doorknob

It's code golf who cares about warnings

Alex A.

@ETHproductions It was the automatically generated short link from the site I was on

ETHproductions

9:53 PM

Oh, cool

Alex A.

@Doorknob :P I don't understand though because I am using the result. I'm printing it.

Doorknob

@AlexA. No you're not.

s isn't the result. It's a mutable argument.

Alex A.

...oh

Doorknob

The result of read_line is probably a Result<something, Error> of some sort.

Alex A.

@Calvin'sHobbies REGGIE, MY ASS CHILD!

ETHproductions

9:54 PM

Sometimes I wish my siblings were mutable

Mauris

dl.dropboxusercontent.com/u/15495351/gamma.html

@Dennis: ^ It's horrible. I feel guilty.

Doorknob

Yeah, I think the shortest you're gonna be able to get is:

pub use std::io::*;let mut s:String=String::new();stdin().read_line(&mut s);

Dennis

@Mauris That's covered by out standard loopholes.

Doorknob

I always write my Rust entries as functions though. Does the challenge require something different?

Alex A.

Full programs

Doorknob

9:58 PM

:/

Alex A.

I'm (foolishly) trying to make my first ever Rust program the golfed primality test.

ETHproductions

When you give up on that, try a "Goodbye, World" program

Alex A.

Who says I'm going to give up? :P

Mauris

I guess MetaGolfScript is. I think I could get away with this if I get very lawyer-y about it. (Not that I want to.)

Doorknob

@AlexA. Anyway, you can use that code, and once you've got the String I think something like (&s).parse::<u64>() would work (not sure if first pair of parens are required or not).

Calvin's Hobbies

10:01 PM

@AlexA. "Ass child" just sounds gross

Mauris

Does Rust still take like five billion hours to compile and get to work?

I kind of want to give it a try.

Doorknob

@Mauris Assuming you're on a Unix OS, it's a single shell script that takes ~2 minutes to run.

Actually, you might be able to just grab the binary.

Rustup is convenient though.

Calvin's Hobbies

@AlexA. But yes, Reggie makes a cameo :P

Mauris

OK, that was fast.

Alex A.

@Calvin'sHobbies I know, I saw. :)

Mauris

10:04 PM

I remember a friend telling me she needed to leave her laptop on overnight to compile the Rust compiler

Alex A.

D:

Mauris

I guess not needing to do that helps!

Doorknob

The Rust compiler is written in Rust. How are you supposed to compile the compiler without a compiler?

ETHproductions

Why does Rust take so long to compile?

Mauris

I remember this was when Rust was still very young, so maybe you had to compile it with a slow, bad version of Rust?

I don't quite remember the history.

Alex A.

10:06 PM

@Doorknob That sentence hurts my brain

ETHproductions

I can't write a JS function anymore without trying to golf on some level :P

Alex A.

@Calvin rofl your reaction on the elevator

xnor

P=n=1
exec"P*=n*n;n+=1;"*~-input()
print P%n

@Sp3000 I think this saves 2 chars

rather that saving the input to a var, you count up to it

@Mauris

Mauris

10:24 PM

Ooh, neat.

Faraz Masroor

hey guys whats up

Mauris

Hi!

Faraz Masroor

where do you guys learn such arbitrary languages

Doorknob

;)

Faraz Masroor

=/

Alex A.

10:39 PM

Haha

It's true

Mauris

Nice, someone submitted in ECMAScript 2015.

What a super-formal way to refer to that language.

feersum

> Brendan Eich reads up on every mistake ever made in designing a programming language, invents a few more, and creates LiveScript. Later, in an effort to cash in on the popularity of Java the language is renamed JavaScript. Later still, in an effort to cash in on the popularity of skin diseases the language is renamed ECMAScript.

Alex A.

Hahaha

@PhiNotPi You should watch Calvin's MC server video

PhiNotPi

10:55 PM

@AlexA. watching it right now

Alex A.

@PhiNotPi Ok

Calvin's Hobbies

@Dennis @MartinBüttner The Dennis challenge is ready: codegolf.stackexchange.com/questions/57719/…

New Main Posts

0

Q: Generate Dennis Numbers

This challenge is a tribute to PPCG user Dennis for winning the robbers' part of The Programming Language Quiz. Looking at Dennis' PPCG profile page we can see some pretty impressive stuff: He currently has over thirty-seven thousand reputation, making him second in rep overall, surpassing th...

code-golf number sequence palindrome

Alex A.

@Calvin'sHobbies It got auto-flagged as excessively long XD

Calvin's Hobbies

Yeah, it's all the tabs that got turned into spaces

Alex A.

10:59 PM

Rightfully so.

Dennis

@Calvin'sHobbies Great challenge!

It will never cease to amaze me how you come up with all these ideas.

^^^

^^^

^.^

It should be the next OEIS sequence

Calvin's Hobbies

11:14 PM

I have a lot of sequences around PPCG that could go on OEIS

Alex A.

OECH: Online Encyclopedia of Calvin's Hobbies

New Main Posts

0

Q: Generate ordered binary combinations without repetitions

Challenge Write the shortest program that receives two signed integers n and i and for each i between 1 and 2^n - 1 returns the next ordered permutation based on the binary representation of the number. There is no specific order of the combinations but the number of 1s in the binary representa...