The Nineteenth Byte

Pavel

8:01 PM

Whoever came up with if/fi, case/esac, while do/done deserves to be shot.

14

FlipTack

harsh

Erik the Outgolfer

but true

muddyfish

@cairdcoinheringaahing here's an 8 byte solution for pokemon: w\x817F~_+.o Try it here!

AdmBorkBork

You mean why isn't it od to end a do block? :) — Wildcard Jan 18 '16 at 22:37

DJMcMayhem

@Pavel endif/endwhile is much nicer if you have to use keywords, however I much prefer either braces (C-style) or indenting (python-style)

Keyword control flow just doesn't feel as nice

Pavel

8:05 PM

@DJMcMayhem Or eve ruby/perl end

muddyfish

I prefer indentation. It's pretty

AdmBorkBork

Indentation is fine up to about four indents

Erik the Outgolfer

@DJMcMayhem that can prove golfier under certain circumstances ;-p

like python's return/break/continue/etc.

DJMcMayhem

Vimscript is even worse. It has endfunction

And endfor

And endtry

AdmBorkBork

> YOU HAVE BEEN TERMINATED

Pavel

8:08 PM

Yep. That's definitely worse.

Both of those.

VoteToReopen

Probably also on Programming Puzzles and Code Golf site — 707 Mar 3 '14 at 13:35

Do we have plain obfuscation challenges? Not very objective, so probably not...

Pavel

@VoteToReopen Used to, underhanded

VoteToReopen

Fair point.

Pavel

Also maybe useful for restricted-source

AdmBorkBork

@Pavel Oh, durr, I forgot about Select-String. You can do something like sls bar foo and get something like foo:33:blah blah bar blah back.

Pavel

8:23 PM

@AdmBorkBork Ah. That's cool, is there also an equivalent to the sed s/foo/bar/ command?

Anonymous

@AdmBorkBork Haha no :P

AdmBorkBork

Not that I can think of, you'd need to gc file |%{$_-replace'foo','bar'}

Pavel

That's not so bad.

I guess

:O Just had a brilliant idea

MinGW has windows executables

AdmBorkBork

Direct text manipulation isn't really PowerShell strong suit, so it's a little clunky

Pavel

I can add MinGW's /usr/bin/ to path

VoteToReopen

8:26 PM

Is there a way to get the feeds of this room exclusively? Without the other messages, that is.

Pavel

And use PowerShell and grep/sed/whatever

AdmBorkBork

lol, you could be sneaky and do something like

caird coinheringaahing

@Mego I'm genuinely interested how you found that, given it was said 4 hours ago. Do you just read through the transcript before joininin chat?

AdmBorkBork

filter sed($before,$after) { %{$_ -replace $before,$after} }

in your profile

Anonymous

@cairdcoinheringaahing A bit. Chat was scrolled up to just above that from leaving the tab open overnight, and the mention of QFT caught my eye.

Pavel

8:30 PM

@AdmBorkBork Too late, already did the MinGW thing

AdmBorkBork

Well, that works too.

Pavel

:P

This was an excellent idea.

Anonymous

@FlipTack Heresy

AdmBorkBork

I don't have any languages installed on my machine except for those that came with Windows -- PowerShell and BATCH

and I suppose VBScript, but who even uses that anymore?

Pavel

C:\Users\Pavel
λ  "bar
>> foo
>> baz" | set-content test
C:\Users\Pavel
λ  if ( $(grep foo test) -eq 'foo' ) { echo 'ye' }
ye

Am I doing powershell right

AdmBorkBork

8:36 PM

lol

Anonymous

@AdmBorkBork I think Windows still has a JScript engine

Anonymous

Though I don't know if that's installed/enabled by default

Pavel

@AdmBorkBork How do I run a command as admin in powershell

Lembik

CMC: sample a multiset of size n from the integers 1...n uniformly at random

Mr. Xcoder

@Lembik Pyth, 3 bytes O^S

S takes the range [1, input], ^ is Cartesian Power with the input, O takes a random elements.

caird coinheringaahing

8:53 PM

Hmm, has anyone else read through the moderator election page? I think it's a fascinating read :P

Lembik

@Mr.Xcoder Wow. Is each multiset really equally likely?

Mr. Xcoder

Yes.

A Python equivalent would be:

lambda k:r.choice([*i.product(range(1,k+1),repeat=k)])
import itertools as i,random as r

Try It Online!

Lembik

what does that header do?

f=\

Mr. Xcoder

Assigns the lambda to a function called f

Lembik

it's very impressive

Mr. Xcoder

8:56 PM

@cairdcoinheringaahing I did. Could you link me to that again, I can't find it for some reason.

Lembik

normally you have to worry about stars and bars

Mr. Xcoder

@Lembik Thanks :D

caird coinheringaahing

@Mr.Xcoder codegolf.stackexchange.com/election/1

Mr. Xcoder

Thanks!

caird coinheringaahing

I think it's very interesting to see the community's reaction to Mego's nomination

Lembik

8:56 PM

how did you avoid having worry about stars and bars?

or is this equivalent?

Mr. Xcoder

What stars and bars...?

Sry I am dumb :D

Pavel

@cairdcoinheringaahing If there was an election now, Mego would totally win.

Lembik

@Mr.Xcoder en.wikipedia.org/wiki/Stars_and_bars_(combinatorics)

@Mr.Xcoder math.stackexchange.com/questions/781125/…

etc.

caird coinheringaahing

@Pavel Yeah, I have a feeling that Alex might lose his place to someone else, most likely Mego

Mr. Xcoder

@Lembik Oh right... No, random.choice handles that very well :)

Pavel

8:58 PM

@cairdcoinheringaahing RIP Alex

Lembik

@Mr.Xcoder nice

Mr. Xcoder

anyway gtg o/

caird coinheringaahing

o/

Pavel

How do you see the list of diamond mods of a site?

caird coinheringaahing

@Pavel Users -> Moderators

Pavel

8:59 PM

thanks

VoteToReopen

Does anybody know at what amount of reputation one can create a chatroom?

caird coinheringaahing

@VoteToReopen 100 rep

VoteToReopen

Oh, huh.

Must've missed it. Thanks!

flawr

@VoteToReopen O_o what did you to to VTC?

Mr. Xcoder

They said they don't know them (and that VTC was their inspiration)

VoteToReopen

9:02 PM

Heh. I did intend to change my name but I haven't come up with anything.

I'm not incredibly creative...

caird coinheringaahing

@VoteToReopen Yeah, I had that problem. I changed my username 6~8 times in a month

Mr. Xcoder

@Mr.Xcoder I don't know them, but my name was inspired by theirs. Is that a problem? EDIT: Never mind, might as well change my name. — VoteToReopen yesterday

Hehe

VoteToReopen

Also, sorry for the late reply. :P

Mr. Xcoder

now I really have to go :D

flawr

@Mr.Xcoder cu!

VoteToReopen

9:04 PM

Bye.

flawr

@Mr.Xcoder every sock says exactly that

VoteToReopen

Who you calling a sock? :P

caird coinheringaahing

CMP: How many 0 voted answers do you have?

22/94 for me

VoteToReopen

0/1.

Mr. Xcoder

@cairdcoinheringaahing 79/426

muddyfish

9:12 PM

@cairdcoinheringaahing ~82/531

flawr

20/314

Adám

@Lembik Dyalog APL, 3 bytes: ?⍴⍨

caird coinheringaahing

ಠ_ಠ I just saw this challenge, and got annoyed because its basically the first challenge I posted in the Sandbox, before abandoning it because it was poorly made.

Anonymous

@cairdcoinheringaahing 68/442

Adám

CMC: Random int in range [1,∞). You may chose distribution, but all values should be possible (at least in principle; only limited by implementation and/or hardware).

VoteToReopen

9:22 PM

import sys,random:print random.randint(1,sys.maxint)?

dzaima

@Adám SOGL, 4 bytes - while random(9) {inc()}

Adám

@VoteToReopen No, that has an upper limit.

@dzaima Ah, I think I get it.

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Square-free Rock-Paper-Scissors tournament On the day of the Codegolf Rock-Paper-Scissors tournament you hear through the grapevine that everybody else is going to play a fix square-free sequence (a sequence made of the letters R, P, S is square-free if it does not contain a subsequence that rep...

Uriel

@Adám is this possible in APL?

muddyfish

@Adám Pyke, 7 bytes, ohHIr)o Try it here!

Adám

9:26 PM

@Uriel Translation of dzaima's SOGL answer: 1+⍣{1=?9}0

@Uriel … but I can do it in 4 too.

Uriel

@Adám oh, I didn't see the "You may chose distribution"

Adám

@Uriel If not for a non-linear distribution it would be impossible in every language.

flawr

@Adám fix(1/rand) should work in MATLAB, up to the floating point limits.

Adám

@flawr fix is floor?

flawr

Right, but it rounds towards zero.

Adám

9:32 PM

@flawr That's the solution I had in mind in APL.

caird coinheringaahing

Would you guys mind giving me a bit of feedback about a challenge idea I have?

Uriel

@Adám so ⌊÷?0?

Adám

@Uriel Yup. Nicely done.

DJMcMayhem

import random
f=lambda:random.randint(0,1)
s='1'
while f():s+=str(f())
print(int(s,2))

@Adám Python ^

Heavily heavily weighted towards 0 1

But in theory, any number is possible

Oh wait

Yeah, just noticed

Fixed

Also, not very golfy

FlipTack

Nice solution though

You could make it a recursive lambda for a nice golf

caird coinheringaahing

9:36 PM

Basically, the cops have to modify their language to make addition non-commutative, but their program has to be able to be changed so that multiplication is non-commutative instead. The multiplication program has to be a levenstein distance of to be decided from the addition program, and is trying to be found by the robbers.

Adám

@DJMcMayhem OK, so same principle as dzaima's SOGL answer, i.e. your Python code is 1+⍣{?2}0 in Dyalog APL, assuming ⎕IO←0.

DJMcMayhem

@FlipTack Actually, There's a much more obvious golf though.

from random import*
f=lambda:randint(0,1)
s=''
while f():s+='0'
print(len(s))

Not as big, but way more obvious

FlipTack

`import random
f=lambda n=0:random.randint(0,1)and f(n+1)or n`

damn how does formatting work here

Adám

@DJMcMayhem Why can't you do while randint(0,1): ?

caird coinheringaahing

@FlipTack single backtick

DJMcMayhem

9:38 PM

Duh

muddyfish

@Uriel that has a limit due to floating point presision

flawr

@Uriel is random.random() never zero?

DJMcMayhem

@FlipTack Just paste it in multiline, then hit the fixed font button

FlipTack

import random
f=lambda n=0:random.randint(0,1)and f(n+1)or n

thanks!

dzaima

@DJMcMayhem why not just an integer counter?

DJMcMayhem

9:38 PM

@dzaima Because I'm an idiot

Adám

@FlipTack Your name is ⊖⊢ in APL.

FlipTack

interesting.

What do Flip and Tack actually do?

Flip is probably as it sounds, but Tack...

DJMcMayhem

I'm guessing flip is reverse, and tack is append?

Adám

@FlipTack ⊖ Flip flips tables (matrixes) upside down. ⊢ Tack is the identity function.

FlipTack

I'm about as useless as a reversed identity function

mbomb007

9:41 PM

Anyone else want to do a bounty on the Tetris in CGOL question? This answer only got 500 rep and the others got 1000.

Crowdfund rep for muddyfish

muddyfish

:P
I thought it would be rude to mention it

DJMcMayhem

@mbomb007 Yeah, sure

Adám

@FlipTack Not useless at all, f←⊖⊢ returns the flipped right argument. Try it!

FlipTack

you learn something new every day

mbomb007

@FlipTack Wait, really?

DJMcMayhem

9:44 PM

@mbomb007 Done

@FlipTack TIL

3

mbomb007

@HyperNeutrino Am I my language's keeper?

dzaima

@Adám Perfect time for errors :p

DJMcMayhem

Underrated comment:

Oh you avoid politics, Star Wars, and religion, you say?! Well let me get you started on the politics of the Star Wars religion! — corsiKa 2 hours ago

Adám

@dzaima Yeah. It is being worked on. Meanwhile, using an incognito tab should fix it until the server resets your session.

@dzaima I just fixed another bug last week which cause the server to hang with certain input…

Luckily it is running in Docker and is being polled for availability, so Docker will kill it and restart it whenever that happens. Anyway, I've fixed that bug – just need to commit.

Lembik

@Adám wow :)

Adám

9:57 PM

@Lembik Well, it is a pretty odd edgecase of various factors clashing, so nobody thought of it during the last 5 years.

ATaco

10:20 PM

I have no idea how to grasp just how incomprehensibly massive TREE(3) is.

Eric Tressler

@ATaco there are easier hard to grasp numbers

Now I remember scottaaronson.com/writings/bignumbers.html, but there's a similar article by someone else that I can't think of

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Self-compiler/interpreter code-golf quine string classification Create a language validator, which takes a string/file that might or might not be syntactically in the language of your choice, and returns a value corresponding to whether or not it is a valid program in the implementing language....

Eric Tressler

Maybe I'm thinking of the Strong Law of Small Numbers (maa.org/sites/default/files/pdf/upload_library/22/Ford/…)

DJMcMayhem

@ATaco The problem with TREE(3) is that I've never seen any lower bound for it or estimate of how big it is. At least with Graham's number, I understand the steps to get to it, even though it's still too big to comprehend

Actually now that I think about it, I know the size it Graham's number, but not what it answers. And I know what TREE(3) answers but I don't know how big it is lol

Eric Tressler

10:40 PM

@DJMcMayhem genealogy.math.ndsu.nodak.edu/id.php?id=145932

I don't actually understand why TREE(3) is big

ATaco

Oh, You've got a Ph.D?

Consider me rather impressed!

Eric Tressler

@ATaco Yeah, but that wasn't what I was trying to show DJMcMayhem; Ron Graham was my advisor.

ATaco

Wait, as in Graham's number Graham?

Eric Tressler

yeah

ATaco

o.O

DJMcMayhem

10:46 PM

@EricTressler woah

Eric Tressler

I worked on the lower bound for the problem related to Graham's number while I was grad school, but this paper (arxiv.org/abs/0811.1055) came out while I was working on it

3

For showing lower bounds to those things, a lot of times you just need to write a better program to find an example. And every 5-10 years, thanks to computing power going up, that can sometimes be easy.

ATaco

My knowledge of large numbers ends with Graham's number's definition and the feeling of "Crikey that's large!"

Wait, if you've worked as a student of Graham, does that mean you have a Graham number of 1?

Eric Tressler

I have never heard anyone refer to that, but I have an Erdős number of 2

ATaco

(It was a bad joke regarding Erdös numbers)

And that's pretty incredible, you should be proud of where you are academically.

Joshua

I agree it was bad.

DJMcMayhem

10:57 PM

Random question: what is 1/3 + 1/5 + 1/7 + 1/9... to infinity?

Eric Tressler

@DJMcMayhem diverges

DJMcMayhem

I think it's supposed to work out to something nice if I remember correctly, but I don't remember what

Joshua

I thought it was pi but I doubt my memory.

Eric Tressler

it diverges for the same reason 1/2 + 1/3 + 1/4 diverges. The alternating sum is nice, though

I think 1/3 - 1/5 + 1/7 - ... is pi/4 or something

DJMcMayhem

@EricTressler Ah, that sounds right

It was a typo in my program that they were all positive, I think it was supposed to be alternating

Eric Tressler

10:58 PM

Leibniz formula for π

See List of things named after Gottfried Leibniz for other formulas known under the same name. In mathematics, the Leibniz formula for π, named after Gottfried Leibniz , states that 1 − 1 3 + 1 5 − 1 7 + 1 9 ...

Joshua

series: sum(cos(x)/x) is really hard to figure out if it diverges or not

ATaco

The series diverges.

Joshua

I constructed it specifically to break all the tests in the book and run off the end of the decision ladder.

ATaco

(I only just realised Eric answered it first)

@Joshua I'm going to guess it converges.

Eric Tressler

@Joshua It depends on n being sufficiently far from a multiple of pi often enough, right?

DJMcMayhem

11:01 PM

Someone sent me this picture, and I was trying to figure out what it works out to:

scontent-sjc2-1.xx.fbcdn.net/v/t1.0-9/fr/cp0/e15/q65/…

Eric Tressler

I think it diverges, it's just a mess to prove it

ATaco

Desmos makes it look like it converges, but that's Desmos.

desmos.com/calculator/rngmofg9h7

Eric Tressler

oh, you're probably right, I was thinking of |cos(x)|

but I don't know how you'd prove it, maybe by expanding cos(x) to a few terms and finding a way to throw out the end of the taylor series

ATaco

It still looks like a pain to prove.

And just as hard is to prove where it converges to.

Eric Tressler

What is the sin() doing there?

ATaco

11:08 PM

Oh, that's sum, not sin...

I am the fool.

Eric Tressler

I've never seen this site, it looks nice

VoteToReopen

Microsoft employee installs Chrome mid-presentation because Edge keeps crashing

Eric Tressler

Wolfram alpha has a little too much random information sometimes to use it as a plotting tool

ATaco

I'm pretty hopeless without it, it's a fairly powerful graphing calculator.

(Plus the code is Valid MathJax)

Taken straight from Desmos code.

HyperNeutrino

well I think my school board's email servers died

or nvm works now :P

Eric Tressler

11:15 PM

I was looking to see if desmos uses D3, but I don't see it there, so maybe not

ATaco

I'm not sure what D3 is.

HyperNeutrino

how to handle git conflicts: 1) sudo rm -rf repo/ 2) git clone https://github.com/username-here/repo.git

VoteToReopen

10/10.

ATaco

> -rf

You madlad

VoteToReopen

Why?

HyperNeutrino

11:17 PM

well ok fine to be fair when I just did that I didn't do the -f part

ATaco

(If you need to sudo to remove a git, you might have something set up wrong)

HyperNeutrino

well there are write-protected files in it

ATaco

Eh, a-ta.co is stored on root without source control.

HyperNeutrino

lol

eaglgenes101

I would type up a clever means of putting "hello world" here, but "hello world" is surprisingly kologmorov complex

VoteToReopen

11:20 PM

$ mkdir git-test/
$ cd git-test/
$ git init
Initialized empty Git repository in ...
$ cd ..
$ rm -rf git-test/
$

Huh.

ATaco

Don't forget my favourite mistake sudo mv / /www

HyperNeutrino

wat

ATaco

(That's how I screwed A-ta.co once)

VoteToReopen

...How did you type that (even accidentally)?

ATaco

I was trying to move my current folder into /www with ./ /www, and must have missed the ..

VoteToReopen

11:24 PM

Ah, that's not as worrisome.

HyperNeutrino

lol

huh for whatever reason I can't open my GH wiki pages locally

Atom just doesn't do anything when I try to open it

and atom . doesn't do anything

VoteToReopen

Have you cloned the wiki as a separate repository?

HyperNeutrino

yes

time for my favorite trick

VoteToReopen

(Also, may I mention that having two repositories for a wiki is a PITA?)

ATaco

(I should probably set up some Source Control for a-ta.co)

VoteToReopen

11:28 PM

(That's usually a good thing.)

HyperNeutrino

lol I just tried to do sudo rm -rf ./ (which didn't work anyway) and almost missed the dot

VoteToReopen

Do you like to live on the edge?

eaglgenes101

Gnu coreutils rm already stops you if you try to delete root

HyperNeutrino

@VoteToReopen yes. also, I just crashed my computer

ATaco

--no-preserve-root

eaglgenes101

11:33 PM

I wonder when they are going to put in other safeguards to make sure you don't delete every directory in root or anything like that

VoteToReopen

@HyperNeutrino How?

HyperNeutrino

atom stopped working so I ps-ef | grep atom its PID and then killed the wrong process :P

VoteToReopen

There really should be a license to operate a Bash shell.

HyperNeutrino

haha yes

VoteToReopen

Or any kind of shell to be honest.

HyperNeutrino

11:38 PM

yes

eaglgenes101

Bash is full of subtleties, which is what you get when you try to have a construct for everything under the sun

Adám

CMC: Given N, return 4 × N terms of 1 - 1/3 + 1/5 - 1/7 + ...

E.g. N=3: 3.466666667

Pavel

@HyperNeutrino 0/10 should have used pkill atom

HyperNeutrino

@Adám oh

Adám

@HyperNeutrino "oh" on the N=3 case‽

Pavel

11:44 PM

pkill means pgrep kill, which bothers me since it's not the p part of pgrep that makes pkill different from kill.

HyperNeutrino

lol

ATaco

I made a website that's powered by Funky

VoteToReopen

...Are you guys trying to make up language names that can be made puns on easily?

ATaco

Funky's name is an accident. I couldn't think of a name at the time so I used it as a placeholder..

Pavel

@ATaco I predicted this happening months ago

Aug 18 at 6:58, by Pavel

@ATaco Nope, you've named the repo 'funky' and you're stuck with it now.

@ATaco did you roll your own stdlib for the server stuff or did you hijack an existing one

Also, are there docs?

Transcript for