The Nineteenth Byte

AdmBorkBork

18:00

@Xanderhall That's kinda neat, actually. Not many languages will have ++ work on non-numbers.

totallyhuman

1 min ago, by totallyhuman

` https://imgur.com/Br00TCn

Business Cat

But -- doesn't work with strings

Cows quack

such consistency

Business Cat

57 secs ago, by Business Cat

Because why not ¯\_(ツ)_/¯

Xanderhall

PHP's consistency is amazing Kappa

Business Cat

18:01

And also echo ("0"?1:0); is 0

Xanderhall

that one's a weird case

AdmBorkBork

Well, given that everything is a string, that kinda makes sense.

Business Cat

But echo ("test"?1:0); is 1

Leaky Nun

@Dennis who deleted it?

AdmBorkBork

Right, because that's non-zero

Xanderhall

18:03

A non-empty string should evaluate to true. But I think PHP treats it as a number, in which case 0 evaluates to false

Business Cat

"Let's just cast everything to a number first for no reason! Sounds like a great idea!"

Xanderhall

Actually, it's "let's cast everything that looks like a number to a number!"

If you manually cast "test" to a number, ie "test" + 0, it will give you 0

AdmBorkBork

[int]$a = 3; [int]$b = 2; $a / $b ... what do you think is the result in PowerShell?

Xanderhall

1?

AdmBorkBork

Nope, a double 1.5

Xanderhall

18:05

Nice, it converts to double

totallyhuman

another challenge copied from some place else

Xanderhall

That can either be a nice feature or an inconvenience

AdmBorkBork

if you then [int]($a / $b) you won't get 1, you'll get 2

Husnain Raza

I am so confused curerntly

currently*

Xanderhall

because (3/2) = 1.5, which rounded is 2

AdmBorkBork

18:07

right

How about this one? [int]$a = [int32]::MaxValue; $a + 1

Xanderhall

No idea

Wouldn't that be language dependant?

AdmBorkBork

Talking PowerShell

Xanderhall

Never used Powershell

AdmBorkBork

You get a double of 2147483648

Xanderhall

so I wouldn't be able to tell you

Husnain Raza

18:09

codegolf.stackexchange.com/questions/133754/… if the terms in the next sequence in the question are not even known, how are you supposed to write a program to calculate it?

Xanderhall

Anyways, I shall leave you with the greatest function ever built into a language, belonging of course to PHP: strtotime

AdmBorkBork

Straight Oti Me?

totallyhuman

string to time

AdmBorkBork

@HusnainRaza I think the intention is to write a program that would calculate the next number in the sequence, i.e., check if it's prime and keep looping until it is.

Xanderhall

Check the examples for some ideas of what you can feed it for input

Leaky Nun

18:12

3

A: Float 754 to Hamming

Python 3, 72 71 bytes 1 byte thanks to Lynn. lambda n:n and(bin(1020+len(bin(abs(n))))+bin(abs(n))).count('1')-(n>0) Try it online! Explanation The binary64 format consists of three components: the first bit is the sign bit, which is 1 if the number is negative the next 11 bits store th...

AdmBorkBork

@totallyhuman Oh, duh. Now it's obvious when looking at it.

Leaky Nun

I remain to be the only answer that doesn't use built-in.

Husnain Raza

oeis.org/A000066 the sequence only has terms from n=3 to 12

only upper and lower bounds are known for higher values of n, could i code those?

AdmBorkBork

@LeakyNun That's because I don't want to do manual float-to-binary conversion in PowerShell.

@HusnainRaza I don't understand your question.

totallyhuman

hard code, you mean?

if so, yeah i think so

New Main Posts

18:17

-2

Q: Given a palindrome write a program to print the sorted list of all palindromes that can be constructed from the alphabets of the given palindrome

THE OUTPUT PALINDROMES SHOULD BE IN SORTED ORDER Output Format: Output containing a sorted list of all palindromes constructed from the given input palindrome string. Example 1 Input NITIN Output INTNI NITIN Explanation There are only two palindromes that can be constructed from NITIN Examp...

Lembik

hi all

Husnain Raza

Yeah, I made a program that hardcoded only the known values of the sequence, and yet it still got deleted

totallyhuman

hmm

there was another answer that hardcoded and it was fine

should probably just ask caird

Husnain Raza

is there a pm system on stackexchange?

totallyhuman

just ask in the comments

or ping them here

Step Hen

18:20

but caird's suspended

totallyhuman

oh?

KSmarts

@Husnain I did the same thing and it got deleted, too.

totallyhuman

horrible timing

Husnain Raza

step hen, can you look at my deleted answer?

oh i think i saw your answer this morning @KSmarts

it was a 2d program iirc

totallyhuman

yeah fission

Step Hen

18:22

I think you need to implement a program that would find more if they existed

or something like that

and that calculates the first ones too

KSmarts

I was told "Just because the next values are unknown doesn't mean that they do not exist."

Step Hen

instead of hard coding

totallyhuman

hang on lemme find the other hardcoded ones

Husnain Raza

my bash answer

KSmarts

But if no-one knows what the next values are, what am I supposed to output?

Husnain Raza

18:23

^

Step Hen

@KSmarts you're not (I think) but you should be doing something that would theoretically find the next one

totallyhuman

-4

A: One OEIS after another

40. Rust, 3416 bytes, A000001 The rules seem to imply hardcoding the first 1001 cases is fine. static COUNT: [i32; 1001] = [0, 1, 1, 1, 2, 1, 2, 1, 5, 2, 2, 1, 5, 1, 2, 1, 14, 1, 5, 1, 5, 2, 2, 1, 15, 2, 2, 5, 4, 1, 4, 1, 51, 1, 2, 1, 14, 1, 2, 2, 14, 1, 6, 1, 4, 2, 2, 1, 52, 2, 5, 1, 5, 1, 1...

Husnain Raza

2

A: One OEIS after another

83. Bash, 53+165265=165318, A000014 #!/bin/sh read index head -"$index" a.txt | tail -1 requires that a.txt (https://hastebin.com/eroterefov) be in the same directory next sequence Also, this is my first post to the Code Golf Stack Exchange, please let me know if I messed up in any way.

totallyhuman

@totallyhuman downvoted bu valid

Step Hen

Rules say it has to work up to n=1000

KSmarts

18:24

How is this allowed?

Step Hen

that's the thing

KSmarts

2

A: One OEIS after another

77. Braingolf, 51 bytes, A004147 m32&gM9784&gM7571840&gM11140566368&gMvvcv1+[R<v]v_; Try it online! Next sequence

totallyhuman

2

A: One OEIS after another

77. Braingolf, 51 bytes, A004147 m32&gM9784&gM7571840&gM11140566368&gMvvcv1+[R<v]v_; Try it online! Next sequence

this one too

Step Hen

I dunno

Husnain Raza

my first answer which hardcoded them got accepted

totallyhuman

18:25

oh up to 1000

KSmarts

Sure, the numbers are uncomputable but that doesn't mean they don't exist!

totallyhuman

but the braingolf answer doesn't..?

Step Hen

I think because you're first version @HusnainRaza didn't work even for the first ones, everything got screwed up

and I don't know anymore

AdmBorkBork

@KSmarts Graham's number is pretty uncomputable, but it's known to exist...

totallyhuman

really have to wait for caird at this point

can't do much

Husnain Raza

18:27

rip

totallyhuman

oh and they're suspended till august 26th

so guess the challenge is over

KSmarts

Or even if Dennis would explain his reasoning and what he expects to see

Business Cat

...or wrote what he expected as an answer himself

totallyhuman

dennis, even as a mod, cannot confirm what the challenge requires you to do

Business Cat

I kinda get the reasoning though. Just because it's not computable past a certain input, doesn't mean it's not computable before that, so you have to actually compute it

Husnain Raza

18:29

I'm sorry

totallyhuman

and writing an answer is practically impossible

KSmarts

I'm tempted to write a program that gives the lower bound for higher-index answers (since there is a known formula for that) and ask anyone who challenges it to prove me wrong.

totallyhuman

so challenge is dead tbh

KSmarts

But I liked that one. I want to play more :(

Business Cat

@KSmarts I would say go for it

totallyhuman

18:30

mm

yeah might as well

you can't prove it wrong

Business Cat

It's not necessarily right, but nobody can validly say it's wrong without a proof

Husnain Raza

i dont know any more programming languages tho :C

totallyhuman

you can do the one you deleted

or whoever deleted

Business Cat

I do want it to keep going

totallyhuman

yeah me too but oeis is weird

KSmarts

18:32

I've had two deleted. One on its own, and one because it followed someone else's deleted answer.

Husnain Raza

that was probably mine :(

totallyhuman

there are erroneous sequences, uncomputable sequences

Business Cat

I have 1 deleted because ETH posted literally 4 seconds before me

Step Hen

I told caird he should make it not include sequences that didn't go to 1000 :(

totallyhuman

and since the challenge does not specify what to do on those

can't do much

KSmarts

18:33

@Husnain was yours the Shakespeare answer?

Husnain Raza

yes :(

totallyhuman

i think though

we could edit the challenge to specify required behaviour on those circumstances

Dennis

@LeakyNun The author. Who else?

Leaky Nun

@Dennis alright

AdmBorkBork

@BusinessCat FWIW, Meta consensus is that the poster must be able to prove their answer correct when challenged.

Business Cat

18:40

Well then how can you compute something with no proven right answer?

Leaky Nun

@BusinessCat it just wouldn't finish before the heat death of the universe

but it would theoretically output the right answer

Business Cat

Hmmm I guess

Maybe I have to learn Perl just to have a useful language to answer this in

totallyhuman

any rules about editing the question?

AdmBorkBork

12

Q: Are challenges that may not be solvable on-topic for PPCG?

Consider the following example challenge: Given a set of integers, output a truthy value if there is a non-empty subset whose sum equals 0, or a falsey value if no such subset exists. Solutions must have worst-case polynomial time complexity or better. This is the subset sum problem, which ...

discussion policy scope

Poke

@AdmBorkBork clogging with pure ecstasy

AdmBorkBork

18:48

There's another answer, I think by Peter, about proving if the program terminates, but I can't seem to find it.

wizzwizz4

@Poke I'll assume you mean the emotional state.

Wheat Wizard

Any name ideas for a language embedded in a hyperbolic geometry?

wizzwizz4

@WheatWizard Smileylang?

Because a hyperbola... never mind.

That's not a good name.

Step Hen

@WheatWizard Exaggeration or some derivative

Husnain Raza

If the sequence is unsolvable, can I add a lower bound and just output that?

Lembik

18:51

when people write C code with SSE/AVX support, how do they specify that the AVX code should only be used if the cpu supports it?

is there some variable you can check in the code?

AdmBorkBork

7

Q: On the subject of testing code

Somewhat related In my Bernoulli Numbers challenge, I've been seeing a lot of invalid answers that fail to compute certain values to the required accuracy/precision, or that take substantially longer to run than the amount of time between the posting of the challenge and the answer (such as this...

discussion policy answers testing

Poke

@wizzwizz4 mmyeah good assumption ;)

i suppose i would have wanted a capital E otherwise

wizzwizz4

@WheatWizard π∫|f(x)|^2

Because that's the volume of...

I'm not good at names.

John Dvorak

@WheatWizard hypercode

Husnain Raza

Just hyper

Step Hen

18:55

hyper diaper

4

Cows quack

I haven't learnt what hyperbolic geometry is yet

John Dvorak

There's already hyperRogue, a roguelike in hyperbolic geometry

@Cowsquack check hyperRogue

Wheat Wizard

I think I might go with Hyperb. Sounds like superb + hyperbola

wizzwizz4

@WheatWizard codeh

A naming convention based on sinh and cosh.

Husnain Raza

So just to clarify, code needs to be written that actually calculates the sequence

like with an actual algorithm or something

Step Hen

18:59

@HusnainRaza believe so

Wheat Wizard

That doesn't seem enforcable

OEIS sequences are not well specced

Cows quack

@WheatWizard is it a 2d lang?

Wheat Wizard

IMO it should just be everything in a b file

@Cowsquack Yes.

John Dvorak

CMC: find an OEIS sequence that cannot be computed by an algorithm

Step Hen

@JohnDvorak by any algorithm?

Lembik

19:01

@JohnDvorak that's not well specified. I mean an algorithm could just read the answers from the page

Step Hen

like what about busy beaver

it's supposed to be uncomputable but I can compute the ones that are there

John Dvorak

Busy beaver is good

Wheat Wizard

@JohnDvorak oeis.org/A060843

Ah I see Stephen and I had the same idea

:39092247 We linked different sequences but the same general idea

Step Hen

@WheatWizard oh wait they are different? 0.o

Husnain Raza

Travelling salesman

?

Wheat Wizard

19:03

Travelling salesman is NP but computable

Poke

Taken down by a font

> Finally, Pakistan is now Sans Sharif.

2

Step Hen

and then there's this: A000017

Lembik

@WheatWizard NP-hard you mean

Wheat Wizard

Oh yeah

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Can you compete with a supercomputer? The challenge is to write super fast code for computing the permanent of a matrix of complex numbers. In a paper from 2016 a team of coders managed to compute the permanent of a 40 by 40 complex matrix on 8192 nodes of what was at the time the world's f...

Lynn

19:08

The downvotes on my A000001 answer are actually pretty ridiculous. I added a little comment saying why that’s so, but there’s another aspect to it, I guess — the fact that I blocked off such a “useful” language like Rust

wizzwizz4

@Lynn ... Sometimes people make me sad.

Lynn

I intentionally picked the most popular, useful language I could think off to block off

Wheat Wizard

@StepHen That is an excellent reason why we should allow hard coding

Lynn

Other people not being able to use it means I’ve slowed down the flow of answers, which is the only strategic angle I can grab in this challenge

Mr. Xcoder

@Lynn Hence the downvotes maybe?

Lynn

19:11

(although I share it with everyone who does know weirder languages)

I mean. It’s a competition with no way to influence “how hard you’re winning” (because the penultimate answer wins for some bizarre reason). Making things hard for others is the closest thing.

Wheat Wizard

Anyone know why this is happening to me?

Lynn

Do you downvote hard cops because they make the robbers’ life tougher?

It makes absolutely no sense.

Mr. Xcoder

No, I don't. I encourage the hard ones with upvotes

Wheat Wizard

I do often downvote artless crypto in cops and robbers

But I will say I see no problem with your answer

Lynn

What did people expect to see, for A000001?

Mr. Xcoder

19:14

But those that use clever techniques (this does not include hardcoding)

Wheat Wizard

In fact I've up voted it

Lynn

An optimized, 10k line algorithm full of heuristics and group theory? Nobody was going to write that. If I didn’t post my Rust answer, someone was gonna post the equivalent of fun (n) -> FiniteGroupCount(n) in some math suite that has a built-in for it like Magma

And I know y’all would have upvoted the hell out of that for some strange reason

even though, internally, it obviously has the first 1001 group counts memoized

Mr. Xcoder

Sorry for that ^

@Lynn I never upvote trivial posts, I don't understand that way of voting either.

CMC: Given a 2D list of positive integers, add the length of the flattened list to all the integers.

[[1,2,3], [1,2,3]] -> [[7,8,9], [7,8,9]]

totallyhuman

+F might work in jelly

i'm not sure about the vectorization depth

Mr. Xcoder

IDK

Business Cat

19:22

FL+ should work but there might be a better way

totallyhuman

whoops

Lynn

FL+ does the trick

totallyhuman

len

huh

Business Cat

+ is commutative why did I say +@ ಠ_ಠ

Poke

the community can be a fickle beast

Mr. Xcoder

19:23

@Poke What do you mean?

Poke

at least we can all agree that java is the best language

totallyhuman

yes

Mr. Xcoder

Sure, sure ಠ_ಠ

Poke

@Mr.Xcoder i was referring to the large push for not upvoting trivial solutions but then doing it anyway or how we have hmmhmm criteria for closing popcons, etc

New Main Posts

1

Q: Continuous Niven Numbers

In recreational mathematics, a Niven number in a given number base, is an integer that is divisible by the sum of its digits when written in that base. For example, in base 10, 18 is a Niven number since 18 is divisible by 1+8 = 9. Also, 12001 in base 3 is also a Niven number since the sum of the...

code-golf

Mr. Xcoder

19:24

Closing bad pop-cons!

Well made Pop-cons are gems of this community

Poke

s/closing popcons/determining which popcons are worth closing/

anyway

totallyhuman

those challenges were clearly copied from somewhere

Poke

as long as we agree with the java thing

Mr. Xcoder

@totallyhuman I feel so too

HyperNeutrino

Wait are there more bad challenges?

Poke

19:25

i'm happy enough to ignore the other arguments and discussions

Wheat Wizard

@Mr.Xcoder too bad they don't exist. :P

Business Cat

@Poke I don't know about that. PowerShell is great and does all things.

Mr. Xcoder

@WheatWizard Yours is one of the best around.

Poke

@BusinessCat check yourself, sir

totallyhuman

reddit.com/r/learnprogramming/comments/6q3rdm/…

Wheat Wizard

19:26

@Mr.Xcoder If thats true then it only proves my point.

Mr. Xcoder

@WheatWizard Unknown Completeness is very interesting :)

totallyhuman

this was quite clearly some kind of task that the OP was supposed to do

isn't there policy on this?

Mr. Xcoder

I have seen those before, I am searching for them now

@totallyhuman I have a meta post on that

totallyhuman

also the reddit post was posted 6 hours ago

Mr. Xcoder

@totallyhuman codegolf.meta.stackexchange.com/questions/13311/…

totallyhuman

19:29

which probably means that the reddit OP and the OP here are the same person

Mr. Xcoder

Too many challenges at once!

totallyhuman

can somebody who knows how to deal with this deal with this? :P

Mr. Xcoder

@totallyhuman Unless they find the original reference, no...

totallyhuman

>_< but it is so obviously copied...

Wheat Wizard

How do I get to the sandbox?

AdmBorkBork

19:33

what part is copied? From reddit to here?

HyperNeutrino

Then use a plagiarism checker to try to find its source.

totallyhuman

@AdmBorkBork reddit.com/r/learnprogramming/comments/6q3rdm/…

HyperNeutrino

@WheatWizard here?

totallyhuman

it has the exact same body

HyperNeutrino

Then flag it.

AdmBorkBork

19:34

Right, who is to say it's not the same author?

HyperNeutrino

Although it is copied, I'm wondering that too. The moderators will have to decide. It seems probable that it's the same author.

Mr. Xcoder

@AdmBorkBork Why place it on two different sites?

Wheat Wizard

@HyperNeutrino No the chat sandbox

Even if it is the same OP its really suspicious that they would post in both places

HyperNeutrino

Sandbox

Where you can play with chat features (except flagging) and ch...

2

AdmBorkBork

@Mr.Xcoder Why not? If it was posted asking for help, the person got the help, then thought it would be a challenge? What's the harm with that?

Wheat Wizard

19:36

Seems to me like they are perhaps fishing for a homework solution

@HyperNeutrino Thanks a bunch I am awful at navigating chat rooms

HyperNeutrino

No problem :)

AdmBorkBork

Whatever code-golf solution gets put out is not going to help with homework, unless the homework is to golf something.

totallyhuman

it's not golf

it's fastest-code

Mr. Xcoder

I flagged it as in need of moderator attention.

AdmBorkBork

@totallyhuman Refresh your page.

totallyhuman

19:38

>_<

but still

Mr. Xcoder

Now it's very clear it seeks hw help. They restrict the output format.

I'll personally just boycott both the challenges.

HyperNeutrino

Restricted output format is typically either inexperience on this site or just homework help.

Mr. Xcoder

> Can we view source code in code-golf?

This convinced me.

New Main Posts

1

Q: Solve the Trolley Problem

Philosophers have long pondered the Trolley problem. Unfortunately, this no human has solved this problem yet. Luckily, as programmers we can use computers to solve the problem for us! Your program will take as input a (finite) directed graph (with at most one edge from x to y, for any x and y...

code-golf graph-theory optimization

totallyhuman

what

what does that mean?

Mr. Xcoder

19:41

Can we view source code in code-golf? — Shobit Mahajan 1 min ago

AdmBorkBork

@Mr.Xcoder bona fides

Mr. Xcoder

English?

totallyhuman

yeah I'm asking what he means by that

AdmBorkBork

That's Latin for "Good Faith." As in, assume that the person is making an honest attempt and something else (likely English-as-second-language) is getting in the way.

Mr. Xcoder

I highly doubt that anyway.

AdmBorkBork

19:43

The amount of editing done, and the fact that the user posted a second (unrelated) question seems to me like they've discovered a new community.

give-me-teh-codez people don't do that

totallyhuman

to get their homework done by?

HyperNeutrino

The only way to determine is to interact with the user and try to either get them to let information slip by or for us to figure out that he's not asking for homework help.

Wheat Wizard

Could I get another user to help me test something in the sandbox?

Step Hen

@WheatWizard I'm up

totallyhuman

@AdmBorkBork why not? they have two assignments so they posted both

wizzwizz4

19:44

@WheatWizard I'll help.

Mr. Xcoder

@AdmBorkBork A new community of unpaid programmers seeking to solve things as fast and as golfier as possible?

Step Hen

@totallyhuman Second one does not seem homeworky to me at least :P

totallyhuman

a possible, most probable, situation

Cows quack

@WheatWizard chatjax?

Wheat Wizard

@Cowsquack No, I'm trying to reproduce a bug.

AdmBorkBork

19:45

@Mr.Xcoder For homework? What homework asks for golfed code?

Step Hen

@AdmBorkBork well he didn't originally wanted golfed code

Mr. Xcoder

@AdmBorkBork None. But fast answers are on trend here

totallyhuman

originally there was no winning criterion

AdmBorkBork

@StepHen Right. I assume that to be an ESL thing.

totallyhuman

and then he accepted fastes-code

Mr. Xcoder

19:46

Now it's 100% up to the mods to decide. I've flagged the first one.

totallyhuman

and then code-golf

wizzwizz4

Anyone know an inverse of Python 3's int.from_bytes

Step Hen

@wizzwizz4 int( ,2) what you want or no? or do I have that backwards

Mr. Xcoder

@AdmBorkBork It's not a new community: Member for 1 year, 11 months

wizzwizz4

@StepHen int(,256) is closer.

But that isn't a thing.

totallyhuman

19:48

oh that's complicated

python doesn't have good base conversion

Step Hen

is your input in binary, hex, or just bytes?

AdmBorkBork

@Mr.Xcoder That doesn't mean it's not a new community. I was a "member" for several months before I actually joined, and another several weeks before I actually contributed.

wizzwizz4

@StepHen My input is a big int.

Mr. Xcoder

@AdmBorkBork Me too. I just made an observation

Step Hen

@wizzwizz4 uh... int.to_bytes?

wizzwizz4

19:50

@StepHen If only... :-/

Step Hen

@wizzwizz4 docs.python.org/3/library/stdtypes.html#int.to_bytes

wizzwizz4

Wait...

That's a thing?

Step Hen

I dunno, docs say so

wizzwizz4

I must've missed that.

No, it's definitely in dir(1).

I missed it when looking for an inverse. :-S

totallyhuman

hmm

hey what does dir(...) output?

wizzwizz4

19:54

@totallyhuman dir(x) is all ys where x.y is a thing.

totallyhuman

ye I know lol

what does it output on ...

on Ellipsis

AdmBorkBork

Oh, lol, the way your question is worded

totallyhuman

['__class__', '__delattr__', '__dir__', '__doc__', '__eq__', '__format__', '__ge__', '__getattribute__', '__gt__', '__hash__', '__init__', '__init_subclass__', '__le__', '__lt__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__']

:/ lame

__doc__ returns None

print(....__doc__)

doesn't look like valid code but y'know

Ellipsis seems like the most arbitrary object they could've implemented

wizzwizz4

It... doesn't disassemble.

import dis

dis.dis(type(...))

... No output.

AdmBorkBork

Can someone smarter than me at JavaScript put a leaderboard of uncracked answers on @WheatWizard 's cops thread?

Wheat Wizard

19:59

robbers thread?

totallyhuman

you could use the one from the halting CnR

Sandbox

Transcript for