The Nineteenth Byte

ASCII-only

1:00 PM

@LeakyNun i'll leave mine running overnight and see if it gets anything

Leaky Nun

@ASCII-only what is the point?

you're just going to get what I got

ASCII-only

1:10 PM

@LeakyNun I said I'll leave mine running

Leaky Nun

@ASCII-only and I asked why

ASCII-only

I'm testing until the sums get too close to ULL_MAX (no idea why)

Can you prove/have you proven one doesn't exist in that range?

@LeakyNun ???

Leaky Nun

@ASCII-only I already ran my program through 2^32.

ASCII-only

Yeah I know

Leaky Nun

so it's kind of a proof

totallyhuman

1:15 PM

Oh no

ASCII-only

Didn't stop people from searching :P

Like Collatz

Okx

https://esolangs.org/wiki/Timesig
> work in progress language
> hasn't been updated since 2009

Step Hen

I really like my newest Sandboxed challenge - check it out once it hits chat

Business Cat

That could be an hour from now though

ASCII-only

Or 1 minute

Anonymous

1:20 PM

@BusinessCat It will probably be within the next 10 minutes

TheLethalCoder

It'll come in, in the range 0<t<=∞

Step Hen

t - x seconds

totallyhuman

a ∈ (0, ∞)

Fancy math notation ftw

feersum

Next time I see someone talking about an MVC app, I'm going to assume it's about cars hitting moose

Step Hen

@feersum Sounds just like deer here 0.o

TheLethalCoder

1:26 PM

I had a deer jump over my bonnet the other day

As I was doing 60mph

Anonymous

@TheLethalCoder This sentence is really strange from an American perspective

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Interpret a Maximal Number of Brainf**k Variants code-challenge interpreter brainfuck A while ago in chat, I had an idea: It was prompty shot down as impossible. Because of that, I'm making it a challenge! Valid Languages The only valid brainfuck variants you can use are the languages appe...

Step Hen

that one

totallyhuman

Yay

TheLethalCoder

@Mego Is it called a hood in the states?

Anonymous

1:27 PM

@TheLethalCoder Yep

TheLethalCoder

Ah what's bonnet?

Step Hen

@TheLethalCoder a hat

Anonymous

Usually an antiquated women's hat

Anonymous

bonnet

TheLethalCoder

Like a milkmaids hat? I think we still call them bonnets just not used very often

Anonymous

1:29 PM

They haven't been in style for a few centuries

TheLethalCoder

I think you should bring it back

Anonymous

@TheLethalCoder Now I'm imagining an old-fashioned hat on a milking machine

Anonymous

@TheLethalCoder What's dead should stay dead

3

TheLethalCoder

Haha

Googled "bonnet on milking machine" was disappointed

You get one or the other apprently

never both

Anonymous

Alright folks, our mission is clear

LiefdeWen

1:32 PM

What would happen if there was a language that trivialized a certain type of code-golfing problem?

Step Hen

@LiefdeWen There are

totallyhuman

You mean every challenge with most golfing languages? :P

Step Hen

Oasis, cQuents, Ceres for sequence

Anonymous

@LiefdeWen You mean like how every golfing language has one-byte builtins for Hello World and Ninety-Nine Bottles of Beer?

LiefdeWen

@mego yes, but like for the latest hello world challenge none of them could be used

totallyhuman

1:33 PM

intrnt for internet based challenges hopefully :P

Step Hen

Bubblegum, SOGL for kolmogorov-complexity

Anonymous

Or Bubblegum and Cinnamon Gum for kolmogorov-complexity

Step Hen

@LiefdeWen yeah, but that's why it was made :P see Help, Wardoq!

Leaky Nun

@Mego it's pretty much just your language (for the latter)

Anonymous

@LeakyNun IIRC Vitsy and either Japt or Jolf also have NNBB builtins

Anonymous

1:35 PM

To be fair, it is also useful for busy beaver challenges - output as many bytes as possible in as few bytes as possible

Anonymous

Since in 4 bytes (N;l*) you can output 138109504 bytes

programmer5000

Is this ready to post?

1

A: Sandbox for Proposed Challenges

code-golf tail: troll: file truncated Background When you run tail -f file in bash, the file is outputted and then any subsequent appends. However, when you remove something that has already been displayed, tail outputs: tail: nameoffile: file truncated Your Challenge When given an input ...

totallyhuman

o0

Anonymous

Actually 11752000000000 is possible with N9╤*

LiefdeWen

Okay so what response from the community could I expect if I made a language all characters except like 9( which I use to make it turing complete) and the rest all output Hello, World, would people start dissallowing that language for hello world restricted source problems?

Anonymous

1:38 PM

@programmer5000 Considering it's been in the Sandbox for 2 hours - no

7

Step Hen

@LiefdeWen as long as it can add and do primality testing you're golden

@LiefdeWen people will just change the capitalization or punctuation though :P

As I said, see Help, Wardoq!, because that's basically what you're trying to do, except not as broad

Anonymous

@LiefdeWen The language itself wouldn't be explicitly disallowed, but the use of builtins like that would likely get disallowed in any language

Anonymous

Also HW restricted complexity challenges are basically exhausted at this point

Anonymous

@Mego That's 11.752 terabytes from 4 bytes of source code

Anonymous

Well, 10.688 tebibytes, which are what count

Step Hen

1:43 PM

@Mego I can output 2e32 bytes with a 9 byte source in cQuents (theoretically)

TheLethalCoder

@Mego It's a simple challenge though, can you find any improvements for it? There's no point a challenge sitting in the sandbox just because of a time restriction if it is perfectly fine.

totallyhuman

@StepHen can't you output infinitely with much less?

Step Hen

@totallyhuman yes, of course

feersum

But that's a zombie beaver.

Step Hen

@totallyhuman I think busy beaver is supposed to terminate eventually though, right?

Anonymous

1:45 PM

@TheLethalCoder Just because I can't find any problems with it doesn't mean that they don't exist. The whole point of letting a challenge sit in the Sandbox for a while is so that many users get the opportunity to review it. The vast majority of the PPCG userbase isn't online during any given 2-hour period.

totallyhuman

Ah ok

Anonymous

Also IMO busy beaver should have consistent output across runs, regardless of available memory, so "output until we run out of memory" isn't valid either

New Main Posts

5

Q: Fake divisor sum polyglots

The task In this challenge, your task is to write a program in a programming language L that takes a positive integer n, and outputs the sum of the proper divisors of n (sequence A001065 on OEIS). It should return the correct output for any 1 ≤ n ≤ 10 000. Here are the first 10 outputs: 0, 1, 1...

TheLethalCoder

@Mego I understand the point of the sandbox, I'm one of the main reviewers in it at the moment. But this challenge is simple and clear. As far as I can tell there won't be any problems with it so no point leaving it in there.

Step Hen

@Mego how's that supposed to work?

Anonymous

1:47 PM

@StepHen By "regardless of available memory", I mean "assume that there is enough memory to produce the output, but no other assumptions"

Poke

@TheLethalCoder If he wants to post the challenge, no one is going to stop him. He asked in chat and Mego gave a response

TheLethalCoder

@Poke Yeah and I gave mine...

Poke

I don't see the sense in arguing

Anonymous

@TheLethalCoder What's the harm in leaving it in the sandbox for a while longer?

Step Hen

@Mego ah gotcha.

Martin Ender

1:48 PM

@Mego are you already playing Pyre?

TheLethalCoder

I'm not I'm giving the opposing, and my, opinion...

Poke

You responded directly to mego

TheLethalCoder

@Mego No harm but as far as I can tell there's also no harm in posting it now

Anonymous

@MartinEnder Not yet - got a few bills coming up that I need to pay first before I can consider buying any new games

Martin Ender

ah, fair enough

TheLethalCoder

1:49 PM

@Poke An opposing point isn't necessarily an argument.

Martin Ender

I'm currently squeezing it in between 100%ing Hollow Knight and waiting for the free content update this week :D

Step Hen

Let's not argue about whether we're having an argument or not. I'm going to link to Monty Python soon.

3

Anonymous

@TheLethalCoder There's the potential for gain by leaving it in the sandbox longer (not having it closed quickly on main, and thus losing out on the FGITW popularity rush). There's no potential for gain by posting it now.

Anonymous

@MartinEnder I've been busy playing Dishonored 2 and CiV

Poke

ar·gu·ment
ˈärɡyəmənt/Submit
noun
1.
an exchange of diverging or opposite views, typically a heated or angry one.

TheLethalCoder

1:50 PM

I give up...

Poke

that's exactly what it is

Step Hen

(no onebox) Monty Python - Argument Clinic

Martin Ender

@Mego ah nice. how is Dishonored 2? I played about 80% of the first part.

Anonymous

@MartinEnder Dishonored 2 feels like a very refined version of Dishonored 1. They kept all the good parts, and fixed the problems. Doing nonlethal/low chaos is much more enjoyable because you have more options than just avoid/sleep dart/choke. The fact that you can knock enemies out during combat is a very welcome improvement.

TheLethalCoder

@Poke typically a heated or angry one. is the key part

Lembik

1:52 PM

hi... not really ppcg but is there a tool to search all the pdfs on a windows machine?

TheLethalCoder

@Mego I got bored of 2 after a bit but I think that's because I had a few games going at the same time

Lembik

I have lost a document :)

TheLethalCoder

agent ransack?

Martin Ender

@Mego oh that sounds good. I prefer games where going the stealth route doesn't feel like missing out on stuff compared to the combat route.

wizzwizz4

@Lembik Yes.

Search.

Step Hen

1:53 PM

@Lembik type the name in the bottom left search thing?

Anonymous

@Lembik You could use Windows Explorer to search for *.pdf on the whole computer, but it will be slow.

wizzwizz4

I prefer the one inside Windows Explorer.

Ninja'd.

Anonymous

@MartinEnder Arguably in Dishonored and Dishonored 2, the stealth route is more rewarding

Anonymous

@Poke There's no need to be petulant

TheLethalCoder

@MartinEnder Have you played Styx?

wizzwizz4

1:54 PM

Everybody calm down.

Anonymous

→ 4 messages moved to Trash

Lembik

@StepHen @Mego.. sorry I want to search all the documents for a particular word

Step Hen

@Lembik oh... you don't remember what you named it?

wizzwizz4

@Lembik Harder.

Lembik

exactly

TheLethalCoder

1:55 PM

@Lembik Use agent ransack

Anonymous

@Lembik Unless the documents actually contain the word and not just an image of the word, you're likely out of luck

Step Hen

@Lembik How many PDFs are on your computer?

And as @Mego said if it's not in document form, but in image form, you're going to have to do it yourself, unless you can afford a natural language Google-type bot

Anonymous

If the PDF in question actually contains text, you can grep the files

Martin Ender

@TheLethalCoder no, it's been in my steam library for ages though

Lembik

@TheLethalCoder agent ransack? Sounds interesting

Anonymous

1:56 PM

@StepHen Well there are open source OCR utilities, but there's no guarantee that they will work

TheLethalCoder

@MartinEnder That's pretty fun for stealth, though I never completed it

Lembik

@StepHen I don't know exactly.. a few hundred maybe

ASCII-only

@Lembik WSL and grep :P

Step Hen

@Lembik ~300? That's not that many, you could probably do it manually :P

TheLethalCoder

@Mego I work for an OCR company, we'll find it for you ;)

Lembik

1:57 PM

@StepHen right but where are they? :)

@ASCII-only pdfgrep you mean :)

wizzwizz4

@TheLethalCoder Could you provide OCR of all of the images posted to TNB?

ASCII-only

@Lembik wait scanned or text

Lembik

@ASCII-only text

wizzwizz4

It's hard for bots to read images. :-)

Martin Ender

@TheLethalCoder I'll probably get around to it eventually. so many games to play though...

Step Hen

1:58 PM

4 mins ago, by Mego

@Lembik You could use Windows Explorer to search for *.pdf on the whole computer, but it will be slow.

Lembik

@wizzwizz4 not in the modern world :)

ASCII-only

@Lembik ah yeah wouldn't that work?

Lembik

@StepHen right but then I have to pipe that list into something

TheLethalCoder

No we're quite fast now :P

Step Hen

@wizzwizz4 That's what captcha's for, helping bots read text

ASCII-only

1:59 PM

@wizzwizz4 *super easy

wizzwizz4

@Lembik It is if they're grainy or fuzzy or use a font that is not installed on the computer running the bot for license reasons.

@StepHen Only certain bots.

TheLethalCoder

@MartinEnder I know the feeling, currently 661 games in my steam library. And I've played a handful of them.

Step Hen

@TheLethalCoder I just play Rocket League, and that only when I have the time, so I don't have to worry about that :P

ASCII-only

@StepHen yeah, in text captcha good bots get it right >99.9% of the time

TheLethalCoder

@StepHen I'm more CS:GO when I can, Rocket League to relax

5

A: How do I rotate individual letters of an image into the right orientation for optimal OCR?

You can use the code in the following code project article to segment each individual character. However, when trying to deskew these characters individually any result you get is not going to be very good because there isn't very much information to go off of. I tried using AForge.NETs HoughLin...

Solving captchas

Anonymous

2:04 PM

That could make for an interesting challenge - a test battery of images with characters rotated/flipped, and the goal is to correctly OCR as many as possible

Anonymous

The only issue is, it wouldn't be impossible to get 100%

TheLethalCoder

What OCR engines would you allow though?

Or would you have to write one of your own?

Step Hen

@TheLethalCoder probably custom

Anonymous

@TheLethalCoder Probably only custom, for fairness

TheLethalCoder

Would probably be shorter to do a custom one as well.

Anonymous

2:06 PM

It's not really interesting to see a bunch of answers all using Tesseract

TheLethalCoder

Tailored to the test images obviously

Anonymous

I've got a great idea: collect handwriting samples from PPCG users, including flipped and rotated characters, and have solutions try to OCR them

2

Step Hen

Would probably be best with a high quantity of images that are all of a certain decently predictable type

TheLethalCoder

ICR*

Step Hen

@Mego Bonus points: cursive

Anonymous

2:07 PM

@StepHen That might be a bit too difficult

TheLethalCoder

So would normal handwriting

Anonymous

Handwriting would be easier because of the space between characters

Step Hen

(although with some younger PPCGers might not even know how to write cursive)

TheLethalCoder

Still be difficult, at least with fonts you know they will "always" be the same

totallyhuman

Why is cursive dying ;-;

Anonymous

2:08 PM

@TheLethalCoder The entire point is the difficulty

TheLethalCoder

I do like the idea of the challenge, though, as with any difficult one, you wouldn't get many answers.

Anonymous

With set fonts, getting 100% isn't terribly hard

Mayube

@TheLethalCoder ICR?

TheLethalCoder

Yeah but you'd do a score like number correct/bytes to make people look for the optimal solution

Step Hen

@TheLethalCoder In my recent Sandboxed challenge I did something like 200*numCorrect - bytes

that way the bytes matter, but not so much

wizzwizz4

2:10 PM

@StepHen What's cursive?

Never mind. According to Wikipedia, that's how I write.

totallyhuman

Lol

Step Hen

@wizzwizz4 it's when your letters mesh together - I learned how to write in two ways, cursive and print, but I think American schools stopped teaching cursive recently

AdmBorkBork

What about this?

Jul 18 at 14:27, by AdmBorkBork

@totallyhuman I like fonts. I'm halfway done with creating my own based on my own personal handwriting.

wizzwizz4

@StepHen I just don't know the terminology.

totallyhuman

Cursive's awesome though

wizzwizz4

2:12 PM

When I first read instructions to write "print", I wrote in Times New Roman.

totallyhuman

You don't have to keep lifting your have every single stupid time

Step Hen

@wizzwizz4 I got confused by that when I was little, a form would say "print your name" and I'd say "mom why are you writing that it says to print it"

AdmBorkBork

Anyone else in here explicitly change your handwriting so that you have a unique style? Or just me?

TheLethalCoder

@Mayube Intelligent character recognition or handwriting recognition

@StepHen Yeah I meant something along those lines

cursive is essentially joined up writing

Step Hen

@AdmBorkBork But... everyone already has a unique style. That's the point of signatures and handwriting identification.

Anonymous

2:15 PM

PPCG Handwriting OCR Challenge

TheLethalCoder

@AdmBorkBork My handwriting is a unique mess that no one else can read. Does that count?

Step Hen

If you're bored, you can look at my Sandbox post and tell me if it's good or not:

0

A: Sandbox for Proposed Challenges

Interpret a Maximal Number of Brainf**k Variants code-challenge interpreter brainfuck A while ago in chat, I had an idea: It was prompty shot down as impossible. Because of that, I'm making it a challenge! Valid Languages The only valid brainfuck variants you can use are the languages appe...

Poke

LinkedIn: It’s illegal to scrape our website without permission

man do i hate the cfaa

totallyhuman

@StepHen what if I'm not bored?

Poke

what a vague law

Step Hen

2:23 PM

@totallyhuman Then you can look at it anyway :P

Poke

@TheLethalCoder my signature is the same thing. i got bored of writing it

HyperNeutrino

@StepHen Is this valid? ::codegolf.stackexchange.com/questions/136901/…

totallyhuman

18

Q: Hello, World! with semi-common characters

As some of you may know, to get a language onto Dennis's wonderful Try It Online!, a Hello, World! program is required. Recently, Dennis pushed up a way to load these programs from the website. Here is a JSON pastebin of the characters used in those programs, by frequency, as of July 27, 2017. S...

code-golf restricted-source hello-world

Can some js wizard put a snippet that checks a solution for validity

Step Hen

@HyperNeutrino Oh, what that means is that you have to use the character's appearance, not its codepoint

IE if Jelly has A in a different place than ASCII, use Jelly's A for Jelly programs and ASCII's A for ASCII programs

You're fine

HyperNeutrino

Oh okay. Thanks.

Step Hen

2:29 PM

@HyperNeutrino SmallF uses UTF 8 right?

HyperNeutrino

Whatever Python's ord uses xD

Mayube

Python3?

totallyhuman

Depends on Python version

... sorta ninja'd?

Mayube

P3 uses UTF-8, P2 complains if you don't specify

HyperNeutrino

Python 3

totallyhuman

2:30 PM

Then you're fine

Step Hen

@Mayube wait you can't just call ord in Python 2?

totallyhuman

You can

Mayube

afaik you can if it's an ASCII character

HyperNeutrino

If the file contains weird characters, Python 2 will scream at you

totallyhuman

It'll be ASCII

HyperNeutrino

2:31 PM

Unless you specify what encoding it is

That's why Jelly won't work in Python 2 :P

Step Hen

More reasons not to use Python 2 for anything except golfing

totallyhuman

That's way past proven

:P

It reaches EOL in... 3 years?

Step Hen

@totallyhuman We'll still have TIO >:)

totallyhuman

4 mins ago, by totallyhuman

Can some js wizard put a snippet that checks a solution for validity

Pls js wizard

TheLethalCoder

Could someone have a look over my sandboxed challenge

1

A: Sandbox for Proposed Challenges

The Double Slit Experiment code-golf ascii-art physics random animation A lazy physicist has the job to perform the double slit experiment. However, they are lazy and can't be bothered to set up all the equipment themselves and so are going to simulate the effects. They can't program though so ...

Nathan Merrill

2:44 PM

why is killing threads/processes so difficult in most languages?

TheLethalCoder

To stop idiots using them and crippling a system?

Nathan Merrill

no, there's something fundamental about it I think

like, in Java, Thread.stop() is deprecated, but Thread.interrupt() waits for the thread to accept the interrupt

and I've had problems with python's subprocess.kill()

IMO, the person who spawns a thread should be in charge of when the thread dies

but I feel like its lifetime management is given to the thread itself

TheLethalCoder

Well in C# you'd do var t = new Thread();t.Start();t.Join(); so that which calls it kills it.

HyperNeutrino

Are a person's parents in control of when their life ends or are they in control of that? (xD) /s

Nathan Merrill

lol

TheLethalCoder

2:47 PM

Though if you don't kill it yourself it only dies when its finished its work

Nathan Merrill

@TheLethalCoder can the thread itself do anything to stop the join?

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Secret Message Part 2, Random Interleaving This is the second part of a series, the first part of which can be found here. Now that our message is slightly encrypted, we want to obfuscate it more thoroughly. We're going to do this by chopping the back half off of the message and interleaving...

TheLethalCoder

@NathanMerrill Not that I know of

Though it probably could in some funky way

Nathan Merrill

oh, Thread.Join waits for the thread to finish

msdn.microsoft.com/en-us/library/95hbf2ta(v=vs.110).aspx

so, joining doesn't actually kill the thread

TheLethalCoder

Yeah, that wasn't the one I meant then. I've been doing more async code recently

There is a way to kill it though

Anonymous

2:49 PM

Nope, usually threading APIs make it hard to kill threads because you usually don't want to do that

Anonymous

You're better off using an event system and sending an event to the thread to make it kill itself

Nathan Merrill

why?

Anonymous

It's much easier for a thread to handle cleaning up and exiting than it is for its parent

Nathan Merrill

what cleaning up are we talking about? closing files?

because the language can handle that

Anonymous

Any resource that needs disposal, so yes files, but also dynamic memory

TheLethalCoder

2:51 PM

Oh there's Thread.Abort() which essentially throws an exception inside the thread.

Nathan Merrill

memory is also managed by the language runtime.

Anonymous

It's almost always safer and cleaner to have the thread kill itself than to kill the thread directly

TheLethalCoder

But when we usually use threads in our code we let the threads handle their own status and have a flag called stopThread or similar that it monitors

Anonymous

@NathanMerrill Yes, but the runtime still has to clean up the memory inside of the thread. With that consideration, it's simpler and cleaner to have the exiting done by the thread instead of by the parent.

AdmBorkBork

@totallyhuman Here's a checker in PowerShell -- Try it online!

Nathan Merrill

2:53 PM

from the language's runtime point of view, I agree. The code to cleanup after killed threads sounds messy and I wouldn't want to write it

however, would the programmer give anything up if killing threads was common?

totallyhuman

@AdmBorkBork I can't run power shell in Java script :P

Step Hen

@Okx I can't see them either

Okx

uh... i don't know if it's just me or have stack snippets been removed?
from PPCG, they work on stack overflow

Business Cat

They work for me

totallyhuman

^

AdmBorkBork

2:56 PM

@Okx this one still seems working

Step Hen

@Okx oh wait they work for me, I couldn't find it 0,o

AdmBorkBork

@totallyhuman lol, but at least you could copy-paste the code to TIO

Okx

@AdmBorkBork Go to edit that, remove one of the characters (such as the ! in a HTML comment) then add them again...

HyperNeutrino

._. lambda stack: ... causes an error saying that there are too many arguments and lambda stack, arguments: ... causes an error saying that there are too few arguments

Okx

no more stack snippet

HyperNeutrino

2:57 PM

Conclusion: My lambda needs to take 1.5 arguments

Nathan Merrill

oooh...this is interesting: docs.oracle.com/javase/1.5.0/docs/guide/misc/…

basically, when you kill a thread you also need to open any locks associated with it

Step Hen

@Okx well of course not, the HTML comment is what makes the snippet a snippet

Okx

no i said remove it and add it again

Nathan Merrill

however, many locks depend on the thread finishing its task

Okx

it's the exact same but the stack snippet preview is gone

Step Hen

2:58 PM

@Okx works for me

AdmBorkBork

@Okx Oh, if that's the case, I don't want to break something

Nathan Merrill

but it didn't, that means that the state of the program isn't what the other threads expect

Okx

you don't need to actually click the finish edit button

totallyhuman

@HyperNeutrino is it part of a class?

Cause then you might have issues with self

2

HyperNeutrino

oh wait the issue was that I had a triply-nested-function, one of which required the arguments as an argument, so the too-few error was coming from that part and the too-many error was coming from the part I was changing

PPCG Handwriting OCR Challenge

Transcript for