The Nineteenth Byte

Downgoat

06:06

Question: are the following assumptions safe even with weird IEEE floating point behavior with edge cases like nan, infinity, etc?

    x >= y -> x > y || x == y
    x <= y -> x < y || x == y
    x != y -> !(x == y)

feersum

I think so...

Downgoat

@ASCII-only why is tokenizer do this :|

vsl::asts> let issub = 1
Syntax Error: expected identifier after assignment (1:4)

    > 1 | let issub = 1
        |     ^^

ASCII-only

:/

idk

oh

because that is "is subclass" operator?

Downgoat

@ASCII-only no

there is only is operator

another example:

vsl::asts> let forgoat = 1
Syntax Error: expected identifier after assignment (1:4)

    > 1 | let forgoat = 1
        |     ^^^

quartata

@Downgoat I think NaN is an exception to that last one?

ASCII-only

06:12

@Downgoat wat.

@quartata Nope

I tried it, NaN !== NaN is true

Downgoat

@ASCII-only pls tell how should fix this problem

quartata

@ASCII-only I wouldn't exactly trust JS on that

ASCII-only

hmm idk

Downgoat

github.com/vsl-lang/VSL/blob/develop/src/vsl/parser/…

ASCII-only

@quartata I'd trust JS to at least follow IEEE though

quartata

06:14

gotta go to the standard

@ASCII-only I wouldn't

spotty track record tbh

feersum

The thing is NaN isn't equal to itself, but still any pair of numbers is either equal or unequal, not both or neither.

Downgoat

@quartata then trust rust

user202729

@quartata I doubt that JS developers reimplement their own number system.

Downgoat

^ they aren't smart enough to do this

@ASCII-only :/ idk how to efficiently fix

DJMcMayhem

Programming, expectation vs reality

ASCII-only

06:26

@Downgoat what type is it parsing as

Downgoat

@ASCII-only meaning?

vsl::lex> forgoat
[ { value: 'for' }, [ 'goat', 6] ]

6 = identifier

ASCII-only

oh.

that should not happen

you probably bork

Downgoat

no it should not

@ASCII-only ಠ_ಠ

i unborked

ASCII-only

because all word literals are replaced with /\bword\b/

Downgoat

idk if i borked while unboroking

@ASCII-only where does this happen?

I don't see that: github.com/vsl-lang/VSL/blob/develop/src/vsl/parser/…

ASCII-only

06:30

@Downgoat github.com/vsl-lang/VSL/blame/…

Downgoat

@ASCII-only where do i look

link the line?

ASCII-only

@Downgoat github.com/vsl-lang/VSL/blame/…

Downgoat

@ASCII-only that won't work anyway since we support unicode Identifiers

ASCII-only

@Downgoat but that is for keyword?

just test for \w+ and add \b at the end

Downgoat

found workaround:

function kw(name) {
    return new RegExp(`${name}(?=$|[^\\p{L}\\d])`, 'u');
}

ASCII-only

06:36

that works too

Downgoat

feels really slow though

ASCII-only

slow?

Downgoat

@ASCII-only slow meaning will take long time to execute

ASCII-only

@Downgoat well isn't identifier regex equally slow

Downgoat

how?

it's not matching against 3KB unicode regex

ASCII-only

06:44

@Downgoat because it is 3kb unicode regex?

3

at least last time i checked

Downgoat

oh i see what you m ean

Οurous

Why is 3kb of unicode regex a thing?

Downgoat

@Οurous because JS doesn't support \p{L}

user202729

@Downgoat ... Why do you have a 3KB regex...?

Downgoat

babel team is so incompetent they seriously think a 200ms speed slowdown per regex is 'not an important change'

@user202729 ...

Qwerp-Derp

07:04

Can anyone check my minesweeper implementation for any bugs? I want to solve this challenge, and I'm implementing my own version of the program given.

user202729

... I can't implement Tarjan's strongly connected component algorithm correctly ...

@Qwerp-Derp (code-review) [0 for _ in range(x)] is the same as [0]*x because 0 is immutable.

ASCII-only

@Qwerp-Derp neighbors looks nasty

user202729

and doesn't exclude self.

Qwerp-Derp

@ASCII-only It does, I used to have this inside two nested for loops (and I didn't want 4 nested for's), will change into two for loops.

Pavel

Hmm, man man reports that man seraches through sections in the order 1 1p 8 2 3 3p 4 5 6 7 9 0p n l p o 1x 2x 3x 4x 5x 6x 7x 8x but then goes on to explain that the sections are 1 2 3 4 5 6 7 8 9.

???

ASCII-only

07:19

@Qwerp-Derp no as in you don't need two ranges for 8 cells...

user202729

@ASCII-only Writing out the 8 cells would be worse.

ASCII-only

@user202729 why?

just store the offsets: [(-1, -1), (-1, 0), (-1, 1), (0, -1), (0, 1), (1, -1), (1, 0), (1, 1)]

longer but doesn't seem as bad imo

Pavel

> Exact rendering may vary depending on the output device. For instance, man will usually not be able to render italics when running in a terminal, and will typically use underlined or coloured text instead.

Qwerp-Derp

@ASCII-only Fixed it to that instead

Pavel

When will man not be outputing to a terminal?

If you pipe it into a process or file, it won't be able to do italics there either.

ASCII-only

07:23

@Pavel when it outputs to a virtual console?

idk

but definitely when outputting to HTML...

Pavel

Oh actually that's a good point

If you look up the online man pages

Qwerp-Derp

@ASCII-only I'm looking at the VSL docs (or at least trying to look at them anyway). Why don't you host this on GitBook or github.io or something like that?

ASCII-only

@Qwerp-Derp wait. VSL docs are in markdown?

Qwerp-Derp

@ASCII-only IDK, did you guys write the docs in pure HTML+CSS+JS?

ASCII-only

@Qwerp-Derp JSDoc

IIRC

DJMcMayhem

07:26

I'm working on a new esolang of mine

Qwerp-Derp

@ASCII-only In any case, it would still be better to host the docs on github.io

I can't access the docs on c9

DJMcMayhem

I just spent that last hour trying to write a divisibility test, before realizing that this language is so terrible I can't even get that to work

ASCII-only

@Qwerp-Derp huhwhat?

what's the problem

Qwerp-Derp

File '/docs/index.html' could not be found

DJMcMayhem

You thought brain-flak was hard to write, just wait until I release this abomination into the world

ASCII-only

07:28

@DJMcMayhem well it can't be worse than malbolge

Qwerp-Derp

@DJMcMayhem Is it minimalistic or just really crazy

DJMcMayhem

Minimalistic

The goal is to provide slightly more features than brain-flak (like integer literals, strings, mathematical symbols, etc.) but then make the memory and control-flow as minimalist as possible

Qwerp-Derp

sigh Studying for a half-yearly exam right now... the subject is so boring that 10 minutes of studying sucks the soul out of me

DJMcMayhem

And un fortunately the control flow and memory are actually the same thing

Qwerp-Derp

@DJMcMayhem So basically BrainFlak + more features + Brian&Chuck (mix of memory and control flow)?

DJMcMayhem

07:31

I've never used brian&chuck

ASCII-only

@Qwerp-Derp what subject is this O_o

Qwerp-Derp

@ASCII-only Information Processes and Technology

ASCII-only

why are you even doing it then

@Qwerp-Derp also what do you mean by boring

Qwerp-Derp

@ASCII-only Doing it because I've already invested a year into this and because it's a HSC subject

@ASCII-only It's boring because the content is outdated, the subject itself is pretty much counterproductive in the outside world in most cases, and the subject is pretty much just rote memorisation

You can't apply any of the knowledge, it's one of those "mindlessly absorb knowledge and spit it out on a test paper" sort of subjects, except the knowledge is business-y stuff which I don't find engaging at all

ASCII-only

rip

Qwerp-Derp

07:37

As well as that, when I came in to the subject I thought it was going to be a cool programming subject (seeing the "information" and "technology" bit) but instead I got bogged down with stupid business-y stuff

So the disappointment of it all is another reason for me hating the subject I guess

user202729

@Qwerp-Derp (another code review) Why do you need (tile[0], tile[1]) while you can write tile?

Qwerp-Derp

@user202729 Did not catch that, fixed!

Mr. Xcoder

08:01

@DJMcMayhem O_o More tarpitish than Brain-Flak?

Downgoat

08:17

@ASCII-only goat news: short circut works :DD

ASCII-only

@Downgoat :D

Downgoat

VSL should be TC and like nicely TC now not like turing tarpit TC

only problem is you can't print integers/floats Q_Q

not sure how to solve problem, feel free to add to STL though: github.com/vsl-lang/libvsl/blob/master/src/print.vsl#L10

ASCII-only

@Downgoat oh yeah qwerp derp can't access docs

user202729

@feersum I think that there is a way to crack it without factoring it. After all, there are 4 users who upvoted it, and proving that it's a composite is so easy...

Downgoat

@ASCII-only what docs?

ASCII-only

08:32

@Downgoat vsl

@user202729 how to prove it's a composite though...

Downgoat

@ASCII-only yeah had to put down server for a little bit while hardware upgrade

user202729

@ASCII-only Try it online!

(why is the cops thread get much more upvotes than the robbers thread? Because it's votes)

Downgoat

@ASCII-only should be back up

@ASCII-only should I add if cond: statement syntax

seems like good idea

that way this:;

func factorial(n: Int) -> Int {
    if n == 1 { return 1 }
    else { return factorial(n - 1) * n }
}

can become:

func factorial(n: Int) -> Int {
    if n == 1: return 1
    else: return factorial(n - 1) * n
}

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

I wrote the following question then found this question. Is it similar enough to mine that mine would be considered a duplicate? Recursive Prime Multiplicative Base I've been thinking about this idea for months now, and, as far as I have been able to find, nobody else has thought of it. Please ...

Downgoat

will add ternary tomorrow very late in #westcoastbestcoast rn

also more good news is that function is very well optimized:

; Function Attrs: nounwind readnone
define i32 @"factorial.$228"(i32) local_unnamed_addr #0 {
entry:
  %1 = icmp eq i32 %0, 1
  br i1 %1, label %true, label %else

true:                                             ; preds = %else, %entry
  %accumulator.tr.lcssa = phi i32 [ 1, %entry ], [ %3, %else ]
  ret i32 %accumulator.tr.lcssa

else:                                             ; preds = %entry, %else
  %.tr2 = phi i32 [ %2, %else ], [ %0, %entry ]
  %accumulator.tr1 = phi i32 [ %3, %else ], [ 1, %entry ]

user202729

08:38

(removed, too)

If your program gets wrong answer on a huge test case, what would you do?

Manish Kundu

08:58

In the recursion challenge I am allowed to use my own created integer series right?

user202729

@ManishKundu Yes.

Cows quack

@betseg I couldn't register on firefox, it works on chrome though

user202729

CMP: Do you prefer 0-indexing or 1-indexing?

NieDzejkob

@user202729 forth?

@user202729 1-indexing? What is this heresy?

user202729

And what would you do if you need to solve a problem that requires the other?

Οurous

09:12

@user202729 0-indexed, and add or subtract one.

user202729

@Οurous Yes I often do that, and it's terrible to debug.

Next question: Then how do you usually debug your program?

Mr. Xcoder

@user202729 0-indexing.

user202729

(as I often find myself forgetting to "add one before print debug information" or "subtract one when set breakpoint"...)

NieDzejkob

@user202729 depends, but in most cases, if I can get away with using printf (i. e. not esolang and not assembly), I use printf

... which is sad if you think about it

DJMcMayhem

@Mr.Xcoder >:D

Mr. Xcoder

09:19

😖

Οurous

@user202729 Clean: runtime heap / stack inspection, edge-case-triggered abort-and-dump-memory events. Other languages: Whatever the print function is. If it's C# / C++ in VS, the tracing / debug harness.

If it's a nasty one, I'll read over the IL files. If it's really bad then I have a friend who I go over the assembly with.

user202729

09:33

(sorry nothing, I just didn't notice the "must", my English is bad)

Manish Kundu

10:09

can this be further golfed? tio.run/##K6gsycjPM9ZNBtP//…

user202729

@ManishKundu 61 bytes

60 bytes

NieDzejkob

[n-1] -> [-1]?

user202729

58 bytes

Manish Kundu

oh yeah just forgot about that :O

user202729

52 bytes (but remove the parameter a)

Manish Kundu

10:12

I am dead now :p

ASCII-only

@Οurous re: C#/C++, basically the IDE is such a huge overhead that you should use it anyway

@Οurous for JS I use DevTools too

Manish Kundu

thankfully 52 isnt a problem cuz my non recursive solution is 44 bytes

user202729

@ManishKundu Cracked

Manish Kundu

thankfully i hadnt posted it :P

you are too fast lol

feersum

10:27

@user202729 Primality tests are much faster than factoring.

user202729

@feersum Yes I know. But if you can use math to find a way to check if it is prime, you can probably use math to (find a way to) check if it has a factor === 2 mod 5?

NieDzejkob

10:39

@user202729 maybe you need to consider a larger mod?

ASCII-only

@user202729 *possibly

New Bounties With No Deadlines

0

A: List of bounties with no deadline

50 - 100 rep for Flattening the Cube I posted this challenge in May 2017, and so far my answer is the only one. I'm offering a 50 rep bounty for any answer that outgolfs mine. In addition, I'll raise the amount to 100 rep if your submission is written in either Jelly or Add++, as I've already t...

ASCII-only

10:56

@cairdcoinheringaahing hmm doesn't seem too hard in Charcoal

Unless I messed something up which I probably did

user202729

Is it possible to see who is typing in SE chat?

Mr. Xcoder

Yes, there is a userscript called "I am Typing "

@user202729 github.com/TehFlaminTaco/TacosUserscripts/raw/master/…

user202729

And (how) does it work?

feersum

11:13

@user202729 Why does that matter?

NieDzejkob

@user202729 new WebSocket("wss://a-ta.co", "typing")

feersum

I haven't actually analyzed the jimmy answer, I just assumed it was standard factoring.

NieDzejkob

@user202729 (you can only see other users of the script type)

user202729

@feersum Why doesn't that matter?

feersum

If I suppose the puzzle is to find the two prime factors, then I wouldn't care what they are mod 5.

NieDzejkob

11:15

@feersum you need to prove that either there is or there's no factor of the number that is congruent to 2 mod 5 to crack the answer

user202729

> there exists a transformation rule if and only if there exists a nontrivial prime factor of X that is congruent to 2 mod 5.

NieDzejkob

it doesn't matter whether you throw math or computing power at the problem :P

feersum

But if there is a factor equal to 2, you're back to square one, right? You still have to know the factor.

user202729

> You can do this either by (...) or by proving mathematically that this does, or does not, exist.

@feersum Then you have cracked the answer, as you've proven that there exists a solution.

feersum

You're allowed to post answers where the solution does not exist?

NieDzejkob

11:17

@feersum yes...

feersum

Huh, interesting.

NieDzejkob

the challenge is to prove it either way

user202729

@feersum ... there had been 2 answers like that.

@NieDzejkob Whichever way that is easier.

NieDzejkob

yafu factors the smaller number in 2 seconds, but it didn't finish overnight with the larger one

user202729

... see? Use math is a better choice.

NieDzejkob

11:25

I'm currently running msieve + ggnfs in the background :P

user202729

Also, even if it's possible, probably not before the deadline passed (~a week left?)

reading math.SE to see if I can find anything

NieDzejkob

If Y mod 5 = Z mod 5 = 1, then Y mod 10 can be either 1 or 6 mod 10, same for Z

feersum

Ooh, better check if the number's even!

NieDzejkob

but the three combinations that have a 6 can't be the solution, because X mod 10 = 1

user202729

@NieDzejkob It's odd anyway...

NieDzejkob

11:32

@feersum if it had any factors less than a 1000000, we would know them already

feersum

@NieDzejkob links @NieDzejkob to a definition of sarcasm

user202729

Yes, some algorithms can detects small prime factors in no time.

ASCII-only

@user202729 link pls

user202729

@ASCII-only I didn't prove it rigorously but I know it's true (feel free to verify)

ASCII-only

@user202729 oh >_> i meant to the submission

user202729

11:35

@ASCII-only Deleted, on the robbers page.

NieDzejkob

@ASCII-only here is the cop

(and the deleted cop)

ASCII-only

@user202729 what's wrong with hyper's robber though

NieDzejkob

@ASCII-only hyper proves the if part, we might need the only if

feersum

Has anyone tried my cop, by the way?

user202729

~~No...~~ <- useless answer, don't read.

5

NieDzejkob

11:49

I need to build a CPU based on the Mornington Crescent execution model that uses trains for data transfer

user202729

12:02

I guess I can have some experiment with small semiprimes to see if they has any interesting properties.

Lembik

12:45

hi.. I have my usual problem.. who should I award the bounty to for codegolf.stackexchange.com/questions/157049/… ?

the first awesome answer or the one that uses their idea but in C?

@ngn what do you think? :)

Mr. Xcoder

13:06

@Lembik Definitely the Haskell one imo

New Main Posts

13:24

0

Q: Can you say the truth?

Your task is to create a function that output the truth table of a given expression. The output will be a multidimensional array. Rules Logic symbols to use: implies → if and only if ↔ negation ¬ and ∧ or ∨ First position of array must be the segmented expression Results must be either 1...

user202729

13:39

@NewMainPosts Deleted...?

flawr

CMS: I heard there kind of batteries are not very well known in the US, is that true? (They used to be quite ubiquitous in europe as flash light batteries.)

mınxomaτ

13:53

I remember those. Powered many of my DIY projects. They also lasted about 0.3s.

flawr

yes, they were great for tinkering:)

ngn

14:34

@Mr.Xcoder @Lembik I agree, the Haskell answer deserves it more than mine. Understanding the Swedish paper was the hardest part and I would have probably given up without working code to look at.

Mr. Xcoder

15:12

The Pyth interpreter seems down :(

DJMcMayhem

@Mr.Xcoder Works for me

:P

Mr. Xcoder

ಠ_ಠ

@Mr.Xcoder Fixed, someone was probably messing around with it

caird coinheringaahing

Probably isaacg :P

Mr. Xcoder

I dont't think so

Dennis

15:38

Clicking run n times produces the same result though — Nathanael C. 4 mins ago

Can anyone reproduce this?

Mr. Xcoder

ಠ_ಠ We had an extra week of winter break due to snow (yay) and now they want to hold additional classes to "recover the lost courses"... And they scheduled all of them from 7 to 8 and from 14 to 15... So I have to wake up at 5:30 AM and stay 8 hours in school tomorrow :/

mudkip201

ew

that sucks

Mr. Xcoder

@Dennis It always produces It is certain for me.

caird coinheringaahing

@Dennis It works for me (random values) (The TIO link, not the GO playground one)

Dennis

@Mr.Xcoder Just checking: the TIO link?

Mr. Xcoder

15:41

No, the Go Playground

And to make things worse... They scheduled German and Romanian classes, which we already had tomorrow... So 3 Romanian classes and 2 German :( yuck

Dennis

Hey, German is nothing to yuck at. :P

2

caird coinheringaahing

2 German lessons would be great for me :P

Mr. Xcoder

Yeah but having 2 of a kind... 4 days in a row does suck

We're going to have 8 German classes instead of 4 this week.

And even worse... On Tuesday we have History...

I think I'll just boycott that class

And lol I just noticed we have Romanian and Maths scheduled at the same time on Wednesday haha

user202729

@Dennis Can't reproduce.

caird coinheringaahing

Anyone up for a game of Spyfall?

user202729

15:49

Reading the rules take time... I won't.

Mr. Xcoder

@user202729 Come on :)... We can wait.

I don't think we have a choice anyway

In fact, we really need more people

caird coinheringaahing

Q: Can you understand what this code does? (Or should do)

user202729

@cairdcoinheringaahing But it's your language...

caird coinheringaahing

@user202729 What's your point?

Mr. Xcoder

@cairdcoinheringaahing I don't

user202729

16:09

Where can I find the source code of the TIO interpreters? (some TIO language links go to esolangs page)

Dennis

@user202729 github.com/TryItOnline/tiosetup/tree/master/languages

user202729

So... a git clone is executed every time someone run something? ...

Sounds terribly inefficient...

Dennis

No, those are setup scripts.

They're executed when the server is built or when a language needs to be updated.

user202729

What does the sss (a function in Jelly source) stand for?

Dennis

I think it was Short String Shortener.

New Main Posts

17:05

0

Q: Make a hammer animation

The Goal Is To Make A Two Frame Animation With ASCII Art of a hammer Smashing something. Lowest Byte Count Wins For Each Language!

ascii-art animation

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Drunkard's Journey Home In this challenge you are to write a program which simulates a drunkard stumbling his way home from the bar. Input: The input will be an adjacency matrix (representing a directed graph) which represents paths the drunkard can take. At each location, the drunkard will ch...

New Main Posts

17:40

0

Q: Fibbonaci Sequence

This challenge is simple. Write a program that prints out the fibbonaci sequence, each number separated by newlines. Shortest answer for each language wins. Ex. Output: 1 1 3 5 8 ...

code-golf math

caird coinheringaahing

17:55

Q: Can you retract a reopen vote? If so, how?

Manish Kundu

The stackexchange app can't open long pages, or the sandbox. That kinda sucks

New Sandboxed Posts

18:14

0

A: Sandbox for Proposed Challenges

The Ascii Labyrinth Challange: Given a map of a Labyrinth consisting of the characters '+', '-', '|' , ';', ' ', '\n' and '>' only (see description bellow), write a program that finds the starting position, prints "I", then walks through the labyrinth until he finds the exit, printing all the di...

caird coinheringaahing

19:09

Q: Should I add a 'version select' flag in a language (like Perl's -M<version>)?

Mr. Xcoder

@cairdcoinheringaahing Make it only support versions without bugs

Riker

> version without bugs

> without bugs

nice one

Mr. Xcoder

@Riker That was a nice way of saying: No :P

caird coinheringaahing

@Mr.Xcoder They're not bugs, they're features :P

Does anyone here know how to use Python's argparse module?

New Main Posts

19:25

2

Q: Decompose Polynomials

Given an integral polynomial of degree strictly greater than one, completely decompose it into a composition of integral polynomials of degree strictly greater than one. Details An integral polynomial is a polynomial with only integers as coefficients. Given two polynomials p and q the composi...

code-golf math number abstract-algebra polynomials

Riker

@cairdcoinheringaahing yes

what do you need help with?

caird coinheringaahing

Is there a way to specify the 'value' of a flag e.g. --version=10? The examples say stuff like --flag value but that doesn't really look as nice

BMO

19:44

@cairdcoinheringaahing: Looking at some old code of mine, I think .add_argument('-v','--version',metavar='VERSION',type=int,help='specify version of foo') should work.

This could be used as ./main.py -v 10 or ./main.py --version=10..

caird coinheringaahing

@BMO Yep, that worked like a charm. Thanks!

BMO

No worries, glad I could help

Christopher

20:49

Hello

Riker

21:11

@cairdcoinheringaahing sorry, left before you responded :/

(still afk)

New Sandboxed Posts

22:06

0

A: Sandbox for Proposed Challenges

Split the wagons! code-golf parsing In some variants of APL, a tacit function, or a train, consists of several functions next to each other. Your task is, given a train, to separate the different functions it consists of. The symbols you will be given and their meanings are: F: Function O: Dy...

Zacharý

22:40

"afk": at f<insert whatever here> keyboard

ASCII-only

22:54

@cairdcoinheringaahing yes Charcoal is 100% argparse (not really but CLI is)

Riker

23:10

@Zacharý "at friend's keyboard"

Zacharý

23:20

Jagony Development

Development for the Object-Oriented-Esoteric Language Jagony: ...

Riker

→ 36 messages moved to Jagony Development

Zacharý

23:39

Quoi?

Riker

meant to use the sandbox to convert imgur to stack.imgur, wrong tab

Zacharý

Ah

Jagony Development

Transcript for