@ATaco do you see any way to add a feasible comment system?

@TrojanByAccident yes

@Downgoat example?

@TrojanByAccident just override the test: section

e.g.

test:
    - make test
    - rake test
    - npm run test

@Pavel Yeah, actually, I could.

I've been thinking, and it would kind of screw up structure, wouldn't it?

Nope, Because it's in groups of three lines.

Note that:

	11
	2
	3
	1
	2
	3

Is the same as

	111
	2 2
	3 3

@Downgoat I have no idea what the different sections are o_o

add that to your YML

How would comments look? /comment/ in one line, and space in the rest?

11
2
3
1
2
3

Which is two chunks of three:

11
2
3

1
2
3

Is then stacked onto eachother in compiling, making:

111
2 2
3 3

@Downgoat I'm not even sure what I put in my .travis.yml at all :/

Okay, first thing first, what is the reason you are using CI

From there we can determine what steps to take to achieve goal

Ok, that makes sense

because I'm not sure what language we should be using as the main language

I've been trying to wrap my head around it for a bit

What if it's not a multiple of three lines?

@Downgoat Because I want to

New lines are appended to the end.

@TrojanByAccident ok well then I'm not sure what you want CI to do

111
22
3
a
b

Becomes

111a
22 b
3

128[ r    ]
    -
1    lco_1

can be golfed to

128[

1
 r
-
 lco_1
]

(The first is 28 bytes, the second, 21)

Wait, how do you move an element from one tape to the other?

l and r

@Downgoat Well, to run tests, I guess, then

okay, so are your tests written in mocha or like your own format?

In what order does subtraction happen? l-r or r-l?

@Downgoat I'm not sure what you mean

l-r

Ok, I think i figured this out. tio.run/nexus/threead#@29ooBCtoFAUy6UABLr5QMrw/38A I fell more proud of myself than I deserve to be.

tio.run/nexus/threead#@29oEK1QFMuloKCgm8@lYPj/PwA works as well.

I'm not really sure how your code for printing every character works though. I'm guessing c converts it to a character, but wehat is _ doing?

It 0's the memory.

Why do you need to 0 it if you immediatly overwrite it with 1?

The 1 is a function that does *10+1, to the current buffer.

So I need to zero it first if I need to set it to 1.

Ok, makes sense.

I found more unicode: ᕕ( ᐛ )ᕗ

tio.run/nexus/threead#@29oZBGtUKSgEMuloKCgy2UIJHOS8@MN//8HAA this works just fine with less space

(I just golfed the same thing)

128[r   ]
   -
1   lco_1

128         | Sets buffers 1 and 3
            | to 1 and 128
1           |

   [        | While this number isn't 0
   -        | Subtract the right from the left.
            |

    r       | This sets all memory cells to the i - 1
            | This is for our loop, and our print.
    l       |

            | Convert to a character, then output it.
            |
     co     |

        ]   | Loop back if larger than 0
            |
       _1   | And reset the third buffer to 1.

Is there any builtin string manipulation? Specifically looking at this challenge

I'll find a nice way to explain these scripts...

Noncompeting, of course.

Not easily, no...

Ok... does + concatenate strings?

+ does.

And is there a way to retrieve a char from a string?

And ^ will get the string at location r of l

Ok

Hmm...

How to get input?

Not implemented yet.

DO tests by just pre-initing a cell.

Ok, in that case standard IO formats permit hardcoding input

Actually, that's too hard. I'll try something less stringy.

I'm gonna try it.

You have the advantage of knowing everything you builtin and how it works.

I'm gonna snipe hello world

At some point, somebody is going to have to write a quine in this language.

I'm rather annoyed that quine commands are forbidden :/

It's for the best.

Yeah, if they were allowed it would make making quines kinda trivial

Got it.

            >r <
    2>^[o<>^< +>  ]
32c>1    l  _1 <o>

Assumes the input is in the first buffer.

This does the A E S T H E T I C thing.

Cool. You can enter that as non competing.

I'll get to that.

You could replace 32c with " ", right?

Some byte count either way.

Yeah.

(Although this errors on tio, I fixed the bug that causes it on mine)

You can move the ] left one to save a byte

You are correct.

On which register does input start?

1.

Ok.

I'm debating whether or not to put my name on the license for my project e.e

So, on TIO, it prints newlines indefinitly.

@TrojanByAccident make it GNU GPL

It's a bug.

Which I've fixed on my end.

No wait those are spaces

I'll ask Dennis

@Pavel It's MIT

I license all my software under MIT

Use the GitHub built in licensing thing

I do

Then you're fine

but I mean the Copyright (c) 2017 XXX part

Oh, yeah, put your name

Or it's meaningless

I see

@ATaco works now

on a side note, repeated copy-paste of similar spec parts is tedious

Create a script to generate documentation for you?

It won't take any less time, but it'll be more fun that way.

@Pavel ...

What language

@Pavel No, because then it won't be explained the way I want it to

No way

use an actual documentation generator

Threead

@Pavel So you're using lua?

also whoops

Actually, ATaco is.

wait

@TrojanByAccident what is this for? PigeonScript?

@ASCII-only yes

@TrojanByAccident Okay, example?

Threead may be the least readable language I've ever written

How many have you written?

I only know RProgN

@ASCII-only Example for what?

@TrojanByAccident similar spec parts

Well I've written other ones, just nothing I've used publicly

@ATaco Mine is rather annoying to read for longer programs, but it follows a structure

Okay @ATaco @TrojanByAccident how readable would you say they are compared to other esolangs?

@ASCII-only are you seeing this?

TFW^ happens

[Math]?

@ASCII-only Well, the individual commands are sorta easy to make out, but stuff like control flow can get confusing

CSS, like JS, never just randomly works.

@ASCII-only replaced in the docs by 'Add', 'Subtract', etc

@Pavel What do the arrows do?

@Pavel ?

ATaco knows

I just copied his stuff

@Pavel nice

@ATaco Any docs yet? What do the arrows do?

See: Brainfuck

Moves the individual tape.

@ATaco okay help i need an explanation for this

like what happens when there are 6 lines in the code

Okay, Explaining time...

Compiles to:

     |
     |
32c>1|
-------
     |
2>^[o|
     |
-------
   |
<>^|
l  |
-------
>r <  |
< +> ]|
_1 <o>|

Tape together these chunks...

     |     |   |>r <  |
     |2>^[o|<>^|< +> ]|
32c>1|     |l  |_1 <o>|

And remove formatting...

             >r <
     2>^[o<>^< +> ]
32c>1     l  _1 <o>

"THREEAD"            >r <
             2>^[o<>^< +> ]
         32c>1    l  _1 <o>

Putting "THREEAD" there wont work, it needs space to fully compile the string.

But I gotta dash, I'll be back in about an hour and a half.

Yeah, oops. Spacing was wrong

That looks better

32c writes a space to the bottom register

_ clears the cell

o outputs

r reads from the right tape and writes to current cell

I can still talk, I'm on my phone now, but I can't exactly code

^ gets reads from r and l and gets the r'th char of string l

[] is same as BF, as is <>

That seems to be it

I may be able to write a more detailed documentation

When I get to a computer

Hello

i'm new to codegolf, can someone tell me how one strikes off the previous byte count in the answer, i searched but did not find exactly how

@AbelTom As in doing the strikethrough? <s>old score</s> new score?

@ASCII-only Yes sir, strikethrough is the word :) Thanks!

@AbelTom To see how someone formatted something in an answer, you can click "edit" on their post and look at the markdown code of their post. Just click cancel afterwards to not change anything

@Zgarb Is it possible in CL to have something where what it does changes depending on the number of arguments after it?

Everything in CL takes exactly one argument.

It's the same as in lambda calculus.

So no, you can't have that because there may not be any more arguments ever.

@Fata

@Fatalize Thanks for the tip! :)

@Qwerp-Derp CL?

@TrojanByAccident Combinatory logic

oh

ḩęļļǫ

Atm I can't get docs generated, so I'm just going to manually write out the specs, and worry about docs later

@TuxCopter greetings

@TrojanByAccident docs are for the code, you probably won't need that

@ASCII-only not sure what you mean

language spec != docs

Will the Clay Mathematics Institute give the $1,000,000 dollars, to someone who can create an algorithm that solves an NP-complete problem in O(n) time? cs.stackexchange.com/a/1880 seems to say that I need an mathematical proof...

yeah, you need to be able to solve every np-complete problem in O(n) time or something iirc

@TrojanByAccident unless you want people to contribute + understand the language source you won't need generated docs

Polynomial (O(n^k)) time, not O(n) IIRC

@ASCII-only oh but I do

Alright, thanks!

I'm thinking of ideas targeting Sudoku. I have an idea of finding a solution to a n*n sudoku board, with a partially filled grid. The math is just too complex to keep in head yet, though.

I love the fact that PLLD was closed as dupe of cs, but cs has neither a tag for esoteric languages nor one for golfing related stuff

@TrojanByAccident ... code golf has nothing to do with plld

@ASCII-only the golfing languages part

code golf != golflangs

plus you can just tag it under language design or whatever

there's no need for narrower tags

there, better now?

@ASCII-only Solving one NP-complete problem in polynomial time means a polynomial-time solution for every NP-complete problem. Problems are typically shown to be NP-complete by proving that there's a polynomial-time transformation from that problem to another NP-complete problem (or vice versa). That is, given an algorithm (any complexity) for one, a second, polynomial-time algorithm can be wrapped around it to solve the other.

To put it more succinctly and to give an example, let's say that we have an algorithm F(X) that solves the Traveling Salesman problem. There exists a second algorithm G(Y) that is polynomial-time and G(F(X)) finds a minimal vertex cover on the same graph (or solves any other NP-complete problem).

@El'endiaStarman Thanks for the response! I recorded an idea about solving TSP in linear time in the last summer (and P=NP overall). A while ago, I got an idea to solve the Sudoku problem in linear time, and listened the recording, which turned out to have the exact approach as the approach for the Sudoku. I've delayed looking at these problems, since my solution model seems to require math I can't just yet figure out. I'll use different methods to grow my brains, so that I can solve it.

@TuukkaX No disrespect, but: the chance that you find a linear time algorithm to solve TSP is almost certainly 0.

@TuukkaX Even though I agree with Fatalize, I'm still curious as to what your idea is.

@Fatalize I'm fully aware, and I agree. @El'endiaStarman When I'm able to get my idea in a presentable form, I can indeed show it.

Well, what's the gist?

@

Oops

@El'endiaStarman I can discuss this in a private chat, is that fine? Maybe you, with more knowledge of these things can show some flaws in the approach.

The gist is the following: Do the calculations in such a way, that they follow the rules that the problem contains (very vague). However, I'm completely unsure whether there are numbers that actually fit mathematically in the calculations that are to be determined.

@TuukkaX That is indeed very vague. That said, I can see there's a possibility that you're thinking of using specially-chosen integers and then applying arithmetic to get the answer. If that's the case, then you haven't actually made progress because you still have the same information and constraints, just packaged in a different form.

For example, checking whether an integer n is prime is O(n) in the magnitude of the number. However, if you care more about the number of digits (which you should), then it's much harder to find a polynomial-time algorithm for primality-checking (but still possible; see the AKS primalty test).

@El'endiaStarman Good point. From a quick look, the AKS primality test seems a little too complicated for a short read. Let's define S, a number that is different for any sudoku board size nn. S contains a solution for the nn board, which is in a format that yields the correct values when (for example) modulated with the board indexes. Let A be the number calculated from the partially filled numbers. Read next message

@El'endiaStarman Now, upon reading the numbers, we modify A in such a way, that when calculating the correct values after fully calculating S & A, we gain the correct values with S & A & the indexes of the non-filled slots in the grid.

@El'endiaStarman That's the kind of fashion I'm talking about. "you still have the same information and constraints, just packaged in a different form" -> in my way of approach, do you suggest that I look for the patterns outside of the numbers?

I'm finding it hard to follow without an example. How would this work for a 2x2 SuDoKu where there's a 1 in the upper left cell?

@El'endiaStarman I should be leaving school, and I'm still stalling. I'll give the example when I'm at home

From what I understand you want to encode the board and the partially filled board into integers, and then find the first integer from the second

If that's what you are saying then this is just moving the problem to the operation that finds that int from the other int

hello can i get js help here?

plnkr.co/edit/?p=streamer&s=EWCrVACGe0MwT2me My cube is spinning in a weird way

@LearnHowToBeTransparent yes, but I would recomment the JavaScript room

@TrojanByAccident they are busy

ehh

Are there any guidelines for editting old submittions?

@LearnHowToBeTransparent That's a kinda cool spinning cube

@ais523 Would you be interested in discussing about what points to update in Brachylog's grammar? Also would you like Brachylog to have a custom code page or would you prefer if it stayed ASCII? (e.g. b;E'L∧▯:Iz{+¹₁|ײ₁|÷²₁}Lt instead of b:[E]cL,?:Iz{:+a|:*a|:/a}Lt)

Hello

Yey I put more stuff in my Processing gist

@Qwerp-Derp Ḩęļļǫ

@TuxCopter Don't zalgo me pls

but it's not zalgo

just regullar cedillas

@TuxCopter yep its cool but that cool rotating cube will harm my project..

@Qwerp-Derp Minor, unobtrusive Zalgo is okay

@TrojanByAccident SO peoples - mainly chat - are salty though, take caution when approaching any one of SO's chat rooms

I made a Button and ProgressBar class