The Nineteenth Byte

Downgoat

00:49

Does anyone have idea for optional syntax to shorten this:

if foo != null: f(foo)

(this is for VSL)

I was thinking like foo ?| f but that's like really esoteric

Destructible Lemon

wait nevermind i didn't read properly haha

Destructible Lemon

01:11

@Downgoat probably some null check syntax

Joshua

@Downgoat: foo?f(foo),0:0

Scrooble

Argh.

When I edit files in GitHub, my typing is reversed!

Why?

Joshua

because somebody stuck a force-RTL character for you to pick up?

"Help sometimes when I run apt-get upgrade dwarf fortress starts half way through and apt-get upgrade won't finish until the game quits?" Man that was spectacular psychic debugging.

muddyfish

I loved that question

Scrooble

Not a trollish RTL.

It works just fine in Internet Explorer

But for some reason, it's backwards in Chrome

This is infuriating. !esaelP !em pleH

Downgoat

01:23

vsl::lex> 1

<--- Last few GCs --->

[75037:0x103800000]     3498 ms: Mark-sweep 1052.9 (1076.0) -> 763.8 (785.8) MB, 160.7 / 0.0 ms  (+ 176.8 ms in 2 steps since start of marking, biggest step 92.7 ms) GC interrupt GC in old space requested
[75037:0x103800000]     6149 ms: Mark-sweep 1298.6 (1320.6) -> 543.7 (567.8) MB, 1409.9 / 0.0 ms  allocation failure GC in old space requested


<--- JS stacktrace --->

==== JS stack trace =========================================

Security context: 0x3460f06c0d11 <JS Object>

TIL tokenizing number 1 is very memory intensive task for node.js

Joshua

somebody's VM host is sick

quartata

01:48

needs spam flags (no delete votes please)

Since people keep doing it: PLEASE don't delete vote spam. Spam flagging spam and getting community to delete it will (eventually) ban that IP

Downgoat

@ASCII-only you can ignore earlier message I made new tokenizer

HyperNeutrino

02:32

@quartata I made a meta post to hopefully make more people aware of how to properly handle spam.

New Meta Posts

0

Q: How do we handle spam posts?

Occasionally, spammers show up on PPCG and post spam, including stuff like promotion of (probably fake) products, (probably fake) offers to transfer money into your account (after you give then $500 dollars), etc. Upon seeing this, it might make sense to downvote, flag, and delete. However, if e...

HyperNeutrino

03:17

Guys help with deleting a few posts? Here, here, and here; please flag them as spam/R/A or something thanks

Eric Tressler

03:37

flagged

ATaco

05:35

Oh neat, I got my silver code-golf badge.

New Main Posts

07:48

1

Q: Type the Alphabet - as fast as you can!

Your task is to make a program that measures how fast you can type the letters of the English alphabet. The program shall only accept lowercase letters a to z in alphabetical order. Each letter is echoed as typed on the same line (without new line or any other separators between letters). If ...

code-golf date alphabet

Martin Ender

@AdmBorkBork is there a way to set up PowerShell so that it sends each character I type to a process's STDIN immediately, instead of buffering it until I send a newline?

apparently this is called raw mode on Unix

Nobody

09:07

Should we create an "anti spam bot"? Or maybe hire one?

user202729

09:51

@Nobody Seems that we already have one. github.com/Charcoal-SE/SmokeDetector

jaytea

10:30

any of you guys play real golf? I played a 9-hole round yesterday and all I could think about was how to hack off strokes. tried telling my friends that by resigning before we even started i will not have a score that is less than anyone else's and therefore i will be the winner

but they weren't having it

caird coinheringaahing

12:25

Any reason for those 6 stars?

Actually, even more single starred messages. This doesn't look good :/

Adám

@MartinEnder @Mego @Dennis Can you clean up the star abuse?

Mr. Xcoder

Was just about to ask too lol

caird coinheringaahing

Just saw a deleted spam post saying "Join the Illuminati", and was disappointed to find out that it wasn't on the "Confirm the Illuminati" question

Mr. Xcoder

I agree 100% with Hyper’s post on meta +1 / +1 from me

Rohan Jhunjhunwala

13:01

@DestructibleLemon for syntax highlighting, you may customize highlighting by word, but there is a documentation window which displays a string based on what regex matches the line you are at.

TreFox

14:00

I'm trying to design my own golfing language (for shits and giggles) but I have absolutely no experience with parsers or tokenizers or any sort of language design. Most of the tutorials I find separate tokens by whitespace or some other separator, which I can't do. Does anyone know a tutorial or some other way to learn how to design these languages?

dzaima

@TreFox That varies a lot for the type of language - for stack-based you don't need anything special, but even other types AFAIK shouldn't be too difficult.

I don't know anything about parsers/tokenizers, but somehow I made this practical-ish language which has quite advanced parsing rules

muddyfish

14:24

For Pyke, I just had the parser go over every node and ask it if it could accept the string and if so return an instance of a node and the rest of the string

For the GOL language I used Python's built in tokeniser and made the rest up as I went along and ended up creating an XML grammar, in the process learning XML has it's place over json in some circumstances

dzaima

@muddyfish that sounds way too complicated for a stack-based language. SOGLs interpreter just has a program pointer and it asks to interpret the character under it, which can modify the pointer. It's a bit more complicated for strings/loops, but that's because I was stupid and made SOGL horribly :p

muddyfish

The second one definitely is. I doubt it would even be useful for a golfing language. The first however works really nicely for me

It means my nodes have full access to the script and can change things as needed and inject new code for example. There's even restoration points for continuing code execution in places

Like save the program counter and current stack and restore it in case of an exception

dzaima

@muddyfish I had the idea of doing something like that, but I realised that making exceptions not happen would be more useful :p

muddyfish

No exceptions makes it hard to debug stuff imo

And hard to work out when a feature breaks

dzaima

14:40

@muddyfish I've had that happen so little that adding any language debugging is useless for me :p

muddyfish

Heh. Stuff breaks often for me, though that's probably been because I've been changing the language core recently rather than the nodes themselves

dzaima

@muddyfish SOGLs "core" is so horrible that I only touch it in emergencies :p

muddyfish

Making one byte and two byte things work at the same time was awful, especially automatically golfing code

I tried auto explanations but was confounded by the halting problem

dzaima

@muddyfish 1) ooo auto-golfing code 2) SOGL as of recently has variable length tokens that change themselves depending on the stack at the moment

muddyfish

I'm thinking of making it so the output types of each node can be determined at compile time

So explanations work automatically

dzaima

14:47

@muddyfish that reminds me that I should work on that golfing language I've been (or,... well not been) working on with ASCII-only, for which knowing the stack item types at compile time is pretty required for better byte counts

muddyfish

Oh, also fixing time inputs and infinite generators

Something useful I've added is going from a list of node objects back to code form

*in the most compact way that complies

AdmBorkBork

15:20

@MartinEnder I'm not sure but possibly? You could try to super-ping Joey; he's actually done work on writing I/O modules for PowerShell, so he may have a better insight.

user202729

15:34

yesterday, by user202729

@NieDzejkob I guess I understood what is A000256.

@NieDzejkob (again)

> A triangulation is simple if no three edges, not all
external, form a simple closed curve enclosing a region which is subdivided
into three or more triangles.

And the sequence is "number of simple triangulations of a triangle with n nodes. "

LegionMammal978

wat

For some reason, Dictionary.com lists two pronunciations of tuple: TUH-pəl, which makes sense, but it also has TYUU-pəl for some reason

I guess I can kinda see the second pronunciation

muddyfish

I pronounce it as t-eww-p-el

LegionMammal978

Yeah, but that would just be TUU-pəl instead of TYUU-pəl

DJMcMayhem

15:59

That's the way I've always said it TUU-pəl

I've never heard TYUU-pəl

Erik the Outgolfer

huh

how is that a pronunciation of tuple

Downgoat

16:29

> i claim that everybody including me is a fucking sh**p

damn

HyperNeutrino

wat

Downgoat

it was a flag

HyperNeutrino

oh I must have missed the blue circle then

Downgoat

@ASCII-only should generic classes not be a class

because they are not really a class

more like a template

HyperNeutrino

17:08

Question: How does Jelly' Bijective Base work?

Erik the Outgolfer

jelly's?

I suppose like bijective base?

not sure what you mean here

HyperNeutrino

I don't quite understand bijective base vs. normal base

Leaky Nun

@HyperNeutrino bijective base uses 1 to n while normal base uses 0 to n-1

HyperNeutrino

and like it appears that using the base->integer function, it works for both types of base conversions

@LeakyNun ohh huh makes sense

how is that useful tho? :P

Leaky Nun

@HyperNeutrino because both follow the notion that (ak,...,a1,a0) represents the number ak b^k + ... + a1 b^1 + a0 b^0

HyperNeutrino

17:13

ok. yeah that makes sense now

Erik the Outgolfer

@LeakyNun and for 0 it returns an empty list, not [1]

Leaky Nun

@EriktheOutgolfer of course it wouldn't

HyperNeutrino

So question, why is bijective base useful? I can't currently see any applications of it but there's got to be some...

caird coinheringaahing

CMC: Sign of the input, without builtins

Erik the Outgolfer

an example would be integer compression

HyperNeutrino

17:18

ah ok

dzaima

@HyperNeutrino lengths for run-length encoding is the first thing that comes to mind

HyperNeutrino

hm so mostly just golfing uses?

Erik the Outgolfer

where you convert a list [1..n] from base n to integer then compress it (ḃ250ịØJ)

dzaima

@HyperNeutrino .. isn't that what Jelly's for?

HyperNeutrino

@cairdcoinheringaahing Proton, untested: x=>x*((x>0)*2-1)

Erik the Outgolfer

17:19

the reverse being to convert to bijective base afterwards

HyperNeutrino

@dzaima yeah but I was just wondering if it had any uses outside of golfing :P

and why is it named "bijective" base?

Pavel

@cairdcoinheringaahing Ruby, with -p: $_=$_[0]==?-

dzaima

@cairdcoinheringaahing SOGL, 3 bytes - input/abs(input), 0/0 errors and outputs input

DJMcMayhem

Is the following sentence true? For any constants a and b, where a > 1 and b > 1, then eventually with large enough x, you'll have a^x > x^b?

(^ means exponents, not XOR :P)

HyperNeutrino

wait if a^x==x^b then increasing x will make it a^x>x^b right? or am I being dumb

DJMcMayhem

17:25

Yes

HyperNeutrino

ok

Erik the Outgolfer

yes

but you don't know if a^x == x^b or even a == b

HyperNeutrino

a^x = x^b => log_x a^x = b

not sure how that helps

or if

DJMcMayhem

@HyperNeutrino How do you solve a^x == x^b?

HyperNeutrino

@DJMcMayhem log

DJMcMayhem

17:26

Oh Ninja'd

HyperNeutrino

log_x a^x = b => x log_x a = b

Erik the Outgolfer

@HyperNeutrino he meant how do you determine if that's true

HyperNeutrino

?

Erik the Outgolfer

a != b can be true

DJMcMayhem

@EriktheOutgolfer No, I meant what x makes it true

HyperNeutrino

17:26

a^x == x^b does not imply a==b

Erik the Outgolfer

@DJMcMayhem oh that's 0

I think

assuming a != 0 and b != 0

DJMcMayhem

Well, yes. But there should be two x's

HyperNeutrino

so just b = x log_x a

DJMcMayhem

For example, if a and b are both 2, then x == {0, 2}

HyperNeutrino

because then x ^ b = x ^ (x log_x a) = (x ^ log_x a) ^ x = a ^ x

and then we need to rearrange it for x

which is starting to look harder than I originally thought :P

DJMcMayhem

17:29

Intuitively, I'm sure that it must be true, but I don't know if it's definitely true or how to prove it

HyperNeutrino

b = x log_x a -> x = b / log_x a

DJMcMayhem

@HyperNeutrino Still has an x in the right hand side

HyperNeutrino

yeah

Erik the Outgolfer

@DJMcMayhem and no, regardless of how many thousands of plots you make that confirm your statement, who knows if there's the one plot that confirms the opposite? ;-p

(struggles of people trying to prove things ._.)

DJMcMayhem

@EriktheOutgolfer That's why I'm asking. To verify if my intuition is correct :P

HyperNeutrino

17:31

assume that for all x, a ^ x <= x ^ b

(when you can't do it normally, do it by contrapositive :P)

oh wait actually

a^x == x^b => log_a x^b = x => b log_a x = x

still has x on both sides grr

Erik the Outgolfer

@HyperNeutrino why

HyperNeutrino

idk ¯\_(ツ)_/¯

Erik the Outgolfer

you can't use unproven statements in a proof ;)

HyperNeutrino

lol

no but you assume it's false and then draw a contradiction

so your assumption is wrong and it's true

Erik the Outgolfer

but that's not how it works I think

HyperNeutrino

17:37

pretty sure it is

Erik the Outgolfer

I think that proof by contradiction is that instead of assuming what you can assume you assume that what you can assume is false, and then try to prove that's false, so the original statement is true

HyperNeutrino

> no but you assume it's false

yeah ik how proof by contradiction works :P

DJMcMayhem

@EriktheOutgolfer I'm pretty sure that is how it works. That's how you prove the irrationally of √2

Erik the Outgolfer

then I think that hyper meant something else than what I thought he means

DJMcMayhem

You start by assuming that √2 is rational, and show that results in a contradiction

HyperNeutrino

17:40

i've done many types of proofs before :P

DJMcMayhem

Really, this is a computer science question disguised as a math question

Erik the Outgolfer

heh

DJMcMayhem

Is there any pair of algorithms, one exponential, one polynomial, where the exponential algorithm is faster than the polynomial for all inputs?

HyperNeutrino

you mean for all or for large enough inputs?

DJMcMayhem

Which intuitively is a resounding no.

@HyperNeutrino I think it's all.

HyperNeutrino

17:43

ok

this sounds a lot like P vs. NP :P

Erik the Outgolfer

@DJMcMayhem well then be sure and ask again after that ;-p

HyperNeutrino

also:

Neutrino Programming Language

General discussion, feature requests, bug reports, golfing tip...

:P

Erik the Outgolfer

otherwise we could be trying to prove something totally wrong and not intended to

HyperNeutrino

yeah

DJMcMayhem

@EriktheOutgolfer Inspired by, but not directly related to

mınxomaτ

17:48

This is hilarious: circleci.com/blog/its-the-future

HyperNeutrino

lol

Erik the Outgolfer

> you know Kubernetes works cause Google built it and it runs on etcd

heh

Neil

18:07

@DJMcMayhem yes. to see this, note that a^(x+1)/a^x =a while the limit of (x+1)^b/x^b is 1 so at some point a^x starts growing faster than x^b, and must therefore eventually overtake it

Eric Tressler

@mınxomaτ reading that reminds me of why I lost interest in whole swathes of software development

Adám

@cairdcoinheringaahing J, 3 bytes: ]%|

mınxomaτ

@EricTressler Don't worry, these are the extremes you can safely ignore.

I routinely see teams shipping and getting their first customers for a CRUD app written as a single binary in an "old-school" language, while devs who chase the newest 'orchestration' framework still struggle to debug their Terraform script.

Eric Tressler

18:22

@mınxomaτ Lately I've been trying to learn how to use JNI with a C++ library. So that's where I am :|

caird coinheringaahing

18:57

@Mr.Xcoder Let me guess, Ṡ?

DJMcMayhem

@Neil I'm not sure I understand. So, you're inferring something about the slope of the f by looking at f(x+1)/f(x)? Does that work? Because the slope is f(x+1)-f(x) so I'm not sure that works

Mr. Xcoder

@cairdcoinheringaahing no, the message was ”Jelly, “

DJMcMayhem

Not saying you're wrong, I just don't understand how the limit((x+1)^b/x^b, x --> infinity) == 1 is meaningful

HyperNeutrino

are you still talking about the a^x>x^b?

DJMcMayhem

Yeah lol

Neil

19:07

if you want to look at the slopes, take the log first

so the slope of log(a^x) is constant while the slope of log(x^b)->0 as x->Infinity

Adám

19:32

@cairdcoinheringaahing J, 2 bytes: %| (any shorter, and it would be a built-in, no?)

Jonah

For those in knowledgeable in PL theory, would it be possible to transpile J into javascript? If not, why not? If yes, how difficult would be?

Mr. Xcoder

PL theory -> programming language theory?

Jonah

Mr. Xcoder: yes

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Are these numbers perfect for each other? code-golf number math A perfect number is a number that is the sum of its divisors. Examples of perfect numbers are: | Divisors | Divisor sum -----+-------------------------------------------+------------- 6 |...

sergiol

19:54

hello

HyperNeutrino

hello o/

sergiol

@Dennis there?

Riker

20:05

@sergiol if you need him, you can always just ping him and he'll get around to it when he comes back

sergiol

how do I ping him?

Mr. Xcoder

@sergiol Exactly how you did earlier (@Dennis)

That is called pinging.

sergiol

ah LOL XD

OK

tks

Riker

@sergiol the other reason why: @pinging leaves a message in the person's inbox, and it's kinda useless to get a "are you there" message, when they're clearly not there :p

Adám

@Jonah ngn/apl is written in JavaScript, so I don't see why not. May not be pretty, though.

HyperNeutrino

20:13

> J

is J similar to APL?

Adám

@HyperNeutrino J is an APL dialect.

HyperNeutrino

oh I'm dumb lol

Adám

@HyperNeutrino No, how would you know?

Riker

CMC: shortest way to map 1 to 0 and 0 to 1

any other inputs can do anything

it must output 1 and 0, not truthy and falsy (though string is acceptable)

H.PWiz

@Riker 1-n

Adám

20:15

@Riker APL, 1 byte: ~

Riker

@Adám is that just bitwise negation?

or standard not

caird coinheringaahing

@Riker Jelly, 1 byte: ¬

Riker

@H.PWiz ah, didn't even think of that

Adám

@Riker Nope, that maps 1 to 0 and 0 to 1. All other inputs cause an error.

Riker

huh

feersum

20:16

Snails, 2 bytes: \0

Riker

huh, I thought more langs would leave as bool

shortest way I could find in python was -~- on the input

Adám

@Riker It penetrates arrays, though, so ~1 0 (1 1 (0 1) 0) gives 0 1 (0 0 (1 0) 1)

Riker

@Adám ah cool

Adám

@Riker Also, APL only has one numeric data type; number.

Riker

ah ok

Jonah

20:21

@Adám, ngn is an interpreter. theoretically speaking, if you can write an interpreter, does it mean you can write a transpiler? i suppose so, since your output could just contain the interpreter itself... but this seems like a cheat. there may or may not be a well-defined question in what i'm asking...

HyperNeutrino

@Adám ¯\_(ツ)_/¯ idk :P

Jonah

@Adám random side question: do you work with roger hui at dyalog?

Adám

@Jonah No, you're fine. If you can write a J interpreter in JS, then you can find a JS expression for every J expression.

@Jonah Yes, I do.

Jonah

@Adám cool

Adám

@Jonah Yeah, my Iverson number is 2.

Jonah

20:25

@Adám do you pick his brain for iverson stories? i know i would...

Adám

@Jonah No need to pick, they come all by themselves…

Jonah

@Adám that's great

@Adám out of curiosity, do you program mostly APL because of work, or do you just prefer it to J and, if so, what are your reasons?

Adám

@Jonah Both really. I grew up on APL, but I've been exposed to J over the years, and always found its bi-glyphs hard to read. I also dislike the inability to manipulate boxed content and have mixed data types in the same simple array. Furthermore, I find J's focus on tacit OTT: Confusing that the REPL is explicit, but functions are tacit (by default). There's probably more.

@Jonah Yeah, I find that J, in its strive for purity, is willing to sacrifice usability.

@Jonah Oh, and [: is an abomination :]

Jonah

@Adám what do you mean by "inability to manipulate boxed content"?

Adám

@Jonah Compare J and APL for [1,2]+[3,4]

@Jonah Don't get me wrong though, a lot of things are obviously better in J, like rank being defined for all functions, with leading axis extension, consistency across primitives, and the broader selection of primitives. No historical syntactic anomalies , etc.

Jonah

20:38

@Adám yeah, of course you can use each but it's a valid complaint. I see no good solution though, since sometimes you do want the verb to apply to the boxes and not their content.

Dennis

@sergiol I am now.

@Mr.Xcoder still pings.

Adám

@Jonah Well, "under unbox" that is. The code just ends up being littered with those. In APL, only scalar functions penetrate enclosures, structural functions work on them. In practice it usually ends up being what the user wants, in my experience. Can you give a counter-example?

caird coinheringaahing

@Dennis would you be willing to test something in the Sandbox with me?

Dennis

If I can do it from my phone, sure. PPCG sandbox or chat sandbox?

caird coinheringaahing

Chat sandbox

Jonah

20:45

@Adám, eg, you might want to ravel , two boxed lists, rather than ravel their contents. You may be right though that the default ("each'd" vs not each'd) could depend on the verb, and therefore be more sensible for the majority of use cases.

Dennis

@cairdcoinheringaahing Well, what is it?

caird coinheringaahing

@Dennis Could you ping me, then after 5 minutes, edit the message and ping me again?

Mr. Xcoder

@Dennis Aww sorry, forgot pinging is enabled in code snippets too! I really didn’t mean to disturb you ;-)

Dennis

No worries.

@cairdcoinheringaahing Done and done.

caird coinheringaahing

@Dennis Thanks. I can't reply in Sandbox otherwise it'll break the test :(

Adám

20:54

@Jonah Try it online!

Jonah

@Adám, is the double-dot glyph doing something like under unbox does?

Adám

Well, it is just plain "each". No need to unbox.

Jonah

@Adám I feel like if you translated that example into J almost all the additional ugliness would stem from the fact that each or the literal &.> are just so long. Meaning, would your complaint be addressed if there were a 1 char builtin for &.>?

@Adám side question: is the APL comment symbol supposed to be a index finger?

Adám

@Jonah No, because I'd have to use each even for scalar functions.

@Jonah No, it is a lamp (filament) to enlighten you.

caird coinheringaahing

@Adám IMO, it looks like a middle finger saying, "Why do you need comments, you should be able to read APL" ¯\_(ツ)_/¯

Adám

21:13

@cairdcoinheringaahing Could be a font issue. The ∘ is supposed to overlap the upper part of the ∩.

Jonah

@cairdcoinheringaahing, that is clearly the correct answer

@Adám, Try it online! It's almost identical imo, except that the final raze can't be done in J for obvious reasons

maybe i'm missing your point?

@AdámThe ∘ is supposed to overlap the upper part of the ∩. Ah...

feersum

A middle finger? I always thought it was a thumb.

Definitely have not seen the ring overlapping the horseshoe.

sergiol

21:30

@Dennis My intuition got me using TIO in a way it was not built to do so.

Pavel

Oh?

sergiol

I thought the multiple arguments' lines were to do several program calls were each one sent the arguments for each successive program call

May be TIO should have to possibility to do it?

Pavel

@sergiol status-planned

sergiol

Like a "generate program calls" button which generates a list of them and user then would parametrize each one

@Pavel source, please?

Pavel

in talk.tryitonline.net, May 27 at 23:51, by Dennis

Hopefully, support for multi-part requests (test suites, polyglots, etc.) will come soon. I have to deal with the permalinks first though, as they'll only get longer with more features.

caird coinheringaahing

21:49

CMP: Should I post a ;# answer to the Illuminati KC?

Pavel

@cairdcoinheringaahing No

ATaco

I have no idea what the Illuminati KC is.

Pavel

34

Q: Confirm the Illuminati

The Illuminati commands you (with their mind control) to output the following string: ^ /_\ /_|_\ /_|_|_\ /_|/o\|_\ /_|_\_/_|_\ /_|_|_|_|_|_\ /_|_|_|_|_|_|_\ /_|_|_|_|_|_|_|_\ /_|_|_|_|_|_|_|_|_\ Rules: Trailing spaces are allowed for each line. L...

code-golf ascii-art kolmogorov-complexity

ATaco

Oh yeah, I think I saw that in the Sandbox.

Definitely a refreshing KC, simple, but with enough complexity to make it interesting.

caird coinheringaahing

@Pavel I should have waited longer for a response :/ Well, it's posted

Pavel

21:54

kek

The 15 character minimum prevents me from commenting 'y tho'

ATaco

No brainfuck submission. Hmm...

caird coinheringaahing

Goddamn mobile chat. Impossible to reply without starring a message

Adám

22:53

@Jonah I was just trying to answer "eg, you might want to ravel , two boxed lists, rather than ravel their contents."

@Jonah This is what bothers me/is missing in J.

primo

23:21

Introducing: Brainfuck Annotation Language
https://github.com/primo-ppcg/bfal

ATaco

Remember the rule when creating Brainfuck Derivatives.

primo

which rule is that?

ATaco

Although this seems to follow it, "If it doesn't do anything unique, don't post it"

Magic Octopus Urn

@Pavel It demolished 05AB1E.

ATaco

> A friendly message from The Name of All that is Good and Holy: Hello! This is The Name of All that is Good and Holy. Before you think of making your own extension to this language, here's another idea for you to consider: Don't! For the love of me, please don't make another derivative of this language! Unless you have something truly new to contribute, it's been done a million times before and will probably make it worse than the original.

Jonah

23:34

@Adám Yeah it's not as nice, but this J version seems slightly better: 1 +L:0 (1;2 3);4 5.

primo

@ATaco I don't know of any other language designed specifically to annotate brainfuck

Magic Octopus Urn

The main point I had was, that was the dude's first catchy question, let him have it haha.

ATaco

Hence I said this seems to follow the rule.

Magic Octopus Urn

We should implement like... a weekly tag cap. You can't post a question in PPCG if the tag exists for more than 10 questions for that week (other than code-golf).

primo

most bf code that is annotated doesn't in any way resemble runnable pseudocode

ATaco

23:37

@MagicOctopusUrn I disagree.

Magic Octopus Urn

@ATaco Knowing SE it's probably not even possible >_>

ATaco

I think you're idea of Annotated more closely resembles C-Like, rather than the particular memory and writing models of Brainfuck.

primo

maybe it's just my coding style. i've found it very helpful to start with a working logical structure, and then just fill in the blanks

ATaco

Is this your first language?

primo

first i've shared with anyone

feersum

23:50

I don't get what it is, a BF code generator or what?

The readme is hard to follow because it dives into details immediately without explaining any overarching goal.

primo

maybe i should move the sample code to the top

the purpose is to annotate brainfuck:

+++>+>++[++[>+++[>+++++++>+++<<-]<-]<]>>>-
>>>---[<]+[>.+.>>>.[<]<]>[.---.>..>.>.[<]] ; what is this

feersum

So it's some executable brainfuck code with comments in it?

primo

; byte sequence
; h l o excl
+++>+>++
[++[>+++[>+++++++>+++<<-]<-]<]
>>>->>>---[<]

n=1
+
if(n)
[
  'hi' excl
  >.+.>>>.[<]<
]>
else
[
  'hello' excl
  .---.>..>.>.[<]
]

ATaco

Is it necessary to include BF? Does it generate BF on it's own?

primo

remove the brainfuck, and it;s still valid

yes, you need to write the BF :p

feersum

23:56

What does the Python code even do then?

If it's just a style of commenting brainfuck

primo

n=1
if(n)
[
  'hi' excl
]
else
[
  'hello' excl
]

Thomas Ward

@Mego i think the laser eyes make it hard to be one of you lol

accidentally slices through the chat room walls with his gaze

primo

this is valid code in bfal. it's an interpreter

the idea is, you create the logical structure in bfal, confirm that it works, and then add bf bit by bit

ATaco

So, it's an interpreted language designed to act as a sort of template to write brainfuck to match..?

primo

exactly, and as you add brainfuck, it remains valid in bfal

Neutrino Programming Language

Transcript for