@Dennis if frontend shows encoding as sbcs

what is sent to arena? still utf-8?

@Dennis there is something wrong with whitespace

this does not work

if you copy and paste this to whitespace.kauaveel.ee

it displays hello world

@AndrewSavinykh For now, yes. The encoding is merely an attempt to get the right byte count for PPCG answers, but it's far from perfect. I plan to offer multiple code pages eventually and actually sending the correct byte stream to the arena, but we're not there yet.

@AndrewSavinykh Unfortunately, not all Whitespace interpreters are compatible. TIO uses this one for now. I'll see if I can compile the official one. Haskell and I had some misunderstandings in the past...

@Dennis well all I want is "Hello, World!"

Do you think it's more fun to compile an official interpreter or write hello world on current TIO's? ;)

If you take a lowercase w, this works.

But even that exits with an error, and it's from the repo I got the interpreter from...

@Dennis I'm guessing I have to compromise here. I just came across trumpscript which just refused to put capitals

so I changed the expected output to lowercase

and once you take this slippery slope

...

TrumpScript is case insensitive.

Do you happen to know Haskell?

sorry, not at all. I know forth ;)

so the goal is to prove that arena works including wrappers. this means it does not have to be hello world to reach this goal. Also with the number of langauges out there if I do not compromize I will never finish

The page offers a compiled binary as well. I sorta works, but... tio.run/nexus/…

@Dennis don;t worry about this. I'll take a lowercase world then

I read yesterday about 7, it properly did my head in

I'm pretty convinced at this point that 8 bits per character is the wrong number for a golfing language

@ais523 oh! oh! it's all your fault!

it's not enough to fit in everything in, say, Jelly, and if you're going to Huffman code (i.e. mix one- and two-char commands) you'd want a smaller character set

I have a suspicion that the optimal number if you're going for a relatively small set of builtins is 6

with more, it's much less obvious

now tell me please, why swapping and adding an empty section in between is uselful? tell me why pacifying and removing the last bar is useful?

most of the operations allow you to reduce the number of operations full stop, or else the verbosity of some common operations

or being useful was not the idea at all?

6, for example, is the simplest operation I found that lets you escape things in the original program arbitrarily

so I'm guessing what I'm trying to understand that the operations that I mentioned they do 2 things at the same time

in the first case 1) swapping 2) adding bar. In the second 1) pacifying 2) removing bar

you can currently escape with 7…6 (thus reusing 7 for two purposes); if it just pacified, you'd need to write some custom stack manipulation code simply to fix the literals

what I'm trying to understand is why doing them together is useful

because it reduces the size of the language's character set

3 is the most obvious example: it a) does I/O; b) discards stack elements

I'd need to have two different commands for those if they weren't combined, as there's no way to do either operation individually

and yet you can get at the two individual operations trivially via 23 and 13 respectively

@AndrewSavinykh I'm nothing if not stubborn. tio.run/nexus/whitespace#XY1RDgAwBEO/n1P0Hu5/… works now.

so with 3 it discards two stack elements

so then you can think about "why does 4 leave an extra bar?", and the answer is "so that it sets up 43 and 40 in the most convenient way"

The source files didn't have the debugging information at the end. I wonder why the compiled binary would then.

if you're swapping something, you're probably intending to treat it like data, and thus the bar below it is helpful because most commands also have a second purpose on literals

@Dennis cool. I understand by the nature of being case insensitive trumscript can't produce cased out in principle?

actually my main regret is that I made 0 a passive 6; active 6 is highly useful but it possibly doesn't even need a passive variant as a literal

@Dennis Thank you so much, I already feel "cleaner"

anyway this is getting offtopic and I should really be in bed

@ais523 good night ;)

@AndrewSavinykh It was bugging me too. :) Anyway, afaict this is the official interpreter for Whitespace 0.3.

@Dennis do you by any chance have hello world for 99?

@Dennis and one more question. For gs2 my test works fine because I send to the backend \x12h, where \x12 is symbol with this code. But is there any way to type it on the web site?

@AndrewSavinykh No, but Martin does.

@AndrewSavinykh The GS2 interpreter supports the glyphs from the original CP437 instead of control characters. en.wikipedia.org/wiki/Code_page_437#Characters

@Dennis Thank you!

More accurately, the wrapper does some conversion before calling GS2.

nice!

So I need hello world for 32 more languages

Here is the list

If anyone wants to donate a hello world from the list, they are most welcome

down to 24

@Dennis Can you please pull Brachylog? (It's a big one). It now uses a different code page but if the files for the code, input and output that you use for Brachylog are encoded in UTF-8 it should work without you having anything to do

(To check that it works if I'm not here, the code :I{ṗ≜}ᶠ⁽ with input 25 and argument Z should output Z = [2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53, 59, 61, 67, 71, 73, 79, 83, 89, 97].)

@Fatalize Seems to be working. tio.run/nexus/brachylog#@2/lWf1w5/RHnXNqH25b8Khx7///Rqb/owA

@Dennis Woohoo! Thanks!

Is it possible to change the way it counts byte on the right?

Sure. Should every character count as a single byte or is it more complicated?

Every character counts as a byte

(Though you can also put characters in strings that are not in the code page. But in that specific case we would use utf-8 byte count)

Alright, changed from UTF-8 to SBCS. Might take a hard refresh to get picked up.

Thanks! (what does SBCS stands for btw?)

@Fatalize Single byte character set.

@ais523 would you mind providing a "Hello, World!" for a pear tree?

@Adnan would you mind providing a "Hello, World!" for oasis? "Hello, world!" works but "Hello, World!" prints three zeroes instead of capital W

@Dennis would you mind providing a "Hello, World!" for M?

@AndrewSavinykh Right now, I don't think there is a way to print hello world.

@Adnan okay, thanks

@AndrewSavinykh tio.run/nexus/m#@/@oYY5Hak5Ovo5CeH5RTori//8A and tio.run/nexus/m#@/@oYY7xwx2tiUemWx@dfGj3//8A both work.

@Dennis thank you!

okay 22 to go. @Denis, would you mind looking over the list and let me know if there are any of these that would be relatively easy for you to dig up? I'm sure there will be some languages, where some work on my part will still be required to figure out how the hell to print hello world, but may be some of those 22 has a published or known solution that I have not found and you are aware about?

@Dennis can you add Vala to TIO practical languages?

@AndrewSavinykh I don't really know any of these. I thought I knew Minkolang, but it's being weird. Brainbool isn't possible atm; it only prints 0's and 1's.

← 1 message moved from The Nineteenth Byte

@betseg Moved here so I can't forget about it.

Whoops I keep forgetting this room.

Alright, this should be easy enough. Vala is in the repos.

Almost half of GNOME is written in Vala AFAIK

@Dennis that's all right, thank you for having a look

@AndrewSavinykh numberwang is randomized IIRC

@AndrewSavinykh pbrain Hello World would be the same as Brainfuck

Package gobject-2.0 was not found in the pkg-config search path.
Perhaps you should add the directory containing `gobject-2.0.pc'
to the PKG_CONFIG_PATH environment variable
No package 'gobject-2.0' found
error: pkg-config exited with status 256

@betseg Help.

That would be dumb, but, y'know, is gobject installed?

Nevermind.

@betseg, nope, does not work unfortunately, I already tried this: tio.run/nexus/pbrain#HYqBCQBACAIHEp1AXCR@/…

Vala seems to have screwed up dependencies. The file I needed was in glib2-devel.

@Dennis wow

@AndrewSavinykh here's a now-working Minkolang answer

@AndrewSavinykh pbrain's tape doesn't wrap like Brainfuck's, that's why that program doesn't work with pbrain. This non-golfed program works with both languages though: tio.run/nexus/pbrain#VYmBCQAwCMMOKvYC6SNj/5/…

@betseg tio.run/nexus/…

Site blames caching, but how do I hard reload on mobile?

Chrome?

Yes

Nvm found on SO

Visit javascript:reload(true).

From the Nexus tab.

Ninja'd

@betseg awesome, now if we could somehow add the comma ;)

@betseg I mean, thank you ;))

@AndrewSavinykh tio.run/nexus/…

@betseg Perfect! Thank you do much!

@Dennis where did you get implementation of glypho from?

@Dennis can you update to a develop version of cheddar on TIO if possible?

By that I mean from github

@AndrewSavinykh From here.

@Downgoat I can try. Just clone the repo and make, yes?

@Dennis actually just clone repo and do npm install -g

Eh, I ran it without -g.

@Downgoat npm install prints a lot of warnings, then gets killed.

Well, 4, not exactly a lot.

@Dennis thank you

Probably requires more memory than the arena has.

@Dennis yeah you'll need about 1GB (don't ask)

I guess that's that then. I only have 512 MiB.

I could try with a swap file.

@Downgoat asking

@ConorO'Brien how do you mean?

no consistency iirc

@Downgoat Do I actually need -g? Will the things npm install still be required after make?