The Third Stack

1000000000

00:23

@DJMcMayhem I'm not sure that it makes more sense in general. It makes more sense when thinking of input but when thinking of about output printing the top first just feels more stacky.

That being said, I might be a little biased since printing the top first would make CraneFlak easier to write.

DJMcMayhem

Haha, what's craneflak?

1000000000

@DJMcMayhem An interpreter for brainflak written in C

DJMcMayhem

Oh cool.

Eamon Olive

@1000000000 I agree with this

DJMcMayhem

I know Downgoat is working on one in C++

1000000000

00:26

@DJMcMayhem Yeah, I heard

Eamon Olive

in The Nineteenth Byte, yesterday, by Fatalize

If you use C++ like C then you don't use C++

DJMcMayhem

@1000000000 Most stack-based esoteric languages I've used output from the bottom up. For example, MATL, CJam, Golfscript

Then again, MATL has first input on the top of the stack

1000000000

@DJMcMayhem So an empty MATL program reverses the input?

DJMcMayhem

No. Maybe I worded that wrong

Inputs aren't put on the stack until you call a function that reads from values on the stack.

For example, in this the + means push(pop() + pop())

1000000000

@DJMcMayhem I'm still not sure I understand exactly how that works, is there some operation to pop from the input stack and push that value to the output stack?

DJMcMayhem

00:38

No, it's called implicit input. + pops the top two values on the stack, and pushes their sum. If there isn't enough already on the stack, then it grabs it from the input.

1000000000

@DJMcMayhem okay I see how that works

DJMcMayhem

3

A: Tips for golfing in MATL

Implicit Input Most functions accept some number of input. These inputs are taken from the top of the stack. If the top of the stack does not contain enough arguments, it will draw the remaining argument from the input. (See Section 7.3 in the documentation) I'd like to cite the original explana...

You should learn MATL. It's the only golfing language I know (other than V), and I think it's pretty fun.

Eamon Olive

I think that we may want to consider reversing the input. It is really stacky.

if you think about it when you push and pop from a stack you get everything back in reverse

It also is the way that you initially had.

DJMcMayhem

I hate to veto you guys, especially since I really appreciate all the work you guys have done on it (especially example code!), but I really like the IO format in it's current state.

Eamon Olive

ok

DJMcMayhem

00:44

It just makes the most sense to me, and I like how the empty program just prints all input in the same order

1000000000

@DJMcMayhem Okay fair enough

DJMcMayhem

If we want to keep that ^^, we could reverse input and output, but that seems like more trouble than it's worth. I don't know if that would provide any benefit.

Eamon Olive

@DJMcMayhem Could you clarify?

DJMcMayhem

So for example, 1 2 3 for input would have the 1 on top of the stack, and output would basically be just while not stack.empty: print(pop())

Eamon Olive

@DJMcMayhem That's the current system

I think

wait nevermind

DJMcMayhem

00:49

No, the current system would have 3 on top of the stack

e.g.

Eamon Olive

@DJMcMayhem Which system do you want? I am really confused right now

1000000000

@EamonOlive Okay lets clarify. Here is an example from the current version on GitHub:

brainflak ([{}]) 10 20 -> -10 20

so the first input winds up on top of the stack

DJMcMayhem

Shouldn't that be negative 20?

brain-flak.tryitonline.net/#code=KFt7fV0p&input=MQoyCjM

Eamon Olive

@DJMcMayhem Not the way it is now on github

The two are out of sync a bit

DJMcMayhem

Oh. I like the way it is on TIO

@1000000000 I suppose that way would be OK too

Eamon Olive

00:54

@DJMcMayhem We both really like the way it is on github.

DJMcMayhem

FWIW, I almost never run brain-flak locally. I usually use TIO

Eamon Olive

It is your language though

DJMcMayhem

@EamonOlive My language that you've written more code in it than I have!

:P

Eamon Olive

Ok so we'll revert it to the way it is on TIO?

DJMcMayhem

I guess now that I think about it, the current format (if I understand it correctly) is good enough.

Just to clarify, if the input was "Hello world!", then {} would print "ello world!"

Is that right?

Eamon Olive

00:56

Yes

DJMcMayhem

Actually, now that I think about it, that's even better since it visually looks like a stack. For example, 1\n2\n3 looks like 1 is on top.

Eamon Olive

@DJMcMayhem Thats why I like it

DJMcMayhem

Alright, sounds good to me

One reason I was opposed to that is since I didn't feel like going to the trouble of implementing it, but if you guys have already done it, I don't see why not.

Eamon Olive

@DJMcMayhem Ok so are you ready for 1.1.0?

DJMcMayhem

Let me check some things real quick

Eamon Olive

00:59

@DJMcMayhem np

1000000000

One other thing, we will soon have 3 (and in the future with JaneFlak, 4) repositories for Brainflak (

DJMcMayhem

Haha, is that bf in J?

Eamon Olive

Javascript I bellieve

DJMcMayhem

@EamonOlive It looks good to me, but I'm gonna make one more push before I ask Dennis to pull.

Just some minor tweaks to make it work better with TIO

1000000000

@DJMcMayhem Make sure that you remove the DEVEL from the string that the -v flag prints

DJMcMayhem

01:03

OK

Are there any uncomitted changes or anything else I should know?

Eamon Olive

nope we are ready to go

1000000000

@DJMcMayhem Not that I know of

DJMcMayhem

Oh one question, do all debug flags go to STDERR?

Eamon Olive

@DJMcMayhem no should we?

DJMcMayhem

Yeah. That way it doesn't officially change the output, since STDERR is ignored.

And it won't show up unless you hit the debug button on TIO

I can get that while I'm at it (it's related to the changes I'm making anyway)

1000000000

01:06

@DJMcMayhem Sounds good

Eamon Olive

Ok

DJMcMayhem

02:02

@EamonOlive @1000000000 I just realized that reversing the input will invalidate my Hello World answer :/

Eamon Olive

02:27

@DJMcMayhem Are we going to fix it or change the language?

DJMcMayhem

Idk. It seems kinda silly to base a design choice around one existing program, but I don't feel like rewriting it... :/

1000000000

@DJMcMayhem Well, its not like @EamonOlive and haven't expressed such reasons

Eamon Olive

@DJMcMayhem If I beat the current score with a backwards version would you change it?

DJMcMayhem

@EamonOlive What if we just add a reverse flag? That way, my byte count won't change (since I used an older version) and we can go ahead with the new order.

Eamon Olive

@DJMcMayhem Seems ok it might resolve this issue

1000000000

02:39

@EamonOlive It would be pretty easy to implement as well

DJMcMayhem

Should it reverse input and output, or just output?

Idrc either way

1000000000

@DJMcMayhem Probably both, right?

Eamon Olive

@DJMcMayhem I say both

1000000000

Okay, I'll implement it

DJMcMayhem

It makes no difference to me, so sounds good

@1000000000 BTW, I've been wondering, how did you first find bf? When you submitted the first issue, I was shocked since it was just this silly language I hadn't finished, and then this random person I had never met started using it, haha.

1000000000

02:47

@DJMcMayhem Well, I stumbled upon your code golf question on implementing it, which was already a bit of a happy accident since I tend not to check the PPCG stackexchange very frequently. I had had some recent interest in esoteric languages when I had discovered Piet a couple months earlier. Upon reading your question I was inspired to try to implement Brain-Flak in Brain-Flak (of course with the caveat that the program was passed in ASCII values).

DJMcMayhem

Hahaha, that sounds like a nightmare. Did you get very far?

Eamon Olive

brain-flak.tryitonline.net/…

1000000000

@DJMcMayhem I wrote a small part of it, ran it to test it, noticed that the interpreter was not running the program correctly, delved into the Brain-Flak source code, and fixed the bug that was causing it. From there I stopped working on the Brain-Flak in Brain-Flak program in favor of helping work on the Brain-Flak interpreter and the rest is history, I guess.

DJMcMayhem

Huh. Now I'm really glad I posted that challenge. :D

1000000000

@DJMcMayhem Someday I'll get around to writing an answer to it in Brain-Flak

DJMcMayhem

02:57

I'll put a bounty on it if you do. Unfortunately it won't be able to take input (since the code is the input)

1000000000

@DJMcMayhem The input will just have to go after the code

DJMcMayhem

Oh, I guess that would work.

You know what would make it way more impressive? Is if it correctly throws errors on invalid code. :P

1000000000

@DJMcMayhem Yeah, but I'll be impressed enough if I manage to convert the ascii values of integers into the corresponding numbers

DJMcMayhem

Why not just take the code points instead?

1000000000

Oh, by the way I added the reverse flag

Eamon Olive

03:02

Hello world golfed: brain-flak.tryitonline.net/…

At least we can rest easy now knowing that our hello world will improve with the new changes

DJMcMayhem

Is that shorter cause you golfed it better or because reverse is more convenient?

Eamon Olive

@DJMcMayhem reverse is more convenient

(for that particular problem)

DJMcMayhem

@1000000000 Huh. The newest version seems to be treating flags as input. Any idea what that's about?

1000000000

@DJMcMayhem where are the flags situated?

DJMcMayhem

ruby brain_flak.rb source -A

If I do ruby brain_flak.rb -A source then it just prints Ā a bunch of times

1000000000

03:10

@DJMcMayhem yeah it treats all flags after the first nonflag argument as input. When it parsed all the arguments for potential flags it would attempt to treat negative numbers as flags

@DJMcMayhem what is the source file

DJMcMayhem

@1000000000 So then what's the correct way to call it?

And the source file is just my hello world.

Eamon Olive

ruby brain_flak.rb -A source

1000000000

@DJMcMayhem ruby brain_flak.rb [options] program [args...] should work

@DJMcMayhem Is there somewhere I can find that?

Eamon Olive

3

A: "Hello, World!"

Brain-Flak, 180, 176, 170 + 3 = 173 bytes ((()()())((((((((({})){}{}){}){}){})(((()()()){}()){}){}())([])[]{})))((((([][]()){}){})[[][]])<>)<>((((((({}))[]{}[][]())[][][])()()())[[]]()()()())[[]]()()())(<>{}()<>) Try it online! This code is 170 bytes long, but adds three bytes for the -a flag...

DJMcMayhem

ty

Eamon Olive

03:17

@DJMcMayhem I noticed that multiple flags don't work in TIO. Is there some sort of reason?

DJMcMayhem

Which flags did you try?

Eamon Olive

-a and -d

DJMcMayhem

It's working for me

Eamon Olive

ah I didn't know how to use flags

thanks

DJMcMayhem

@1000000000 I think I have fixed it. Something was weird in stack.rb but it should be better now

I simplified the print function, so there is less ternary now

1000000000

03:27

@DJMcMayhem Probably for the best

DJMcMayhem

Haha, I am now the contributor with the least contribution in my own language.

:P

Eamon Olive

Still have more than PeteyPii

DJMcMayhem

That's cause he just fixed a typo

So is 1.0.0 official now? Any thoughts before I ask Dennis to pull?

Eamon Olive

@DJMcMayhem Seems so

DJMcMayhem

Sounds good

Eamon Olive

03:37

1000000000 approves too. Our power went out so he can't say anything

DJMcMayhem

Hahaha sounds good

Eamon Olive

03:49

Already found a bug.

DJMcMayhem

What is it?

Eamon Olive

@DJMcMayhem the #<debug flag> goes to STDOUT

minor

but still a bug

DJMcMayhem

Huh. Link?

I actually know absolutely nothing about the debug flags. >_<

Eamon Olive

brain-flak.tryitonline.net/…

Its hard to demostrate whats going on but things are going to the wrong places and it becomes a mess in the output

I pushed a commit fixing it

DJMcMayhem

No, I see it.

Silencing STDERR makes it more obvious what the problem is

Eamon Olive

03:56

Ah yes

DJMcMayhem

OK. Do you wanna ask Dennis to pull?

Eamon Olive

We might find more bugs I think it should wait so we don't bother him too much

Transcript for