« first day (2338 days earlier)      last day (2506 days later) » 

10:01 AM
>4+3→a will assign the memory location at (4,0) to the variable a
 
oh you mean for the BF+ side?
because that won't reduce to BF1 either
 
now when you access a, you go back to (4,0) using < and > to return back
I think it can reduce
 
oh that's what you mean
yes, but where the pointer is is non-deterministic
 
I don't know how the assignment will actually look like, but otherwise yeah
@HyperNeutrino ah, I see
 
The problem is let's say I do ,[>]. Then, the pointer could either be at ->1 or ->0
I wanted to do that but it wouldn't work :(
 
10:04 AM
Try it online! Why is this programming giving random outputs on TIO while works fine in python on windows?
 
aha! there you have it, a shortcut for "read all input" that compiles into ,[>,]
 
well no
,[>,]
 
Maybe assignment could use something like a:3?
 
@HyperNeutrino that's what I meant
 
lol
I thought so
 
10:05 AM
you saw nothing
 
But then where would you store the variables
 
but yeah adding that to BF3
@Qwerp-Derp that's the issue. the pointer location is non-deterministic so I can't reduce it down to < and > to return to the location :(
 
@Qwerp-Derp even if they could be stored somehow, the variable can be overridden quite easily
 
Or maybe you could set up aliases for cells?
So maybe a^1 means a is now an alias for cell 1
and if you do a:3 you change cell 1 to the value of 3
 
but that won't reduce to brainf**k
> The purpose of this project is to try to build up many layers of abstraction, all compiling down to Brainfuck, with the ultimate intention that Brainfuck will be able to be used for more complex things.
 
10:08 AM
oh right, you said you can't determine the IP location
 
can -> can't?
xD
 
anyways you need comments
 
it's called everything that doesn't match [+-\[\],.<>0-9]
 
so that comments can even have otherwise valid BF symbols
 
oh
should be easy enough
I might make that BF2 and shift the other ones over
I also need an alias to shift over everything to essentially insert a 0
 
10:13 AM
do you mean [5,1,5,99,132,5] => [5,0,1,5,99,132,5]?
 
yeah
I mean, that's just <
because it autoexpands
 
so you shift everything from cell 2 (1-indexed) 1 to the right
 
But I mean something like [>]<[[>+<-]<]<
yeah
 
ninja'd
 
lol
ಠ_ಠ my attempt failed
grr it keeps skipping the zero but if I make it not skip the zero then it also doesn't loop
oh wait i'm dumb of course it does that
 
10:18 AM
what zero?
 
the gap created
 
How feasible would be a processor/computer performing only 8 brainfuck operations () provided we have a smart high_level_language-to-brainfuck translator?
 
but the way i did it is move to end and keep pulling right, which obviously pulls everything to the right >_>
need halp
 
you need to set your start cell to 0, I think
or actually before the start cell
 
but that overwrites something
 
10:20 AM
How abstracted from BF are we talking about right now?
 
not at all
this is meant to be a BF code thingy
 
So currently, BF3 is just BF, but with some builtin functions?
 
mostly, but I'm trying to make a BF thing here, not a BF3 thing.
 
Oh wait the conversation drifted off
Uhhh can I give a suggestion for an abstraction thing?
 
I had sort of an idea for what you're doing
the compiling from high level to lower level over small parts that you suggested
 
10:24 AM
@Qwerp-Derp sure
 
plus_five=+++++
plus_five+++.
That would output 8 (or the byte equivalent)
 
Or just +8.
because that's part of BF2
 
Yeah, but this should theoretically work with any BF string on the other end, as long as it's on the same line
 
I mean I thought up a challenge where you compile your language into bf, the next person compiles their language into that language etc.
 
@HyperNeutrino but this (what Qwerp said) could be more useful, you can have templates for if-statements and for-loops and stuff
 
10:26 AM
but I got stuck on how to make sure there is an actual challenge to doing it
 
@DestructibleLemon You'd have a hard time trying to enforce the creation of "interesting" languages, as opposed to trivial extensions
 
I would never stipulate that
 
@KritixiLithos that would be quite useful; though quite challenging, I think I could probably do it
 
I was thinking that you couldn't use the same datatype as the previous answer
 
Ooh, what if you had to make a compiler for language A to language B, but the compiler must be written in language A?
 
10:27 AM
@Qwerp-Derp this is exactly my idea
just saying
 
@DestructibleLemon Oh :P
How are you stuck then?
 
just making the same language over and over again
 
ban that then???
 
but how would you ban it?
 
There doesn't appear to be any loopholes - if language A is a trivial extension, then it's more difficult to come up with a compiler (theoretically speaking).
 
10:28 AM
@Qwerp-Derp how so?
actually I misread the thing
I thought it was to language A
 
ಠ_ಠ comments not working
 
if you do that, then you can program language A with a specific builtin
 
oh wait it's because BF2 removes comments, but BF3 just deletes the #
 
@DestructibleLemon But then language A has to implement that builtin into the compiler
 
I had the opposite idea @Qwerp-Derp because your idea allows builtin abuse
@Qwerp-Derp yes, it does. by just using the builtin
also it works better if the compiler can be used before the lang has already been compiled
 
10:31 AM
Make it so that the language has to have a previous entry on PPCG (and the language has to be Turing-complete).
 
so how about this: language A can only have stacks, language B can only have queues, Language C can have neither stacks nor queues, language D can have only stacks, etc.
@Qwerp-Derp but that takes the language design out of it
 
@DestructibleLemon Hard to enforce
 
I feel like it could be
 
@DestructibleLemon Wait they have to design the language?
 
@Qwerp-Derp yes
or no
they could do either
 
10:33 AM
@DestructibleLemon glhf then
 
I mean technically it is unobservable, but I mean, you make a spec, and if the interpreter doesn't follow the spec, it is invalid?
I mean if that is unobservable, every answer is unobservable
do you see my point?
Also I like the idea of the compiler chain
maybe not using brainfuck? IDK. also you probably need a TC proof
the only issue is that you still have the resubmitting languages issue...
maybe we could make it somewhere else because it won't work on site
@KritixiLithos delet this
 
@DestructibleLemon You can just say "no reuse of languages"
 
who here would like to contribute to a github of compiler chains?
@Qwerp-Derp of course. but then you still have trivial variations
 
done, sorry
 
10:38 AM
@DestructibleLemon Wait how is that offensive in any way though, it was a misunderstanding
 
it's probably a good idea not to have them anyway. let's move on now
ok this probably looks weird out of context but this message makes it less out of context
again, who would like to contribute to a github of a compiler chain?
like at the bottom it is brainfuck. like the idea you had @HyperNeutrino was it?
guise? ooh my rep in chat is palindromic
 
ooo yay
but yeah that was my idea
 
I had the idea first
for the record
in mine the emphasis was more on the compiler chain's links than the ends of it
 
There is a similar idea though which is a program that when run in language A outputs program B, when run in language B outputs program C, etc. and eventually loops back to the original after 150 languages
 
that's rather different
but anyway
 
10:44 AM
s/similar/related/
 
anyway, should I make a group or something?
 
group?
also yay functions work
 
push
once you're ready
 
updating README first
 
is this that high level bf thing?
 
10:50 AM
pushed
 
ok I tried to make an organisation but the billing thing has to be there, even though there shouldn't be any billing
 
@HyperNeutrino but your main program can't be multiline ;_;
 
How should I change it then
there's thing wrong with ^^ though
other than missing comments
 
use Qwerp's idea of name:function_body
 
who wants to be part of this organisation?
 
10:52 AM
oh wait the objective wasn't to golf whoops. yeah that works better
 
where the name can be more than one character long if needed
 
@HyperNeutrino do you want to be part of this organisation I am making?
 
for compilerchaining?
 
sorry no thanks. i have other projects i'll be working on soon
 
10:53 AM
aw...
@Qwerp-Derp?
 
@KritixiLithos Should I impose any restrictions on the function name?
 
@DestructibleLemon Maybe maybe not, I might have to wait until school holidays
 
might as well add you now...
 
And I'm making my own (non-golfing) language with a friend
 
Also, when did everyone here wake up, out of curiosity? It seems like most of you are probably not American teenagers because no teenager could be up this early.
(oh wait... i'm here)
 
10:54 AM
you are Qwerp-Derp on github?
 
@DestructibleLemon what is the org for?
 
compiler chain project but instead of it being slighly by slightly higher level bf it's about the different langs
 
do you think [A-Za-z_]+ is a good restriction for function names?
 
@HyperNeutrino IMO there shouldn't be any spaces for colons or {set of BF characters} in the name
 
@HyperNeutrino for your high level bf, one of your design goals is backwards compatabilitys?
 
10:55 AM
or digits
 
Shameless plug for Aether, a programming language I'm making
 
@HyperNeutrino \w+?
 
@DestructibleLemon sorry, I don't know exactly what backwards compatibility is :P
@KritixiLithos I guess then (\w|_)+
 
any program in brainfuck works in your BFomega
 
Oh. Yes.
 
10:56 AM
\w includes _ ;)
 
bleh
i don't regex
 
unless it has nops
 
actually I think that might not be possible to do?
 
Any program in a version without no-ops will work in a higher level
 
10:57 AM
ok
you may or may not need to be careful with the kinds of things you allow just in case you make it impossible to make a higher level language with that capability
can I join your project?
 
@isaacg, I'm changing this to fastest-code... Do you mind sharing the specs (and limitations) of your computer here (if it's still OK for you to do the testing).
 
sure. you can PR me if you want on the repo
anyway will be back soon gotta go take a break from this for a bit cya o/
 
wut is bfcore?
 
@HyperNeutrino yee, done, it was just a formatting PR :/
 
11:14 AM
actually now that I think about it, you have plenty of unicode if you exhausted ascii
why do you have a bunch of layers if you compile straight to bf?
 
Done, posted the challenge
Thanks for the feedback guys :P
 
11:31 AM
@LeakyNun "How will I obtain the browny?" -> I'll send you one soon
 
@Mr.Xcoder thanks
 
@LeakyNun Here you have it: 🍪
 
@Mr.Xcoder thanks
 
Wow, PPCG is quite inactive this time of the day
9 views in 18 mins.... 1view/2mins
@LeakyNun Can you remove that comment now?
 
@Mr.Xcoder done
 
11:37 AM
@LeakyNun Thanks
Wow, I don't even know 4 different languages :)
 
1
Q: Proper Divisor mash-up

Mr. XcoderA proper divisor is a divisor of a number n, which is not n itself. For example, the proper divisors of 12 are 1, 2, 3, 4 and 6. You will be given an integer x, x ≥ 2, x ≤ 1000. Your task is to sum all the highest proper divisors of the integers from 2 to x (inclusive). Example (with x = 6): ...

 
11:50 AM
who else wants to join the compiler chain organisation?
 
Nebodee if they nose see, come to hair :P
(Translation: Anybody if they knows C, come to here)
 
somebody say they want to join?
do you want to join the compiler chain thing? @AlexKChen
 
What's that ?
 
so, the first lang is bf
 
I suddenly got 2 ups on the build the worst hexagon challenge
 
11:56 AM
@Mr.Xcoder You know C ?
 
the next lang is made, and someone makes a compiler in bf that compiles from the new lang to bf
 
Nearly at the same time... After it was sattled
@AlexKChen Only C++
Sorry
 
@DestructibleLemon Nope sorry I don't know any language even that properly.
 
the next lang after that is made, and a compiler written in the lang1 compiles lang 2 to lang1
etc.
@AlexKChen you can make a lang?
 
@DestructibleLemon Is it code-golf, or just simply build a functional compiler?
@DestructibleLemon If it's the latter, maybe I'll consider joining it
 
11:58 AM
build a functional compiler
it's going to be on github
because it would be offtopic on ppcg
github account?
 
Yes
Never really used it
and never really used github at all
 
you should make a github account if you want to join
also github is good for if you ever make a programming language
 
I have one
 
what is your github username?
 
@DestructibleLemon I'm updating my profile right now on github. My username will be set to Mr. Xcoder after the update
 
12:02 PM
can you have spaces?
 
dunno
Mr_Xcoder then
@DestructibleLemon Done, I'm Mr_Xcoder
No, I'm
no
 
@DestructibleLemon I have a problem, and cannot update the profile
I'll be back in a few mins
 
oh, so that wasn't you then
 
yup
 
12:12 PM
any c# devs?
 
what is happening now?
@Mr.Xcoder am I gonna invite you?
@Mr.Xcoder are you changing your name?
 
12:29 PM
@DestructibleLemon Sorry for the delay: I have changed my username to MRXCODER123 and I have the Swift logo as my profile picture
@DestructibleLemon But I don't be online for the next few hours. I have some family stuff to do...
Are you Destructible-Watermelon?
I have followed you
sya guys :)
 
lel
 
@Mr.Xcoder I won't be online for ~10 hours because of sleeping
maybe you should get a more original name than mr xcoder
maybe
ok bye
 
12:47 PM
@HyperNeutrino iirc he's generally on in early morning/late night
so inverse of US, broadly
 
lel
 
1:28 PM
1
Q: Plant trees on a golf course!

Stewie GriffinThis challenge is inspired by this app. This is a much easier version of this challenge. This challenge is code-golf, while the other one is fastest-code. You'll be given a square input grid, of dimensions 6-by-6 which is divided into 6 areas, where the cells of each area has a unique ident...

 
2:10 PM
@officialaimm because dicts in python are unordered
and the order of replacement keeps changing
which messes it up
 
@totallyhuman Yea.... But it is returning the same output on my desktop.... that is strange...
 
0
A: Sandbox for Proposed Challenges

Comrade SparklePonyIncremental Ranges! code-golfsequencelist Your task is to, given a positive integer n, return the first 100 (or more) numbers in the incremental ranges sequence, where each range is length 4. The incremental range sequence first generates a range from one to n inclusive. For example, if n was ...

 
@officialaimm try using a list of tuples
 
2:27 PM
0
Q: Knight-fill a grid

DaveA knight fill is a flood fill using the connectivity of the knight chess piece. Specifically: 1 1 1 1 0 1 1 1 1 (0 is the initial point, 1s show the connected cells) Challenge Given a 2D grid of spaces and walls, and an initial location, perform a knight-fill on the grid. Shortest co...

 
I tried to invent a sequence for a code golf challenge, but every time I came up with one, I accidentally created an alternate definition of Fibonacci numbers. >_>
4
 
how about "print the last of the the nth fibonacci number"?
1, 1, 0, 1, 1, 0, 1, 1, 0, ...
oh wait, that's trivial pattern
n=>n&3>0 if the sequence is 1-based
or the last 2 bits
 
@KritixiLithos The last what? Bit? Digit?
 
bit
I think 2 bits might be more interesting
 
Hi @KritixiLithos !
Did you check my question at "The C programming laanguuage" ?
 
2:36 PM
1, 1, 10, 11, 1, 0, 1, 1, 10, 11, ... okay that's another repeating pattern
I pinged you in that room
pattern formed last 1 bit of fibonacci sequence => repeats every 3
pattern formed last 2 bits of fibonacci sequence => repeats every 6
 
2:53 PM
Oh Yea... Thank you @totallyhuman
 
@KritixiLithos next is 12
 
ಠ_ಠ no interesting pattern there
 
ok so i just made a golf lang clone
but i'm sure i can put the commands in an interesting order
or something
 

« first day (2338 days earlier)      last day (2506 days later) »