Cheddar

Sorry i have no idea what im saying

Quill

It's a good idea, but you just need to choose whether a) it belongs in stdlib (meaning any frameworks you use for gui become core dependencies) b) it belongs outside in an external framework like Electron (which means dependencies are less of a problem)

what i meant was i want GUI toolkit to just need cheddar code

12:57 AM

make your own

are you kidding ._.

that is very difficult

i will need to learn nan

and C++

@Downgoat ?

btw: any suggestions for image proccessing lib for cheddar?

@Downgoat image processing?

DJMcMayhem

@Downgoat PIL. :P

1:01 AM

@DJMcMayhem I was gonna but I couldn't find any node port

this?

i want to make Python interfaces for cheddar but that seems difficult

Are there any good resources for learning cheddar?

DJMcMayhem

I thought you hate snek?

@Zwei docs.cheddar.vihan.org

1:03 AM

@ASCII-only that's a good one

btw, i wanna add an op that'll take a function and reduce it's arity to a specified one. Any ideas for what op symbol should be?

@Downgoat what on earth does that mean

also, should we do this in cheddar room

@Downgoat currying?

(+) = (a, b) -> a+b but.... (+) <op> 1 = a -> +a

@ConorO'Brien btw +1 for BrainFuck-dervative HTML

@Downgoat not a good idea

1:10 AM

y?

it's probably better to make a way to create unary functionized operators

Its helpful for things like negating an array: a.map((-)) will result in (-)(item, index)

@ASCII-only how?

idk, some syntax

something like (@+) or something?

@ConorO'Brien 0/10 too slow to run on TIO

@ASCII-only it's not. it's a repl

@ASCII-only thanks!

@Downgoat Cheddar functions have side effects no

1:17 AM

^

@Downgoat oh, you mean limiting

@ConorO'Brien oh wat why is it even on TIO then

@ASCII-only no, it's a repl made in reticular

@ConorO'Brien ohh

@ASCII-only not a good idea--we're planning to add a behaviour for multiple ops in functionized op expressions

@Downgoat @ASCII-only perhaps talk in here

we're kind of flooding TNB

@mod pls move to Cheddar room

1:20 AM

doorknob is most recent

(+).unary?

perhaps

@Quill @Doorknob pls move messages?

I think it might be useful to have a "limit" behaviour for functions

@ConorO'Brien btw pls update your GH website

1:22 AM

Like, f << n = (...a) -> f(...a.slice(0, n)

@ASCII-only which one?

@ConorO'Brien why

@ConorO'Brien yourusername.github.io?

@ASCII-only kk, will do

@ASCII-only well, negate = (-) << 1

say we have a hypot function

like js's

oh

no wait wat

ternary_hypot = hypot << 3

can't the user just give two/three args in the first place?

1:25 AM

well, the user isn't always passing the args

Quill

← 75 messages moved from The Nineteenth Byte

but when would it be useful?

map

@Quill thanks

@Quill you moved some general content as well

1:26 AM

@ConorO'Brien wat

8 mins ago, by ASCII-only

@ConorO'Brien ohh

9 mins ago, by Conor O'Brien

@ASCII-only no, it's a repl made in reticular

9 mins ago, by ASCII-only

@ConorO'Brien oh wat why is it even on TIO then

35 mins ago, by Steven H.

Challenge idea: implement Strong Array Notation?

10 mins ago, by Conor O'Brien

@ASCII-only it's not. it's a repl

18 mins ago, by ASCII-only

@ConorO'Brien btw +1 for BrainFuck-dervative HTML

@ConorO'Brien pls stop spam

not spamming

1:28 AM

Going through the docs

I don't see the reasoning behind things like 123_456

@Zwei ?

@ASCII-only numeric placeholders

123_456 is a valid numeric literal, but I don't see what the underscore is there for

@Zwei wait

both // and # for comments?

why not also allow --, if you're already allowing both?

1:33 AM

@Zwei oh :/ that's outdated. # for comments is not allowed

can't find it

oh

but C# allows _ in integers, makes it much more readable

@Zwei its mostly for clarity. When you write normal numbers you write 123,456,789; instead of , in cheddar you'd use _

@Zwei why, what language uses that

1:33 AM

so are they ignored?

@ASCII-only SQL

@Zwei yeah

@ASCII-only lua

is 1_2_3_4_5 a valid numerical constant?

yeah :/

position is not enforced

1:34 AM

ah

partly because in future it will be much easier to allow both 0x_7f_ff and 0x_7fff

so it's just to add spacing for readability and serves no practical purpose?

yes

Nested comments seem likea cool idea but I'm honestly not sure where I'd use them

commenting out code with comments in them

1:38 AM

ah

does 'this is a double quote: "' work?

@Zwei yes

The neatest feature I've seen so far is the "formatting" in strings I guess

?

like "5 plus 5 is #{5+5}"

psst that's string interpolation and perl has that

1:43 AM

Hi @quartata

ah

I knew formatting was an inaccurate term for that

come to the perl side we have cookies and working recursion

@quartata do does cheddar...

@ConorO'Brien @Downgoat Do you have Array.shuffle yet by the way?

@quartata idk, downgoat decided to do it

1:46 AM

I did?

I haven't looked at perl yet

@quartata perl side also has weird syntax and an odd mix of half golf-lang and half real-lang

Oh, and are list comprehensions on the todo list

oh gosh

@Downgoat yeah

1:47 AM

perl is sounding tempting now

I think so

@Zwei ಠ_ಠ don't do it

does perl have pattern matching?

cheddar is a tasty, wonderful food while perl is a glorified rock

^^

pattern matching for cheddar is in-dev atm and should be out within some time

also, I do notice the "half golf-lang" thing from cheddar too

1:49 AM

._.

how is cheddar golf-lang ._.

@Zwej in all seriousness the book Higher Order Perl is an excellent read if you have basic Perl knowledge, really good stuff

@Downgoat btw en.wikipedia.org/wiki/…

@Zwej Not by default but wirh tricks yes

It performs better than it should in many code golf stuff

@Zwei Same as Python/JS

1:50 AM

true

@Zwei that's because its designed to have the least amount of overhead while being as clear as possible

There's a difficult balance between conciseness and readability

@Zwei give an example of unreadable cheddar

a->b->(g->g((a-b+1)|>a)/g(1|>b))(n->n.map(Math.fib).reduce((*)))

I may have been wrong

of course everything looks this bad when golfed

@Zwei yes

1:54 AM

[1, "2", "3", 4, "5", 6] I think I'll pass...

@Zwei ?

@Downgoat 1. functions can still have side effects right 2. plans for concurrency?

@quartata define side effects

As a person that currently uses haskell and occasionally rust, I don't like that

then don't?

1:57 AM

Side effects are when a function changes the state

@quartata side effects?

```int a = 5;
int side_effects(int b){
a = b;
}```

oops

If I run a function twice will the output change

which languages don't have side effects?

Haskell doesn't

1:58 AM

does haskell have a random function?

(with the exception of IO)

@quartata i mean you can have it that way but you don't have to

haskell uses generators to deal with randomness

@quartata J has side effects, right?

@quartata concurrency? Node is designed for concurrency

2:00 AM

learnyouahaskell.com/input-and-output and scroll down to randomness

@ConorO'Brien Yes.

@quartata that's what I thought

@Downgoat I mean in Cheddar

@ConorO'Brien J at its core is a function level (not functional) language so functions are meant to be changed

@quartata ooo cool

2:12 AM

Trying to put an example for my own edification together. You have .filter right?

@quartata yes

e.g. [1,2,2,3,3,3].filter(is 2) == [2,2]

@Downgoat is 2?

@ASCII-only (is 2)(2) == true 1 is Number == true

is RHS is basically (==)&RHS

[2] is [2] == ?

@Downgoat TIL 1 == Number

@ASCII-only [2] is not a class....

@ASCII-only I meant unary is

2:17 AM

is x is n->n==x then

binary is checks instanceof

2:38 AM

@Downgoat getting array has no property filter on tio

tio is not updated

install cheddar

Jul 26 at 23:48, by Downgoat

npm install -g cheddar-lang

Gotta boot up the VM I guess

Should tell Dennis to pull :P

@quartata while not integral to a functional language, do you think forks & hooks are a good idea to have?

2:58 AM

Situational but useful to have is my guess

Dunno you're talking to the developer of a language with ~> and $

3:13 AM

?

which lang?

1:56 PM

@Downgoat complex number support?

2:14 PM

@Downgoat == has higher precedence than &

Is that intentional?

Makes this ugly:

let left = (0|>16).filter(n -> (n & 2) == 0);

6:05 PM

@quartata use !(n & 2)

6:20 PM

@quartata and (|>17)

^

speaking of that lemme get RHS bonding done

actually i gotta fix npm ASAP

How does (n->n)(n->n)(n->n)(10) get parsed?

@LeakyNun syntax error

@Downgoat function application...

(n->n)(n->n)?

@LeakyNun wat

6:25 PM

oh, right

(n->n->n->n)(n->n->n)(n->n)(10)

ohhh

it get's parsed as:

so it's already from left to right

because the first one assumes from right to left

(
  n->
    n->
      n->
        n
)(
  n->
    n->
      n
)(
  n->
    n
)(
  10
)

no, i was asking about left to right

??

gaot is confused ._.

6:27 PM

eh

never mind

7:22 PM

@Downgoat This isn't golf

In most languages & has higher precedence which is why I was asking

@Downgoat I could make complex lib, seems easy enough

cheddar> Math
< Instance of "Namespace" >
cheddar> Math.sin(0)
Syntax Error: Unexpected token at 1:0
1 | Math.sin(0)
  | ^
cheddar> Math.sine(0)
Runtime Error: Math has no property sine

@Downgoat what?