@Doorknob So how would I say it?

what are you trying to say?

what language do you speak

speak as in "are fluent in"?

yes

or speak as in "are speaking in right now"

what language do you use < baupli

@QPaysTaxes me too :D

@KennyLau "what language do you use" is {ma bangu do}

@Doorknob ma bangu do

"what language are you fluent in" is something like {do certu tu'a ma poi bangu}

Cool, I figured out how to solve D-large (by testing my large solution on the small sample)

Even though D-small is actually really easy in itself, I figured I should do it the "proper" way.

@Doorknob (s/tu'a/tau/ if you're a proponent of cekitaujau)

@Doorknob What languages do you speak?

I only speak English fluently

but you also speak Lojban?

not really, but I'm learning still :P

I see

But, can you speak it?

as in, I can kind of hold a conversation like the one we just had without having to consult a dictionary, but I can't speak it comfortably

I meant with actual sound.

@PhiNotPi it's not like pronunciation is an issue :P

.i .ie mi kakne lo nu sance bacru

Someone crack mine?

I got a decent percentage of IMPOSSIBLEs for D-large.

I'm worried that I didn't do it right.

But, nothing I can do about it now.

spoiler were impossible for me.

unspoiled

I got a lot more impossible than that.

Are all the large testcases the same?

yes

afaik

Welps...

As long as I get above 30 points, I'm good, right?

yeah

anything past A and B is just fun :)

I just discovered the compose key function :D

oooh, I found an awesome way to golf D

spoiler

add 1 too

yeah, all of their 1-indexing XP

besides it doesn't work if spoiler

oh, you're right

:(

That is what I basically did though for mine :)

I just spoiler

Hi xsot. You are currently winning the rest of us at GCJ.

I messed up D-large

I'm too lazy to do C and D, when I could just be coding Minesweeper instead. :-P

@ChrisJester-Young What is your username for CGJ?

@TheNumberOne cky

Ok, thanks :)

@TheNumberOne Sure. I'm also cky on a number of other sites, including GitHub, GitLab, Medium, Wikipedia, etc. ;-)

how's code-bowling?

@Doorknob What is Rust's OO like? I seem to recall something about structs

@KennyLau bad don't post any challenges with it

@quartata lol

The problem with code bowling is that there's no way to stop people from arbitrarily padding their source code

(at least that we've discovered)

Might possibly work as some kind of atomic-code-bowling but it would have to be language specific then

finally beat D small!

@TheNumberOne I got 8 more than you, so either I missed some or the larges are different

out of curiosity, for those that have done D-large, is the 98th test case 89 7 44 ?

@NathanMerrill no

huh, so they are different

@NathanMerrill no

1 59 1 here

9 10 1 here

okay then

1 59 1 looks really easy D:

I know I missed D-large though.

well, I'm not going to try C

I don't like the brute force way

and I can't think of a better one

same here

You know it wants composite numbers right?

Not prime?

I know. Still, brute forcing, finding a single factor isn't fun

@NathanMerrill It's better to use a library for that kind of thing.

but I don't want to brute force

I'd prefer to find some generator of 01 strings

and, if that point, I need to find a factor, fine

but simply iterating through all of them is dumb

If the number takes too long to find a factor, just generate a new number and try again?

I'd saying that I'd prefer to find of sequence of 01 strings that always is factorable

basically, if you are doing itertools.product("01",n), then its bruteforcing it

It doesn't want you to find all jamcoins ... it just wants you to find J jamcoins.

you can break out of loops :)

oooh, I wish python had an inline yield

It has inline generators :)

do you need a def?

no

For example: (i for i in range(10))

oh, somebody else's generator

I was thinking something like:

while (len(arr) > n):
    for i in itertools.product("01",m)
        arr.yield("1"+i+"1")

not sure if arr.yield is the syntax, but you get the gist

@NathanMerrill I'm doing brute force at the moment because I'm lazy, but I'm doing it Monte Carlo.

Man, rvm sucks.

I'm using a login shell and it still says not a function.

@quartata I no understand.

Yeah well me neither.

Soo I'm just going to not bother with that today and concentrate on Racket a.k.a the one true language

(I wanted to update my version of Ruby because I have a super old version on here)

Hehehehehe.

@quartata All my Google Code Jam solutions are in Racket.

I'm currently running my Monte Carlo solution to C.

does anybody remember if gcj scores are posted immediately?

@NathanMerrill Yes for small. For large, they assume optimistically that you got it right, and it updates at the end of the contest.

I was meaning for large, and immediately after the contest is over

so I assume that's a yes

Yep.

@ChrisJester-Young Is there anything in Racket for overlapping slices? I could foldl cons over it, but I was wondering if there was a built-in so that I could do slices larger than 2

Actually wait.

folding wouldn't work.

I want to do [a,b,c,d] -> [[a,b],[b,c],[c,d]] basically

Folding cons would give [[[a,b],c],d]

Definitely not right

(define (fun-times lst)
  (for/list ([a (in-list lst)]
             [b (in-list (cdr lst))])
    (list a b)))

> (fun-times '(a b c d))
'((a b) (b c) (c d))

@quartata ^ like this?

in p6 we have .rotor but you don't want p6 I assume :p

@ChrisJester-Young Yeah, that works.

What does for/list do again?

@quartata It's a list comprehension.

> ('a','b','c','d').rotor(2 => -1)
((a b) (b c) (c d))

er

@ChrisJester-Young How is it different from for?

@LockOpeners Markdown doesn't work in multiline messages.

Hit fixed font instead

there

@quartata for does not collect results; you use it to do side-effecting stuff only. for/list collects the results into a list.

that's the p6 way

@ChrisJester-Young Oh, I see.