Julia

« first day (1 day earlier) ← previous day next day → last day (92 days later) »

Dennis

2:49 AM

@AlexA. Is there any way to vectorize character subtraction?

Alex A.

Like subtracting the code points?

Dennis

Yes. 'a'-'b' works, but ['a']-'b' doesn't.

Alex A.

Try .-?

Dennis

I did.

ERROR: MethodError: `-` has no method matching -(::Array{Char,1}, ::Char)

The worst part is that I'm certain it worked in a previous version.

Alex A.

Yeah, they got stricter with how Chars work in more recent versions

map(i->i-'b',['a']) works

Dennis

2:53 AM

Yeah, I was kinda trying to avoid the map... :P

Hm, I should be able to use digits instead.

Alex A.

Oh were you doing a ["$n"...] kind of deal?

Just note that digits returns them in reverse order

Dennis

No, the challenge allows taking a string or a number.

Ah okay

It worked! \o/

Nice!

2:57 AM

A: A Numpad's Knight Numbers

Julia, 51 bytes n->diff(["_1238_4765"[digits(n)+1]...]).^2⊆[9,25] Verification julia> f=n->diff(["_1238_4765"[digits(n)+1]...]).^2⊆[9,25] (anonymous function) julia> map(f, (1, 2, 3, 4, 5, 6, 7, 8, 9, 16, 18, 38, 61, 81, 294, 349, 381, 383, 729, 767, 38183, 38383, 18349276, 183492761, 618349...

Alex A.

Once I go back to being less busy with things I'll start ninja'ing the Julia answers :P

Dennis

There's absolutely no rush. :D

Alex A.

ಠ_ಠ

My mistake was trying to force you to use Julia however long ago...

Dennis

If you wanted to keep her for yourself, yes.

Alex A.

Haha

Dennis

3:04 AM

I'm still baffled by the inconsistent behavior for character subtraction. Even diff takes characters...

Alex A.

diff should be calling - behind the scenes, so as long as -(::Char, ::Char) is defined, you're fine

Dennis

You can't deny that's inconsistent though.

I mean, you can. But you shouldn't.

Alex A.

Idk, anything that uses - should have access to all of the methods defined for -. ¯\_(ツ)_/¯

Dennis

Precisely. Well, it would have saved only one byte, so I'll stop complaining now.

Alex A.

I guarantee the Julia devs don't give a hoot about your precious bytes. :P

Dennis

3:14 AM

They should give one about consistency though.

Alex A.

I wonder if I can stalk them, get their contact info, and convince them to join PPCG...

Dennis

Also, I very much preferred it when character arithmetic worked. If 'a' - 'b' is defined, there's simply no good reason not to define (let alone, deprecate) 'a' $ 'b'.

Alex A.

I wouldn't be all that surprised if 'a' - 'b' goes away in a future version. I think I read a thread of the devs discussing removing it.

Dennis

17 mins ago, by Alex A.

ಠ_ಠ

I should probably install 0.4 and 0.3 as well, and choose the golfiest one for each task.

Alex A.

59 secs ago, by Dennis

17 mins ago, by Alex A.

ಠ_ಠ

Dennis

3:19 AM

I do it with Python. Everyone does!

Alex A.

That's because there's no current consensus on what constitutes "Python": 2 or 3. For comparison, note that the same is never done for Ruby, for example.

Dennis

We do it with C as well.

Most answers are in C89, since it's the golfiest.

Alex A.

Oh really? I don't recall having seen C versioned in a submission.

Dennis

That's because everyone uses C89, so we just call it C. :P

Alex A.

._.

Dennis

3:39 AM

Why does Julia have iseven and isodd functions? o_O