I thought we liked our monadic functions here.

And this is a byte shorter.

(taking the gcd with 9 instead of 3)

@BMO why is it necessary?

@Laikoni: Do you mean the space?

(If that's the case, it is because otherwise the length would be divisible by 3 and the gcd would always be 3 as well.)

@Nitrodon: Well caught! Not sure how the two of us missed that -.-

Yes, I was referring to the space and thought there might be a weird syntax rule which enforces it front of the comment.

damn Data.Char doesn't have a swap case function

@totallyhuman More than half of my code is implementing one. :(

91 bytes btw

did it on my own and saw that you had exactly the same approach

Not exactly the same. You are using a list comp instead of a map

You can golf yours I'm pretty sure

the code might not be the same, but the algorithm is :P

quick 90 :P

@Laikoni: Luckily no :) Thinking about it now, we should have provided some context..

Yeah, that was my thinking

if only there was a pointfree ternary that's short

if'?

I believe that requires imports though

yeah and that's not short :(

89, you made a pretty simple errror

oh right

What kind of chars are allowed in the string?

damn spaces..

All types

I had a similar answer earlier

That's annoying :(

@WheatWizard that was a really clever answer:)

Which one?

Another approach for a case swap function: f x=last$x:do(a,b)<-zip['a'..'z']['A'..];[b|x==a]++[a|x==b]

Though it's 97 bytes for the challenge: Try it online!

@WheatWizard the euler poincaré

Oh. It just seemed like the way to do it.

It never occured to me to just count the the vertices in the triangulation and divide by 2:)