Better explanation of flip function

[a1300]

As a complete beginner I had a very hard time to understand the example for flip. I didn't saw any practical use in the example.

I found a great example and explanation on stackoverflow which opened my eyes on the use of flip

The licence for the stackoverflow post would be CC BY-SA 3.0 according to stackoverflow/licensing. Is this compatible with the current licence? Any thoughts on including the stackoverflow post?

[milesfrain]

Excellent feedback. Thanks.

I recall flip being pretty mind-melting for me when learning Elm the first time, but having paid those FP dues, I didn't catch this as being confusing in the PureScript book.

I think the solution is to fix the library docs and perhaps link to the Pursuit entry from the book. I'm not a fan of the existing Pursuit example, since const is pretty weird too. Opened purescript/purescript-prelude#237, which is up for grabs. Let me know if you want to tackle it, otherwise I can make the PR.

Also, great workaround for linking to lines in markdown. Annoying that isaacs/github#297 has been on the wish list for the past 6 years.

[milesfrain]

I think flip would make more sense to cover in Ch4 when discussing map.
It's also a good compliment to operator sections, which is not covered yet in the book. Not sure if the book should cover this syntax in detail, or if we can just link to the language reference.

Something like this could be a good sequence:

---

Let's say we want to perform truncating division by 5 on each element of an array of integers.

> arr = range 0 20
> arr
[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20]

This first attempt is not quite right:

> map (div 5) arr
[0,5,2,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]

We're mixing up the argument order and dividing 5 by each element of the array instead.
We can fix that by indicating that each array element needs to be the first argument to div:

> map (\n -> div n 5) arr
[0,0,0,0,0,1,1,1,1,1,2,2,2,2,2,3,3,3,3,3,4]

It's more concise with div's infix operator /:

> map (\n -> n / 5) arr
[0,0,0,0,0,1,1,1,1,1,2,2,2,2,2,3,3,3,3,3,4]

But most concise by using _ as an operator section:

> map (_ / 5) arr
[0,0,0,0,0,1,1,1,1,1,2,2,2,2,2,3,3,3,3,3,4]

flip is another way to swap the ordering of a function's arguments:

> map (flip div 5) arr
[0,0,0,0,0,1,1,1,1,1,2,2,2,2,2,3,3,3,3,3,4]

[danielrlc]

+1 on Chapter 3 suddenly getting very difficult when flip got introduced (rather arbitrarily?). Suddenly being confronted by this: forall a b c. (a -> b -> c) -> b -> a -> c was tough! Some kind of encouraging comment like: "Don't worry about understanding all of this code right now" might help there? (But only if you really DON'T need to understand it all at that early stage, of course!)

[JordanMartinez]

Would this orientation help?

flip :: forall firstArg secondArg returnType
   . (firstArg -> secondArg -> returnType)
  -> secondArg
  -> firstArg
  -> returnType
flip functionABToC b a = functionABToC a b  

append "hello" "world" == "helloworld"

flip append "hello" "world" == "worldhello"

-- flip's type signature can be simplified
flip :: forall a b c.
   . (a -> b -> c)
  -> b
  -> a
  -> c
flip f b a = f a b  

[danielrlc]

I like the first two lines or so:

flip :: forall firstArg secondArg returnType
  . (firstArg -> secondArg -> returnType)

... but then I'm getting lost in the rest. For me at least, flip feels too complicated to introduce so early in the book. Maybe something simpler like a sum function that just added two numbers could work better?

As a PureScript beginner coming from JavaScript, I'd like to spend a bit more time getting used to the syntax this early on, as it's pretty different to JavaScript. Getting the hang of the functional nature of PureScript could maybe come a bit later? Just me thinking out loud though... :)