Pattern Matching

  factorial :: Int -> Int
  factorial 1 = 1
  factorial n = n * factorial(n-1)

All cases must be covered, using a placeholder at the end of the function definition will cover all other possible cases. Cases are checked from top to bottom of function definition, so the wildcard entry should be at the bottom.

  charName :: Char -> String
  charName 'a' = "Albert"
  charName 'b' = "Billy"
  -- x acts as a universal matcher and will match anything
  charName x = "Sorry, I don't know what that is"

This wildcard matching pattern works very well as you can define functions very cleanly:

  addVectors :: (Double, Double) -> (Double, Double) -> (Double, Double)
  addVectors (x1,y1) (x2,y2) = (x1+x2, y1+y2)

_ represents a generic variable in function definitions, something that we don’t care about.

  head' :: [a] -> a
  head' [] = error "Can't call head' on an empty list"
  head' ( x: _ ) = x -- there should be no spaces between the x and underscore, this is to fix md rendering

++ is used for concatenation of lists

As Patterns

As patterns allow you to refence a whole list easily when you’ve broken it down into constituent elements in a function definition. Syntax: name@(list), name is used to refer to the list, and list is usually written like x:y:xs, allowing easy access to the first and second items of the list.

  firstLetter :: String -> String
  firstLetter "" = "Empty string, whoops!" -- Match any empty strings
  firstLetter all@(x:xs) = "The first letter of " ++ all ++ " is " ++ [x]

Guards

Guards are used for checking whether parameters satsify some predicate, they’re like having a whole load of if and else statements, just must much prettier, | marks the start of a guard, \n ends it:

  bmiTell :: Double -> String
  bmiTell bmi
    | bmi <= 18.5 = "You're underweight"
    | bmi >= 30.0 = "You're fat"
    | otherwise = "You're intermediate"

Here otherwise is used (It’s a wildcard matcher) and will handle anything that doesn’t match the predicates above it. Guards are evaluated like patterns, from top to bottom.

Storing intermediate results

Where

In python you’d calc something once like: x = x * 10.0/13.0, then use x for all our further calculations. Haskell doesn’t have variable, only functions, or only immutable variables, depending on how you want to think of it.

  bmiTell :: Double -> Double -> String
  bmiTell weight height
    | bmi <= skinny = "You're underweight, you emo"
    | bmi <= normal = "You're supposedly normal"
    | bmi <= fat = "You're fat!"
    | otherwise = "Blimey, surgery needed ASAP"
  where bmi = weight/height^2
        skinny = 18.5
        normal = 25.0
        fat = 30.0

It’s very simple really, don’t overcomplicate it in your head. Just make sure all vars are located in same column like above. The vars lexical scope is the function.

Let expressions

let expressions are very local, they’re pretty much self-contained: let <bindings> in <expression>, they can be used anywhere other expressions can be used. The use of where just binds values to names whereas let actually is an expression itself.

  calcBmis :: [(Double, Double)] -> [Double]
  calcBmis xs = [ bmi | (w,h) <- xs, let bmi = w / h^2, bmi > 25.0]

Case expressions

case expressions follow the pattern:

  case express of pattern -> result
                  pattern -> result
                  pattern -> result

A more concrete example:

  describeList :: [a] -> String
  describeList ls = "The list is " ++ case ls of [] -> "Empty."
                                                 [x] -> "Singleton list"
                                                 xs -> "Long list."