[Haskell-tut] Follow-up: Combining Functions

[wtd]

In my introduction to Haskell I createded a simple function greeting, which takes a name as a string and formulates a greeting.

code:

greeting :: String -> String greeting name = "Hello, " ++ name ++ "!"

And then, I created a greet function which takes a name and prints the greeting for it.

code:

greet :: String -> IO () greet name = putStrLn (greeting name)

The latter function I rewrote as:

code:

greet :: String -> IO () greet name = putStrLn $ greeting name

Now, the argument "name" should appear quite redundant in that last example. It is. What if, instead, we could simply combine the two functions, "greeting" and "putStrLn".

Haskell provides an easy mechanism for doing so.

code:

greet :: String -> IO () greet = putStrLn . greeting

Now, we can make a similar observation about the greeting function.

code:

greeting :: String -> String greeting name = "Hello, " ++ name ++ "!"

This is really two functions, since operators are just functions. Given the order of evaluation, this could be rewritten:

code:

greeting :: String -> String greeting name = "Hello, " ++ (name ++ "!")

Now, since we can partially apply functions - give them one argument, and get back a function which takes another argument and gives us the rest - we can rewrite this as:

code:

greeting :: String -> String greeting = ("Hello, " ++) . (++ "!")

Just as in the original, the function takes a string, appends "!" to it, then prepends "Hello, " to that.