Welcome to part two in my series of Ocaml tutorials! There was an overwhelming response to the first tutorial in the series and I have taken plenty of suggestions on board for this and future tutorials. In this instalment I’ll be introducing the list data type and the very useful List.iter function from the Ocaml standard library.
An Introduction to List Initialization and List.iter
Here’s the source code we’re going to be breaking down:
let main () = let seq = ["a"; "b"; "c"] in List.iter print_endline seq ;; let _ = main ()
Looks scary, doesn’t it? It’s actually quite simple, but we’ll get to the source code in a moment. First, let’s see the program’s output. Save this file as tmhoo02.ml and run it like so:
$ ocaml tmhoo02.ml
You should see the following:
a b c
Now, let’s take a look at the interesting bits of the source code:
let seq = ["a"; "b"; "c"] in
As you (hopefully!) learned in part 1, the let syntax binds the variable seq to some expression. The in keyword indicates that this binding is a named local expression whose value to be discarded when the function exits (think “local variable”). But what exactly does that expression mean?
["a"; "b"; "c"]
This is the most direct way to initialize a list in Ocaml. Each element in a list is separated by a semi-colon. This particular list has three elements: the strings “a”, “b” and “c”. As our list comprises of string elements, we can say that this is a string list. Note that unlike some dynamically typed languages (e.g. Python and Ruby), every Ocaml list element must be of the same type: you have a list of string elements or a list of integer elements, but you simply can’t mix the two. Honestly, it won’t compile.
List.iter print_endline seq ;;
Here we’re calling the iter function of the List module – a module which is bundled as part of the Ocaml standard library. This function takes two parameters: the first is another function that accepts a single parameter, the second is a list.
As you might suspect, List.iter iterates over the list, passing each element in the list to the function one by one. To word it another way, inside List.iter we’re calling print_endline for each element in our list to the equivalent of: print_endline “a”, print_endline “b”, print_endline “c”. This explains the output of our example program.
let _ = main ()
main is evaluated here and its return value (the unit value) is ignored by the underscore wildcard. You’ll notice in part 1 I omitted everything up to the equals operator but, at the behest of others who know better, I’ve included it here in part two. I’m not going to say anything more about it, other than the fact that the underscore wildcard is a simple but powerful construct you’ll likely become very familiar with at a later date. If you’d prefer to use the syntax from part 1, that should work too (but don’t come crawling to me when other Ocaml programmers come down on you like a ton of bricks ;) ).
Advanced Discussion of the Type System
As a sneak preview into Ocaml’s type system, I can tell you that List.iter isn’t restricted to string lists. In fact, it can work with lists containing elements of any type. However, because each element of the list is passed to the function given to List.iter as the first parameter, that function must accept a parameter the same type as a single element in the list. Since print_endline accepts a string, we must use a string list. Or: since we’re passing in a string list as the second parameter to List.iter, the function passed as the first parameter must itself accept a string as its sole parameter.
Confused? Read that last paragraph again carefully. If you still don’t understand it, don’t worry too much: we’ll cover Ocaml’s type system more in the next few tutorials.
**Until Next Time …
So that wraps up tutorial number two. I’ve already got parts three and four in the pipeline and I can tell you they introduce a fair bit of new material. Take your time trying to understand what was covered in this tutorial and post a comment or two if you have any suggestions, questions or improvements. :)