B.6 Lists
The following are the some important points about lists in ML.
Unlike in Scheme, lists in ML are homogeneous, meaning all elements of the list must be of the same type. For instance, the list [1,2,3] in ML is homogeneous, while the list (1 "apple") in Scheme is heterogeneous.
In a type-safe language like ML the values in a tuple (Section B.7) generally have different types, but the number of elements in the tuple must be fixed. Conversely, the values of a list must all have the same type, but the number of elements in the list is not fixed.
The semantics of the lexemes nil and [] are the empty list.
The cons operator, which accepts an element (the head) and a list (the tail), is :: (e.g., 1::2::[3]) and associates right-to-left.
The expression x::xs represents a list of at least one element.
The expression xs is pronounced exes.
The expression x::nil represents a list of exactly one element and is the same as [x].
The expression x::y::xs represents a list of at least two elements.
The expression x::y::nil represents a list of exactly two elements.
The built-in functions hd (for head) and tl (for tail) are the ML analogs of the Scheme functions car and cdr, respectively.
The built-in function length returns the number of elements in its only list argument.
The append operator (@) accepts two lists and appends them to each other. For example, [1, 2] @ [3, 4, 5] returns [1, 2, 3, 4, 5]. The append operator in ML is also inefficient, just as it is in Scheme.
Examples:
