Course of Raku / Functional, concurrent, reactive, and web programming / Functional programming / Iterators

The iterator protocol

Every value you can loop over — a list, a range, an array — can hand you an iterator with the .iterator method. An iterator has one essential method, pull-one, which returns the next value each time you call it:

my $it = (10, 20, 30).iterator;
say $it.pull-one; # 10
say $it.pull-one; # 20
say $it.pull-one; # 30

When the values run out, pull-one returns a special sentinel, IterationEnd, instead of a real element. Asking the same $it for one more value shows it:

say $it.pull-one =:= IterationEnd; # True

The =:= operator tests for the same object, which is how you recognise the sentinel. That detail matters inside a loop, where you need to keep the pulled value and test it. If you assign it with =, the value lands in a container and the comparison ends up looking at the container rather than at IterationEnd. The remedy is to bind with :=, so the variable simply is whatever pull-one returned:

my $it = <a b c>.iterator;
loop {
    my $v := $it.pull-one;      # bind, not assign
    last if $v =:= IterationEnd;
    say $v;
}

The loop prints a, b, c and then stops. This is exactly what a for loop does for you under the hood: it calls .iterator on the thing you are looping over and keeps calling pull-one until it meets IterationEnd. You rarely need to write it out by hand, but doing so once shows that iterators — not lists — are what Raku actually loops over.

Course navigation

← Iterators   |   Quiz — The iterator protocol →

💪 Or jump directly to the exercise to this section.