Course of Raku / Functional, concurrent, reactive, and web programming / Reactive programming / await

Awaiting a promise

await takes a promise, waits for it to complete, and returns its result:

my $p = start { 7 * 6 };
say await $p; # 42

The program pauses at await only as long as needed, then continues with the value the promise produced.

Given several promises, await waits for all of them and returns their results in the same order as the promises in the list:

my @jobs = (start { 5 }), (start { 10 }), (start { 15 });
say await @jobs;     # (5 10 15)
say [+] await @jobs; # 30

The jobs run concurrently, but await @jobs does not return until every one is done. Each result then sits at the same position as its promise in @jobs — no matter which job happened to finish first.

If an awaited promise was broken — its block threw an exception — await rethrows that exception at the point of the await. This means errors in background work surface where you wait for it, so you can handle them with the usual try / CATCH from the part on exceptions:

my $p = start { die 'no such file' };

my $result = try await $p;
say $result // "the job failed: $!";

This prints:

the job failed: no such file

The die happens on a background thread, but the exception travels with the promise and pops out at await. The try catches it there, leaving $result undefined and the error in $! — so a failure in concurrent work is handled exactly like any other exception. Awaiting is therefore not just about getting a value; it is the moment where concurrent work rejoins the main flow, results and errors alike.

Course navigation

← await   |   Quiz — await →

💪 Or jump directly to the exercises in this section.