It's nice that Go has a Duration type -- having explicitly defined units can prevent real-world problems.
And because of Go's strict type rules, you can't multiply a Duration by an integer -- you must use a cast in order to multiply common types.
/*
MultiplyDuration Hide semantically invalid duration math behind a function
*/
func MultiplyDuration(factor int64, d time.Duration) time.Duration {
return time.Duration(factor) * d // method 1 -- multiply in 'Duration'
// return time.Duration(factor * int64(d)) // method 2 -- multiply in 'int64'
}
The official documentation demonstrates using method #1:
To convert an integer number of units to a Duration, multiply:
seconds := 10
fmt.Print(time.Duration(seconds)*time.Second) // prints 10s
But, of course, multiplying a duration by a duration should not produce a duration -- that's nonsensical on the face of it. Case in point, 5 milliseconds times 5 milliseconds produces 6h56m40s. Attempting to square 5 seconds results in an overflow (and won't even compile if done with constants).
By the way, the int64 representation of Duration in nanoseconds "limits the largest representable duration to approximately 290 years", and this indicates that Duration, like int64, is treated as a signed value: (1<<(64-1))/(1e9*60*60*24*365.25) ~= 292, and that's exactly how it is implemented:
// A Duration represents the elapsed time between two instants
// as an int64 nanosecond count. The representation limits the
// largest representable duration to approximately 290 years.
type Duration int64
So, because we know that the underlying representation of Duration is an int64, performing the cast between int64 and Duration is a sensible NO-OP -- required only to satisfy language rules about mixing types, and it has no effect on the subsequent multiplication operation.
If you don't like the the casting for reasons of purity, bury it in a function call as I have shown above.