== is often used to check if two errors equal in golang, but in some cases, errors.Is is needed for checking if an error is wraped by another error, in which case == returns false, but errors.Is returns true.
Let's look at the below code,
var ErrNotFound = errors.New("not found")
func main() {
err := someFunction()
if err != ErrNotFound {
fmt.Println("err != ErrNotFound")
}
if errors.Is(err, ErrNotFound) {
fmt.Println("The error is 'not found'")
} else {
fmt.Println("The error is not 'not found':", err)
}
err = someFunction1()
if errors.Is(err, ErrNotFound) {
fmt.Println("err is ErrNotFound")
}
}
func someFunction() error {
return fmt.Errorf("this function failed: %w", ErrNotFound)
}
func someFunction1() error {
return ErrNotFound
}
the output,
err != ErrNotFound
The error is 'not found'
err is ErrNotFound
ErrNotFound is the base error, someFunction returns an error which wraps ErrNotFound.
The err!=ErrNotFound returns true, but errors.Is(err, ErrNotFound) returns true because err wraps ErrNotFound. So if you want to check if err is another error, errors.Is should be used.
