Re: [RFC] Enumerations, Round 2

On 04/01/2021 22:04, Mike Schinkel wrote:

I don't quit get how you are thinking of with ": string"; can you give an example?

The current RFC requires the declaration of a Scalar Enum [1] to include the scalar type, so it looks like this: enum BookingStatus: string {

      case PENDING = "PENDING";
      case CONFIRMED = "CONFIRMED";
      case CANCELLED = "CANCELLED";

} The simplest opt-in for default scalar values would be to declare the type but no values, so the above could be abbreviated this way: enum BookingStatus: string {

      case PENDING;
      case CONFIRMED;
      case CANCELLED;

} However, it's probably better to be a bit more explicit, e.g.: enum BookingStatus: auto string {

      case PENDING;
      case CONFIRMED;
      case CANCELLED;

} Either way, we'd have to decide if it was allowed to specify some values but not all, e.g.: enum BookingStatus: auto string {

      case PENDING;
      case CONFIRMED;
      case CANCELLED = 'TERMINATED';

} A more verbose but possibly clearer syntax would be a token on the case statements themselves, such as "= auto", to mean "set the value for this case based on its constant name": enum BookingStatus: string {

      case PENDING = auto;
      case CONFIRMED = auto;
      case CANCELLED = 'TERMINATED';

} There's also the question of whether an automatic value for ints should also be allowed, e.g. incrementing in declaration order. This is more dangerous, as adding, removing, or re-arranging cases can affect the values of *all* the cases in the enum, but is done in some other languages. [1] https://wiki.php.net/rfc/enumerations#scalar_enums Regards, -- Rowan Tommins [IMSoP]