[swift-users] Protocol with instance var that's set on construction, otherwise read-only
David Sweeris
davesweeris at mac.com
Tue Aug 2 20:48:50 CDT 2016
Oh, it's a computed property! Got it, I thought you meant its value was computed in the init and never changed again.
Sent from my iPhone
> On Aug 2, 2016, at 19:01, Rick Mann <rmann at latencyzero.com> wrote:
>
> It complains if I make it a let because computed properties must be var. Because it's a protocol, it can't be stored (even though it can be stored in the conforming type).
>
> If I make it { get }, I can't set it in the extensions init() method.
>
> I guess I could make it private set (not sure of the syntax for that), but it still doesn't have let semantics.
>
>> On Aug 2, 2016, at 16:28 , David Sweeris <davesweeris at mac.com> wrote:
>>
>> If I understand things correctly, you *can* make uuid a let because you’re allowed to set them (once) during init functions.
>>
>> - Dave Sweeris
>>
>>> On Aug 2, 2016, at 6:22 PM, Rick Mann via swift-users <swift-users at swift.org> wrote:
>>>
>>> I'm trying to define a protocol that has a read-only, immutable member "uuid" that can be set in the init() method, but I'm having trouble. I have this:
>>>
>>> protocol
>>> Element
>>> {
>>> var uuid : { get }
>>> }
>>>
>>> extension
>>> Element
>>> {
>>> init(...)
>>> {
>>> self.uuid = ...
>>> }
>>> }
>>>
>>> I can't make it let, because they're computed.
>>>
>>> I'm realizing from other stuff that I really can't have the init(...) method in the extension, anyway. But I'd really like to be able to specify a let member in the protocol. What's the best way to have that effect?
>>>
>>> In my semantics, an Element has a uniquely-assigned uuid. It might be generated when the object is instantiated, or it might be deserialized from disk, but once that's done, it can never change. How do I express that?
>>>
>>> Thanks,
>>>
>>> --
>>> Rick Mann
>>> rmann at latencyzero.com
>>>
>>>
>>> _______________________________________________
>>> swift-users mailing list
>>> swift-users at swift.org
>>> https://lists.swift.org/mailman/listinfo/swift-users
>
>
> --
> Rick Mann
> rmann at latencyzero.com
>
>
More information about the swift-users
mailing list