[swift-evolution] Class and Subclass Existentials (Round 2)

Adrian Zubarev adrian.zubarev at devandartist.com
Mon Feb 6 15:57:59 CST 2017


Looks really promising to me. :)

Aside the class existentials, I still miss generalized struct and enum constraints (this is out of scope of this proposal, and also not meant to be a request). Sometimes I really want to tell someone who’s using my code base “no you’re not allowed to create a class type here, because the way you’re doing it is ugly, as if Objective-C, Java and C# had a child called quasimodo”. By default such constraints would not enforce value semantics, but at least they’ll do some good to some codebase.

Speaking of existentials again, struct & ProtocolFoo & ProtocolBar would be the same as if we had protocol FooBar : struct, ProtocolFoo, ProtocolBar {}. Such an existential/protocol will reject any class instance.

Alternatively there could exist AnyValue which would cover any value type including tuples. (As a sidenote: IMHO AnyValue should not conform to something like ValueSemantics by default, because there might be some point where you don’t want value semantics to apply.)



-- 
Adrian Zubarev
Sent with Airmail

Am 6. Februar 2017 um 22:18:19, David Hart via swift-evolution (swift-evolution at swift.org) schrieb:

Hello mailing-list,

I rewrote the draft proposal concerning the class and subclass existentials. Please let me know what you think, especially concerning the class and AnyObject conundrum.

Regards,
David.

https://github.com/hartbit/swift-evolution/blob/e6411d8a9e7924bbd8a48fc292bf08d58a8d1199/proposals/XXXX-subclass-existentials.md

Class and Subtype existentials
Proposal: SE-XXXX
Authors: David Hart, Austin Zheng
Review Manager: TBD
Status: TBD
Introduction

This proposal brings more expressive power to the type system by allowing Swift to represent existentials of classes and subtypes which conform to protocols.

Motivation

Currently, the only existentials which can be represented in Swift are conformances to a set of protocols, using the &protocol composition syntax:

Protocol1 & Protocol2
On the other hand, Objective-C is capable of expressing existentials of classes and subclasses conforming to protocols with the following syntax:

id<Protocol1, Protocol2>
Base<Protocol>*
We propose to provide similar expressive power to Swift, which will also improve the bridging of those types from Objective-C.

Proposed solution

The proposal keeps the existing & syntax but allows the first element, and only the first, to be either the class keyword or of class type. The equivalent to the above Objective-C types would look like this:

class & Protocol1 & Protocol2
Base & Protocol
As in Objective-C, the first line is an existential of classes which conform to Protocol1 and Protocol2, and the second line is an existential of subtypes of Base which conform to Protocol.

Here are the new proposed rules for what is valid in a existential conjunction syntax:

1. The first element in the protocol composition syntax can be the class keyword to enforce a class constraint:

protocol P {}
struct S : P {}
class C : P {}
let t: P & class // Compiler error: class requirement must be in first position
let u: class & P = S() // Compiler error: S is not of class type
let v: class & P = C() // Compiles successfully
2. The first element in the protocol composition syntax can be a class type to enforce the existential to be a subtype of the class:

protocol P {}
struct S {}
class C {}
class D : P {}
class E : C, P {}
let t: P & C // Compiler error: subclass contraint must be in first position
let u: S & P // Compiler error: S is not of class type
let v: C & P = D() // Compiler error: D is not a subtype of C
let w: C & P = E() // Compiles successfully
3. When a protocol composition type contains a typealias, the validity of the type is determined using the following steps:

Expand the typealias
Normalize the type by removing duplicate constraints and replacing less specific constraints by more specific constraints (a class constraint is less specific than a class type constraint, which is less specific than a constraint of a subclass of that class).
Check that the type does not contain two class-type constraints
class C {}
class D : C {}
class E {}
protocol P1 {}
protocol P2 {}
typealias TA1 = class & P1
typealias TA2 = class & P2
typealias TA3 = C & P2
typealias TA4 = D & P2
typealias TA5 = E & P2

typealias TA5 = TA1 & TA2
typealias TA5 = class & P1 & class & P2 // Expansion
typealias TA5 = class & P1 & P2 // Normalization
// TA5 is valid

typealias TA6 = TA1 & TA3
typealias TA6 = class & P1 & C & P2 // Expansion
typealias TA6 = C & P1 & P2 // Normalization (class < C)
// TA6 is valid

typealias TA7 = TA3 & TA4
typealias TA7 = C & P2 & D & P2 // Expansion
typealias TA7 = D & P2 // Normalization (C < D)
// TA7 is valid

typealias TA8 = TA4 & TA5
typealias TA8 = D & P2 & E & P2 // Expansion
typealias TA8 = D & E & P2 // Normalization
// TA8 is invalid because the D and E constraints are incompatible
class and AnyObject

This proposal merges the concepts of class and AnyObject, which now have the same meaning: they represent an existential for classes. They are four solutions to this dilemna:

Do nothing.
Replace all uses of AnyObject by class, breaking source compatibility.
Replace all uses of class by AnyObject, breaking source compatibility.
Redefine AnyObject as typealias AnyObject = class.
Source compatibility

Leaving aside what is decided concerning class and AnyObject, this change will not break Swift 3 compability mode because Objective-C types will continue to be imported as before. But in Swift 4 mode, all types bridged from Objective-C which use the equivalent Objective-C existential syntax could break code which does not meet the new protocol requirements. For example, the following Objective-C code:

@interface MyViewController
- (void)setup:(nonnull UIViewController<UITableViewDataSource,UITableViewDelegate>*)tableViewController;
@end
is imported into Swift-3 mode as:

class MyViewController {
    func setup(tableViewController: UIViewController) {}
}
which allows calling the function with an invalid parameter:

let myViewController: MyViewController()
myViewController.setup(UIViewController())
The previous code continues to compile but still crashs if the Objective-C code calls a method of UITableViewDataSource or UITableViewDelegate. But if this proposal is accepted and implemented as-is, the Objective-C code will be imported in Swift 4 mode as:

class MyViewController {
    func setup(tableViewController: UIViewController & UITableViewDataSource & UITableViewDelegate) {}
}
That would then cause the Swift code run in version 4 mode to fail to compile with an error which states that UIViewController does not conform to the UITableViewDataSource and UITableViewDelegate protocols.

Alternatives considered

None.

Acknowledgements

Thanks to Austin Zheng and Matthew Johnson who brought a lot of attention to existentials in this mailing-list and from whom most of the ideas in the proposal come from.
_______________________________________________
swift-evolution mailing list
swift-evolution at swift.org
https://lists.swift.org/mailman/listinfo/swift-evolution
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.swift.org/pipermail/swift-evolution/attachments/20170206/0f16bd86/attachment.html>


More information about the swift-evolution mailing list