How is a type-erased generic wrapper implemented?

Question

I need to implement a type-erasing wrapper for my own structure, very similar to SequenceOf, GeneratorOf, etc. So I started by trying to just re-implement the standard SequenceOf myself.

I just copied & pasteed the the declaration for SequenceOf, renamed it to MySequenceOf, and filled in some stubs to get:

/// A type-erased sequence.
///
/// Forwards operations to an arbitrary underlying sequence with the
/// same `Element` type, hiding the specifics of the underlying
/// sequence type.
///
/// See also: `GeneratorOf<T>`.
struct MySequenceOf<T> : SequenceType {

    /// Construct an instance whose `generate()` method forwards to
    /// `makeUnderlyingGenerator`
    init<G : GeneratorType where T == T>(_ makeUnderlyingGenerator: () -> G) {
        fatalError("implement me")
    }

    /// Construct an instance whose `generate()` method forwards to
    /// that of `base`.
    init<S : SequenceType where T == T>(_ base: S) {
        fatalError("implement me")
    }

    /// Return a *generator* over the elements of this *sequence*.
    ///
    /// Complexity: O(1)
    func generate() -> GeneratorOf<T> {
        fatalError("implement me")
    }
}

I get the compiler error: "Neither type in same-type refers to a generic parameter or associated type". So I assume that the Xcode-generated declaration of SequenceOf's "where T == T" constraint really means "where G.Element == T", which gives me the following compilable struct:

struct MySequenceOf<T> : SequenceType {

    init<G : GeneratorType where G.Element == T>(_ makeUnderlyingGenerator: () -> G) {
        fatalError("implement me")
    }

    func generate() -> GeneratorOf<T> {
        fatalError("implement me")
    }
}

So now, easy enough, I just need to hang on to makeUnderlyingGenerator from the initializer and invoke it from generate():

struct MySequenceOf<T> : SequenceType {
    let maker: ()->GeneratorOf<T>

    init<G : GeneratorType where G.Element == T>(_ makeUnderlyingGenerator: () -> G) {
        self.maker = { return makeUnderlyingGenerator() }
    }

    func generate() -> GeneratorOf<T> {
        return self.maker()
    }
}

but that gives me the error: "'G' is not convertible to 'GeneratorOf'"

It does compiles if I force a cast:

struct MySequenceOf<T> : SequenceType {
    let maker: ()->GeneratorOf<T>

    init<G : GeneratorType where G.Element == T>(_ makeUnderlyingGenerator: () -> G) {
        self.maker = { return makeUnderlyingGenerator() as GeneratorOf<T> }
    }

    func generate() -> GeneratorOf<T> {
        return self.maker()
    }
}

But then it crashes at runtime from the dynamic cast.

So how can type-erasure like this be implemented? It must be possible, because the Swift standard library does it a bunch (SequenceOf, GeneratorOf, SinkOf).

Answer 1

Try:

struct MySequenceOf<T> : SequenceType {
    private let _generate:() -> MyGeneratorOf<T>

    init<G : GeneratorType where G.Element == T>(_ makeUnderlyingGenerator: () -> G) {
        _generate = { MyGeneratorOf(makeUnderlyingGenerator()) }
    }

    init<S : SequenceType where S.Generator.Element == T>(_ base: S) {
        _generate = { MyGeneratorOf(base.generate()) }
    }

    func generate() -> MyGeneratorOf<T> {
        return _generate()
    }
}

struct MyGeneratorOf<T> : GeneratorType, SequenceType {

    private let _next:() -> T?

    init(_ nextElement: () -> T?) {
        _next = nextElement
    }

    init<G : GeneratorType where G.Element == T>(var _ base: G) {
        _next = { base.next() }
    }

    mutating func next() -> T? {
        return _next()
    }

    func generate() -> MyGeneratorOf<T> {
        return self
    }
}

The basic strategy of implementing ProtocolOf<T> is, like this:

protocol ProtocolType {
    typealias Value
    func methodA() -> Value
    func methodB(arg:Value) -> Bool
}

struct ProtocolOf<T>:ProtocolType {
    private let _methodA: () -> T
    private let _methodB: (T) -> Bool

    init<B:ProtocolType where B.Value == T>(_ base:B) {
        _methodA = { base.methodA() }
        _methodB = { base.methodB($0) }
    }

    func methodA() -> T { return _methodA() }
    func methodB(arg:T) -> Bool { return _methodB(arg) }
}

---

Added to answering @MartinR in comment.

Is there a special reason that _generate is a closure and not the generator itself?

First of all, I think, It's a matter of specification or semantics.

Needless to say, the difference is "when to create the generator".

Consider this code:

class Foo:SequenceType {
    var vals:[Int] = [1,2,3]
    func generate() -> Array<Int>.Generator {
        return vals.generate()
    }
}

let foo = Foo()
let seq = MySequenceOf(foo)
foo.vals = [4,5,6]
let result = Array(seq)

The problem is: result should be [1,2,3] or [4,5,6]? My MySequenceOf and built-in SequenceOf results the latter. I just matched the behaviors with built-in one.