[swift-evolution] Default Generic Arguments

Tue Jan 24 13:41:23 CST 2017

It’s worth noting that the question of “how do these defaults interact with other defaults” is an issue that has left this feature dead in the water in the Rust language despite being accepted for inclusion two years ago. See https://internals.rust-lang.org/t/interaction-of-user-defined-and-integral-fallbacks-with-inference/2496 <https://internals.rust-lang.org/t/interaction-of-user-defined-and-integral-fallbacks-with-inference/2496> for some discussion of the issues at hand.

For those who don’t want to click that link, or are having trouble translating the syntax/terms to Swift. The heart of Niko’s post is the following (note: functions are used here for expedience; you can imagine these are `inits` for a generic type if you wish):

// Example 1: user supplied default is IntegerLiteralConvertible

func foo<T=Int64>(t: T) { ... }

foo<_>(22)
//  ^
//  |
//  What type gets inferred here?

// Example 2: user supplied default isn't IntegerLiteralConvertible

func bar<T=Character>(t: T) { ... }

bar<_>(22)
//  ^
//  |
//  What type gets inferred here?

There are 4 strategies:

(Note: I use “integer literal” here for simplicity; in general it's “any kind of literal, and its associated LiteralType”. So this reasoning also applies to FloatLiteralType, StringLiteralType, BooleanLiteralType, etc.)

* Unify all: always unify the variables with all defaults. This is the conservative choice in that it gives an error if there is any doubt.

* Prefer literal: always prefer IntegerLiteralType (Int). This is the maximally backwards compatible choice, but I think it leads to very surprising outcomes.

* Prefer user: always the user-defined choice. This is simple from one point of view, but does lead to a potentially counterintuitive result for example 2.

* Do What I Mean (DWIM): Prefer the user-defined default, except in the case where the variable is unified with an integer literal and the user-defined default isn't IntegerLiteralConvertible. This is complex to say but leads to sensible results on both examples. (Basically: prefer user, but fallback to IntegerLiteralType if the user default doesn’t actually make sense)

| Strategy       | Example 1 | Example 2 |
| -------------- | --------- | --------- |
| Unify all      | Error     | Error     |
| Prefer literal | Int       | Int       |
| Prefer user    | Int64     | Error     |
| DWIM           | Int64     | Int       |

Personally, I’ve always favoured DWIM. Especially in Swift where IntegerLiteralType inference is so frequently used (you don’t want adding a default to cause code to stop compiling!). In practice I don’t expect there to be many cases where this ambiguity actually kicks in, as it requires the user-specified default to be a LiteralConvertible type that isn't the relevant LiteralType, and for the type variable to affect an actual Literal. So <T=String>(x: T) never causes problems, but <T=StaticString>(x: T) does.

As for the matter of “what if I want the other one” — you clearly know the actual type you want; just name it explicitly.

> On Jan 24, 2017, at 2:59 AM, Srđan Rašić via swift-evolution <swift-evolution at swift.org> wrote:
> 
> > If the answer to the above question is "yes, T is inferred as Int" then we need some way to express "give me the default for T, which is Float."
> 
> I don't think that we need that. It would introduce a new level of explicitness, "I want the default, but I don't care what the default is", that is not really useful. If you don't care what the default type is, you probably also don't care that you are defaulting. If you do care what the default type is, you would explicitly sepecify it as `X<Float>`. 
> 
> 
> > If the answer to the above question is "no" then we need some way to express "don't give me the default; rather, infer type T from the right hand side."
> 
> That would be preferred behavior. Infer from the context if possible, use default otherwise. 
> 
> 
> tir. 24. jan. 2017 kl. 05.11 skrev Xiaodi Wu <xiaodi.wu at gmail.com <mailto:xiaodi.wu at gmail.com>>:
> While it looks nicer without the angle brackets, that suggestion is unresponsive to David's point that we need some way to distinguish defaulted generic arguments from inferred generic arguments.
> 
> Consider:
> ```
> let a: Optional = 1 // Optional<Int>
> 
> enum FloatPreferringOptional<T = Float> {
>   case some(T)
>   case none
> }
> 
> let b: FloatPreferringOptional = 1
> // Does this give you an FloatPreferringOptional<Int>?
> ```
> 
> If the answer to the above question is "yes, T is inferred as Int" then we need some way to express "give me the default for T, which is Float." If the answer to the above question is "no" then we need some way to express "don't give me the default; rather, infer type T from the right hand side."
> 
> 
> On Mon, Jan 23, 2017 at 6:30 PM, Matthew Johnson via swift-evolution <swift-evolution at swift.org <mailto:swift-evolution at swift.org>> wrote:
> This proposal looks good to me.  I have been looking forward to more flexible generic arguments for a while.
> 
> I agree with previous commenters who prefer the option to leave off the angle brackets when all parameters have defaults.
> 
> The proposal specifically mentions that the syntax is inspired by that of function arguments.  This is good, but I wonder if maybe we should draw further inspiration from function arguments and also add parameter labels for generic arguments.  Both feel like low hanging fruit in the generics area (correct me if I’m wrong about that) and it would be great to see both enhancements make it into Swift 4.
> 
>> On Jan 23, 2017, at 9:55 AM, Srđan Rašić via swift-evolution <swift-evolution at swift.org <mailto:swift-evolution at swift.org>> wrote:
>> 
>> Hi Everyone,
>> 
>> I've opened a PR (https://github.com/apple/swift-evolution/pull/591 <https://github.com/apple/swift-evolution/pull/591>) proposing default generic arguments which I think would be nice addition to the language. They are also mentioned in "Generic manifesto". 
>> 
>> The proposal is focusing around generic types. Generic functions are not coved by the proposal and I don't think that we need default generic arguments in generic functions as all the types are always part of the function signature so the compiler can always infer them. One corner case might be if using default argument values in which case support for default generic arguments in functions might be useful.
>> 
>> It would be great to hear your opinions and suggestions so I can refine the proposal.
>> 
>> 
>> _______________________________________________
>> swift-evolution mailing list
>> swift-evolution at swift.org <mailto:swift-evolution at swift.org>
>> https://lists.swift.org/mailman/listinfo/swift-evolution <https://lists.swift.org/mailman/listinfo/swift-evolution>
> 
> 
> _______________________________________________
> 
> 
> swift-evolution mailing list
> 
> 
> swift-evolution at swift.org <mailto:swift-evolution at swift.org>
> 
> 
> https://lists.swift.org/mailman/listinfo/swift-evolution <https://lists.swift.org/mailman/listinfo/swift-evolution>
> 
> 
> 
> 
> _______________________________________________
> 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/20170124/12cbdb13/attachment.html>