[Pitch #2] Add `isKnownIdentical` Method for Quick Comparisons to `Equatable`

[vanvoorden (Rick van Voorden)]

@tera

@vanvoorden

Hmm… I would maybe start with a comment from the previous pitch:

@lorentey

First, the bitwise equality of two objects does not necessarily imply that they are identical. An object's identity sometimes includes the location of its bits in addition to their value. For example, this is the case for noncopyable synchronization constructs like struct Atomic and struct Mutex -- two distinct atomic integers are never identical, even if at some point they both happen to have the same integer value, i.e., their storage consists of the same bits. Some Swift types are not bitwise comparable.

I must admit I do not fully understand this bit. Is it only applicable to non copyable types? How would I be able to even attempt to compare the two (presumably identical) variables if the type is non copyable to begin with? (without always getting a known result of false, I mean). Could you give an example?

I am not sure I have enough context on this right now to give a good example. Maybe @lorentey can help with some more information on that?

@tera

So Int and other similar types would return nil in their isKnownIdentical implementation?

Not to sound like a "broken record"… but my intention is that this pitch-proposal would ship without making its own strong opinions about specific concrete types from standard library that should override isKnownIdentical. My preference is for that decision to remain the domain of the library maintainers outside of this specific pitch-proposal design review cycle.

But if you are asking me for a "weak" opinion on this topic… then I would say that these are examples of standard library types that should return nil for isKnownIdentical:

• Bool
• Int
• InlineArray

Again… I want the library maintainers to use their domain-specific knowledge and context to help them make those decisions on their own under the semantic and performance contract of the pitch-proposal. But my weak opinion on this topic is that Int should not expose a concept of identity equality.