Conceptually, maybe? But of course the sequence is lazily evaluated, so the only state stored in an instance of StrideTo is one value for the current iteration; all the other properties in a StrideTo are just the inputs to the stride function, and no length is computed let alone stored. I get that you want something that represents the input instead of the output, but what do you gain from that type not achievable otherwise? On Thu, Apr 7, 2016 at 6:59 PM Brent Royal-Gordon <[email protected]> wrote:
> > I'm confused. There is an instance owning the start and end. It's called > StrideTo or StrideThrough, conforms to Sequence (with a FIXME comment that > it should conform to Collection) and is distinct from Range and from the > Strideable bounds themselves. Is that different from what you're describing? > > Yes, it is different. StrideTo and StrideThrough represent the sequence > resulting from the striding operation. If there was only one of them, you > would probably call it StrideSequence. They are the *result* of the > `stride` function or `striding` method. > > What I'm talking about is a single instance which represents what you are > striding *over*, but *not* the length of the stride. It is an *input* to > the `stride` function or `striding` method. > > Does that make sense? > > -- > Brent Royal-Gordon > Architechies > >
_______________________________________________ swift-evolution mailing list [email protected] https://lists.swift.org/mailman/listinfo/swift-evolution
