Yes, I've had occasion to sample every n deciRuncibles as well. Although there are difficulties with edge cases, I think it would be throwing out the baby with the bath water (so to speak) to forbid striding from 1.0 to 10.0 by 0.5, which I think most users would expect to be able to do if they can stride from 1 to 10 by 1 (and which they can do at present, just with accumulating error).
By contrast, I really can't imagine that too many intentional use cases would push up against any possible upper iteration limits, whether it's 2^64 or 2^53, but the Swifty thing would be to have predictable behavior if those limits are reached. On Thu, Mar 31, 2016 at 1:40 PM, Erica Sadun via swift-evolution <[email protected]> wrote: > > On Mar 31, 2016, at 12:29 PM, Stephen Canon via swift-evolution > <[email protected]> wrote: > > I also don’t think that such a loop is particularly useful. For > floating-point types, something like stride(from: T, to: T, steps: Int) > seems safer and more workable to me (this is just my immediate reaction, I > haven’t thought this through in detail, and am likely to change my mind if > someone makes a good case). > > > While I do not object to n-steps, allowing a step difference is equally > important in my opinion. > "I want to sample every n deciRuncibles" versus "I want to sample 8 times > between 1 deciRuncible > and 10 deciRuncibles" > > I disagree that the former is less useful than the latter. > > -- E > > > _______________________________________________ > swift-evolution mailing list > [email protected] > https://lists.swift.org/mailman/listinfo/swift-evolution > _______________________________________________ swift-evolution mailing list [email protected] https://lists.swift.org/mailman/listinfo/swift-evolution
