On Fri, Sep 7, 2018 at 9:19 PM, Lucio <lucio.d...@gmail.com> wrote: > > PS: Maybe I should add a thought to clarify what I'm imagining. Am I > mistaken in my belief that user-defined methods and functions are much less > likely to appear within contracts than operators on intrinsic types? And > that the "generic" nature of operators on intrinsic types lies at the core > of the problem contracts are trying to solve? To some degree, restricting > the use of such generic operators (and Stroustrup also foresaw that indexing > is one such operator, despite its use of two, separate characters, give him > some credit) is the final purpose, much more so than the "extension" of > functions to some concept of "compatibility" between functions.
I think it's difficult to say. I think most generic functions and types will be written without any contracts at all. I think there are clear needs for generic functions that require operators, but I also think it's likely that all such functions will be written fairly quickly. Once we move past a set of standard generic functions, I think that functions that require operators will be relatively rare special cases. On the other hand, I think that there will always be a need for generic types that are slices/maps/chans of types that support specific methods, where people want to, in effect, use interface types without requiring boxing. But I'm just guessing. Only time and experience will tell. Ian -- You received this message because you are subscribed to the Google Groups "golang-nuts" group. To unsubscribe from this group and stop receiving emails from it, send an email to golang-nuts+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
