On Tue, Jul 15, 2014 at 10:30 PM, Jonathan S. Shapiro <[email protected]> wrote: > I don't agree with you about punting to dynamic checks, since (e.g.) the > range check on array bounds is defined to be part of the accessor operation.
This is actually important. If you're OK with regularly punting to dynamic checks then completeness isn't hard after all. I guess I assumed (yes) you considered implicit dynamic checks to be undesirable. Does BitC actually have an effect or mode or something to rule out dynamic checks? I seem to remember some discussion of a no-throw effect. How do you deal with the [range check is part of the operation] approach there? _______________________________________________ bitc-dev mailing list [email protected] http://www.coyotos.org/mailman/listinfo/bitc-dev
