My thought on this is definitely "yes". If Parrot provides an operation, it should produce the correct result over the largest possible input range. We do allow users to override the default types and many default behaviors, but for people who don't take that step we should still provide the best possible default. We do offer a built-in Complex PMC, so there's no extra overhead in producing the correct results.
--Andrew Whitworth On Sun, Sep 26, 2010 at 1:58 PM, Paul C. Anagnostopoulos <[email protected]> wrote: > Folks, > Should PMC functions with partial complex ranges promote their result to a > Complex? For example, should sqrt(negative Float) produce a Complex? This > issue does not pertain to native numbers, since we never promote them to > PMCs. > There is precedence for doing this. For example, Integer promotes to BigInt > when necessary. > ~~ Paul > > ---------------------------------------------------------------- > Windfall Paul C. Anagnostopoulos > ---------------------------------------------------------- > Software 978 371-2316 > www.windfall.com > ---------------------------------------------------------------- > Metaphorical invocations ... often suffer from the > weakness of giving such satisfaction to the human > mind that they tend to be mistaken for incisive and > illuminating observations. ---Torkel Franzen > _______________________________________________ > http://lists.parrot.org/mailman/listinfo/parrot-dev > > _______________________________________________ http://lists.parrot.org/mailman/listinfo/parrot-dev
