On 29 Sep 2009, at 5:14 am, Brian Harvey wrote: > But what gives /me/ the heebie-jeebies is having to have a .schemerc > file > full of things like > (set! fixnum-overflow-mode 'bignum) > (set! lg-of max-bignum 400) > or whatever.
Yes, that'd be nasty. I don't like the idea of this being a whole- program choice. What if some DSP library expects fixnums and nasty ratios to overflow to flonums, while a crypto lib needs exact 2048-bit integers, and somebody wants to write an application that sends compressed speech down an encrypted tunnel, hmmm? Since my proposal to have specialist arithmetic procedures that embody different semantics didn't go down well (but it's what everyone seems to do currently!), how about we make it part of the dynamic environment? (with-fixed-size-numbers-that-overflow-to-inexacts ...stuff...) Efficient implementation thereof is left as an exercise to the reader, mind. ABS -- Alaric Snell-Pym Work: http://www.snell-systems.co.uk/ Play: http://www.snell-pym.org.uk/alaric/ Blog: http://www.snell-pym.org.uk/archives/author/alaric/ _______________________________________________ r6rs-discuss mailing list [email protected] http://lists.r6rs.org/cgi-bin/mailman/listinfo/r6rs-discuss
