Your bitwise-nand etc takes more arguments than they have to. They are 
2-argument procedures according to the spec, which gives you better performance 
than the apply-dance you are doing now. Maybe have a bitwise-nand and a 
bitwise-nand*?
-- 
  Linus Björnstam

On Thu, 9 Jan 2020, at 10:10, Frank Terbeck wrote:
> Hi Linus!
> 
> Linus Björnstam wrote:
> > I have a port of the SRFI code as well, using renaming of guile and srfi-60
> > procedures as necessary.
> 
> I see! I did mine from scratch, while reading the spec.
> 
> 
> > It has the make-bitwise-generator from srfi-151.
> 
> Interesting. I thought  the generators were part of  some disjoint type.
> After seeing  the iterator you  implemented, I  actually took a  look at
> SRFI-121 and indeed: “They are just procedures that conform to a calling
> convention, so you can construct a generator with lambda.”
> 
> I guess I'll do the same for now! Thanks for letting me know. :)
> 
> 
> Regards, Frank
>

Reply via email to