Dan Sugalski wrote:
<>Cool, go for it. I'd think that for the set_(integer|number) vtable slots
we'd set the real part and make the imaginary part 0, while the string
version'd look for the "x + yi" version.
And have set_num_keyed set the real and the imaginary part (indexed as strings, say "real" and "imag") and maybe change the modulus and the argument independently. Also get_num_keyed will return the real part, imaginary part, modulus and argument (and maybe cache the last two).
And I'm now thinking that we want to do mmd for assignment. Dammit. :(
You'll have to explain that to me. :)
I've done some more work on the String PMC and its tests, I'll send a patch later today.
Danalexm
