Raymond L. Buvel <[EMAIL PROTECTED]> wrote: ... > This is a bit off topic but I would like to know how you would go about > doing an implicit operation with an mpz and Decimal becoming a Decimal.
I would tweak Decimal.__new__ to accept an mpz and perform an int() on it before proceeding as it does now, for example. > I would like to use something like this to make the float and complex > types more interoperable with mpq and cmpq in my clnum module > (http://calcrpnpy.sourceforge.net/clnumManual.html). In those cases, I > would like the mpq to degrade to float and the cmpq to degrade to > complex (right now they just raise exceptions). Seems like you need to > either modify the target type to recognize the new one or the code has > to get very complex to handle all the combinations. Modifying the target type is indeed what I had in mind (for Decimal). As long as the target-type is Python-coded, you can substitute its __new__ with a wrapped version. Unfortunately, float and complex aren't; but fortunately, Python does have for these cases appropriate special methods/slot that your SOURCE type can implement. In gmpy, I have float 'upgraded' to mpq (with a Stern-Brocot heuristic), but it wouldn't be any harder to have the mpq 'degrade' to float instead. Alex -- http://mail.python.org/mailman/listinfo/python-list
