[Python-3000] Extension: mpf for GNU MP floating point

[Rob Crowther]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I've uploaded the latest code to http://umass.glexia.net/mpf.tar.bz2

It's been cleaned up, implements a little bit of the abstract number interface, many very repetitive function declarations were turned into macros making it far easier to maintain, and it now has a printable representation like you'd expect from a float. At this point, I'm able to use it as a stripped down drop in replacement for Decimal. It's also much, much faster.

One question I was asked in IRC was if it was possible to change the precision. Currently, that's only implemented during initialization of an instance, by passing the prec keyword. It defaults to 128 bits, what looks to me to be about double the precision of a builtin float. 

Included in this is a copy of my git repository since I don't have it online. I'm going to be away for a while, and someone else may find it useful if they want to hack on it. (I have occasionally been known to screw up =P)

Well, that's all for today's update. Hopefully, more to come soon.

Rob
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFG+REZqR5p8HaX4oURAnJMAKCDxjO2YUnNrJVClujA0l8+wKSLkACeOc5F
097rKqJO6DoaLShfpA3oPsU=
=lB+r
-----END PGP SIGNATURE-----