On 6 December 2011 09:58, John Maddock <boost.regex@virgin.net> wrote:
There is ISO/IEC TR 24733 draft on decimal floating-point arithmetic. I found old thread by Stefan Slapeta: Decimal floating point library [1] I wonder, what is status of this idea in Boost?
I don't think anyone in Boost is working on it (that I know of) - it appears there was a C++ reference implementation available on the IBM website, but it looks like it might have been removed :-(
decNumber++ is still listed here http://goo.gl/mOqSI but the download is missing indeed If I'm looking in the right direction, decNumber++ was developed by Robert Klarer and implements the interfaces specified in the ISO/IEC TR 24733: http://speleotrove.com/decimal/ I found some references to Robert's activity here: http://lists.boost.org/Archives/boost/2003/07/50268.php I have no idea if the decNumber++ is the mentioned GPL'ed reference implementation, but perhaps it is not and Robert would be willing to contribute it to Boost.
In the mean time, some of us have been working on extended precision arithmetic (see "big_number" directory in the sandbox) using Christopher Kormanyos' e_float code for decimal arithmetic.
Thanks, I'll take a look.
Now that I've been reminded of IBM's C lib reference implementation of decNumber, I'll add support for that as well.... could make a good test case.
So, it is the decNumber. Best regards, -- Mateusz Loskot, http://mateusz.loskot.net