Using mpf_class values from GMP library

source.C
#include <x/sql/decimal.H>

mpf_class value;

conn->execute("INSERT INTO table VALUES(?)", value);
Makefile.am
app_LDFLAGS=-lsqldecimal

decimal SQL columns can always be INSERTed or SELECTed as text string values, floats, doubles, or maybe even ints or long longs. LIBCXX and unixODBC handle the data type conversion automatically. The GMP library implements arbitrary precision math that does not suffer from floating point round-off errors, and is a natural fit for the SQL decimal datatype.

To use GMP library's mpf_class values:

This enables mpf_class to be used directly with execute() or execute_vector(); or receive fetched column values with fetch() or fetch_vectors().