Pre-1.5 versions of Money.h++ included two multiple precision integer classes: Int64 and Int96. Versions 1.5 and higher of Money.h++ contain a single multiple precision integer class, RWMultiPrecisionInt<n>. This class uses a specified number of integer types, represented by the integer template parameter n, to represent a single integer value. For example, on a 32-bit operating system, RWMultiPrecisionInt<2> uses two 32-bit integers, for a total of 64 bits, to represent an integer. Similarly, RWMultiPrecisionInt<3> uses three 32-bit integers, and so forth, up to RWMultiPrecisionInt<16> which provides 16*32, or 512 bits for integer representation. Class RWMultiPrecisionInt<1> has been specialized to use the built-in type double, which provides 52 bits for representing an integer.
NOTE: In Money.h++ 1.5 and higher, class RWMultiPrecisionInt<n> is private. While it has been certified for use as a mantissa for the RWDecimal classes, we do not recommend other uses. Its interface is not documented in the Class Reference portion of this manual.
For convenience, Money.h++ provides typedefs for instantiation of RWMultiPrecisionInt<n> for n=1, n=2, and n=3. These typedefs are RWMP1Int, RWMP2Int, and RWMP3Int, respectively.
©Copyright 1999, Rogue Wave Software, Inc.
Contact Rogue Wave about documentation or support issues.