This unit provides an arbitrary precision BCD float number type. The usecase is performing a few arithmetic operations with the maximal possible accuracy and precision, e.g. calculating the sum of numbers from a text files, where the conversion from decimal input to binary floats would take more time than the calculation.

It can be used like any numeric type and supports:

• At least numbers between 10^-2147483647 to 10^2147483647 with 2147483647 decimal digit precision
• All standard arithmetic and comparison operators
• Rounding functions (floor, ceil, to-even, ..)
• Some more advanced operations, e.g. power and sqrt
• Accurate and precise binary float (single/double/extended) to BCD float and string conversion
• ..

See my webpage for the detailed bigdecimalmath documentation