Python 3 - Second Edition

26.4    Intuitive Decimal Numbers: decimal

You’ll certainly remember the following example, which shows that calculation errors occur when using the built-in float data type:

>>> 1.1 + 2.23.3000000000000003  

This is because the internal memory model of float cannot represent every decimal number; it can only approximate them to a certain accuracy.[ 121 ] These calculation errors are usually accepted for reasons of efficiency. In some cases, however, such as when calculating monetary amounts, it is particularly inconvenient that even simple floating-point numbers such as 1.1 and 2.2 cannot be represented exactly, and calculations with currency amounts therefore quickly become error-prone.

The decimal module contains the Decimal data type, which ...