Truncation

Truncation
A Programming Pattern

Prof. David Bernstein
James Madison University

Computer Science Department

bernstdh@jmu.edu

Motivation

The Idea:
- Sometimes integers have more digits of accuracy than are needed
An Example:
- Employees are paid a fixed amount per piece, but only for multiples of 10 pieces (e.g., an employee will get paid for 520 pieces whether they make 520, 521, or 529 pieces)

Review

A Related Situation:
- Employees are paid per batch of ten (rather than per piece)
Solving this Problem:
- We can find the number of batches by dividing by 10 (using integer division)
An Example:
- 526 pieces is 526 / 10 or 52 batches

Thinking About the Problem

The Pattern

Notation:
- place denotes the place to truncate to (10 for the 10s place, 100 for the 100s places, etc...)
- number denotes the original number

The Process:

javaexamples/programmingpatterns/Truncation.java

(Fragment: pattern)

                truncated = (number / place) * place;

Examples

The Situation:
- Someone was born in 1996
- We are concerned with the decade and century when they are 87 years old (i.e., in 2083)
Truncating:
- To the decade: ((1996 + 87) / 10) * 10 or 2080
- To the century: ((1996 + 87) / 100) * 100 or 2000

Some Warnings

Be Careful with Integer Arithmetic:
- In general, (a / b) * b does not equal a
Be Careful about Terminology:
- Truncating floating point values to integer values requries a different process (i.e., type casting)
Don't Confuse Accuracy of Calculations and Output:
- In some situations it is necessary to perform calculations using truncated values
- In some situations it is necessary to perform calculations using all of the digits of accuracy available and truncate at the end
- In some situations it is necessary to perform calculations using all of the digits of accuracy available and then format the output with fewer digits of accuracy when it is displayed