Data Manipulation

Introduction

Drills are very short exercises designed to exercise the most basic building blocks of modules and programs.

Problem Description

Data manipulation applications almost always include tranformation of data from one or more forms into other forms. In this drill you will practice transforming data stored in data structures into equivalent data stored in different data structures. Such transformations are often done to make the data easier to process for a particular application.

Solution Description

names = ["Peter", "Lois", "Meg", "Chris", "Stewie", "Brian"]
ages = [41, 40, 16, 17, 1, 18]

Assume that the ages in ages correpond to the names in names by index. Write a Python module which: