Scripting and Programming

for Technical Computation

Bruce Char
Copyright Bruce Char
2012
Scripting and Programming for Technical Computation
by Bruce Char
All rights reserved. Permission is given to StaII and students oI CS 121/122/123, 2011-2012 to reproduce these notes Ior their personal use only.
iii
Contents
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
1 Introduction -- Technical computing in the 21st century . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Chapter synopsis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 What is technical computing? What kinds oI technical computing are there? What are we studying in this course? . . . . . . . . . . . . . . . . . . . . . . . 1
1.3 What are the advantages oI doing technical calculation with a computer? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.4 The spectrum oI devices Ior technical computing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Personal computers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
High perIormance computers, also known as "supercomputers" . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Hand held mobile devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Dedicated controllers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.5 Maple, a system Ior technical computing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.6 What about Systems X, Y, Z, ...? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.7 Why pick Maple as the Iirst system? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.8 Using more than one system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1 Chapter synopsis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2 Starting up Maple, getting a Iresh start . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.3 Evaluating an expression involving exact arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Grade school arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Diagnosing typographical mistakes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Correcting typographical mistakes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Exponentiation (powers). Numbers with lots oI digits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Detecting and Iixing semantic and "logic" mistakes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.4 Saving and retrieving your work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.5 Retrieving backups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.6 Algebra, plotting and mouse-clickable operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Algebraic expressions and equations. Solving equations. Working with pieces oI expressions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
For those with previous experience on other systems: some things are diIIerent, Ior a reason . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Plotting, approximate numerical solutions through cursor position . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
The Expressions Palette and the Common Symbols Palette: entering Trigs, logs, roots, exponentials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Approximate numerical (calculator - type) arithmetic in Maple . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Evaluation, and selection oI pieces. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.7 A quick-reIerence summary to this chapter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3 Chapter 3 Technical word processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.1 Chapter Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.2 Maple as a word processor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.3 Shortcuts to entering math symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.4 Other word processor Ieatures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.5 Troubleshooting word processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.6 Summary oI Chapter 3 material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
4 Chapter 4 Assignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.1 Chapter Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.2 Assignment: remembering results Ior Iuture use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.3 : is a keyboard shortcut Ior assignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.4 The Variables Panel: a way to know the current value oI your variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.5 DiIIerences between the appearance oI the worksheet and the state oI a Maple session . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.6 restart causes all variables to be deassigned . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
4.7 Evaluation and chains oI assignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
4.8 Troubleshooting assignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.9 Summary oI Chapter 4 material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
5 Chapter 5 Building scripts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
iv Contents
5.1 Chapter Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
5.2 Getting inIormation Irom solve, data structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
5.3 Finding simultaneous solutions, constraining solutions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
5.4 Scripting: making computational work easy to reuse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.5 Rewriting the script by assigning parameters at the start oI the script . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
5.6 Summary oI script writing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
5.7 Troubleshooting scripts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
5.8 Attachments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Attachment: Version 2 oI sheep script without parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Attachment: Version 2 oI Sheep Script, with parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Attachment: Version 3 oI Sheep Script, with parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
5.9 Summary oI Chapter 5 material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
6 Chapter 6 More sophisticated scripting: textual entry, multiplots, point plots, character strings, lists, solving equations, and evaluation. . . . . . . 77
6.1 Chapter Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
6.2 Textual entry oI operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
6.3 Why are there two diIIerent styles Ior entering operations? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
6.4 Plotting a list oI expressions (multi-plots), plotting lists oI numbers (point plots) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
6.5 Character Strings and putting titles and labels into plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
6.6 Troubleshooting with strings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
6.7 Learning through on-line documentation and experimentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
6.8 More operations on lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
6.9 solve, lists and sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
6.10 Evaluation, eval, and assignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
6.11 Summary oI Chapter 6 material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
7 Chapter 7 Using and DeIining Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
7.1 Chapter overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
7.2 Functions in computer languages: a way oI producing an output Irom inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
7.3 The textual names oI common Iunctions: doing math calculations using the keyboard. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
7.4 A Iunction name to commit to memory: exp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
7.5 How can I remember so many Iunctions? Use command completion and on-line documentation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
7.6 Being savvy about using on-line documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
7.7 DeIining your own Iunctions with -~ (arrow) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
7.8 Using Iunctional evaluation instead oI using assignment or eval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
7.9 Troubleshooting Iunction deIinitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
7.10 Using Iunctions Irom library packages, with . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
7.11 Attachment: some built-in problem-solving Iunctions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
7.12 Summary oI Chapter 7 material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
8 Chapter 8 Programming with Iunctions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
8.1 Chapter overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
8.2 Designing Iunctions Irom context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
8.3 Function composition: daisy-chaining Iunctions together . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
8.4 Expressions with units oI measurements: convert . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
8.5 User-deIined Iunctions whose inputs and outputs aren't numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
8.6 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
9 Chapter 9 Visualization, modeling, and simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
9.1 Chapter Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
9.2 plot structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
9.3 plots|display | and combining plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
9.4 plottools, lines, and other shapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
9.5 Mathematical models and simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
9.6 Drawing x-y position as a Iunction oI time through parameterized plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
9.7 A Iirst animation example using animate, animation controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
9.8 The Iirst animate example, part 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
9.9 Designing a Iunction to use in animate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
Contents v
9.10 Designing a more elaborate animation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
9.11 Solving a problem with the help oI animation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
9.12 Exporting animations and non-animated plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
9.13 Summary oI Chapter 9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
10 Moving into programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
10.1 Chapter Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
10.2 Code edit regions: entering a block oI code, executing it all at once . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
10.3 Troubleshooting common errors in entry into code edit regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
10.4 print and annotating an execution trace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
10.5 The CarSimulator package: a Iirst view . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
10.6 A second task with the CarSimulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200
10.7 Programming to solve a class oI problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
10.8 Reviewing old and introducing new programming concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
10.9 Summary oI Chapter 10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
11 More on proc and while . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
11.1 Chapter Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
11.2 Review oI Iunction terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
11.3 Creating a short Iunction with proc... end . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
11.4 Creating a Iunction with encapsulated (local) variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
11.5 Conditional repetition through while... do... end do . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
11.6 Example: car control procedures with conditional expression built Irom Boolean operations and multiple loops . . . . . . . . . . . . . . . . . . . . . . 217
11.7 Example: car control procedures with a local variable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
11.8 Example: the Iour corner problem with a shorter program using a nested loop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
11.9 Example: a procedure that does a calculation instead oI controlling a car. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
11.10 Troubleshooting small Iunction deIinitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
11.11 Troubleshooting local variable declarations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
11.12 Troubleshooting while loops: loops that don't stop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
11.13 Summary oI Chapter 11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
12 More control: Ior and iI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
12.1 Chapter Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
12.2 Ior and indexed repetition Ior... Irom ... to ... do... end do . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
12.3 variations on Ior: to ... do ... end do, Ior... Irom ... by ... to ... do ... end do . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
12.4 Choosing alternatives: iI...then...end iI, iI... then... else...end iI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
12.5 Coding situations with three or more cases using iI-then-eliI-eliI...else-end iI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
12.6 A CarSimulator scenario solved with iI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
12.7 For the curious: comparing control statements in conventional programming languages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254
12.8 Troubleshooting iI statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
12.9 Summary oI Chapter 12 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260
13 Looking inside execution oI a procedure or script . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
13.1 Chapter Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
13.2 More on printing: print, printI and sprintI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
13.3 For the curious: why printI is so cryptic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
13.4 Suppressing execution trace output Irom loops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
13.5 Looking inside oI procedure invocations (Iunction calls) with trace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
13.6 Summary oI Chapter 13 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
14 Tables and other data structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
14.1 Chapter Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
14.2 A review oI Maple data structures and properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
14.3 Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
14.4 Troubleshooting with tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
14.5 Using sequences in multiple simultaneous assignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
14.6 Using multiple assignment with procedure results: return sequences oI results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
14.7 A pattern oI looping: accumulation oI results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
14.8 Application oI tables in accumulative looping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
vi Contents
14.9 Summary oI Chapter 14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
15 More on looping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
15.1 Chapter Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
15.2 Animations, revisited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
15.3 Combining Ior and while . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
15.4 Patterns oI looping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
15.5 Constructing iterations and the roles oI variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305
15.6 Implicit looping: seq, list addition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
15.7 An example that uses Ior...while... looping and display(..., insequencetrue) Ior animation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312
15.8 Summary oI Chapter 15 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
16 Calculus and optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
16.1 Chapter Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
16.2 Learning about additional kinds oI mathematical computations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
16.3 DiIIerentiation, simpliIication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
16.4 Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324
16.5 Finding minima and maxima by using Maple as a calculus calculator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
16.6 One-step extrema-Iinding using maximize and minimize . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331
16.7 One-step approximations to extrema using Optimization|Maximize| and Optimization|Minimize| . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
16.8 Multivariate operations in Maple . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344
16.9 Chapter summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 346
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351
vii
List of Tables
Table 1.1: An early computer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Table 1.2: A high-end calculator in 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Table 2.1: Maple started up with new document in Windows XP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Table 2.2: Maple document with Iirst entry area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Table 2.3: Maple input using 2d math . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Table 2.4: Maple input with labeled result . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Table 2.5: Arithmetic Operations in Maple . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Table 2.6: Examples oI Maple error messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Table 2.7: Create Document Block to Iorce a Math input area wherever the cursor is placed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Table 2.8: Example oI a vocabulary mistake . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Table 2.9: Maple save menu operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Table 2.10: Maple save dialog box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Table 2.11: The Maple state display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Table 2.12: Example oI Plotting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Table 2.13: Plot created by right-click -~ Plot -~ 2DPlot (with cursor positioning) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Table 2.14: User-conIigured plot using PlotBuilder instead oI 2DPlot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Table 2.15: PlotBuilder Dialog box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Table 2.16: The Expression palette . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Table 2.17: Examples oI palette-driven computation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Table 2.18: Examples oI computing with approximate solving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Table 2.19: Evaluate at a point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Table 2.20: Operations on equations, multi-part expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Table 3.1: Maple in math entry mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Table 3.2: Document aIter control-T (or Insert-~Text) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Table 3.3: Document with a mixture oI text and math . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Table 3.4: control- puts the results oI a calculation in the midst oI text . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Table 3.5: Keyboard shortcuts in math mode through the escape key . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Table 4.1: Assignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Table 4.2: : is another way to do assignment oI a name . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Table 4.3: Assignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Table 4.4: The variables panel shows currently assigned variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Table 4.5: The state oI a Maple Session . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Table 4.6: Evaluation oI expressions involving assigned variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Table 4.7: BROKEN - MISSING TITLE! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Table 4.8: Assignment : is not symmetric . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
Table 4.9: Some Names can't be assigned to . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Table 4.10: Can't solve Ior a variable with an assigned value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Table 5.1: Selecting parts oI a solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Table 5.2: Basic data structures in Maple and operations to extract parts oI them . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Table 5.3: Solving simultaneous equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Table 5.4: Executing a clone oI a script through copy-paste . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Table 6.1: The textual Iorm oI equation solver solve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Table 6.2: The textual version oI plotting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Table 6.3: Plotting oI multiple expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Table 6.4: Plotting points with lists oI numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Table 6.5: Plot options and labels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Table 6.6: What happens when you Iorget to use the " delimiter in strong, or use the wrong character Ior the delimiter . . . . . . . . . . . . . . . . . . . . . . . . . 87
Table 6.7: Plot command help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Table 6.8: Examples oI plot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Table 6.9: Copying the example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
Table 6.10: ModiIying the example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
viii List oI Tables
Table 6.11: Plot option help page . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Table 6.12: Operations on lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Table 6.13: Solving systems oI equations with solve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Table 6.14: Evaluating an expression using a particular value oI one oI the variables in the expression, and then plotting . . . . . . . . . . . . . . . . . . . . . . . 96
Table 6.15: Evaluating an expression using a particular value oI one oI the variables in the expression, and then solving . . . . . . . . . . . . . . . . . . . . . . . 97
Table 6.16: Evaluating an expression using a particular value oI one oI the variables in the expression using eval, and then solving . . . . . . . . . . . . 98
Table 6.17: An example using assignment and eval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Table 7.1: Function results in Maple . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Table 7.2: Textual names Ior common math Iunctions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Table 7.3: exp is the name oI the exponential Iunction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
Table 7.4: Command Completion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
Table 7.5: A Iunction deIinition in a math textbook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Table 7.6: User-deIined Iunctions through the Expression Palette . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Table 7.7: Comparing evaluation using eval and Iunctional notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Table 7.8: Troubleshooting Iunction deIinitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
Table 7.9: with, the package loading operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Table 7.10: Textual names oI common operations in Maple . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Table 8.1: A script that uses Iunctional (chaining), and its use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Table 8.2: Examples oI Unit Conversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
Table 8.3: A problem solved with a script using Iunction deIinition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
Table 8.4: Solving multiple versions oI a problem through Iunctions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
Table 8.5: Unit Converter Assistant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
Table 8.6: A Iunction that takes a list as its input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
Table 8.7: A Iunction that returns a plot as its output, rather than a number . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
Table 8.8: A problem solved with a Iunction that outputs a sequence oI two numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
Table 8.9: several versions oI a Iunction with Iunction deIinitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
Table 9.1: Plot results, displayed and not displayed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
Table 9.2: Display combines plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
Table 9.3: BROKEN - MISSING TITLE! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Table 9.4: Combining a line with other plots using display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
Table 9.5: Examples oI plots where the x and y positions are parameterized by t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Table 9.6: First animation example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
Table 9.7: Animation controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
Table 9.8: Animation pop-up menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
Table 9.9: Frames 1, 5, 10, 20, and 25 oI the animation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
Table 9.10: A rough-draIt plot oI ball versus time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
Table 9.11: A plot that draws a ball as a circle at (0,6). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
Table 9.12: A plot that draws a ball as a time t0.5. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
Table 9.13: Trying out a Iunction that draws the ball at a particular time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
Table 9.14: Animating the ballFrame Iunction to produce a movie oI the ball in Ilight. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
Table 9.15: A ball with a vertical target line. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
Table 9.16: A Iirst try at shooting the ball up 30 Ieet: 15 miles per hour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
Table 9.17: Second try: shoot upwards at 22 Ieet per second . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
Table 9.18: Exporting animations through the pop-up menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
Table 10.1: BROKEN - MISSING TITLE! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
Table 10.2: First code edit region Ior CarSimulator work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
Table 10.3: On-line help Ior CarSimulator aIter package is loaded into worksheet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
Table 10.4: Second code edit region with Car Simulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
Table 10.5: First control program: move, turn leIt, move twice. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
Table 10.6: Second code edit region with Car Simulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
Table 10.7: Fourth code edit region with Car Simulator animation result . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
Table 11.1: Demoonstrate oI assignment oI local variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
Table 15.1: Several Irames oI movie1 animation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300
Table 16.1: Symbolic diIIerentiation using the expression palette . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320
List oI Tables ix
Table 16.2: Symbolic diIIerentiation and evaluation oI derivatives, textually . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320
Table 16.3: Plotting a Iunction and its derivative together . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321
Table 16.4: Plotting a Iunction and its derivative together . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
Table 16.5: Clickable interIace version oI limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324
Table 16.6: Textual version oI limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
Table 16.7: Finding the minimum and maximum using textually-speciIied commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
Table 16.8: Writing a procedure to calculate minimum oI a Iunction and produce a plot that will help understanding oI it . . . . . . . . . . . . . . . . . . . . . 327
Table 16.9: minimize and maximize . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331
Table 16.10: Solving a problem with minimize and maximize . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333
Table 16.11: Using optimization in a chemical production problem with minimize and maximize . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
Table 16.12: Finding an approximation to the minimum with Optimization|Minimize| . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
Table 16.13: An example where the approximate minimizer gets Iooled . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
Table 16.14: Multivariate operations in Maple . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344
x List oI Tables
xi
Dedication
To our students, who learn how to work with the new and diIIerent.
To my Iamily, whose support is unwavering.
Acknowledgements
Jeremy Johnson started the Computation Lab course. He chose the initial direction and approach oI the course. Frederick W. Chapman worked
with us in 2006-2008 in the development oI the course.
We are grateIul to our colleagues David Augenblick, Jeremy Johnson, Mark Boady, and Brian Dolhansky Ior their help in the development oI
this text. David was a careIul prooIreader and useIul sounding board. Jeremy's shrewd counsel and Ieedback oIten led to improvements.. Brian's
expert and speedy work in tagging and repairing the manuscript kept us on schedule. We also appreciate the assistance oI MaplesoIt Ior technical
advice and support, especially that oI Karishma Punwani, Bill Laarakkers, and Christina Spirou.
xii
1
1lntroduction -- Technical computing in the
21st century
1.1Chapter synopsis
1. We explain what technical computing is about, why it's done with a machine and what kinds oI devices are used to do it.
2. The advantages doing calculation with a computer are surveyed.
3. Maple, a system Ior technical computing is introduced and contrasted with some oI the other choices available today.
1.2What is technical computing7 What kinds of technical
computing are there7 What are we studying in this
course7
Well, obviously, it's "computing done Ior technical work" -- primarily science and engineering, but any other Iield where mathematical and
scientific reasoning is used. Nowadays this can include Iinance and business, health care, or digital media as well as the traditional math/science/
engineering domain. You are probably Iamiliar with numerical computing since it's done even without computers, with paper and pencil or
with calculators in high school or even earlier. You probably have also seen graphing calculators do simple kinds oI graphical computing, also
sometimes known as mathematical visuali:ation. In addition to this, computers can and are used to do svmbolic computing -- computing where
the answer is a Iormula rather than a number, as well as deductive or logical computing -- using logic or deduction to Iind answers. In this course,
we will concentrate on numerical, symbolic and graphical computing, although we will see some instances oI deductive computing as well.
The success oI such reasoning in these Iields is well-known. Students should aim Ior proIiciency in technical computing to be able to be able to
obtain success. Because oI the many applications, languages, and soItware tools used nowadays Ior successIul technical computing, the learning
task is not "learn one programming language that will serve all needs". They need to learn the terminology and conceptual approaches used
so that they can use their Iirst experiences with technical computing as leverage to continuing selI-taught learning and proIiciency. Being able
to build on the initial experience, rather than being deIined and limited by it, is what will allow access to the stream oI many applications and
soItware tools that will be in play and in vogue during a career.
1.3What are the advantages of doing technical calculation
with a computer7
One advantage is greater quantitv of calculation-- computers can do calculations billions oI times Iaster than humans and thousands or millions oI
times Iaster than calculator-driven computation.. Using them makes some things Ieasible that are not possible any other way:
As you've seen with your high school physics, chemistry, and math courses, scientists and engineers oIten work with mathematical models
-- systems oI equations, symbols, and mathematical relations that try to describe key aspects oI a situation. Calculation with models helps
answer questions about the situation being modeled: How much Iuel will be used in this operating scenario? How long will it take to heat up
the Iurnace to 265 degrees? Computer simulation loads a model with some initial conditions, and then through calculation produces predictive
results. Simulation can be used to make Iorecasts such as: how Iast will the vehicle be moving aIter 5 seconds? How Iar will the pollutants
move underground aIter being buried Iive years? Computer simulations can sometimes generate predictions even when standard techniques oI
"mathematical solution" are not adequate to Iind an answer.
Computers make it possible to use models that require much more extensive calculations to reach conclusions. The payoII may be better
predictions. Extensive calculation also makes it possible to generate and present more inIormation. Computer visuali:ation goes beyond that to
highly detailed pictures or animations oI a situation described through a mathematical model.This can lead to better insight and understanding.
2 1 Introduction -- Technical computing in the 21st century
Another advantage is that it is easier to make a lasting written record of the work. One oI the big diIIerences between proIessional technical
work and homework is that the modeling and calculation has lasting value. it matters to more than one or two people, and it is being used in
an on-going project. In a proIessional situation, a worker may need to go back and review the work a year later, long aIter the details are Iaded
Irom easy recall. Other people may want to reuse the programming and will need to be given an explanation about how to use it. Since they
have long-term value, it is beneIicial to put them into a Iorm suitable Ior easv future reference and reuse. Thus the work typically includes both
programming and documentarv explanation.
1.4The spectrum of devices for technical computing
Back in the days oI the original electronic computers developed during and immediately aIter World War II, the only kind oI device you could
use Ior technical computing (which was the only computing that could be justiIied in those days) Iilled up a whole room and needed an extensive
support staII.
Table 1.1: An early computer
ENIAC, one oI the Iirst electronic computers, being programmed by Herman Goldstine
at the University oI Pennsylvania circa 1946. (US Army photo) Originally intended to do
artillery calculations, it was used Ior the design and development oI the Iirst thermonuclear
weapons.

See http://www.seas.upenn.edu/~museum/.

Today's users have a choice oI a wide variety oI devices:
Personal computers
Typically a computer Ior individual use can be expected to have the Iollowing Ieatures:
1. Processing capability oI a billion or more arithmetic or memory operations per second. SigniIicant amounts oI calculation are used in doing the
graphics involved in processing digital media or supporting the artiIicial worlds presented in computer games. This same power can be used to
do the simulation and exploration in many common kinds oI scientiIic and engineering situations. This course is about using that calculation
power.
2. Standard selection and pointing devices, keyboard and mice. Some computers support "tablet" operation which allow use oI a pen writing on
the display. The display area and input devices make it easy to enter inIormation via text or by pointing/selection. While the World Wide Web
has made much processing possible with just "pointing and clicking", the complexity and mathematical sophistication oI the models used in
current technical computing outstrip the capabilities oI current "point and click" technology. Some (computer programming) language-based
interaction is necessary to handle things deItly. This requires a more knowledgeable user, but many people Iind that the extra power is worth
the learning eIIort.
3. A screen capable oI displaying inIormation equivalent to one or more 8 1/2 x 11 inch pieces oI paper. This helps support the development oI
documentation, or oI more complex visualizations.
4. Local storage capable oI storing a signiIicant Iraction oI the text oI the books in the Library oI Congress. OI course, it's easy Ior a personal
media collection to Iill up this kind oI storage, but it should not be Iorgotten that the billions oI characters that this storage represents can store
years or decades oI eIIorts oI computer programmers.
5. Connection to the internet, which makes it easier to communicate and share work with others, and to download new programming. From the
viewpoint oI technical computing, that it's easy to import and run substantial amounts oI programming.
High performance computers, also known as "supercomputers"
1.4 The spectrum oI devices Ior technical computing 3
While hand held devices and personal computers are used widely by the general public Ior non-technical purposes, there are a class oI computers
that are used primarily Ior mathematically-based computing. Typically instead oI billions oI numerical operations per second ("gigaIlops"
-- billions oI floating point operations), they employ multiple processors in parallel to do trillions oI operations ("teraIlops") or quadrillions
("petaIlops"). The hardware, personnel and energy costs oI such computers are signiIicant, so typically they are available through centers used
through the Internet Irom many locations. The kinds oI technical problems such devices are used Ior would include: simulation oI complicated
physical situations, such as Ior climate prediction, astrophysics, or engineering design (aircraIt, groundwater, large building, automobile).
Using such large devices incur signiIicant expense, Irom hardware, soItware and support staII costs, as well as electrical power consumed.
For example, the U.S. Department oI Energy has an IBM computer at its Los Alamos National Laboratory in New Mexico, USA
capable oI 1 quadrillion ( ) numerical operations per second. It consumes 2345.50 kilowatts when it is running. (reIerence: http://
blog.enterpriseitplanet.com/green/blog/2008/06/green-petaIlop-ibms-roadrunner-wins-supercomputer-top-spot.html) The National Center Ior
Computational Sciences (NCCS) at Oak Ridge National Laboratory in Tennessee, USA which has several large supercomputers, has a 2008
annual budget between $80 million and $100 million. (reIerence: http://news.cnet.com/8301-137723-9985500-52.html). A typical personal
computer might be capable oI approximately 4 billion numerical operations per second, so the Blue Gene computer at Los Alamos provides
roughly more computing power.
Typically the programming is developed on personal computers, then moved to the larger devices. Usually the expensive supercomputer time is
spent primarily on numerical computation, rather than on providing a nice-to-use interIace Ior users. However, the results oI a supercomputer may
be shipped over the Internet to a personal computer so that a scientist or engineer may mull over the results in a more contemplative way without
incurring additional supercomputer costs.
Multiple computers may be linked together over the Internet to get all the pieces oI a particular elaborate computation done. This may have the
supercomputers perIorming the massive numerical computations, while other smaller or personal computers linked in may be displaying the
results to a distributed team oI investigators who are collectively digesting results and steering the on-going work.
Hand held mobile devices
Calculators are useIul Ior casual computation, where one wants to Iigure out the solution to a small problem once. It's easy to punch in a Iew
numbers and operations and to read the answer on the display. As inexpensive and small mobile devices, these are typically more limited in:
1. memory (limits to computation size and to built-in Ieatures)
2. energy consumption (slower processor speed)
3. Iorm Iactor (keyboard and display too small Ior more than casual technical use)
Typically calculators are not networked, making it harder to share results to transIer them elsewhere to continue the work.
Table 1.2: A high-end calculator in 2009
The TI-Nspire with CAS is a recent generation calculator Irom Texas Instruments. It can do numerical
and symbolic calculations, as well as graphing. According to ticalc.org (http://www.ticalc.org/basics/
calculators/ti-nspire-cas.html) it has 16Mb memory, 20Mb storage and has a 150MHz processor. This
makes it have about 100 times less memory, 8000 times less storage, and makes it run twenty times
slower than a low-end laptop with a dual core 1.5GHz processor, 2Gb memory, and 160Gb disk. Its
screen is 240 x 320 pixels, giving it about twenty times less display area than a typical laptop. This is one
oI the Iirst generation oI calculators with limited wireless networking available as an add-on.

See http://education.ti.com/educationportal/sites/US/productDetail/usnspirecas.html?
subid1&topid350.
4 1 Introduction -- Technical computing in the 21st century

Smartphones, personal digital assistants (PDAs), media plavers, and tablets can have processing capabilities almost as powerIul as small
personal computers. One advantage that these kinds oI devices hold over calculators, is that they are typically networked so that it's possible
to use them as a Iront end to a more powerIul computer somewhere else in the Internet "cloud" oI computational resources. Mark Dean oI
IBM, part oI the engineering team that developed the original PCs Ior that company in the '80s, has written that the wave oI popularity Ior
personal computers has crested, "they're going the way oI the vacuum tube, typewriter, vinyl records, CRT and incandescent light bulbs." (http://
asmarterplanet.com/blog/2011/08/ibm-leads-the-way-in-the-post-pc-era.html). However the limitations oI the small display and keyboard with
current mobile devices will continue to constrain the appeal oI such devices Ior extensive technical work Ior the near term.
A high-end tablet in 2011
The iPad2 is a recent generation networked mobile device Irom Apple. According to http://
www.apple.com/ipad/specs/ and http://www.engadget.com/2011/03/09/ipad-2-review/, it can be
conIigured with 512Mb memory, 64Gb storage, and a 1GHz dual core processor. Its screen is 1024 x 768
pixels. This makes it slightly less capable than a low-end laptop with a dual core 1.5GHz processor, 2Gb
memory, and 160Gb disk.

Photo Irom http://4.mshcdn.com/wp-content/uploads/2010/01/ipad-3g.jpg
Dedicated controllers
Even some kinds oI toasters have microprocessors in them nowadays. "Smart homes" (see Ior example http://home.howstuIIworks.com/smart-
home.htm and http://www.drexelsmarthouse.com/) may network appliances and home Ieatures such as HVAC and lighting control. While these
devices are merely conIigured rather than programmed by end users, students oI technical computing should not Iorget that this is another place
where programming is necessary. The economic reasons Ior switching to computer control oI devices appear to be the greater Ilexibility and
variety oI control that can be developed at modest cost through programming Ior many kinds oI devices. The kind oI programming done Ior
device control oIten has a mathematical basis. Although the processors in dedicated devices are typically a Iew orders oI magnitude slower
than those oI personal computers or even smartphones, the programming languages used Ior them are oIten the same or similar to those used in
personal computing.
1.5Maple, a system for technical computing
In this course, you will learn how to do various kinds oI technical tasks using Maple. The Iirst version oI Maple was developed in the early 1980s
at the University oI Waterloo in Ontario, Canada, but has since undergone many reIinements and extension. (In Fall 2011, we will be using Maple
15.) Maple's original emphasis was on algebraic computing, so it was used by those who needed Iormulas as results Ior their work. In the early
'90s it was used extensively in calculus classes as a way oI supporting exploratory experimentation and more extensive experience with "applied"
problems. More recently it has expanded its domain into the symbolic, numeric, and graphical calculations done more generally in scientiIic and
engineering modeling and simulation.
Today, Maple supports numerical computing and graphical visualization about as well as it does symbolic computing. It has its own programming
language but also has ways oI doing calculations through the graphical user interIace (GUI) that is a kind oI augmented "point and click". The
deIault way oI interacting with Maple also allows you to mix documentation and computing instructions and results. Thus it's particularly easy in
Maple to produce documents that produce a well-documented solution -- a description oI the problem, the explanation Ior how to solve it, and the
computations that produce the details oI the answer and evidence to justiIy its correctness.
1.6What about Systems X, Y, Z, ...7
There are a variety oI systems and programming languages used Ior technical computing nowadays. In the TIOBE Programming Community
Index Ior August 2011, Java, C, C, PHP, C#, Objective C, Visual Basic, Python, Perl and JavaScript are, in that order, the ten most popular
programming languages (see http://www.tiobe.com/index.php/content/paperinIo/tpci/index.html). In a large technical establishment such as
1.7 Why pick Maple as the Iirst system? 5
a university or research lab, one might Iind, in addition to the generally popular languages and Maple, systems such as Matlab, Mathematica,
Octave, Macsyma, Sage, Axiom, REDUCE, Mupad, and Fortran. Maple and other "niche" languages are used despite the much greater popularity
oI the "top 20" languages because they specialize in addressing tasks common in technical Iields.
The vast array oI tasks computers can be used Ior, the multiplicity oI philosophies oI system design, and the rapid and unrelenting nature oI
progress in computing means that there is no "best system" Ior any situation. Each system has its technical strengths. There may also be historical,
cultural, or economic reasons why certain systems are used in certain places which co-exist with the technical justiIication. As with all popular
and demanding activities, there are many products in use.
One important ramiIication oI this is that students in technical Iields must expect to become Iamiliar and proIicient in more than one
programming svstem. Studying more than one brings greater knowledge oI diverse ways oI achieving a goal. It also allows one to Iunction
eIIectively as a team member in larger cross-organizational eIIorts and to better to cope with the diversity in computing cultures that's out there.
1.7Why pick Maple as the first system7
1. It is an interactive system, facilitating quick exploration of new ideas. Compared to languages such as C or Java, one can immediately
start up an interactive system and calculate results through "point and click" and a little typing. It's also possible to enter computational scripts,
which are sequences oI steps that are less elaborate to set up than programs. The scripts can be easily re-run on variants oI the original situation
just by changing a line or two in the script. This allows convenient "what-iI" exploration, where a number oI diIIerent scenarios are explored
through computation. The worksheet interIace to Maple allows sophisticated mathematical typography and graphics as well as mouse- and
palette- driven input.
2. It can handle calculations with formulas. Rather than figuring out the formula yourself, you can even get Maple to calculate the
formula for you. While many mathematical calculations produce numerical results, Iormulas are needed to speciIy how those results are
produced. Designers or developers oIten need to Iind and create the Iormulas rather than to just copy them Irom a book. A system where Iormulas
are easy to represent and easy to create supports this kind oI work. A system that allows you to represent and calculate both numbers and
Iormulas makes it easier to do and document technical calculation than a system that just works with numbers.
3. It supports a variety of data structures that support technical computation: Iormulas, equations, Iunctions, sets, lists, tables, vectors, and
matrices. Rather than having calculations done with a number oI digits that's Iixed Iorever, it is easy to change the precision iI more (or less)
precision is needed. Having higher-level interaction with the computer is usually more productive oI human time. Having all oI these entities as
"Iirst class entities" in the system requires less mental eIIort by the user, since they do not have to translate what they are thinking about (e.g. a
table oI Iormulas) into the terms that the computer is using. Rather, the computer language accommodates the human style oI thought.
4. It supports documentation as well as calculation. From the instructor's point oI view, it's easier to create documents that explain Maple
because we can use Maple both to handle the calculation and the presentation oI results. For students, having a Iile with both the directions that
cause the computation and the results makes it easier to present the work cleanly Ior grading, and Ior Iuture reIerence. For proIessionals doing
technical work, having an integrated environment where text, programming and results can be combined together can be a convenience.
5. It has a "conventional programming language". An objective oI this course is to make you become Iamiliar enough with some oI the
standard elements oI programming (e.g. assignments, conditionals, loops, procedure deIinitions) so that you can use them creatively to handle
certain common situations that can't be handled through point-and-click or built-in operations. These elements will be Iound in highly similar
Iorm in the languages used by many other systems used Ior technical work.
6. The mathematics of modeling and simulation is an explicit feature of the language. While its programming language is conventional,
Maple's language has an important added plus -- you can work over expressions, equations, Iunctions, and other mathematical objects just as
conveniently as iI they were numbers or characters. Conventional languages (e.g. Java) can represent arithmetic operations -- multiplication,
addition, subtraction, and division -- but aIter that there are special tricks and conversions that you must perIorm to bridge the gap between what
is written in the program, and the mathematical ideas that you are trying to use in the model. It's more straightIorward in Maple to represent the
mathematical model and the computation based on it. This ease oI expression and comprehension by programmers has a hard-headed dollars and
sense payoII. Less programmer time can be spent developing a computation. There is also an eIIiciency advantage to explicit representation oI
Iormulae: it is possible to use systems such as Maple to automaticallv improve the quality and eIIiciency oI simulation calculations. This again
leads to lower costs oI doing the work.
We think these things provide a soItware engineering advantage that will lead most technical computation systems to eventually have such
Iunctionality built-in into them.
6 1 Introduction -- Technical computing in the 21st century
1.8Using more than one system
Any user oI computers who expects to use them proIessionally Ior design and investigation must expect to eventuallv learn multiple svstems.
Using computer applications Ior work is like using tools in a workshop-- you would not expect to use one tool to do all tasks, even iI the tool, like
Maple, has "Swiss Army KniIe" capabilities. Features that make it easier to do certain kinds oI things may slow down doing other things. Even
when systems overlap in Ieatures, they typically have diIIering philosophies and diIIerent technical strengths, which means that certain kinds oI
work may be signiIicantly easier in one system than another. For example, developing something in Mathematica or Maple may be Iine and quick
Ior a personal computer, but making the same programming work on a supercomputer may take a lot oI eIIort in a diIIerent language. Yet a work
environment with multiple languages need not be overwhelmingly complex. Most systems with major development eIIort behind them (such as
Maple and those mentioned in the "section above) have many similarities.
What makes things work out is this: at the introductory level, the diIIerence between casual computing and proIessional technical computing is
the style oI working (higher emphasis on documentation, justiIication and ease oI reuse), and the use oI language-based commands/programming
needed to do the more sophisticated operations in technical work. "Crossing over" to the proIessional mode oI operation means getting over the
hurdles oI learning the new style oI work, and learning how to interact with computers in a typical computer language. Once this hurdle is passed,
it should take only incremental eIIort to acquire expertise in the second, third, or th technical system. Having Iormal instruction on the Iirst
system should provide an explicit introduction into the concepts and the work processes to Iacilitate this.
Most systems realize that they cannot be the sole provider oI technical computing services. II Institution A uses Mathematica and Institution B
uses Python Ior their work, then iI they expect to use each other's eIIorts, there has to be a way oI interconnecting programs written in one system
with that oI another. Thus most systems have interconnections. For example, the Matlab Symbolic Toolkit allows Matlab users to call Maple to do
Iormula manipulation. Similarly, Maple users can link to Matlab and run a Matlab program they've gotten Irom a colleague as part oI a problem-
solving process written in Maple.
Knowledge oI basic programming and the concepts oI software development make it possible to switch between systems with only a modest
amount oI additional eIIort. SoItware interconnection allows one to reuse programming done in another system without having to translate into
another language. Symbolic computation systems like Maple also have the additional bonus oI being able to translate some oI their programming
between languages. There is a "convert-to-Matlab" Ieature Ior example, or a "convert-to-C" Ieature Ior computations involving just numbers or
text.
7
2Getting started with Maple's Document Mode:
doing technical work with a clickable interface
2.1Chapter synopsis
1. How to start up Maple and perIorm simple calculations, algebra, and plots within it.
2. Detecting and Iixing typographical mistakes.
3. Detecting mistakes caused by misunderstaplondings and incorrect decision making by the user.
4. How to save Maple work so that you can reIer to it or resume working on it later.
5. How to recover a Maple worksheet iI it or your computer crashes.
2.2Starting up Maple, getting a fresh start
Start up the Maple application (this varies on the type oI computer system you have, typically it involves clicking or double-clicking on the
Maple 15 icon, but iI you can't Iigure it out yourselI ask Ior a demo Ior someone who knows). Once the Maple application window appears, a new
document will appear in the main working area oI the Maple application. A Ilashing cursor will appear with the outline oI a small rectangle with
dashed lines. The entry mode will read "Math" and "2D Math". You can make the "quick help" black box disappear by clicking on the "close box"
circle-X in the upper right hand corner.
8 2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace
Table 2.1: Maple started up with new document in Windows XP
AIter you close the quick help box, you will see the Maple cursor in a small rectangle with a dashed line outline
2.2 Starting up Maple, getting a Iresh start 9
Table 2.2: Maple document with first entry area
At this point, what you type will appear in the small rectangle and be regarded as a mathematical expression. In the next section, we describe
what to type in order to get something useIul to happen.
10 2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace
2.3Evaluating an expression involving exact arithmetic
Grade school arithmetic
In the math area, type 2 + 3. "2 3" is regarded as a mathematical expression by Maple when you type it into the dashed rectangle. As you are
typing, the input indicators should say that you are entering Math mode, in "2D Math" input, using "Times New Roman" Iont:
Table 2.3: Maple input using 2d math
This expression should show up in the work area. When you hit the enter key, then Maple will evaluate the expression. AIter the expression is
evaluated, you should see the result displayed below the input, as in the Iigure below:
Table 2.4: Maple input with labeled result
Maple has automatically calculated the answer and given it a label (1). AIter Maple calculates the answer, the cursor should appear below the
result in another dashed-line rectangle. This indicates that Maple is ready to do another calculation.
Maple supports all the basic arithmetic operations in a Iashion similar to many other programming languages. One thing that takes some getting
used to is that * (asterisk) is used to input multiplication rather than "x" or a centered dot. Another thing that takes getting used to is that what you
type is Iormatted to look like math notation. Thus iI you type a /, Maple understands that you are talking about division and immediately starts
Iormatting your input as iI it were a Iraction. II you type an asterisk, Maple Iormats that into a dot
().
There is also Iormatting that occurs with
caret (`) since that is the way you enter an exponent (power) in Maple.
Table 2.5: Arithmetic Operations in Maple
Operation Character to type/
character name
Notes Example
addition
2.3 Evaluating an expression involving exact arithmetic 11
(2.1)

multiplication* ("asterisk") Typing an asterisk makes a center
dot( ) appear in the displayed
expression.

(2.2)

division / ("slash") Typing a slash draws a baseline
and then positions the cursor in the
denominator. Subsequent typing
appears in the denominator. To get
out oI the denominator and return to
normal typing, use the right-arrow key
().

Multiple divisions are by deIault
conducted leIt-to-right.

(2.3)
(2.4)
subtraction - ("dash" or "hyphen",
typically on the same
keyboard key as the
underscore. Dash and
underscore do not mean the
same thing.)
Multiple subtractions are conducted
leItmost Iirst.

(2.5)
(2.6)
parentheses ( , ) ("leIt parenthesis", "right
parenthesis")
Use parentheses to change the order
oI calculation. They are also good Ior
removing any guesswork by the reader
as the order oI operations.

(2.7)
(2.8)
negation - ("dash" or "hyphen",
typically on the same
keyboard key as the
Put a dash in Iront oI a number or
parenthesized expression to negate it.

12 2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace
underscore). This is the
same symbol as used Ior
subtraction
(2.9)

power ` ("caret", typically on the
same keyboard key as the
number 6)
Typing a caret moves the cursor to the
exponent position. Subsequent typing
appears in the exponent. To get out oI
the exponent and return to the baseline,
use the right-arrow key ().

(2.10)
(2.11)
(2.12)
(2.13)
Iactorial ! ("exclamation mark",
typically on the same key as
the number 1.)
is the product oI all the integers
between 1 and n. It is useIul in
computations that compute the number
oI possible ways that something could
happen. For example, the number oI
possible orderings oI a deck oI playing
cards is 52!

(2.14)
(2.15)
(2.16)

Let's try out some oI these operations. We can enter a sum oI two Iractions by using "" and "/". II we type "2/35/6 enter", We intWthis is
what we see:
(2.17)
2.3 Evaluating an expression involving exact arithmetic 13
The way to get a Iraction in is to type a slash (/). As soon as you do so, Maple draws an underscore and positions the cursor underneath
the Iraction line. The next characters you type appear as the denominator. II you type the "" right aIter the "3", the plus will appear in the
denominator which is permitted by Maple but not what we want in this situation. To get the "" to appear outside oI the Iraction, we type the right
arrow kev (the key with on it). This moves the cursor out oI the Iraction back into the baseline oI the expression. Then we can enter the Ior
addition, and another Iraction. AIter we hit the enter key, Maple will simpliIy the result into a single Iraction with any common Iactors removed
Irom the numerator and denominator
Now let's do a multiplication. The Maple programming language (like most) uses an asterisk * as the symbol Ior multiplication. However, Maple
displays the expression with a centered dot. This may be disconcerting -- what you type is not what you see. When you are in "math mode" Maple
will be using Iancy typography to display whatever math you are entering. We'll see more oI this shortly. See iI you can reproduce this result::
(2.18)
We can mix operations. Try to enter and calculate the Iollowing:
(2.19)
In order to get that last denominator, we had to select the expression we had entered Ior the numerator with the mouse, so that the entire contents
oI the entry rectangle were blue. Then we typed a slash and the denominator appeared beneath it all.
An alternative to using the mouse to enter expression (1.2.1.18) would be to use parentheses. II we type "(12/345*67)/8 enter" we will
see this:
(2.20)
This allows you to enter complicated expressions without having to use the mouse. OI course, the mouse is still necessary iI you want to go back
and edit.
We observe in passing that a distinctive Ieature oI Maple is that Maple does exact arithmetic with integers and fractions. It keeps Iractions as the
ratio oI two integers. It will, however, automatically simpliIy such ratios to lowest terms:
(2.21)
Diagnosing typographical mistakes
Making mistakes is a normal part oI using any tool, be it a computer or otherwise. You'll probably make as many mistakes learning how to use
Maple as you would make when learning a new sport, a musical instrument, or when learning how to write a good essay. Some oI them will be
obvious as soon as you make them, others will be subtle or harder to Iigure out how to Iix.
When you make some kinds oI mistakes, the computer may give you an error message For example, iI you make a typo and Maple doesn't
recognize what you enter as being a valid command, it will complain. Here are some typical error messages. At this stage, some oI the messages
14 2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace
will make sense and you can react appropriately. Others will use vocabulary that is unIamiliar. For those, the best thing to do is to have a clear
idea oI what you want to enter (by closely imitating examples that are known to work) and checking careIully that what you have typed was
accurate.
Table 2.6: Examples of Maple error messages

Error, invalid sum/diIIerence
We intended to enter "2 4" but Iorgot to type the "4" beIore we hit enter (return). The appropriate thing to do here is to correct
the expression and hit enter again.

(2.23)

Error, missing operation
This time we mistakenly enter the expression with the symbols in the wrong order. Maple complains that it expects some
operation to be entered between the two numbers but there isn't one.

Error, invalid matrix/vector product
We intended to enter "24" but typed a period instead oI a 2 by mistake. Even though we are trying to do the same thing, the
error message is diIIerent because a diIIerent symbol (the period) in this context suggests to Maple that we are trying to do
linear algebra. The appropriate thing to do here is to correct the expression and hit enter again.

(2.26)

Error, numeric exception: division by zero
2.3 Evaluating an expression involving exact arithmetic 15

II we ask Maple to do an impossible operation, it sometimes gives an error (depending on the operation). The appropriate
question to ask yourselI here is "what should I be dividing by instead oI zero?".


Error, unable to match delimiters
We started a sub-expression with a parentheses but Iorgot to Iinish it. In Maple, a delimiter reIers to a parenthesis -- ( or ) -- a
bracket | or |, or a brace or }. Delimiters are symbols that mark the beginning and end oI an expression. In many instances
they are necessary to unambiguously indicate meaning. For example evaluates to 40, where as the expression
without parentheses means 20 because multiplications are always done beIore additions unless the parentheses
indicate otherwise.

Error, unable to match delimiters
This is another instance oI the same mistake. We wanted to enter but Iorgot a parentheses.

Error, invalid sequence
We intended to enter "13" but typed the extra comma in by mistake. We get an error message that talks about sequences,
a concept in Maple that we haven't discussed yet (that's coming in the next Iew chapters). Maple thinks that the sequence is
"invalid" because it usually expects commas to be between items, such as "1,2,x".
Maple's language has many other elements in it that we haven't gotten to discuss yet. II you use any oI them by mistake then you
will see messages with vocabulary we haven't discussed yet.
It's a Iairly typical experience Ior new users to see some messages that you won't be able to gather much intelligence Irom other
than the Iact that you made a mistake that you should Iix,. With knowledge-rich systems such as Maple, you may blunder into
sections oI the system that you haven't learned yet. The best strategy is to back out oI the situation by editing the expression so
that it is exactly like something that is known to work.
16 2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace


Error, invalid Iraction
This one is Iairly obvious. In order to Iix it though, we need to know what numerator we intended to enter. We can Iix this by
editing the expression with the missing numerator and hitting enter again, as below:


(2.29)

MaplesoIt, the producer oI Maple, provides an on-line list oI all the errors in Maple 15 at http://www.maplesoIt.com/support/help/category.aspx?
CID2544. The leIt column oI the page lists all messages. Clicking on a particular message link will produce an explanation that is oIten helpIul
in Iiguring out what has caused the message.
An on-line list of error messages at http://www.maplesoft.com/support/help/category.aspx?CID2544. Clicking on a
message in the left column will display a page about that message.

2.3 Evaluating an expression involving exact arithmetic 17
Correcting typographical mistakes
The standard procedure Ior Iixing a mistake is as you would in a word processor: and re-execute the
computation.
1. Using the mouse, position the cursor where the mistake is. Then use the backspace key to erase the characters you want to get rid oI. Type in
more characters to replace it..
2. Use the leIt arrow key () to back up. Typing aIter backing up then inserts the new typing at the point where the cursor was positioned.
3. Use the mouse or other "pointing device" oI your computer to select a section oI what you typed. New typing then replaces the selection oI
what you typed.
4. Use the mouse to select a region, then "cut", which you can do through the Maple menu Edit -~ Cut. OI course most people use the keyboard
shortcut Ior cutting, which Ior Windows or Linux is control-X while on the Mac it's command-X.
5. Copying and pasting (control/command-C and control/command-V) also works in Maple.
18 2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace
You may Iind that sometimes you attempt to create a 2D Math input area but Maple does not compute a result Ior the input aIter you hit the enter
key. To create a "clickable math" input area iI this happens, place the cursor where you want the input area to be, and use the Format-~Create
Document Block Menu item oI the Maple window:
Table 2.7: Create Document Block to force a Math input area wherever the cursor is placed
Exponentiation (powers). Numbers with lots of digits
Use a caret ` to speciIy an exponent (a "power"). In math mode, Maple will position the cursor so that the next things you enter will become the
exponent. As with Iractions, you can get out oI the exponent by using the right arrow key when you want to go back to non-exponent numbers.
Try entering these expressions
(2.30)
(2.31)
2.3 Evaluating an expression involving exact arithmetic 19
We note that Maple does integer and Iraction operations exactlv. It will not introduce any rounding error into a computation as a calculator would
when the answer requires more than ten decimal digits to write down.
There are limits to the number oI digits Maple will use Ior integers or Iractions, but they have to do more with limits oI the computer hardware
and memory rather than a "pre-ordained" decision about how many digits might be useIul to keep. II you type kernelopts(maxdigits) into a Math
input area on the worksheet, Maple will print out a number which is the maximum number oI digits it can handle in any integer or Iraction. On
the author's computer in 2011,
. This says that the largest number that can be computed by Maple has over 268 million digits.
Even iI we wanted to use the answer, we wouldn't want to look at
For example, Maple can compute the result oI exactly (try it!).
Exact computation is useIul not only Ior doing algebra but also Ior things such as computing probabilities through counting, or in deriving
mathematical Iormulas Ior use in simulation or prediction programs.
Detecting and fixing semantic and "logic" mistakes
There will be other kinds oI mistakes where there is no error message, but the response is not what you want. Sometimes this happens because
you say something that while grammatical, means something entirely diIIerent Irom what you meant. This kind oI mistake is called a semantic
error. This could be as simple as entering "2-3" where you meant to type "23". Sometimes it is less obvious, such as mistyping the 23-rd digit oI
a 55 digit number or the 12th term oI a long sum. Or worse yet, it could be because vou are using the incorrect vocabularv so what you think you
are saying does not have that meaning to the computer.
Table 2.8: Example of a vocabulary mistake
(2.32)
Suppose we were under the (mistaken) impression could use "x" in Maple to stand Ior multiplication. We might use the above as
a way to calculate "two times three, plus Iive". However, while there is no error message, what is calculated is not the number
we were expecting. The Iirst step towards proceeding to Iix this is to realize that it's not what we want. To Maple, this is a
legitimate calculation -- you want to create a Iormula that's "two times the variable x3, plus Iive".


(2.33)
Knowing that the proper way to enter multiplication is through a palette, or symbol "*" (asterisk) as explained in XXXX.

(2.34)
20 2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace
(2.35)
(2.36)

Maple allows you to write any oI these things, but they mean diIIerent things to it. Maple's grammatical rules requires you to
speciIy multiplication by use oI the explicit mark Ior multiplication, or the use oI the Palette item as described in the
section "The Expression Palette" later in this chapter. The use oI "x" or just juxtaposition doesn't mean multiplication in Maple
-- these are not errors, they mean something else other than multiplication (Iunctional composition, to be precise, although Ior
our purposes it's easier to just think "doesn't mean multiplication").
Finally, there are mistakes made because vou ask Maple to do the wrong calculation. Even though Maple does what you want, it turns out that
what you wanted was wrong!
For example, you may read a word problem and decide to solve the equation whose solution is x4/3. But when you plug the
solution into the circumstances oI the problem, you Iind that it doesn't solve the word problem. The problem may be because your decision that
was the equation when it should have been This is known as an "error in logic" or just a "logic error".
Sometimes it's obvious that you made a mistake because the answer is obviously wrong. For example, seeing "-1" as the answer to "how many
eggs did the chickens lay in three days?" is obviously wrong since there's no such thing as "negative eggs". Sometimes you can plug your solution
into the original situation to see iI it works asit should.
2.4Saving and retrieving your work
You can save your work in a Maple worksheet Iile through the File -~ Save (keyboard shortcut: control/command-S) menu item oI the Maple
application. A dialog box appears allowing you to type in the name oI the worksheet (we've typed in "test" into the box in the .)
The Iile is saved with a ".mw" suIIix (e.g. test1.mw).
To retrieve the Iile in a subsequent Maple session, start up the Maple application as beIore and then use the File-~Open menu item to initiate the
dialog that retrieves the Iile.
2.4 Saving and retrieving your work 21
Table 2.9: Maple save menu operation
22 2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace
Table 2.10: Maple save dialog box
2.5Retrieving backups
The "state oI Maple" display appears on the bottom leIt hand side oI the Maple window. Usually this display says "Ready", which means
that Maple is awaiting your next command. Sometimes it reads "Evaluating..." which indicates that Maple is actively computing an answer.
Sometimes it says "Autosaving worksheet", which means that it is saving a copy oI the present state oI your open worksheets into temporary
storage on your computer. The amount oI time Maple spends autosaving becomes noticable in longer Maple sessions when the worksheet
contains a lot oI results.
Should your computer suddenly lose power or should Maple crash, you can retrieve the last autosaved worksheet by selecting the File -~ Recent
Documents -~ Restore Backup menu item. This will Iill your Maple with copies oI all the autosaved worksheets. You may then delete them or
save them to permanent Iile space as you wish.
2.6 Algebra, plotting and mouse-clickable operations 23
Table 2.11: The Maple state display
2.6Algebra, plotting and mouse-clickable operations
Algebraic expressions and equations. Solving equations. Working with
pieces of expressions.
We don't have to limit ourselves to just numerical calculation in Maple (even iI the ability to use as many digits as we wish and exact Iractions
allows us to do arithmetic more like the math books do). Another distinctive Ieature oI Maple is that you can do algebra by entering expressions
with symbols -- the x,v,:, i, and n that we see in algebra books. Maple will automatically collect terms and do some simpliIications Ior us
automatically
(2.37)
We can even enter equations:
(2.38)
(2.39)
Note that while Maple automatically collected the x terms on the leIt hand side oI the equation, it does not try to do the more interventionist
operation oI moving all the x terms to the same side oI the equation.
Now, enter the Iollowing expression, but rather than hitting the enter key aIter you've entered it, do a right-click. On the Mac, instead oI right-
clicking hold the control key down then click on the mouse button (this is reIerred to as "control-click".) A menu oI algebraic operations will pop
up. Select Factor and see how Maple can Iactor the polynomial:
24 2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace
Note that this line does not have a (XX) label Ior it.
To Iurther demonstrate the right-click (control-click) operations available, enter the Iollowing equation. Right click on the expression and a pop-
up menu should appear that includes a "solve" item. Select the solve and a submenu will appear where you can speciIy that you want to solve Ior
x.
For those with previous experience on other systems: some things are
different, for a reason
(This section is Ior people who already know a programming language and are noticing that Maple does some things diIIerently. Other readers
can skip this section.)
Most mainstream programming languages work in a similar way. That is why learning Maple has value beyond just being able to use the Maple
system -- once you learn Maple, learning Matlab or Java or C is a matter oI incremental adjustment.Noticing the diIIerences is educational in that
it makes you become aware oI the arbitrariness oI some oI Ieatures oI languages. II you have used another programming language such as Java or
Visual Basic (VB), you will notice that some things do not work the same in Maple as they do there. Some oI this can be explained iI you know
something about the history oI programming languages. Some oI it is explained by being aware that the designers oI Maple had diIIerent goals
than those who invented Java or VB.
One thing that you have undoubtedly noted is that symbols do not have to have a value associated with them. For example, in Java iI you said
k5;
Then iI you were to create another expression in Java such as System.out.println(k^2 + k + k + 3); then "5" would be used
as the value oI k in the expression and you would end up printing 38. In Maple, you do not have to associate k with a numerical value beIore
you use k in an algebraic expression. II there is no prior association, Maple just treats the expression as a Iormula with symbols in it. It may do
some algebraic simplification on what you entered, but it does not need to get a number as a result. Since there was no prior assignment Ior the
expressions in section 2.5.1, the calculation done with them just keeps the Iormula..
Another thing that is diIIerent is that in Maple "" is used for equations, not assignment.. The operator in Maple corresponding to "" in Java or
VB is ":" (a colon immediately Iollowed by an equals, with no spaces inbetween). In Maple, iI we wanted to associate "5" with the symbol k,
then we would do:
(2.40)
(2.41)
People who know more than one programming language have a better understanding oI which Ieatures are change a lot between languages, (such
as whether or : is used Ior assignment), and which ones are Iairly uniIorm ( being used Ior addition, or the use oI parentheses in Iunctions
and expressions).
Maple does not use "" Ior assignment because, being a mathematically oriented language that can handle algebra, it wants to make the entry oI
equations a natural thing. Its use oI ":" Ior assignment is a Ieature borrowed Irom the Algol/Algol 68/Pascal Iamily oI programming languages,
which picked this operator to make it clear that the assignment operation is diIIerent Irom algebraic equality.
2.6 Algebra, plotting and mouse-clickable operations 25
Is ":" better than ""? That's a kind oI question that is about as hard to answer as trying to decide whether "uno" is better than "one". II one had
a language where you had to do ":**&`#*" instead oI "" or ":", you could be critical oI the choice because it takes
much more eIIort to enter a 35 character operator than a one or two character one. But the Algol-Iamily choice oI ":" has reasonable motivation
-- studies oI novice programmers have shown that beginners using languages where "" is the assignment make more mistakes because they
conIuse its use in mathematics with its use in programming. Novices have been observed to write things like "5k" which does not work as an
assignment, even though mathematically the equations "k5" and "5k" mean the same thing.
Just as with architectural design oI a building, each Ieature oI a programming language is typically careIully considered. Many Ieatures
are borrowed or copied Irom predecessor languages, where they have already been subjected to the test oI many people using the Ieature.
II you are not Iamiliar with Algol Iamily languages, you can see examples oI them at various educational web sites, such as http://
www.engin.umd.umich.edu/CIS/course.des/cis400/algol/average.html and http://portal.acm.org/citation.cIm?id154766.155365.
Plotting, approximate numerical solutions through cursor position
The right-clickable interIace can also activate a plot oI a Iormula you may have entered. Enter this Iormula, then right-click and select Plot -~2d
plot. The automatic deIaults Ior plotting this produce this result.
Table 2.12: Example of Plotting

26 2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace
Table 2.13: Plot created by right-click -> Plot -> 2DPlot (with cursor positioning)

The user has clicked on the plot and positioned the cursor at the coordinate (-4.12, 61.60). The cursor was not captured by the
screenshot although it is visible under ordinary use.

The 2DPlot operation makes pre-set decisions about the plot, such as the range oI (-10 to 10), the color oI the line, axes labelling, etc. Users
can inject their own preIerences about these things about this by selecting right-click -~ Plot -~PlotBuilder and Iilling in the dialog box with
their choices.
2.6 Algebra, plotting and mouse-clickable operations 27
Table 2.14: User-configured plot using PlotBuilder instead of 2DPlot

Table 2.15: PlotBuilder Dialog box



28 2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace

The Expressions Palette and the Common Symbols Palette: entering Trigs,
logs, roots, exponentials
It's possible to get the common Iunctions oI high school algebra, pre-calculus, and calculus by using the Expression palette oI the Maple Window:
Table 2.16: The Expression palette


For example, to enter the square root oI 36, click on the palette entry Ior That expression will appear in the document, with the "a"
selected. II you then type 36 on the keyboard, that number will replace the selected text. II you then hit the enter key, Maple will evaluate the
expression and produce the exact result "6".
2.6 Algebra, plotting and mouse-clickable operations 29
(2.42)
You can use the palette multiple times, to create more complicated expressions. Just continue to the use mouse or the arrow keys to move around
in the expression. Selecting and typing, backspacing or deleting are all ways oI replacing or correcting pieces oI the expression.
The Common symbols palette, two panels below the Expression palette, can be used to enter a and , the base oI the natural logarithm system.
The palette does not have the inverse trig Iunctions, so you have to enter them through typing. Their names are: arcsin, arccos, arctan,
etc.
Table 2.17: Examples of palette-driven computation

(2.43)
(2.44)
(2.45)

Approximate numerical (calculator - type) arithmetic in Maple
II you enter expressions with integers, exact Iractions, and symbols such as a and . then Maple will perIorm exact calculations rather than give
approximate answers as a conventional calculator would. You can get approximations by selecting the "numerically solve" instead oI the "solve"
option Irom the right-click pop-up menu.
Table 2.18: Examples of computing with approximate solving

Exact solution oI an equation using the "solve" Ieature oI the
pop-up menu.
30 2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace



Using the "numerically solve" Ieature oI the pop-up menu
II you have an exact expression, you can ask Maple to approximate it to 5, 10, 20, or more digits. In this mode, Maple can be used as a super-
accurate calculator.
Examples of numerical computation
1. Enter Iraction, select approximate-~20 Irom right-click pop-up menu.
2. Enter exact expression, select approximate-~5 Irom right click pop-up menu
3. Enter equation. Then solve-~solve, then select Element-~1, then right hand side, then approximate-~10


Evaluation, and selection of pieces.
Sometimes you wish to evaluate an expression for a particular value of a variable. There is a right-click operation that does this.
Table 2.19: Evaluate at a point





This operation will give a pop-up menu that will allow us to choose values Ior all the variables.
In the Iirst example, we we picked 1/2 Ior a value oI x. Note that the pop-up menu will show
what you typed rather than displaying 2D math. In the second example, we speciIied "3*y`2" as
the value Ior a. In the third example, we picked 3 as the value Ior y.
2.7 A quick-reIerence summary to this chapter 31

Using the right-click menu, it's possible to select or extract a portion oI an expression Ior Iurther work.
Table 2.20: Operations on equations, multi-part expressions
right hand side, leIt
hand side





One oI the options in the right-click menu
is "right hand side". It only works Ior
equations.
Operations on multi-
part expressions


Example

select entry





Solving this quadratic equation reveals
that there are two solutions. Right-
clicking on the solution and then
selecting entry 1 gives the Iirst solution,
enclosed in brackets | |. Right-clicking
on that and again selecting entry 1 gives
the Iirst solution, without the brackets.
Right-clicking on that and selecting
approximation gives a calculator
approximation to the root.
2.7A quick-reference summary to this chapter
Example Description Notes
Arithmetic

(2.46)


Use , *, -, /, ` Ior arithmetic. Hitting the
Enter key produces a labelled result.
2 D Math input mode displays the
textbook-like version oI what you input.
Maple's simplification automatically
combined Iractions and places things in
lowest terms.
Use ! Ior Iactorial Do you know what 5!! (double Iactorial)
means?
32 2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace
(2.47)

Making mistakes

Error, unable to match delimiters


Error message mistakes (Irom typos or
mistakes in intensions)
The easiest ones to detect. You have to
Iigure out what you are doing wrong,
though. The error message may not
always be helpIul in advising you on this,
although it oIten is.
A Iarmer plants a Ience post every Ioot,
Ior 1250 Ieet. At that point, he switches
to planting the Ience posts every 4.7 Ieet
Ior another 940 Ieet. How many Ience
posts does he need in all?


(2.48)

Maple did the arithmetic in the above
calculation correctly. But it's not the
solution to the problem. Do you see how
to get the right answer?
"Logic errors" OIten you can Iind these kinds oI
mistakes by looking at simpler versions
oI the problem where the answer can be
Iigured out with paper and pencil. For
example, sit down and draw a picture oI
the posts and their positions when the
initial distance is 10 Ieet, and the second
distance is 9.8 Ieet. Count the posts in
your picture. Then you "scale up" the
calculation to handle the large situation.

It's easy to compute the wrong answer in
these kind oI "oII by one" situation.
Editing (fixing mistakes)
backspace, delete erase starting Irom
current cursor selection

Arrow keys move cursor within
current selection

Select with mouse/type replaces selected
text

Cut, copy and paste oI a selection works
as it does with a text processor

File saves, opens
2.7 A quick-reIerence summary to this chapter 33
Save Iiles with File -~ Save or File -~
Save As. Open a saved Iile with File -~
Open. Other File operations are similar
to that oI standard word processors.

Functions and math symbols

(2.49)


Insert math into an expression by using
the Expression Palette. You can enter
using the Common Symbols Palette.
(the natural logarithm base) can also be
entered this way. Note: typing e Irom
the keyboard does not enter this symbol.


Chapter 2 demonstrated the Iollowing
Iunctions and symbols
Square roots, n-th
roots
natural logarithms
(base )
Trig Iunctions: sin,
cos (trig Iunctions
Base 10 logarithms
all use radians, not
degrees)
arcsin, arccos,
arctan
sec,, csc
summation
a,




II you are entering a Iunction by the
keyboard rather than the palette, you
must enclose the Iunction's argument in
parentheses.

Algebra









Right-click (control-click on Mac) on
an entered expression to get the pop-up
menu.
Chapter 2 demonstrated examples oI the
Iollowing operations:
Iactor solve
solve numerically right hand side (oI
an equation)
leIt hand side (oI
an equation)
select (n-th part) oI
an expression
approximate
numerically (to 5,
plot (two
dimensional) --
many plot options
10, 20, etc. digits'
accuracy)
to determine range
and domain oI plot,
color, captions, etc.
evaluate at a point
(choose values

Ior variables in an
expression)

34 2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace
Plotting
Plots-~2d plot


The expression must be something that
involves a single variable and will result
in a number when a value is used Ior
that variable. Thus you can't do a 2d plot
oI because you wouldn't get a
number iI you picked a value just Ior x
(or just Ior a).

Maple uses deIaults Ior the plot range,
and the plot color.

Trying to plot an equation produces an
implicit plot (see next appendix).
Plots-~plot builder -~ 2d plot


A dialog box appears that allows you
the select Irom many more options, such
as plot color, the line style, vertical and
horizontal ranges, captions, etc.
Limited precision (decimal point)
numbers




Exact numbers in Maple have no decimal
points.
Symbolic constants such as a and
entered Irom the Common Symbols
Palette are also exact.
Use oI limited precision numbers in
algebra (e.g. Iactoring, diIIerentiation,
solving) may not produce good results.

2.7 A quick-reIerence summary to this chapter 35
(2.50)
(2.51)


Numbers with decimal points in Maple
cause arithmetic calculations to be done
approximately.

solve-~numerically solve produces
approximate solutions

.right-click-~approximate-~n takes
an exact numerical expression and
approximates it.

Use them in Maple only when an
approximate result is desired.

Numbers like .25 or .6015 are limited
precision. II you want exact algebra done,
use 1/4 or 6015/10000, etc.

In very large calculations, limited
precision calculations may be noticably
Iaster than those with exact arithmetic.
Most oI the time there isn't an appreciable
diIIerence.
Evaluate at a point Example







This operation will give a pop-up menu
that will allow us to choose values Ior all
the variables. In the Iirst example, we we
picked 1/2 Ior a value oI x. Note that the
pop-up menu will show what you typed
rather than displaying 2D math. In the
second example, we speciIied "3*y`2" as
the value Ior a. In the third example, we
picked 3 as the value Ior y.
Operations on equations Example
right hand side, leIt hand side




One oI the options in the right-click menu
is "right hand side". It only works Ior
equations.
Operations on multi-part expressions


Example

select entry
36 2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace



Solving this quadratic equation reveals
that there are two solutions. Right
clicking on these selections and then
select entry -~ 1 produces the Iirst
solution. We can then approximate it by
using the
Enter an expression in a document , then right-click (control-click on Mac)
followed by:
Commentary
Operations on
symbolic expressions
Example
solve-~solve


solve-~solve Ior a
variable



solve-~numerically
solve


The thing to try when there is a numerical
answer but the exact solution is too
complicated to understand or Maple can't
Iind an exact solution.
Factoring



Factoring can simpliIy an expression
sometimes. Factoring doesn't know the
trig simpliIication rules, though.
Plots-~2d plot

The expression must be something that
involves a single variable and will result
in a number when a value is used Ior
that variable. Thus you can't do a 2d plot
oI because you wouldn't get a
number iI you picked a value just Ior x
(or just Ior a).

Maple uses deIaults Ior the plot range,
and the plot color.
2.7 A quick-reIerence summary to this chapter 37

Plots-~plot builder -~
2d plot



A dialog box appears that allows you
the select Irom many more options, such
as plot color, the line style, vertical and
horizontal ranges, captions, etc.
38 2 Getting started with Maple's Document Mode: doing technical work with a clickable interIace
Operations on
equations
Example
right hand side, leIt
hand side





move to right,
move to leIt


This moves the entire side oI an equation
to the other side.
Operations on
constant expressions
Example
approximate-~5
(or 10, 20, 50)





Uses approximation methods to turn
constants into an approximate (decimal
point) number with 5 (10, 20, 50) digits.
39
3Chapter 3 Technical word processing
3.1Chapter Overview
In this chapter, we learn how to use Maple as a word processor. This allows us to write reports that combine technical writing with math
Iormulae, calculated results, pictures, tables, etc. Many oI the Ieatures are highly similar to MicrosoIt Word or similar WYSIWYG (what you see
is what you get) word processors. The strength oI Maple's word processing is that it makes it easy to enter technical Iormulae, and that the word
processing and calculation can be done in the same document.
3.2Maple as a word processor
Maple documents allow a mixture oI text and mathematics. By deIault, Maple expects that when you position the cursor by clicking somewhere
in the document, you will be entering math and be wanting it do to a calculation. The document is in what is called math entrv mode.
Table 3.1: Maple in math entry mode

You can tell whether the document is in math entry mode because the Math button on the Maple toolbar will be gray, and the "C" menu item says
2D Math.
40 3 Chapter 3 Technical word processing
Another mode oI operation Ior Maple documents is text mode. When in text mode, Maple has the behavior oI a word processor. It just shows what
you typed. Hitting enter while you are in text mode just causes text entry to move to the next line. It does not cause any calculation to be done
with what you typed.
You can switch to entering text in the Iollowing way:
1. Position the cursor at the spot where you want to enter text.
2. Click on the Text button on the Maple toolbar. This places the Maple document in text entrv mode. Alternative you can switch to Text mode by
typing control-T (on Macintosh, command-T) or by using the Maple menu bar Insert-~Text. You can tell when you've switched to text entry mode
because the Text button will be gray, and the "C" menu item says Text..
3. With the keyboard, enter your verbiage. When you are in text mode, you will also see that the menu bar will be enabled Ior boldIace, italics,
underlining, leIt-/center-/right- justiIied text, colored text, and colored backgrounds, and bulleted/numbered text.
4. To switch back to math,click on the Math button on the Maple tool bar. Alternatively you can type control-R (on Macintosh, command-R), or
use the Maple menu Insert-~2-D Math.
Table 3.2: Document after control-T (or Insert->Text)


3.2 Maple as a word processor 41
A Maple worksheet in text mode in OS X. Although it is hard to see, the cursor is positioned at top leIt oI screen.
You can do mathematical word processing without any computation by switching between text and math modes, using the Palettes to help you
enter the math. As long as you don't hit the return (enter) key, the math will not cause any calculation.
Table 3.3: Document with a mixture of text and math




The user typed the text, then went into math mode by typing control-R (command-R). They then entered the "F G..." Iormula
in math mode. The user got the subscripted by typing an underscore _ aIter the m to get Maple to descend to subscripts, then
used the right-arrow key to ascend back up to the main level oI the expression. The other symbols in the midst oI the rest oI the
narrative are entered in a similar way.


It is possible to mix text and the results oI calculations in a paragraph. Typing control- (command-) when the cursor is in a math expression
will cause Maple to print an "" and then the result oI evaluating the expression on the same line. This is an alternative to hitting the enter key
and allows those kinds oI calculations to be mixed with text.
42 3 Chapter 3 Technical word processing
Table 3.4: control- puts the results of a calculation in the midst of text

The user entered text, then did a control-R (command-R), then entered the math expression and then typed control-. AIter the
calculation result appeared, they typed control-T and entered the remainder oI the sentence.
3.3Shortcuts to entering math symbols
Using the Palettes, we can enter a wide variety oI mathematics -- expressions, math symbols, Greek letters (using the Greek Palette), arrows, etc.
There are additional Palettes not shown by deIault, which you can get by View Palettes Show All Palettes. However, you can enter many
symbols in math mode Irom the keyboard through "shortcuts". Most oI the shortcuts consists oI typing the textual name oI the symbol or some
abbreviation oI it, and then hitting the escape kev -- the key labelled Esc on many keyboards.
For example, to enter the symbol while in math mode, you can type infin and then hit the escape key. A pop-up menu oI choices will appear
to allow you to complete entry oI the symbol. With practice, this can be a Iaster way oI entering than using the Palettes.
Table 3.5: Keyboard shortcuts in math mode through the escape key
In math mode, we type infin.
AIter hitting the escape key, a menu oI completions appears.
We pick the Iirst alternative (either by hitting the return key or
by operating the mouse to select the Iirst option) and what we
typed is replaced by the selection.
Greek symbols can be entered by typing the romanized name oI the letter, Iollowed by escape. For example, in math mode, typing omega
Iollowed by escape produces . Typing Omega Iollowed by escape produces (the upper case version oI the Greek letter).
3.4 Other word processor Ieatures 43
A shortcut to entering the svmbolic constant e (the base oI the natural logarithm) is to type e, then hit the escape key, then return.
You can see a summary to all oI the conveniences Maple oIIers through Help Quick Reference .
3.4Other word processor features
Inspection oI the worksheet toolbar reveals many more word processing Ieatures: line justiIication, bold Iace and italics, numbered items, colored
letters or backgrounds, Iont sizes, and Iont types. The Insert operation on the Maple toolbar allows creation oI Tables and Images (graphics
Iiles). Rudimentary drawings can be inserted through Insert Canvas. We encourage you to explore and make use oI the Ieatures on your own.
3.5Troubleshooting word processing
A phenomenon that you may encounter is not being able to switch back to math mode from text mode, even aIter perIorming the operation that
should do so (clicking on the Text button oI the document toolbar, typing control-T, perIorming Insert 2DMath, etc. This may be due to the
worksheet losing track oI where you are in the document. A "sure-Iire" cure Ior switching modes is to position the cursor at the point where you
want to enter math, then do Format Create Document Block. A dashed box will appear at the location oI the cursor, indicating that it is again
in math mode.
Tools-~Spellcheck (alternatively, the F7 key) will run a spelling check on the non-math part oI your document.
3.6Summary of Chapter 3 material
Name Menu operation Key short cut
Important word processing operations in a Maple worksheet
Switch entry to 2D Math mode Insert-2D Math
Click on "Math" oval in menu bar just
below names oI worksheets.
control-R (command-R on Mac)
Switch entry to Text mode InsertText
Click on "Text" oval in menu bar just
below names oI worksheets.
control-T (command-T)
Use a keyboard shortcut in Math mode Type the shortcut, then hit the escape key.
For example, typing omega and then
escape will turn the text into . Typing e
and then escape will allow you to turn the
text into the symbolic constant without
needing the Expression Palette.
44 3 Chapter 3 Technical word processing
45
4Chapter 4 Assignment
4.1Chapter Overview
We learn how to label results with symbolic names through the operation oI "assign to a name", sometimes called assignment. This allows us
to reuse the results in subsequent steps oI a multi-step calculation without re-entering or copying it. Assignment causes a portion oI computer
memory to store the computed result, so that it can subsequently be reIerred to by name. Reassignment changes what's stored.
The keyboard operation : provides a keyboard shortcut Ior assignment. : will be used heavily in later work in programming as we shiIt Irom
mouse/menu operation to textual speciIication oI calculations.
unassign causes Maple to Iorget the association oI value with name. All assignments can be Iorgotten through restart.
4.2Assignment: remembering results for future use
We can compute a result and label it with name. This action is called assignment.. We can do this with the clickable menu by the action right-click
(control-click on Macintosh) assign to a name . A pop-up menu will appear asking us to Iill in the name that we want to use.
Once we have assigned a result to a name, we can use the name, and Maple will use the assigned value.
A name can be any sequence oI upper- or lower-case letters, digits and the underscore character _ . It must start with a letter. Maple distinguishes
between upper and lower case letters, so result and Result are considered diIIerent names.
In programming, the term variable is used interchangeably with name. Both reIer to an identiIier which the action oI assignment associates with
a computed result. Computer books oIten talk about "assigning the result to a variable" which means the same thing as "assigning the result to a
name".
Table 4.1: Assignment
(4.1)
(4.2)

AIter computing a result, we assign it to the name
trigResult by clicking on the result, selecting the menu
item assign to a name and then typing in the name.
(4.3)
(4.4)

We can then use the name instead oI repeatedly entering or
copying expressions.
4.3:= is a keyboard shortcut for assignment
The Maple operator : (colon immediately Iollowed by an equals) also perIorms assignment
46 4 Chapter 4 Assignment
Table 4.2: : is another way to do assignment of a name
Form Examples Commentary
name : expression whose value will be assigned x : 5 Assigns the name x the value 5.
Assigns the name polv the value consisting
oI whatever the assigned
value oI : is. II no value has been assigned
the name :, then the result is the algebraic
Iormula:
When you enter an expression in Math mode (even iI it's just a name, without any arithmetic), you ask Maple to evaluate what you have entered.
Maple also (beIore it does anything else) Iigures out the values assigned to names that appear in the expression. Then it does arithmetic, Iunction
calculation, and any other operations you've described in the expression.
The only exception to the "Maple always evaluates beIore assignment" occurs when with the name that appears on the leIt hand side oI :. That
name is not evaluated, it literally is used as the name assigned to.
Assignment really requires two steps. The Iirst is Iiguring out the result. The second is assigning the result to the name. The "Iiguring out the
result" step is called evaluation. II any oI the symbols in the expression being evaluated have an assigned value, then that value is used. .II those
values involve other symbols, those are in turn checked Ior values, etc..
Symbols without an assigned value have their own names as their value. This allows you to enter an expression such as in x
and use the x's as symbols in the normal mathematical style as long as you don't assign x a value.
II you use a name/variable in an expression, and it has no assigned value, then Maple uses the rule that the value oI an name with no assigned
value is just the name itselI.
To deassign a variable use unassign. For example, unassign(x) causes any value oI x have no assigned value, just as iI it had never been
assigned. This does not undo the eIIects oI other computations done while the assignment was in eIIect , as the Iollowing examples below
illustrate.
Table 4.3: Assignment
Examples of assignment with : Commentary

(4.5)
We assign the name p the value oI the expression. Note that
since x and a have not been assigned values, the results oI
evaluation just leaves them as symbols.
(4.6)

II we enter an expression containing p, its value is plugged in
Ior the calculation oI the result.


(4.7)
Here we assign the name x the value 3.

4.3 : is a keyboard shortcut Ior assignment 47
Examples of assignment with : Commentary

(4.8)
(4.9)

II we now do a calculation with p, the value oI x is used since
p's value mentions x. There may be a chain oI assignments that
Maple must look at to evaluate an expression.

We can solve the result 1.3.4 Ior a by right clicking that
expression.

(4.10)
(4.11)
(4.12)

We change the value oI x by assigning it a diIIerent value.


When we do another calculation with p, the most recent
assigned value oI x is used.

(4.13)

This a way oI assigning 4 -- the current value oI x to the name
v.

(4.14)

We can undo the connection between x and any value by
unassigning x. This operation produces no output, so no label.
We can barely tell that it has happened. The quote marks
surrounding the x -- 'x' are mandatory, otherwise x would
be replaced by its value and Maple would try to unassign the
corresponding value rather than x itselI.

p still has a value, but since x no longer has a value, we are
back to the original result.

(4.15)

We may have unassigned x, but y's unassigned value -- 4 --
is unaIIected. The assignment just connects the name and
the value determined when the assignment was perIormed
(back at (4.13)). The inIormation about which variables or
expressions were used to Iigure out what the value was is not
retained.
48 4 Chapter 4 Assignment
4.4The Variables Panel: a way to know the current value
of your variables
The operation oI assignment uses part oI the computer's memory to remember an association oI a name with a result. II you are doing several
assignments, you can see the current state oI the Maple session -- the associations oI names with values currently in eIIect -- by using the
Variables Panel. It will list all oI the variables assigned, and their present value.
A useIul way to think about assignment is to think oI the computer creating a memory slot containing the result, labeled with the name being
assigned to.
Table 4.4: The variables panel shows currently assigned variables
AIter assignment,
the Variables
Panel shows that
the variable p
is assigned an
expression.
II we then assign x
a value, then there
are two variables
listed. Notice that
the value oI p is
unchanged.


Changing the value
oI x doesn't change
the value oI p
either. However, in
evaluating p (which
is what happens
when you enter "p"
into the worksheet,
Maple will always
use the current
value oI whatever
4.5 DiIIerences between the appearance oI the worksheet and the state oI a Maple session 49
symbols are used.
II the symbol has
no assigned value,
then the name oI the
symbol is its value.
II we unassign
a variable, then
we can think oI
the slot as being
deleted Irom the
computer's memory.
Unassigning x does
not unassign p or
change p's assigned
value. However, the
results oI evaluating
p changes due to
the unassignment
since it reIers to
a variable whose
assigned value has
been removed.



4.5Differences between the appearance of the worksheet
and the state of a Maple session
The state of a Maple session consists oI all the variables that are currently assigned, and what their values are. The state changes every time we do
another assignment or unassignment. Only an assignment operation (or an unassignment) can change the state oI the session. The Variables Panel
displays the state.
Typically work in a Maple document progresses in a top-to-bottom Iashion -- additions are placed at the end oI whatever is already there. This is
also true Ior an ordinary word processor such as MicrosoIt Word. However, a key Ieature oI a word processor is that iI you need to go back and
Iix something, you can do so. In a Maple worksheet, iI you go back and change a line. However, while this will change the instructions, Maple
will not automatically update the results due to the changes. You have to hit enter on the lines you want to be re-calculated.
When you save a Maple worksheet as described by Saving and retrieving vour work (page 20)does not save the state oI assigned variables.
This means that when you reopen a saved worksheet, no variables are assigned. This can be conIirmed by looking at the Variables Panel aIter
reopening a worksheet, as illustrated by the "The state oI a Maple Session" example, below . When you reopen a saved worksheet, it's not "what
you see is what you get" -- at least, not right away.
To repeat: a saved worksheet does not save the state oI the session (the variable assignments). Opening a saved worksheet Iile does not cause it to
automatically execute the operations in the worksheet. This gives you a chance to edit the worksheet to change the calculations speciIied, beIore
carrying out the instructions.
We know that we can jump back and execute a line in the worksheet a second time, just by positioning the cursor there and hitting enter (return).
Thus the way that the document looks is not an accurate reIlection oI variable assignments that current exist in the session.
50 4 Chapter 4 Assignment
The Variable panel oI the Maple worksheet indicates which variables are assigned, and what their assigned values are. Use it to determine the
current state iI you are not sure what it is.
Table 4.5: The state of a Maple Session
Example


We took the work oI the previous section and saved it as a worksheet named Start1.mw. We then closed the worksheet. In a
Iresh Maple, we opened up the worksheet again. Note that while all the work we did is displayed, the Variable Panel indicates
that no variables are assigned when the worksheet is read in.
When the Start1.mw, the cursor is positioned in the Iirst line. II we hit enter, we re-execute the Iirst line again. Note that the
Variable Panel now has the result oI the Iirst line -- assigning p a value.

4.5 DiIIerences between the appearance oI the worksheet and the state oI a Maple session 51
Example
AIter hitting enter the Iirst time, the cursor moves to the second line -- "x : 3". Hitting enter again, causes the second
assignment to occur.
II we now go back and position the cursor on the Iirst line and hit enter again, p is assigned a value that reIlects that Iact that x
has an assigned value.

52 4 Chapter 4 Assignment
Example
II we then execute the rest oI the lines oI the worksheet by hitting enter some more. But the results are not the same as in the
previous example. This because the state oI variables was diIIerent when we executed the Iirst line the second time. In general,
re-evaluation oI an expression will yield diIIerent results iI the variables it mentions have diIIerent values than the Iirst time it
was evaluated.
4.6restart causes all variables to be deassigned
We've seen that it's possible to erase an assignment using unassign. II you want to Iorget all assignments we've made so Iar, then you can use
restart. This can be useIul in situations where you've done some work and made some assignments, but now want to switch to working on a
diIIerent problem and would like Maple to Iorget about the assignments you made beIore. It is generally a good idea to restart at the beginning oI
unrelated sections oI the worksheet just in case variables were previously assigned values that might not be related to their use in the new section.
Table 4.6: Evaluation of expressions involving assigned variables
Examples of assignment with :

4.6 restart causes all variables to be deassigned 53
Examples of assignment with :
We saved a worksheet similar to that oI the previous example, but we placed a restart instruction at the beginning oI it. We
executed all the instructions oI the worksheet, in order, by hitting the "!!!" icon at on Maple toolbar. Note that only p is assigned
a value at the conclusion because we unassigned x.

We got the worksheet to look like this by executing the x :3 line, then every line starting with (but not
the restart). The eIIect oI executing the assignment to p once x had a value causes p to be assigned the result oI evaluating the
expression with x's value as 3: . Since this Iormula no longer has any x in it, reassigning x to have a value oI 4 doesn't
really change the result oI evaluating p, unlike the previous examples.

54 4 Chapter 4 Assignment
Examples of assignment with :

However, iI we re-execute the entire worksheet, including the restart, we have results identical to what we had when things were
Iresh. restart removes any inadvertent side-eIIects that old assignments may have on evaluation.

4.7Evaluation and chains of assignments
Evaluation always uses the current assigned value oI a variable. Evaluation will Iollow the chain oI assignments to Iind values that does not
mention any assigned values.
Table 4.7: BROKEN - MISSING TITLE!
Examples of assignment with : Commentary
We start
Iresh through
restart. .
We assign the
name p the value
oI the expression,
then x. When we
assign result2,
it evaluates p,
and Iinds it to
have the value
. But
x's value is 3 and
a's value is y. It
uses 3 and y in the
4.8 Troubleshooting assignments 55
Examples of assignment with : Commentary
evaluation. This
is why result2's
assigned value is
not the same as that
oI p.

II we then a : z
1, then Variable
Panel shows that
the value oI a has
changed, but the
values oI p, result2,
and x have not.
However, iI we ask
Maple to evaluate
p, the evaluation
uses the value
oI the symbols
mentioned in the
value oI p. This is
why the evaluation
oI p is diIIerent
-- ,
even though the
assigned value
oI p continues to
be what it was
originally.


4.8Troubleshooting assignments
Equations are not the same as assignment
Assignment is an operation that many programming languages have. In some languages (e.g. Maple, Pascal, EiIIel, MuPad) := is used Ior the
assignment operation. In others (C, Java, Matlab) = is used as the symbol Ior assignment. Maple uses : because it uses = Ior equations. It would
be conIusing to computers and to human readers to use the same symbol Ior two common but diIIerent operations in a single language.
and : mean different things in Maple
Example

56 4 Chapter 4 Assignment
We assign a the value 3. However, Ior the next line we type "x4". This is an equation and it doesn't assign x any value. The
Variable Panel shows this because it lists a as the only variable that has an assigned value.


We edit the x4 line into x :4, and hit the enter key. This time the assignment goes through.
The name to be assigned always goes on the left hand side of the :
Since 5 x and x 5 mean the same thing as mathematical equations, some people think that this should mean that x : 5 and 5 : x should both
assign the value 5 to x. However, only x : 5 does the assignment.
Table 4.8: Assignment : is not symmetric
Error, illegal use oI an object as a name

This doesn't mean anything to Maple. The name is supposed to
be on the leIt hand side.
(4.17)

This assigns x the value 5.
This assigns z the (symbolic) value y. It doesn't assign y any
value.
4.8 Troubleshooting assignments 57
(4.18)
(4.19)

(4.20)

To undo all assignments, use restart
Sometimes you want Maple to Iorget all the assignments you have made in a session. You can get this to happen either by using unassign on
each assigned name, or by entering restart in Math mode and then hitting enter. This will unassign everything,undoing all the assignments.
restart does not erase the worksheet, however. The worksheet still looks the same, including the written record oI the assignments you had
previously done. What the restart does is to delete all the slots you have set up in your mental model.
Some names are already used by Maple. You will get an error message if you try to assign to them yourself.
When you Iirst start up Maple, the names that you would ordinarily think oI using to assign to are not assigned. However a Iew are, such as the
symbolic constants Pi and I. Maple will tell you that such names are reserved Ior system use. You need to pick another name.
Table 4.9: Some Names can't be assigned to
Example Commentary
Error, attempting to assign to `Pi` which is protected

You can't redeIine a symbolic constant.
Error, controlling variable oI Ior loop must be a name

for isn't a symbolic constant but it is used in Maple's
programming language. So we can't use it as a variable. We
can tell that something Iunny is going on because the Ior is
automatically turned into a bold for, and a red box appears
around the initial part oI what we typed.
Error, attempting to assign to `solve` which is protected

solve is the name oI operation that solves equations. You can't
change its meaning by using it as a variable to assign to.
Can't solve for a variable with an assigned value
The solve operation takes an equation, and solves with respect to a variable. II that variable becomes assigned a numeric value, then solving
again won't work because evaluation causes Maple to see only the number, not the symbol being solved Ior. Problems occur with other Maple
operations that need a named variable, such as plotting.
58 4 Chapter 4 Assignment
Table 4.10: Can't solve for a variable with an assigned value
(4.22)
(4.23)
(4.24)


(4.25)

At this point, right-clicking on the equation x`25*x60
won't reveal Solve as an operation that can be perIormed:

This is because the equation is evaluated beIore the right-click
menu is activated, producing the equation 300. Since 300
has no variables, Solve isn't permitted since it makes no sense.
II we do a restart and then enter the equation, we can solve it
because x's assignment is gone.
4.9 Summary oI Chapter 4 material 59
(4.26)
(4.27)

(4.28)

iI we assign x a value, then plotting the expression produces
a straight line 30. This is because the expression is evaluated
(producing the constant 30) beIore it is handed to the plot
operation. Plotting a constant is what produces the straight line.
4.9Summary of Chapter 4 material
Assignment
General form Examples

Assignment is perIormed by using the assign to a name
operation oI the clickable menu.
(4.29)
(4.30)
60 4 Chapter 4 Assignment
Assignment also be perIormed by typing in the name, Iollowed
by :, Iollowed by the expression whose value will be the
result to be assigned.

svmbol name : expression


(4.31)
(4.32)

Unassignment
General Iorm Examples
unassign(' svmbol name )
(4.33)

(4.34)


Undoes all assignments made by the user in the session so Iar.

The state of the Maple session
The Variable Panel indicates the variables that are
currently assigned in the Maple session, and their
assigned values.
61
5Chapter 5 Building scripts
5.1Chapter Overview
We brieIly discuss a Iew extra concepts useIul with solve: how to use a combination oI relations rather than just a single equation, and how to take
apart or combine by the various Iorms oI solve.
We then explore the concept oI script. a sequence oI operations useIul Ior solving a problem. Having a script makes it easy to solve diIIerent
versions oI a problem by editing the parameter values and re-executing the script. OIten, the justiIication oI the eIIort to program a computation
is driven by its reuse -- repeating the same computation with slight alterations each time. A Irequently recurring scenario is a parameteri:ed
computation: 1) use variables to assign values to the parameters and 2) have subsequent steps oI the computation reIer to the parametric variables.
Maple is well-equipped Ior reuse oI parameterized scripts, since it has an !!! button on the Maple menu that will execute in sequence all the
operations in a worksheet. Re-execution can also be speciIied by Edit Execute Selection or Worksheet.
5.2Getting information from solve, data structures
The result oI the solve operation can have multiple parts. When there are multiple solutions to the equation, the result oI solve is a sequence, list,
or set oI solutions. We can select a part by giving an index (1 or 2, typically).
Table 5.1: Selecting parts of a solution



We enter the equation, then right-click on it and select the
solve item Irom the menu. This produces a sequence oI two
solutions. We right-click on that and pick the Select Element
menu item to get the submenu that allows us to pick the
Iirst entry.Selecting the Iirst element oI the set gives us the
equation. Selecting the right hand side oI the equation gives us
the Iirst root.


II we give solve a linear equation, it has only one solution. We
can still select the Iirst entry Irom the set, and then the right
hand side oI that to get the root.


The right-click menu also oIIers the solve for variable item,
which allows us to solve Ior x explicitly. The answer comes
back in a slightly diIIerent Iorm, but it still has parts that we
can select Irom.
Maple (as well as many other programming languages) can compute with objects that have structure. Here are Iour diIIerent kinds oI structures
that Maple can handle:
Table 5.2: Basic data structures in Maple and operations to extract parts of them
Type of structure What they look like Examples
Equations, inequalities What math books do Ior equations and
inequalities. The clickable menu allows you
to get at the left hand side and right hand
side oI an equation or inequality.
(5.1)
(5.2)
62 5 Chapter 5 Building scripts
Type of structure What they look like Examples
(5.3)
(5.4)
Sequences Values or expressions separated by a comma
(5.5)
(5.6)
(5.7)

Sets A sequence surrounded by curly braces }
(5.8)
(5.9)
Lists A sequence surrounded by square brackets
| |
(5.10)
(5.11)
Note that the result oI this solve is a set oI
equations.
For the time being, we just want you to recognize the diIIerent kinds oI structures that are output by solve and other Iunctions and be able to select
parts Irom them. Later on we will get a lot oI work done by perIorming more sophisticated operations with them.
5.3Finding simultaneous solutions, constraining solutions.
Suppose we want to solve the system oI equations and . This means Iinding values oI x and v that simultaneous
satisIy both equations. We can do this in Maple by typing in the Iirst equation and then the second, separated by a comma. This is called entering
a sequence oI equations. Right-clicking (control-click on Macintosh) on the sequence will allow you to solve the system.
5.4 Scripting: making computational work easy to reuse 63
In Lab 1, you discovered that solve could also handle inequalities as well as equalities. You can enter a sequence oI equations and inequalities to
solve. This can be used to limit solutions to a particular range oI values.
Table 5.3: Solving simultaneous equations
(5.12)
(5.13)

The result oI this solve is a set oI solutions.
(5.14)
(5.15)

We assign an equation to p, and solve it Solving this equation
produces 4 roots. Two oI them are complex numbers (since
they have I in them). The others are real.
(5.16)
(5.17)
(5.18)

We enter a sequence consisting oI the equation named p, and
the inequality . When ask Maple to Iind values oI x that
satisIy both the equation and the inequality simultaneously
-- roots oI the equation p that are non-negative. We might be
interested in only non-negative real solutions, iI, Ior example,
we were using the equation to model a physical situation and x
was measuring weight. Negative values Ior weight wouldn't be
a solution we'd be interested in.
5.4Scripting: making computational work easy to reuse
Consider the problem you did in Lab 1, along with a solution:
Version 1 and solution
From Anton, Calculus, 8th edition, ch. 1 review exercises, problem 37, p. 99.
A breeding group oI 20 bighorn sheep is released in a protected area in Colorado. It is expected that with careIul management the number oI
sheep, , aIter years will be given by the Iormula:
64 5 Chapter 5 Building scripts
and that the sheep population will be able to maintain itselI without Iurther supervision once the population reaches a size oI 80.
(a) Graph versus .
(b) How many years must the state oI Colorado maintain a program to care Ior the sheep? Show your work.
(c) How many bighorn sheep can the environment in the protected area support? (hint: examine the graph Ior large values oI .)
Solution to (a)
We can enter the equation Irom the problem, but we really need the right hand side oI it because the plot operation can't plot equations, it can only
plot things that evaluate to a number (rather than an equation).
Once we have the expression Ior the sheep population, we need to play with the plotting ranges to see when the leveling oII occurs. We'd have to
think about it and experiment a bit -- but the computer makes the replotting easy to do once we make our decisions about what to try.
(5.19)
5.4 Scripting: making computational work easy to reuse 65
Solution to (b)
(5.20)
(5.21)
We can read the leveling oII point Irom the plot, assuming that we have Iigured out the appropriate time range in (b). Alternatively, we can do a
little calculus and take the limit oI the expression as t goes to inIinity.
(5.22)
We can imagine ourselves working as a environmental engineer Ior the National Forest Service and being very pleased with ourselves Ior solving
the problem with Maple. But now we are handed two more wildliIe management problems to do, Irom other regions in our territory:
Version 2

A breeding group oI 33 bighorn sheep is released in a protected area in Colorado. It is expected that with careIul management
the number oI sheep, , aIter years will be given by the Iormula:


and that the sheep population will be able to maintain itselI without Iurther supervision once the population reaches a size oI 85.
(a) Graph versus .
(b) How many years must the state oI Colorado maintain a program to care Ior the sheep? Show your work.
(c) How many bighorn sheep can the environment in the protected area support? (hint: examine the graph Ior large values oI .)


Version 3

A breeding group oI 45 bighorn sheep is released in a protected area in Colorado. It is expected that with careIul management
the number oI sheep, , aIter years will be given by the Iormula:
66 5 Chapter 5 Building scripts


and that the sheep population will be able to maintain itselI without Iurther supervision once the population reaches a size oI 90.
(a) Graph versus .
(b) How many years must the state oI Montana maintain a program to care Ior the sheep? Show your work.
(c) How many bighorn sheep can the environment in the protected area support? (hint: examine the graph Ior large values oI .)

We have the Ieeling that we will shortly be handed problems Ior a number oI other locations as well. How can we reuse our original work with
minimal eIIort?
II we had the Iirst solution, we could produce the second solution through copv-paste-edit-re-execute.
Table 5.4: Executing a clone of a script through copy-paste
1. Copy and paste the original solution into a new document, or to a spot later in the same document. You do this as with a
word processor -- select a region oI the worksheet with the mouse, type control-C (command-C on a Mac). Then position the
document cursor in the destination, and type control-V (command-V on a Mac). A duplicate oI the solution should appear there.
2, Using the mouse, edit the Iormulae as needed. You only need to edit the black (input) Iormulae, not the blue ones.
3. Position the mouse at the Iirst computation and hit enter. Continue to work your way through the sequence oI the commands.
4. Alternatively, select the entire region containing the edited version oI the solution and hit Edit-~Execute-~Selection.
5. II the region to be executed is the entire worksheet, then you can hit the !!! button oI the Maple toolbar, or you can do Edit
ExecuteWorksheet.
The results oI executing the edited script are not totally correct. We will have to change a Iew things by hand: the clickable operation "evaluate
at a point" in step (b) will use N80 (which is what the copy says) instead oI N85, so we will have to redo it. Also, the last plot will not show
an appropriate vertical range unless we do that. We'd like to do a little less oI this hand-tuning in re-execution. Also, we'd like to make it easier to
remember what to change between multiple versions.
To see the edited version oI the script look at A breeding group (page 68).
5.5Rewriting the script by assigning parameters at the
start of the script
While copying and editing is probably a little Iaster than typing in the whole script again, we would also spend time hunt around Ior assignments
to edit each time we wanted to re-run the script. We can reduce the "hunting around" by adopting the Iollowing working rule: identiIy the
parameters oI the problem, and change the script so that it assigns values to all parameters at the beginning oI the problem.
This does not mean "put all assignments at the beginning oI your script". Only the assignments involving parameters -- values that are speciIied
by the problem description -- need to be relocated to the start oI the script.
5.6 Summary oI script writing 67
Finding and naming parameters
First, solve at least one version oI the problem. Then, imagine what would need to be changed iI you were trying to solve alternative versions
oI the problem. You can Iind parameters iI you have several versions oI a problem by looking at what changes in the worksheet Irom version to
version.
For example, in the sheep problem, we note the Iollowing things changing in diIIerent versions oI the problem.. We pick names Ior these.
1. the numerator oI the "sheep equation" (P)
2. the coeIIicient in the denominator oI the equation (c)
3. the value oI the stable population (s)
The other number in the script is the original population oI the sheep. We realize that it can be derived by evaluating the "sheep equation" at t0.
Thus this value does not need to be a parameter as we can derive it Irom the other inIormation. It might be a good check though.
We then write the script to assign values to the parameters at the start oI the script, and then write the other operations and expressions in terms oI
the symbols.
This allows us to redo the script just by changing the values at the beginning oI the script. We use the word processing Ieatures to add extra
directions to make the script easier to use, basically saying "here, change these things", and "this is the end oI where you should stop copying".
To see the edited version oI the script look at Attachment. Jersion 2 of Sheep Script, with parameters (page 70).
Having created this script, we can handle the third version oI the problem by editing the values oI the parameters and executing the script again
by selecting the entire script with the mouse, and then doing Edit-~Execute-~Selection. We get this result:
5.6Summary of script writing
Script writing is appropriate when vou expect to handle several different versions of the same problem. In proIessional work, this is oIten the case
-- iI it's worth doing at all, it's probably going to occur more than once.
Figure out how to solve the problem first. Then write the script. There's really not much point in writing the script iI you don't have some idea oI
the sequence oI operations in it.
Once you have a worksheet oI instructions Ior solving one version oI the problem, look at it and the other versions and find the parameters. Set
up a new worksheet where the Iirst thing you do is to assign values to the parameters. Then work through the rest oI the instructions and rewrite
them to use the parameters instead of the fixed values from one version of the problem.
5.7Troubleshooting scripts
Programming books have the tendency to show things that work. They say less about what to do about the things that don't work. However, it is
usually the case that computer users oIten spend more time "getting things to work" than "working". Learning how to get out oI jams is at least
as valuable as knowing how to enter operations and what they mean. We oIIer this advice, which is "commonsense about work" as applied to
interactive soItware development:
1. Solve one version oI the problem beIore you try to start scripting. You can use Maple to experiment -- enter and edit snippets oI operations
that try out the solution technique Ior part oI the problem. Eventually edit them together so that they solve the whole problem. II you have only
Iuzzy notions about the math or the operations you want to proceed, your computer work will just ampliIy that. Having a worksheet that solves
one version oI the problem can remove a lot oI the Iuzziness.
Where does the inspiration Ior solving the problem come Irom? II you are lucky, the solution may be told to you. Or you may Iind a description
oI a similar problem as a starter. But the big bucks, as they say, go to those who can devise the solution plan themselves.
2. Limit each step so that it is a small step. II you get into trouble, you will be able to nip it at the bud. This becomes a more prominent tactic in
later work when we are tempted to construct long-winded one-line expressions that do everything at once.
68 5 Chapter 5 Building scripts
3. Test pieces individually, then put them together. For example, iI you don't really understand how to make a dotted line plot in teal with a title,
you should try to do that with a simple plot (such as ) rather than an expression whose shape you aren't that Iamiliar with it. Then take what
works and substitute the real expression you want to plot in a copy oI the plot operation. The complexity oI troubleshooting is reduced iI you
have Iewer lines oI untrustworthy code to look through Ior problems. This can be called divide and conquer troubleshooting.
4. II what you have doesn't work, Iind something similar which does work, and then incrementally edit it. For example, iI you can't get
in, then Iirst see whether you can get to work. Once you succeed with that, edit the expression to
and so Iorth.
5. II strange things continue to happen despite your best eIIorts to troubleshoot, it may be that previous settings in the Maple session are
interIering with your current work. Recall that some people in the ENGR101 lab couldn't get their oscilloscopes to work because oI settings
changed in the oscilloscopes by groups earlier in the day. This can be particularly true iI you are developing scripts and are assigning parameter
values, then switch to development oI another script in the same document. The values you assigned will not magically unassign themselves
when you start working on something new in the same worksheet.
The remedy Ior this is to put a restart in as the Iirst operation in your script, then re-execute the worksheet.
5.8Attachments
Attachment: Version 2 of sheep script without parameters
Version 2 of sheep problem, with edited script
A breeding group oI 30 bighorn sheep is released in a protected area in Colorado. It is expected that with careIul management the number oI
sheep, , aIter years will be given by the Iormula:
and that the sheep population will be able to maintain itselI without Iurther supervision once the population reaches a size oI 85.
(a) Graph versus .
(b) How many years must the state oI Colorado maintain a program to care Ior the sheep? Show your work.
(c) How many bighorn sheep can the environment in the protected area support? (hint: examine the graph Ior large values oI .)
Solution to (a)
5.8 Attachments 69
(5.24)
Solution to (b)
(5.25)
(5.26)
We can read the leveling oII point Irom the plot, assuming that we have Iigured out the appropriate time range in (b). Alternatively, we can do a
little calculus and take the limit oI the expression as t goes to inIinity.
(5.27)
70 5 Chapter 5 Building scripts
Attachment: Version 2 of Sheep Script, with parameters
Version 2, with use of parameters
Start of parameters -- change these for each version of the problem
(5.28)
(5.29)
We call the size oI the stable population s.
(5.30)
End of parameters
(a) Graph versus .
(b) How many years must the state oI Colorado maintain a program to care Ior the sheep? Show your work.
(c) How many bighorn sheep can the environment in the protected area support? (hint: examine the graph Ior large values oI .)
(a)
(5.31)
Note that sheepPopExpr is not a parameter since assignment is always the same Ior all versions oI the problem.
To make the graphing work all the time, we set the vertical axis to "P30" rather than a Iixed value. This will set the vertical axis so that it will be
30 more than the value oI P which Irom the Iirst problem we have realized is the top oI the graph.
(5.32)
5.8 Attachments 71
(5.33)
This is the time in years that model predicts it will take Ior the sheep population to reach selI-sustaining status, allowing the wildliIe managers to
move onto another job.
(c) A little thinking reveals that the leveling oII value (as indicated Irom the graph), is the numerator oI sheepPopExpr, since the denominator
goes to 1 as t goes to inIinity.
(5.34)
72 5 Chapter 5 Building scripts
End of script
Attachment: Version 3 of Sheep Script, with parameters
Start of parameters -- change these for each version of the problem
(5.35)
(5.36)
(5.37)
We call the size oI the stable population s.
(5.38)
End of parameters
(a) Graph versus .
(b) How many years must the state oI Colorado maintain a program to care Ior the sheep? Show your work.
(c) How many bighorn sheep can the environment in the protected area support? (hint: examine the graph Ior large values oI .)
(a)
(5.39)
Note that sheepPopExpr is not a parameter since assignment is always the same Ior all versions oI the problem.
To make the graphing work all the time, we set the vertical axis to "P30" rather than a Iixed value. This will set the vertical axis so that it will be
30 more than the value oI P which Irom the Iirst problem we have realized is the top oI the graph.
(5.40)
5.9 Summary oI Chapter 5 material 73
(5.41)
This is the time in years that model predicts it will take Ior the sheep population to reach selI-sustaining status, allowing the wildliIe managers to
move onto another job.
(c) A little thinking reveals that the leveling oII value (as indicated Irom the graph), is the numerator oI sheepPopExpr, since the denominator
goes to 1 as t goes to inIinity.
(5.42)
End of script
5.9Summary of Chapter 5 material
Type of structure What they look like Examples
Basic data structures in Maple
74 5 Chapter 5 Building scripts
Type of structure What they look like Examples
Equations and inequalities
Expression related by , ~, , , , or
.
(5.43)
(5.44)
Sequences Values separated by a comma
(5.45)

Lists A sequence surrounded by square brackets
| |
(5.46)

Sets

A sequence surrounded by curly braces }
(5.47)

Solving simultaneous equations
(5.48)
(5.49)

The result oI this solve is a set oI solutions.
(5.50)
(5.51)

Solving this equation produces 4 roots. Two oI them are
complex numbers (since they have I in them) the others are
real.
(5.52)
This gets the solver to list only the positive real solutions. We
might be interested in only non-negative real solutions, iI,
Ior example, we were using the equation to model a physical
5.9 Summary oI Chapter 5 material 75
(5.53)
(5.54)

situation and x was measuring weight. In this case the other
values oI x wouldn't be relevant to our situation.
Scripts
Creating a script Ior a
problem
In a Maple worksheet, take a version oI a problem and create a sequence oI operations in the
worksheet that solve it.

Note similarities and diIIerences between diIIerent versions oI the problem. Envision what you'd have
to change in the worksheet in order to solve a diIIerent version oI the problem, and what would stay
the same. You may have to rewrite some oI the expressions to reIer to the parameter rather than the
value.

Assign the parameters at the beginning oI the script. Rework the rest oI the script so that the Iormulas
reIer to the parameters by name, rather than the values used in the original version oI the problem.

For example, iI the value 42 appears in several places in your script, deIine a parameter p : 42 at
the start oI the script and edit the other occurrences oI 42 to be p instead. When you have a diIIerent
version oI the problem, you can edit just the single line p : 42 into say p : 47 and won't need to edit
any other lines oI the script.

Using a script Copy and paste the script to a new location.

Edit the assignments to reIlect the new version oI the problem.

Click on the !!! button oI the Maple toolbar to execute the entire worksheet. Or select a portion oI it
with a mouse and hit Edit-~Execute-~Selection. As a last resort, just hit enter (return on Macintosh)
multiple times to perIorm the operations in the new version oI the script.
Rationale Ior using
scripts
More work to do than clickable interIace the Iirst time. Saves time iI you expect to want to reuse the
operations on multiple versions oI the same problem. Also it is less error prone.
Divide and conquer
troubleshooting
Test each piece oI the script as soon as soon as you build it, rather than waiting to test it only aIter you
have entered it all. It's easier to test incrementally, a piece at a time, because iI there are Iewer lines oI
code to suspect oI causing problems.
76 5 Chapter 5 Building scripts
77
6Chapter 6 More sophisticated scripting:
textual entry, multiplots, point plots, character
strings, lists, solving equations, and evaluation
6.1Chapter Overview
We introduce textual entry oI solve and plot operations, where the operation is speciIied by keyboard alone, without the use oI the mouse or
palettes. Textual entry is oIten preIerred by programmers because it is easier to edit scripts written with text. We will retain use oI the clickable
interIace Ior doing quick one-time calculations, or Ior developing ideas the Iirst time beIore we start script-writing.
We being to introduce additional concepts in Maple, to enhance what we can solve and plot:
1. In Maple, a character string is a collection oI characters delimited by "s: "This is a string." We see how strings are used in the
textual entry oI labels and colors in plots.
2. We discuss lists e.g. [1,2, x, 3.5]. We Iirst saw them in Getting information from solve, data structures (page 61) provide a way oI
organizing multiple results in a single "data container", making it to operate on the whole collection oI results while retaining the ability to getting
at individual results Irom within the collection. Lists are used in both solve and plot.
3. solve uses two other types oI data containers: sequences and sets. We describe how to recognize them. and how to extract inIormation Irom
them.
4. Programmers rely on the on-line documentation to manage the complexity oI remembering the details oI a knowledge-intensive system such
as Maple. They learn/remember how to use a Ieature by looking up the description, Iinding an example close to what is desired, and then actively
experiment with the example in a Iresh worksheet. Reading without experimentation is usually not very productive.
5. In Chapter 4 Assignment (page 45), we explained assignment and how Maple uses assigned values whenever it sees a name in an expression.
We introduce the eval operation, which allows variables to take on values during evaluation without assignment. Another way to view eval is that
it is a way Ior assignments to be done temporarily and immediately Iorgotten. This can be an attractive alternative iI you are concerned about
situations where you'd be making many assignments and then undoing them through unassign. eval allows you to evaluate an expression Ior a
particular value oI a variable in one line, rather than having to type in the assignment and unassignment as well.
6.2Textual entry of operations
The textual Iorm oI an operation in Maple has the general Iorm:
operationName ( sequence of values )
The operationName can be something like solve or plot. By sequence of values, we mean one or more items, each item separated Irom the next by
a comma. "Sequence" here is the same kind oI sequence that was Iirst seen in the previous section on solve (page 62).
Maple will evaluate what vou enter in the same way that was described Ior mathematical expressions possible (page 41). II there are assigned
variables mentioned in the sequence, then their values will be used. II Maple knows the operationName (e.g. sin, solve, plot), then it will perIorm
the calculation speciIied by the built-in programming. Otherwise, the result will be more or less what you typed in.
The technical term Ior the "values" in this situation is actual parameter or argument. Note that the parentheses around the sequence oI values are
mandatorv -- you will either get an error or a result that's Iar Irom what you want iI you omit the parentheses.
This style oI writing things is sometimes called functional notation. In mathematics examples oI Iunctional notation are f(x) or g(3,5) . In these
examples, the name oI the operation is the Iunction name f or g, while the actual parameters or arguments would be x or the sequence 3,5.
Evaluating the function is what causes Maple to perform the action.
Table 6.1: The textual form of equation solver solve
The Iorm oI the answer returned by the textual version is
slightly diIIerent Irom invoking solve through the clickable
78 6 Chapter 6 More sophisticated scripting: textual entry, multiplots, point plots, character strings, lists, solving equations, and evaluation
(6.1)






interIace. The Iormer is typically a Iorm that is easier to work
with in scripts.

In this, the operation Name (or Iunction name) is solve.
There are two arguments. The Iirst argument is the equation
. The second argument is the symbol x. In this
situation the result oI perIorming the solve operation is a
sequence oI two numbers, the two roots oI the equation.


(6.2)

What you get iI you leave out the parentheses -- not much!
Because you leIt out the parentheses, Maple does not think you
are asking Ior any Iunction to be evaluated, which is how the
work gets done.

Error, unable to match delimiters

You get another one oI those "unable to match delimiters"
messages iI you Iorget one oI the parentheses.



Error, invalid power

This is what happens iI we Iorget to use the right arrow key
to descend Irom the exponent oI . Maple thinks that we are
talking about x to a power called 2-2,x, which understandably
it doesn't make sense to it as a power.
6.2 Textual entry oI operations 79
Table 6.2: The textual version of plotting

This is a textual Iorm oI plot. The Iirst argument is an
expression, the second argument is an equation naming a
variable and a range oI a plot. Note that iI we wanted to change
the range Irom -3..3 to -5..2 then we would just edit that line
oI the worksheet and hit enter again. II we wanted to redo the
plot in the clickable interIace, we would have to right-click and
enter all the inIormation all over again.

II the second argument is just the variable, then plot uses
deIault values Ior the range.
80 6 Chapter 6 More sophisticated scripting: textual entry, multiplots, point plots, character strings, lists, solving equations, and evaluation
(6.5)


No plotting happens iI we Iorget the mandatory parentheses.

Suppose we entered this instead oI
. There is no error message, but
the picture is not like the Iirst plot. Why did we get something
diIIerent?

The only way that we would discover that a mistake is iI we
already had an idea oI what the graph should look like, and
noticed that the result diIIered signiIicantly Irom what we
expected. II you haven't Iormed a basis Ior expectations, you
won't discover such mistakes.
Once you realize that the picture is incorrect, you would
be spurred to search Ior the cause. Since there are no error
messages and a plot (albeit a weird one) was produced, the
most likely cause is that the arguments to the plot Iunction are
wrong. The obvious place to look Ior correct examples is the
on-line documentation II you look at the on-line documentation
Ior help, you will see that giving plot three arguments means
something diIIerent -- the third argument can be taken as
the value oI the vertical range oI a plot. Evidently what is
happening is that you are seeing only the tiniest top slice oI the
plot produced above because oI the inadvertent speciIication oI
the vertical range.

This example illustrates the Iact that just because there is
no error message, it does not mean that all is correct. The
computer needs to be told what to do in a way that it can
Iollow. The rules are typically more restrictive than those oI
ordinary human language.
An attachment at the end oI the chapter shows the textual Iorm oI common Iunctions, subscripts. These textual Iorms can be entered Irom the
keyboard wherever the palette entry would work.
6.3Why are there two different styles for entering
operations7
The clickable interIace is a good way to get a calculation done quickly, but the actions speciIied in this way are hard to edit when building scripts.
Maple, like most languages, has a textual version oI all operations it perIorms. The editing involved in scripting development can oIten be easier
6.4 Plotting a list oI expressions (multi-plots), plotting lists oI numbers (point plots) 81
to do on the textual version. In other words, the clickable interface is good for a one-time calculation, but not so good for the editing and re-
execution involved in script reuse.
Another advantage oI the textual mode oI operation is that the number and variety oI operations available in textual Iorm is Iar greater than what's
available in the clickable interIace. Maple has several thousand operations. Building a clickable interIace to all oI them would result in tedious
navigation through menus that would either be huge or involve many sublevels.
The downside oI using the textual entry is that the developer must spell the text correctlv, with the right number and placement of parentheses.
Experienced users Iind that the textual interIace is Iaster to deal with Ior scripting, while the clickable interIace is Iaster Ior short, more casual
use. Fortunately, in either case one can edit Iailed attempts and retry, so perIect entry is not necessary to be productive.
Becoming proIicient with textual entry oI operations is part oI the transition technical users make in going Irom reuse oI other's work to routinely
creating their own programming. Without such proIiciency, it is hard to realize the Iull power oI the computer in modeling and simulation
situations.
6.4Plotting a list of expressions (multi-plots), plotting
lists of numbers (point plots)
Recall that lists in Maple are a way oI collecting expressions together into a single object, as discussed in the previous table describing lists and
other data containers (page 62). You speciIy a list by listing the items in the list , enclosed in square brackets | |.
II the Iirst argument to plot is a list oI expressions, then plot will on a single graph display the plots oI all the expressions in the collection. By
deIault, each one will be displayed in a diIIerent color.
Table 6.3: Plotting of multiple expressions
Example

Plot two expressions on a range where both
have comparable-sized results. We use the list
to indicate the two expressions that should
be plotted.












82 6 Chapter 6 More sophisticated scripting: textual entry, multiplots, point plots, character strings, lists, solving equations, and evaluation
Example














Problem: approximately where is the expression
equal to 5? While we could use solve
to tell us exactly, it's oIten worthwhile to draw a picture
and process the situation visually. II we give plot the
two expressions "-5" and then plot will
plot not only the parabola (the second expression, it
will also plot the expression that is always -5 Ior any
value oI x. This corresponds to the horizontal line drawn
on the plot. Visually we can see that the parabola is -5
at roughly -.8 and 1.2. We can even get a little more
precise by clicking on the plot and using the coordinate
window as was discussed in Plotting, approximate
numerical solutions through cursor position (page 25).
6.4 Plotting a list oI expressions (multi-plots), plotting lists oI numbers (point plots) 83
Plotting multiple expressions simultaneously can be useIul when you want to compare them. Assuming that the scales are comparable, one can
get a sense oI similarity or dissimilarity "at a glance".
We can plot data points rather than smooth curves, iI we give the textual Iorm oI plot separate lists oI x and v coordinates. can be used with the
textual version oI plot . II we give the textual version oI the plot operation two lists oI numbers that have the same length, then plot will regard the
Iirst list as a list oI x-coordinates, and the second list as corresponding v-coordinates. II you provide plot with the third argument stvlepoint, then
it will produce a point plot. Otherwise, it will draw lines connecting each point.
Table 6.4: Plotting points with lists of numbers
(6.6)
(6.7)

We use the textual Iorm oI plot to plot the points (1,5), (2,6),
(3,7) and (4,-1).






84 6 Chapter 6 More sophisticated scripting: textual entry, multiplots, point plots, character strings, lists, solving equations, and evaluation

Without the third argument, plot will try to connect the points
with a curve.
Error, (in plot) unexpected option: Style point

Maple cares about whether things are capitalized or not. Stvle is
not the same as stvle.
6.5Character Strings and putting titles and labels into
plots
A string in Maple is something enclosed in double-quotes: "red", "this is a string?", "GregoryBrothers4" are all strings. The
double-quote symbol is mandatory Ior a string. Single-quotes ' (also known as apostrophes), backquotes ` (also known as acute accent marks) are
not substitutes Ior double-quotes in writing Maple strings. Characters enclosed by apostrophes or backquotes mean something diIIerent to Maple.
Use oI the wrong punctuation marks will lead to undesired results or error messages.
In addition to being used to input data points in plot, lists also can be used in speciIying colors and axis labels.
6.5 Character Strings and putting titles and labels into plots 85
Table 6.5: Plot options and labels

II one oI the arguments to the plot operation is oI the Iorm
then those colors will be used.
Most reasonable names will work, but the Iull list can be seen
in on line help (search Ior colornames). Note that we are using
the textual version oI the symbolic constant a. II you can
remember how to spell it, it can be easier than selecting it Irom
the Common Symbols Palette.

Note that because the expressions being plotted are given in a
set, the color assignment is not in the same order that they were
typed in. II we wanted the same order, we should give the plot
expressions in a list.





Error, unable to match delimiters
Forgetting brackets Ior the list oI colors.
Error, invalid interval

One oI the proIiciency issues with textual input is that you have
to remember all the ( ) parentheses and | | brackets. Can you
Iind the the missing delimiter(s)?
86 6 Chapter 6 More sophisticated scripting: textual entry, multiplots, point plots, character strings, lists, solving equations, and evaluation
Error, (in sin) expecting 1 argument, got 2

II you leave enough delimiters out, you get error messages
that don't complain about missing delimiters. You have to
Iigure out what the problem is, which might involve a missing
parentheses even iI the message doesn't say so.
The "Error, (in sin)" is a cue that you should look at the places
where you included sin in your text and inspect it Ior problems.
It doesn't take too much eIIort Ior you to notice that there's no
Iinishing parentheses in the Iirst sin(x .
One reason why there was no error message about delimiters is
that there are multiple missing parentheses. Because there are
equal numbers oI missing leIt and right parentheses, there was
no alarm Ior missing delimiters.
Error, (in plot) expecting a real constant as range endpoint
but received -4*pi

The problem with this plot is that pi doesn't mean the
same thing to Maple as Pi. Maple is case-sensitive. Only
Pi means the symbolic math constant having to do with the
circumIerence oI a circle.
II one oI the arguments to the plot operation is oI the Iorm
then those will be used. Each title
needs to be a string.
In subsequent work, we will see strings used in other situations within Maple other than Ior plot titles.
6.6 Troubleshooting with strings 87
6.6Troubleshooting with strings
The most common mistakes with strings is to leave out the delimiting "s, or to use the wrong kind oI delimiters. While the similar-looking
keyboard characters ' (single quote or apostrophe), and ` (acute accent or backquote) look like would be equivalent, they are use Ior other
purposes in Maple.
Table 6.6: What happens when you forget to use the " delimiter in strong, or use the wrong character for the delimiter
Error, invalid in

Forgetting to include "s around one oI the titles gives a
cryptic error message about an "invalid in". II you were an
experienced Maple user, you'd know that in is part oI the
Maple programming language, but including "in" within a
string would never be Ilagged as an error. This would be a clue
that there's something wrong around where you entered the
Iirst label.

The message is not very helpIul about telling you how to Iix
the mistake, though. This unIortunately is typical in most
computer programming languages, despite several decades'
eIIort in building systems soItware to help people program.
Error, invalid in

Putting the wrong kind oI quote -- ' instead oI " didn't make a
string. We got the same indication oI a problem as beIore even
though the problem is "wrong kind oI quote" rather than "no
quote".
6.7Learning through on-line documentation and
experimentation
All the options available in the Plot Builder available through the right-click (control-click) interIace are also available in the textual version oI
plot. In Iact, there are many additional options and varieties oI plotting available. The way to Iind out what the Ieatures are and how to invoke
them is to consult the on-line documentation.
88 6 Chapter 6 More sophisticated scripting: textual entry, multiplots, point plots, character strings, lists, solving equations, and evaluation
We can Iind out more about the textual Iorms oI plotting by invoking Help -~ Maple Help and typing plot into the search Iield. When we do so,
we see the inIormation in the Iigure below:
Table 6.7: Plot command help
6.7 Learning through on-line documentation and experimentation 89
We scroll to the bottom oI the page and Iind an example oI this. We are looking Ior a version oI plots where v1 and v2 are lists. We don't see
something exactly like that but we do see something with Vectors which are similar. Since the document says this should work Ior lists or vectors,
we take the example and see iI we can modiIy it Ior our own purposes:
Table 6.8: Examples of plot
Evidently, the the Iirst list is the values oI the x (horizontal) coordinates, and the second list the values oI the v (vertical) coordinate.
We copy and paste the example into a Maple worksheet and then see iI we can get it to work.
90 6 Chapter 6 More sophisticated scripting: textual entry, multiplots, point plots, character strings, lists, solving equations, and evaluation
Table 6.9: Copying the example
According to the documentation, we should be able to get this to work iI the Iirst two arguments are lists or vectors. So we edit the example to do
lists instead and re-execute the line to see iI it works in the same way.
6.7 Learning through on-line documentation and experimentation 91
Table 6.10: Modifying the example
To learn about plot options such as colors and labels, we click on the plot,options item under the search results Ior plot (see green oval in the
Iigure). Clicking on that item produces this inIormation. We see inIormation about color (with another link to see colors), along with possibilities,
Ior labels, symbols, styles, etc. Again, the way to learn the options is through copying and pasting the examples into a Iresh worksheet, getting
them to work, and then modiIying them to suit your own purposes.
Table 6.11: Plot option help page

92 6 Chapter 6 More sophisticated scripting: textual entry, multiplots, point plots, character strings, lists, solving equations, and evaluation
Note the "colour" -- this is a Canadian product. You will see things like this as well as other indications that it's not an all-
American world out there. For example, you can convert pints into Imperial Gallons through Tools -~ Assistants -~ Units
Calculator.
6.8More operations on lists
So Iar we have talked only about creating lists, and assigning lists as the value of a variable. You can also generate a sublist of a list, find a
particular item in a list bv its position index, count the number of items in the list, and convert a list into other types oI data.
Table 6.12: Operations on lists
Operation Example Commentary
Create a list
(6.12)
(6.13)

Lists can contain symbols, numbers,
expressions -- anything, even other lists.


6.9 solve, lists and sequences 93
Operation Example Commentary
SpeciIy a sublist oI values
(6.14)

II a list is Iollowed by another pair oI
braces with a range inside, then a sublist
is computed as a result. Here we have the
list that's the Iirst through third items oI
s1.
SpeciIy one item Irom the list
(6.15)

SpeciIy the last item in the list
(6.16)

SpeciIy a sublist with one item
(6.17)

SpeciIy a sublist Irom the 3rd Irom
the end to the end
(6.18)

Count the number oI items in the list
(6.19)

Add together all the items in the list
(6.20)

Compute the average oI all the
numbers in the list.
(6.21)

Convert a list into a sequence
(6.22)
op has an idiosyncratic name Ior an
operation. It stands Ior "operands".
Convert a list into a string.
(6.23)

6.9solve, lists and sequences
To solve a system oI equations, use a list oI expressions or equations Ior the Iirst argument to solve. Use a list oI variables as the second argument.
solve will return a sequence oI lists as the result.
When solve Iinds two solutions Ior an equation (such as iI the equation is quadratic), it will return a sequence oI solutions. You can recognize a
sequence and distinguish it Irom a list because, the sequence is missing the enclosing brackets| | that a list has.
Part-selection operations work in sequences in a similar Iashion as they were described here (page 92).
94 6 Chapter 6 More sophisticated scripting: textual entry, multiplots, point plots, character strings, lists, solving equations, and evaluation
In solve, lists and sequences look as iI they are almost interchangable. Later on we will see situations where lists and sequences must be handled
diIIerently.
Table 6.13: Solving systems of equations with solve
(6.24)
(6.25)


We get a sequence oI solutions since there is a double root.




(6.26)
We evaluate the equation at the Iirst root and see that it does
satisIy the equation.
(6.27)
Same Ior the second.
Error, attempting to assign to `system` which is protected





We want to assign a set as the value oI the variable system.
Maple tells us that the name is already in use as a built-in
Iunction, so it won't let us do that.

(6.28)

We choose a diIIerent variable to assign the set to.
(6.29)


(6.30)

We speciIy the set oI variables we want to solve Ior, then call
solve. We get a list with one element (which itselI is a list) as a
solution.
6.9 solve, lists and sequences 95
(6.31)
We extract the Iirst element oI the list. Notice that that are
Iewer | |s.
(6.32)
We evaluate the system at the solution that solve has Iound and
veriIy that this really does satisIy the system oI equations.
(6.33)
(6.34)
This system oI equations has two distinct solutions, so we get a
list with two elements in it. Each element is a distinct solution.
(6.35)
(6.36)
(6.37)

(6.38)
(6.39)
(6.40)

We can Iind the non-negative roots oI an equation by including
the appropriate inequality in the list oI relations given to solve.
The result by deIault is a set. however, iI the second argument
to solve is a list oI variables, then the result will come back as a
list. You can select items Irom a set using the same notation as
with lists.
96 6 Chapter 6 More sophisticated scripting: textual entry, multiplots, point plots, character strings, lists, solving equations, and evaluation
6.10Evaluation, eval, and assignment
Here's a problem to consider:
We have an expression relating time t to voltage registered by a capacitor as it is being charged by a battery. In a mathematics or electrical
engineering textbook, we might see this written as . We are interested in taking this expression
Ior voltage and doing several calculations with it -- plotting it Ior a range oI t, Iinding values oI t that correspond to a speciIied voltage (e.g. "Iind
the time t when the voltage reached 55 volts"), or Iinding a voltage corresponding to a speciIied time (e.g. "Iind the voltage at t2.5 minutes aIter
the start").
II we set up an assignment in Maple ,, then we could calculate the voltage at t2.5 minutes by
assigning t the value 2.5 and then evaluating V. The second evaluation will cause the current value oI t to be used. However, iI we wanted to plot
the expression J aIter that, then we'd have problems because whenever we would type t, Maple would use the value oI t rather than the svmbol t.
Table 6.14: Evaluating an expression using a particular value of one of the variables in the expression, and then plotting
(6.41)
(6.42)
(6.43)


Error, (in plot) unexpected option: 2.5 0 .. 10

Rather than t0..10 Maple is seeing 2.50..10 because it is
using the value oI t when evaluating what we typed.
We can clear the path Ior plotting by unassigning t Iirst. Note
that iI we did a restart instead oI an unassign we would lose the
assignment to J. restarting at this point is a bad idea, since it
Iorces us to redeIine V as well as t.
6.10 Evaluation, eval, and assignment 97

The same problem would happen iI we tried to solve an equation involving V iI we had already assigned t a value.
Table 6.15: Evaluating an expression using a particular value of one of the variables in the expression, and then solving
(6.44)
(6.45)

We set up an expression and then evaluate it at t4.7 seconds
Warning, solving Ior expressions other than names or
Iunctions is not recommended.
Error, (in solve) a constant is invalid as a variable, 4.7

Because Maple is evaluating the names
V and t in the solve operation, it is seeing
which it cannot
solve because there are no variables in the equation to solve
Ior.
We can clear the path Ior solving by unassigning t Iirst. Doing
a restart would not work, because that would also unassign
98 6 Chapter 6 More sophisticated scripting: textual entry, multiplots, point plots, character strings, lists, solving equations, and evaluation
(6.47)

everything, including V. We would lose the expression we want
to solve Ior.
It can be tedious to have to remember to unassign variables iI we want to go back to using them as symbols in the expression. We recommend
using the eval operation (also available in the clickable menu as ) instead oI assignment, iI you are switching back and Iorth
between using values Ior a variable and using it as a symbol. eval returns the same result as iI you had done assignment temporarily, but there is
nothing to be undone aIterwards.
You can evaluate using values Ior several variables by giving a list oI equations instead oI a single equation as the second argument to eval.
Table 6.16: Evaluating an expression using a particular value of one of the variables in the expression using eval, and then
solving
(6.48)
(6.49)

We set up an expression and then evaluate it at t4.7 seconds.
(6.50)


Evaluating V produces the same result as beIore (an expression
with the symbol t in it), so we can immediately set up an
equation and solve it Ior t.


(6.51)
Note that the eval did not assign a value to t. It's still just a
symbol at this point.
Begin parameters
(6.52)
(6.53)
This is an example oI how to set up a script using parameters
while taking advantage oI eval. Several symbols in the
expression Ior voltage are set up and assigned as parameters.
The expression that describes how voltage changes over time is
not a parameter, although it is assigned a name Ior easier use in
subsequent steps oI the computation.


6.10 Evaluation, eval, and assignment 99
(6.54)
(6.55)
End parameters
(6.56)




(6.57)
(6.58)
(6.59)
(6.60)
Table 6.17: An example using assignment and eval
Begin parameters
(6.61)
(6.62)
This is an example oI how to set up a script using parameters
while taking advantage oI eval. Several symbols in the
expression Ior voltage are set up and assigned as parameters.
But we use eval to maintain and t as symbols in the
expression Jprime.
100 6 Chapter 6 More sophisticated scripting: textual entry, multiplots, point plots, character strings, lists, solving equations, and evaluation
(6.63)
(6.64)
End parameters
(6.65)







(6.66)
(6.67)


We use the inIormation that the capacitor is observed to at vt
volts at time tt0 to Iind the value oI that is consistent with
this.
(6.68)

(6.69)
tExpr is the value oI the expression with the values oI the
parameters and the calculated value oI plugged into Jprime.
We can solve an equation based on this Iormula to Iind the time
when we achieve a voltage oI 55 volts in this conIiguration.
6.11 Summary oI Chapter 6 material 101
6.11Summary of Chapter 6 material
Troubleshooting textual input in Maple
Remember to... Examples with error(s)
Supply a Iunction name and arguments (parameters)
should be
Match delimiters (parenthesis and brackets)
should be
Press the right arrow key to exit Irom variable exponents
should be
Set ranges correctly when plotting
should be
Examples
Plotting multiple expressions
Plotting with lists
(6.70)
(6.71)
Using multiple colors in a multi-plot
Set the titles oI the axes
Using Maple's built-in help
Use Help~Maple Help or press Ctrl-F1 (Command-F1 on a
Mac)
Remember that you can click on related topics when viewing
the help Ior a particular command
Operations on lists Examples
Create a list
(6.72)



102 6 Chapter 6 More sophisticated scripting: textual entry, multiplots, point plots, character strings, lists, solving equations, and evaluation
SpeciIy a sublist oI values
(6.73)
SpeciIy one item Irom the list
(6.74)
SpeciIy a sublist with one item
(6.75)
Count the number oI items in the list
(6.76)
Add together all the items in the list
(6.77)
Compute the average oI all the numbers in the list.
(6.78)
Convert a list into a sequence


Convert a list into a string
(6.79)

(6.80)
Solving a system of equations using lists Examples
Create the system oI equations
(6.81)

Set the variables oI the system
(6.82)

Solve the system and extract the Iirst solution oI possibly many
solutions
(6.83)
Evaluating an expression with eval instead of assignment Examples
6.11 Summary oI Chapter 6 material 103
eval allows you to substitute a value Ior a variable within an
expression, without assigning that value to the variable.
(6.84)
(6.85)
(6.86)

104 6 Chapter 6 More sophisticated scripting: textual entry, multiplots, point plots, character strings, lists, solving equations, and evaluation
105
7Chapter 7 Using and Defining Functions
7.1Chapter overview
Functions occur so much in mathematics that it's natural that Maple knows a lot about them and how to compute with them.
You can also deIine your own Iunctions in Maple, as with most other programming languages.
There are a Iew pitIalls in the use oI Iunctions in Maple:
a) The names oI common mathematical Iunctions used in Maple may diIIer Irom what you are used to Irom high school math.
b) Some Iunctions in Maple do non-mathematical things, such as solve, and plot. Others take novel arguments -- lists, equations, and ranges,
rather than numbers.
c) Maple Iunction deIinition uses : and -~ rather than equals () as Iound in conventional math textbook notation. This is because the "context-
Iree" language processing oI Maple thinks an equation is being deIined whenever it sees an equal sign. It's currently inIeasible Ior conventional
programming language-processing technology to use context.
7.2Functions in computer languages: a way of producing
an output from inputs
Everyone is introduced to the idea oI a function in secondary school mathematics: Calculators can compute many oI the common Iunctions Iound
in high school algebra and pre-calculus: sin, cos, ln, , etc.
Maple can evaluate these Iunctions. The common ones are Iound in the Expression palette but there are hundreds more.
When entering a Iunctional expression, the syntax used is:
function name ( sequence of arguments )
The parentheses are mandatorv in Maple. Forgetting parentheses can be irksome because oIten the computer leaves you to Iigure out yourselI that
the answer you're getting is wrong.
Table 7.1: Function results in Maple
Example Commentary
(7.1)
When we enter a sequence oI Iunctional expressions, we get a
sequence oI results that come Irom evaluating the expressions.
All the common Iunctions know how to give limited-precision
results given limited precision inputs.
(7.2)
Maple's programming will return an exact number iI the
Iunction has that kind oI result Ior the given input. However,
iI a simple exact result can't be Iound, Maple will return
what you typed in. Sometimes there is a little simpliIication
that goes on so what comes out is not literally what comes
in, although it usually obvious that it is mathematically
equivalent. For example, Maple will always simpliIy Iractions
by eliminating the greatest common divisor Irom the numerator
and denominator.
106 7 Chapter 7 Using and DeIining Functions
Example Commentary
Error, invalid input: cos expects its 1st argument, x, to be oI
type algebraic, but received |a, b, c|
This doesn't work because cosine expects an algebraic
expression, not a list.


Error, (in cos) expecting 1 argument, got 2
This mistake might come about iI you typed a comma instead
oI a *.
Error, missing operation
What was this person thinking? Whatever it was, Maple doesn't
know what to do with it.
Error, unable to match delimiters
Another delimiter message. Look Ior extra or missing
parentheses.
(7.5)
(7.6)

This is what happens iI you Iorget the mandatory parentheses.
There is no error message, but what Maple is giving you is the
product oI Pi, the symbol "cos", and 1/4. One oI the ways you
can tell that this isn't is what happens when you ask
Maple to numerically approximate this expression.
(7.7)
(7.8)

This is the correct way to compute the cosine oI radians.
The attachment at the end oI this chapter shows some oI the many other Iunctions available in Maple. Some oI them work on lists, sets, equations
rather than on numbers or expressions. However, the same principle applies: they have a rule Ior taking the value oI their inputs (also known as
arguments) and computing a result Irom them.
7.3 The textual names oI common Iunctions: doing math calculations using the keyboard. 107
7.3The textual names of common functions: doing math
calculations using the keyboard.
Some built-in mathematical Iunctions have textual names that are already quite Iamiliar Irom their use in mathematics textbooks: sin, cos, ln.
Some are used so oIten that the most convenient thing to do is to remember their names. sqrt, abs, min, and max are straightIorward -- they are
naturally thought oI as Iunctions and have names that are abbreviations oI the standard nomenclature. . Others, such as arcsin, or log10 have
names that make sense but you'd have to look them up in the Maple on-line documentation to know.
Table 7.2: Textual names for common math functions
(7.9)


Maple knows "log10" is the base 10 logarithm. It automatically
converts it into the equivalent expression using ln, the natural
(base e) logarithm.



Many users will Iind it Iaster to type in the textual version oI
plot than to use the plot builder. Furthermore, iI we want to
change the expression or the range or labels oI the plot, it's
easier to edit the text and re-execute the region than it would be
with the clickable interIace.

(7.10)
(7.11)
max can Iind the maximum when all oI its arguments are
numbers. max and min can do limited reasoning iI some oI the
arguments are not numeric. For example, in the third example,
Maple knows that abs(x)-2 is not going to be the maximum
because abs(x) is must be larger.
108 7 Chapter 7 Using and DeIining Functions
(7.12)

7.4A function name to commit to memory: exp
The name oI the exponential Iunction in Maple is exp. Instead oI writing e`x, use exp(x).
The textual doppelganger e`x does not work as a way oI calculating a power oI e, the base oI the natural logarithm (where "e" is just the
letter typed at the keyboard, not augmented by command completion as described in the section on command completion (page 108) ). The
orientation in college-level mathematics to view "a power oI e" as a Iunction is a pervasive change in point oI view Irom what you may have seen
in high school. Making a point to use the new notation Irequently is the best way to make the switch.
Table 7.3: exp is the name of the exponential function
(7.13)
(7.14)

exp(x) is the textual way oI writing "the symbolic constant e
raised to the power x. Maple knows how to simpliIy symbolic
expressions with the simplify assuming real operations in
the clickable menu.
(7.15)
(7.16)

We can calculate powers oI e Irom the keyboard. For
comparison we do the same calculation using the Common
Symbols and Expressions Palette. While more oI a "sure
thing", proIicient users would be able to get the keyboard
version entered more quickly.
(7.17)

We can enter an expression involving exponentials using
only the keyboard. Maple will regard it as meaning the same
thing as the expression entered using the combination oI the
keyboard and the Palette.
7.5How can l remember so many functions7 Use command
completion and on-line documentation.
A well-developed system such as Maple, Matlab, or C# has thousands oI built-in Iunctions. You may need to use on a regular basis the names oI
at least several dozen Iunctions to get your work done. The good news is that learning about Iunctions is not that taxing -- iI you own a scientiIic
calculator you've already dealt with a situation where you can operate a dozen Iunctions.
7.6 Being savvy about using on-line documentation 109
A reasonable stance to take is to be Iamiliar (i.e. know "by heart") Iunctions and symbols that you oIten use and to be adept at using
documentation to look up the details oI the ones that you need only occasionally. In exams and tests that let you use a computer, quick recall oI
the most common will be useIul and expected. However, you should also expect to need to use additional Iunctions and inIormation that you
haven't committed to memory. No one will expect you to memorize it all -- but they will expect you to be able to use it quickly anyway with the
help oI on-line documentation.
A quick-recall method Ior access to common Iunctions with textual entry is to type the escape key (Esc) aIter typing the Iirst Iew characters oI
the Iunction's name. A pop-up menu will appear that will list possible ways oI completing what you typed. This is called command completion.
Recall that you've already used this Ieature to enter symbolic constants such as e, the natural logarithm base. II what you are entering is a
Maple operation such as solve, command completion will provide a textual template to Iill in the rest oI the arguments. It will also provide the
parentheses required when entering Iunctional notation textually.
Table 7.4: Command Completion

We type sol and then type the ESC key. A pop up menu shows
option, including several Iorms oI solve.
Clicking on one oI the options provides something to Iill out.
We are not compelled to have a second variable oI x, it's just
short-hand reminder that the second argument is the variable
and the Iirst one is the equation. We can then go on to edit the
"auto completed" template so that we turn the general Iorm into
exactly what we need.
7.6Being savvy about using on-line documentation
Experienced users reIer to the on-line documentation to help remember details about Iunctions. As has mentioned earlier, this is available through
Help -~ Maple Help menu Ieature (key shortcut: press the control key and then the F1 key). II you recall a phrase or a name oI a Iunction, you
can type it into the search Iield and the on-line help system will, like Google, produce the pages it has about your text entry. You can then explore
Iurther by pressing links. Trying the examples typically given at the end oI the description is a good way to get a Iorm that you can use Ior your
own purposes.
110 7 Chapter 7 Using and DeIining Functions
Using on-line help

We want to Iind inIormation about how to use the
inverse sine Iunction in Maple. We start up on-line
help and type in "inverse trigonometric" into the
search Iield, then hit the "search" button. The page we
see does tell us that it's probably called "arcsin" but
we'd like to see more. We see a link in the "see also"
which we click on.
This uncovers more links. The one that says "invtrig"
seems promising so we click on that link.

This seems to be the page we want to spend time
looking at. There are plenty oI examples at the
bottom.
7.7 DeIining your own Iunctions with -~ (arrow) 111
7.7Defining your own functions with -> (arrow)
Maple allows you to deIine simple Iunctions with the use oI -~. The general Iorm is
function name : ( sequence of arguments) -~ expression that describes result .
These can be entered through the Expression Palette, or textually. The arrow is entered textually by typing a - and then a ~, with no spaces
separating them. We give an example oI Iunction deIinition in the Iollowing example.
Table 7.5: A function definition in a math textbook
Fruit Ilies are placed in a halI-pint milk bottle with a banana (Ior Iood) and yeast plants (Ior Iood and to provide a stimulus to
lay eggs). Suppose that the Iruit Ily population aIter t days is given by


(a) What is the population aIter Iive days?
(b) How long does it take Ior the population to reach 180?
Analyzing the text, we see that it deIines a Iunction named P. It takes one input (argument), t, and produces as output whatever you get Irom
evaluating the expression . Assuming that Maple understands the deIinition oI P like it does a built-in Iunction, then the
description oI what happens when you evaluate P(5) would be:
"Substitute 5 Ior wherever you see t in the expression . This gives you . PerIorm all the
arithmetic and relevant simpliIications in this expression and return that as the result oI the Iunction."
Even though 5 is an exact number, because there are limited precision numbers in the expression we expect that the result will be a limited-
precision number. II the expression had only exact numbers in it, then the calculation would be done exactly.
What we would like to do is to tell Maple about the deIinition oI P and use it in our work.
We can do this through the clickable interIace. We anticipate reuse oI this Ior other days and population levels, and turn it into a parameterized
script:
Table 7.6: User-defined functions through the Expression Palette
User-defined functions through the Expression Palette
Commentary
Start of parameters

The general Iorm is Irom the in the Expression
palette. We alter slots in the template to mention the
speciIic Iunction by name, the name oI the input, and the
expression that describes how to calculate the output.
112 7 Chapter 7 Using and DeIining Functions
User-defined functions through the Expression Palette
(7.18)

(7.19)

End of parameters

(7.20)







Compute the number oI Ilies aIter days.

(7.21)


With this deIinition, we can compute P(5) using the natural
mathematical notation.
We plot the Iunction to see how the population grows. This is not
needed by the problem but it helps us understand the situation
better.

We can plot P(t) like we would any other expression. One
the options to plot (see plot options in on-line help) is the
ability to speciIy the title oI the graph by giving titlestring
as an additional argument to the plot Iunction. Note that we
are not using the clickable interIace to do plotting.
7.8 Using Iunctional evaluation instead oI using assignment or eval 113
User-defined functions through the Expression Palette

Compute when the population reaches desired level by solving the
equation P(t)popLevel.

(7.22)

End of script

In anticipation oI using the value in later work, we assign
it to the variable soln. We would then intend to take Iurther
steps using soln.
We could have deIined the Iunction textually just by typing P := (t) - > ... Iollowed by the rest oI the expression instead oI using the
Expression Palette
7.8Using functional evaluation instead of using
assignment or eval
The problem with assigning variables is that it may cause unwanted side-eIIects, such as when trying to solve or plot with expressions involving
those variables. We have seen in the section on eval (page 94) that we can use eval instead oI assignment, which saves us the chore oI having
114 7 Chapter 7 Using and DeIining Functions
to assign and then unassign variables so that they remain as symbols when solving or plotting. We can more succinctly avoid this problem by
deIining a Iunction based on the expression, and then causing evaluation to occur with standard Iunctional notation.
Table 7.7: Comparing evaluation using eval and functional notation
(7.23)
(7.24)
(7.25)

We set up an expression and then evaluate it at t4.7 seconds.
(7.26)
(7.27)
(7.28)

We create a Iunction oI t using the arrow notation. We can then
evaluate the Iunction at any value oI t that we want through the
standard notation. This takes a little less typing than using eval.
7.8 Using Iunctional evaluation instead oI using assignment or eval 115


We can plot the Iunction Jf in a natural way. plot also has an
abbreviated way oI plotting Iunctions -- just give the name oI
the Iunction and the plotting range, and omit the name oI the
Iunctional argument. You can read more about this in the on-
line help Ior plot.
116 7 Chapter 7 Using and DeIining Functions
7.9Troubleshooting function definitions
Function deIinition in Maple diIIers Irom the standard notation in high school mathematics This is because computers Iind it hard to detect the
context where language is being used, so the same notation has to work both Ior mathematics and programming. The standard notation can't made
be made to do this, so extensions and modiIications to the notation need to be made. (In other words, it's not a conspiracy to make your liIe harder
gratuitously.)
The technology that is standard in most computer language understanding systems needs to assign a unique meaning to input Irom the way it
looks. Equations already use "" so iI you use I(x) ... Maple will understand you to be talking about an equation, not a Iunction deIinition.
Use : when assigning a Iunction a name.
Table 7.8: Troubleshooting function definitions
(7.29)
(7.30)
(7.31)
This deIines a Iunction Ior the gravitational attraction Iorce.
m1, m2 and r are the arguments. g is another symbol used in
the expression but it isn't included as an argument.


We deIine the value oI g. We expect not to change this while
using the Iunction. We Iound the value by typing "gravitation"
into Maple on-line help.

This calculates the attraction Iorce between a 1 kilogram and 2
kilogram mass that is 10 meters apart. The Iorce is in units oI
Newtons.
Error, invalid operator parameter name

This doesn't work at all. Do you see the diIIerence between the
deIinition oI F and Fbad? One uses assignment : the other is
either mistyped or mistaken.






(7.33)

This tries to use "" instead oI "-~" to deIine the Iunction FF.
There is no error message, however, because this assigns FF
an equation rather than a Iunction deIinition, which is perIectly
valid, even iI it isn't what we were thinking oI.
7.10 Using Iunctions Irom library packages, with 117

(7.34)

We can tell that this supposed Iunction deIinition isn't working
when we test it. When we try to use the Iunction in the usual
way we don't get the number we were expecting, as with
F(1,2,10). With testing, you veriIy, looking Ior evidence
stronger than "no error message" that what you entered is what
you want.
(7.35)

This illustrates an alternative way oI deIining a Iunction,
although it is not the Iorm recommend by these notes. Rather
than "" as would appear in a math textbook, the assignment
operation ":" is used instead. This produces the Iollowing
pop-up:

Clicking "ok" to Iunction deIinition will create the proper
Iunction deIinition, as the subsequent line oI the computation
indicates. .
(7.36)

Testing this Iunction deIinition, we see that we get the same
result as with F
7.10Using functions from library packages, with
As with other systems Ior technical computation, Maple has thousands oI built-in Iunctions. The deIinitions Ior some oI them such as cos or
solve are automatically loaded into Maple when it starts up. However, iI all deIinitions were loaded, Maple would take a long time to start up and
would require large amounts oI memory even beIore you had done any work in it. To avoid this, most Iunctions are loaded only when the user
requests it.
Most built-in Iunctions are organized into collections called packages. The general way to access a Iunction belonging to a
package is through package[function]. The least squares Iunction belongs to a package named CurveFitting, hence its Iull name is
CurveFitting[LeastSquares].
The with(package name) operation in Maple will load all the Iunctions in the speciIied package into Maple. AIter this operation, Iunctions can
be reIerred to with just their "short name", e.g. LeastSquares rather than CurveFitting|LeastSquares|. Doing a with(package) can save you typing
iI you expect to use a Iunction, or several Iunctions, Irom a package several times during a Maple session. Ending the line with a colon (:) will
suppress printing oI all the Iunctions in the package that usually occurs.
118 7 Chapter 7 Using and DeIining Functions
Table 7.9: with, the package loading operation


(7.37)
(7.38)
(7.39)

The CurveFitting package has a number oI Iunctions
Ior data Iitting. One is called LeastSquares. However,
LeastSquares(....) does nothing. Although there is
no error message, this is a mistake -- we didn't load
the package in with a "with" so the least squares data
Iitting Iunction isn't known. We get the same kind oI
behavior as iI we had typed in f(1,2,3) with f undeIined
-- Maple just outputs what we typed in.


(7.40)
Doing a "with" gives the names oI all the Iunctions in
the package.


(7.41)

We can Iit a line to temperature as a Iunction oI pressure.
(7.42)
Once we do the with, LeastSquares works.
7.11Attachment: some built-in problem-solving functions
The Iunctions in the Expression palette have the same name and work similarly to those described in math textbooks. The operations discussed
in this attachment are also Iound in math textbooks, but they are usually not given Iunction names. It may seem novel to you that the rules Ior
solving equations, Iactoring polynomials, or plotting can be collected together and given a Iunction name. Yet this way oI writing about such
actions allows us to combine mathematics and programming. Thus solve, plot, factor, etc. are true Iunctions -- they have names, they are invoked
with arguments, and return results that can be assigned to a variable.
7.11 Attachment: some built-in problem-solving Iunctions 119
Table 7.10: Textual names of common operations in Maple
Function Textual name of
function
Example Commentary
solve an expression or an
equation
solve
(7.43)
(7.44)

II there is more than one root,
returns a sequence oI values
solve an expression or an
equation numerically
fsolve
(7.45)
(7.46)
Error, (in Isolve) a, b, c} are in the
equation, and are not solved Ior

II there is more than one root,
returns a sequence oI values



fsolve won't work iI the answer
is not a number.
plot an expression plot

Plot takes two inputs. The Iirst
is an expression, the second
is an equation naming the
variable and the horizontal plot
range.










120 7 Chapter 7 Using and DeIining Functions










plot may take more than two
arguments, optionally. The rest
oI the arguments are reIerred
to as plot options.
evaluate at a point
(7.47)
(7.48)

eval takes two arguments
(inputs). The Iirst is an
expression, the second is an
equation or a list oI equations
indicating what values to use
Ior one or more variables in
the expression.
right hand side, leIt hand
side oI an equation
rhs
lhs
(7.49)
(7.50)
(7.51)







The third example shows that
Maple thinks that "x0..10" is
an equation even iI it isn't one
in the standard mathematical
sense.
7.12 Summary oI Chapter 7 material 121
approximate with a
limited-precision number
evalf
(7.52)
(7.53)
(7.54)
evalI has an optional second
argument. II it's not there,
Maple will compute a 10
digit approximation. II the
second argument provided is
a positive integer, then Maple
will compute that many digits.
convert between units convert
(7.55)
(7.56)
(7.57)

convert does many things.
When it is given Iour
arguments and the second
argument it units, then it
expects the Iirst argument to
be a number, and the third
and Iourth to be expressions
describing the units being
converted Irom and to. Note
that the units can be ratios
or products rather than just
names.
7.12Summary of Chapter 7 material
Common mathematical functions (see on-line help for index of functions)
sin, cos, tan, sec, csc, cot trigonometric Iunctions
exp
exponential Iunction (exp(x) means )
arcsin, arccos, arctan, arcsec, arccsc, arccot inverse trigonometric Iunctions
sinh, cosh, tanh, .... hyperbolic trigonometric Iunctions
ln, log10, log|b| logarithm, log base 10, log base b
min, max minimum and maximum
abs, sqrt absolute value, square root
ceil, Iloor, trunc, Irac ceiling, Iloor, truncation, Iractional part
Command Completion
When entering math, type the Iirst part oI the name oI the
Iunction and then hit the escape key. A pop-up window will
appear with a list oI alternative competitions oI what you
typed. Pick one, and a template will automatically be entered
Ior you. Edit the spots oI the template to Iit your situation.

(then hit the ESC key)

122 7 Chapter 7 Using and DeIining Functions
This is a way to get the computer to automatically type the
right kind oI delimiters.

Defining custom functions with the arrow (->) notation Examples
Create a custom Iunction by naming it and its parameters. A
custom Iunction deIined using arrow notation is oI the Iorm:


In this case the Iunction name is P and its sole parameter is t.
For multiple parameters, use the Iorm:

(7.58)
(7.59)
To call the Iunction, use the Iorm:
(7.60)
(7.61)
To plot the Iunction, set the range oI an independent variable,
and use it as a parameter to your custom Iunction.
To solve the Iunction Ior a particular value, use the solve
Iunction.
(7.62)
(7.63)
Troubleshooting function definitions
Error Examples
7.12 Summary oI Chapter 7 material 123
Troubleshooting function definitions
Forgetting to use the assignment (:) operator.
Error, invalid operator parameter name
Using the equality operator () instead oI the arrow in Iunction
deIinition.
(7.65)
Although technically not an error, deIining a Iunction without
using the arrow notation is not prescribed by these notes.
(7.66)
124 7 Chapter 7 Using and DeIining Functions
125
8Chapter 8 Programming with functions
8.1Chapter overview
In computer languages, Iunctions or procedures are the building blocks Ior programming. An individual Iunction can be programmed to do a
lot, but a single Iunction with a lot oI programming can be expensive and time-consuming to build. OIten a large program is built by combining
Iunctions through daisv chaining -- by making the output oI one Iunction the input oI another. This is sometimes reIerred to as functional
composition. Frequently used daisy chains can themselves be deIined as a Iunction, making them easier to reuse.
Most computer languages extend the concept oI Iunctions to go beyond the numbers or Iormulas that "mathematical Iunctions" provide. In Maple,
as in most languages, a Iunction can return other kinds oI results. It is Iairly common in Maple and other languages to have Iunctions that produce
as output a list, an equation, or a string as a result. As we shall see, Iunctions can even return a plot as the result oI a Iunction. It is also possible
Ior computer Iunctions to have lists, equations, plots, or strings as inputs.
8.2Designing functions from context
In doing technical work, we oIten see Iunctions deIined as an equation relating the name oI the Iunction, its argument(s), and the Iunction
deIinition.. Those are easy to translate into Maple's notation and use. For example iI we see in a mathematics book "deIine I(x)
" then we can just transcribe it into the Maple Iunction notation: .
In word problems, we have to "read between the lines" and design the Iunction. This requires answering the questions:
a) What will the inputs to the Iunction be? Try to give symbolic names Ior it.
b) What will the output be? Sometimes to realize what the output is, you can create a worksheet with several steps. II there is only one Iinal result,
then that should be the output.
c) How do you calculate the output Irom the inputs? HopeIully, there's a simple Iormula that describes this.
We illustrate this process with an example:
Designing a function for pressure/temperature problems
Problem p. 180 in Introduction to Engineering by Jay Brockman, Wiley, 2009
The Ideal Gas Law, as stated in Introduction to Engineering ) is:
(8.1)
where
P is pressure in Pascals (Pa)
J is volume in .
n is the amount oI gas in moles (mol),
T is the temperature in degrees K,
R is the gas constant, approximately .
126 8 Chapter 8 Programming with Iunctions
We want to solve the Iollowing problem (actually, various versions oI it):
Problem
We measure the temperature and pressure oI a gas. It has a pressure oI 100 and a temperature 473 degree Kelvin. Then cool it so it has a
pressure oI 89.8 . What is its temperature?.

Source: Introduction to Engineering by Jay
Brockman, Wiley, 2009 (p. 180)
Finding the answer
First, we do the mathematical thinking and inIormal calculation that allows us to build a Iunction that will solve all problems oI this type:
For a Iixed cylinder volume, according to the Ideal Gas Law:
, so . so

Function design
We see that the problem wants us to calculate a temperature given the atmospheric pressure , the internal pressure , and the Iirst
temperature, .
The output is , the inputs are , , and . We are Iree to name the Iunction anything we want since the problem statement doesn't name
this. We decide to call it something that reminds us oI the purpose.
(8.2)
Testing and troubleshooting the function
We see whether we get the intended result with the numbers we've already worked out. Note that iI we hadn't done the analysis, we wouldn't have
any way oI testing what we designed.
8.2 Designing Iunctions Irom context 127
(8.3)
That isn't the same result. What did we do wrong? The Iormula 1.2.2 seems like the right thing. What else could go wrong? Close inspection
indicates that the Iirst argument to internalTemp is P1, which appears in the denominator oI the Iormula Ior the output. In (1.2.3), that would put
the "89.8" in the denominator, but our example had 89.8 in the numerator. Oops, we gave the values in the wrong order Ior the Iunction. There's
nothing wrong except that we should invoke the Iunction with the inIormation given in the correct order:
(8.4)
This result agrees with what we think the result should be Irom our previous calculations. Note that in testing Iunctions, it may be necessary to
work through at least one calculation on your own in order to have a way to veriIy that the Iunction is working.
Using the function
We are given a diIIerent version oI the problem:
We measure the temperature and pressure oI a gas. It has a pressure oI 2000 and a temperature 473 degree Kelvin. Then cool it so it has
a pressure oI 53.6 . What is its temperature?.
Answer:
(8.5)
Since the answer is in degrees Kelvin, this is only about 14 degrees above absolute zero. That's pretty cold!
The usefulness of alternative function designs
Suppose we had this new problem:
We measure the temperature and pressure oI a gas. It has a pressure oI 100 and a temperature 473 degree . We then heat it to 512
degrees Kelvin. What is its pressure then?
Another function designed
A little thought produces the calculation:

This leads to the Iunction deIinition:
(8.6)
128 8 Chapter 8 Programming with Iunctions
We test this (remembering what happened beIore about the order oI arguments)
(8.7)
Conclusion
To develop Iunctions, it helps to have worked through some typical calculations interactively. Once you have realized which quantities you are
starting with and named them, and have developed the Iormula Ior the calculation using those names, you can create a Iunction deIinition. You
can use the names given in the problem description, or you can make up names based on their purpose. Unlike mathematics, you are not limited
to single letters Ior names oI variables or names oI Iunctions. Computer programmers know that longer names are oIten easier to remember or
understand.
8.3Function composition: daisy-chaining functions
together
In the scripts we have developed so Iar, we have developed a result through a sequences oI actions. These sequences can oIten be described
through Iunctional composition -- an expression that chains together several actions. Consider the Iollowing example:
Problem
On November 1, 2007, one Euro was worth US dollars. We are buying widgets that cost 30 Euros each and importing them into
the US. We then put the widgets into packages that cost .075 dollars. How much should we sell the product in the US iI we want a 10 markup
Irom our costs?
Finding the solution: step 1, first do the calculation interactively.
Doing this in the style oI scripts, we Iirst assign the values to variables, and then do the calculational steps.
(8.8)
(8.9)
(8.10)
(8.11)
(8.12)
(8.13)
We Ioresee using this calculation several times as the conversion Rate, the manuIacturing cost in Europe, and the packaging cost change. We even
see that the markup might change. We can try to boil down these steps into a Iew Iunctions.
8.3 Function composition: daisy-chaining Iunctions together 129
Designing the solution: step 2, design functions to do the calculational steps
(8.14)
(8.15)
Note that the way that the third Iunction sellingPriceFunc is deIined, it daisy chains totalCostFunc and makes the output oI its invocation into one
oI the inputs to priceFunc.
Testing the solution: step 3, test the building blocks in the order that they are used
We test the Iirst two Iunctions. AIter we see that they agree with our preliminary version oI the calculations, we test the third Iunction that
depends on the correctness oI the Iirst two.
(8.16)
(8.17)
(8.18)
We could write a script that just used totalCostFunc and priceFunc, but by designing and using a third Iunction, we reduce the work oI handling
an instance oI the problem to just pasting in the values Ior the Iour parameters in one line. This is probably less work than changing Iour lines oI
parameters that we have had with our previous approach to scripts.
While Iunction composition is a succinct way oI ordering many operations, its advantages are apparent only after the chain is built and tested as
working correctlv. It maybe easier to develop the chain as a script oI assignments and then refactor -- rewrite without changing the meaning --
the script so that it uses user-deIined Iunctions to replace some oI the chains oI assignments. In this situation, the advantage oI reIactoring is that
the user-deIined Iunction is much easier to reuse than a sequence oI assignments (which must be copied and edited Ior reuse).
Using the solution: step 4, present a script that defines the functions, then invoke the "answer function" repeatedly to handle various
versions of the problem
Table 8.1: A script that uses functional (chaining), and its use
Begin function definitions
(8.19)
130 8 Chapter 8 Programming with Iunctions
(8.20)
(8.21)
End function definitions

Problem solving

Version 1

On November 1, 2002, one Euro was worth US dollars. We are buying widgets that cost 30 Euros each
and importing them into the US. We then put the widgets into packages that cost .075 dollars. How much should we sell the
product in the US iI we want a 10 markup Irom our costs?

(8.22)

(We got the number Iormatted to currency by right-click-~Numeric Formatting-~Currency.)

Version 2

On November 1, 2007, one Euro was worth 1.4487 US dollars. We are buying widgets that cost 33 Euros each and importing
them into the US. We then put the widgets into packages that cost .09 dollars. How much should we sell the product in the US iI
we want a 10 markup Irom our costs?

(8.23)
Version 3
8.4 Expressions with units oI measurements: convert 131

On November 1, 2010, one Euro was worth 1.4728 US dollars. We are buying widgets that cost 35 Euros each and importing
them into the US. We then put the widgets into packages that cost .10 dollars. How much should we sell the product in the US iI
we want a 8 markup Irom our costs?

(8.24)

By deIining our solution as a sequence oI Iunction deIinitions, we can solve several versions oI a problem just by invoking the
Iunction with diIIerent parameters, without needing to copy lines oI calculations.
8.4Expressions with units of measurements: convert
Maple has Iacilities Ior converting between various English and metric units. It is useIul Ior doing multi-step calculations because the conversions
happen automatically.
In the Iirst way oI using convert, one thinks oI a value as implicitly expressing a number oI units and wants another number expressing those
number oI units converted to another unit. One uses convert( value, units, fromUnit, toUnit).
Table 8.2: Examples of Unit Conversion
Examples of unit conversion
How many inches in a meter? How many Ieet in a kilometer?
How many millimeters in a mile?

(8.25)
(8.26)
(8.27)










Note that the answer to this was expressed as a Iloating point
number rather than a Iraction because the input was Iloating
point (1.0)
(8.28)
Maple can convert between most compatible units.
132 8 Chapter 8 Programming with Iunctions
Examples of unit conversion
(8.29)
(8.30)
(8.31)
Sometimes units are expressed as ratios oI other units. Maple
can handle such conversions as well
In many examples in the Maple documentation, some oI the arguments to convert are quoted -- surrounded by apostrophes -- to prevent
evaluation Irom using the value oI the names oI the units. For example, iI you have assigned a value to the variable s, then you cannot convert to
seconds with this name without quotation.
Troubleshooting unit conversion
(8.32)


As long as the various names used as arguments to the convert
Iunction don't have values, things work Iine.
Error, (in convert/units) unable to convert `min` to `259200`

Maple perIorms evaluation oI names as it Iigures out what the
inputs to convert is. Since seconds has a value, Maple tries
to compute convert(4, units, minutes, 259200). Since the 4th
argument to convert has to be a name, an error results.
(8.33)

Quoting the 4th argument causes the name seconds to be given
as the 4th input to convert. This works.
(8.34)

Quoting all the names as a prophylactic measure is acceptable.
You see this in a lot oI the Maple on-line documentation.
In Star Wars Episode IJ. A New Hope, Han Solo says that the
Millennium Falcon made the Kessel Run in "less than twelve
parsecs". We want to know how many days a parsec is.

Error, (in convert/units) unable to convert `pc` to `d`

This is the error message you see when you are trying to
convert between incompatible units, e.g. trying to convert a
gallon into a meter. pc seems to be Maple's internal name Ior
parsec, d the name Ior days.
8.4 Expressions with units oI measurements: convert 133
Troubleshooting unit conversion

(8.35)
A parsec is a non-Iictional unit oI distance, not time, so we can
convert 12 parsecs to miles, kilometers, inches.... But we can't
convert it to days any more than we can convert inches to volts.
Table 8.3: A problem solved with a script using function definition
A problem solved, a script built using function definitions
A car travels a 45 miles per hour. How many minutes does it
take to travel 900 kilometers?

(8.36)
(8.37)
(8.38)
(8.39)
(8.40)

We build a sequence oI calculations to understand how to solve
this problem. This is the inIormal phase oI development, while
we are trying to understand what to do. Once we have an idea,
we start designing Iunctions and testing them.
It would pretty obvious how to make a script out oI this to
handle any problem oI the Iorm: A car travels speed miles
per hour. How many minutes does it take to travel distance
kilometers... With a Iew user deIined Iunctions, we can get the
answer with less typing/cutting/pasting


(8.41)
The Iirst step was to convert the distance Irom kilometers to
miles. We create a Iunction that does it.

We test it to the result that we got in the script above and see
that it agrees.

134 8 Chapter 8 Programming with Iunctions
A problem solved, a script built using function definitions
(8.42)
(8.43)
(8.44)
The next step was to calculate the time (in hours) Irom the
distance in miles and the speed in mph.



The test shows that tCalc seems to be built correctly.
(8.45)
(8.46)
The third step was to convert the time Irom hours to minutes.

The test oI this step agrees with what the script does, too.
(8.47)
(8.48)
The solution Iunction chains together the three Iunctions we've
developed. This concludes the development and testing. We
present a script and several solved problems in the Iigure
below.
The above table showed the thinking behind the design and testing oI the multi-step calculation. However, unless the reader wanted to see how
the programming was developed, we wouldn't show the testing or the initial script. We would just the deIinition oI the Iunctions, and the results
oI repeated invocation oI the "solution Iunction".
Table 8.4: Solving multiple versions of a problem through functions
Solving multiple versions of a problem through functions

Begin function definitions

(8.49)
The travelSoln Iunction indicates the parameters oI the script,
speed and distance. In setting up the solution method as a
Iunction, we lose some intelligibility because oI the "inside
out" style oI Iollowing the operation oI daisy-chained Iunction
composition.

However, we gain convenience using the script. Using this
Iorm makes it easier to see several solutions, because you don't
have to wade through all the lines oI script that work out the
8.4 Expressions with units oI measurements: convert 135
Solving multiple versions of a problem through functions
(8.50)
(8.51)
(8.52)
End of function definitions

Problem version A
A car travels at 45 miles per hour. How many minutes does it
take to travel 900 kilometers?

(8.53)
Problem version B
A car travels at 45 miles per hour. How many minutes does it
take to travel 452 kilometers?

(8.54)
Problem version C
A car travels at 65 miles per hour. How many minutes does it
take to travel 1500 kilometers?

(8.55)
answer, just a single line setting up the values oI the parameters
and printing out the answer.
136 8 Chapter 8 Programming with Iunctions
For casual unit conversion, it can still be useIul to rely upon Maple's encyclopaedic knowledge oI how to convert units. You can access this
through Tools-~Assistants-~Unit Calculator
Table 8.5: Unit Converter Assistant
8.5User-defined functions whose inputs and outputs
aren't numbers
Although you don't see much mention oI this in mathematics texts, it is Iairly common while programming to deIine and use Iunctions that take
inputs and produce outputs that are not numbers. For example, iI we have a list L oI numbers, we can create a Iunction that takes a list as input
and produces the average oI all the numbers as its output.
Table 8.6: A function that takes a list as its input
A function that takes a list as its input
Example Commentary
(8.56)

We are expecting the input to be a list oI numbers. As
explained in More operations on lists (page 92), L|i| uses
indexing to get the i-th value oI the list, and nops(L) is the
number oI elements in the list.

(8.57)

When we invoke the Iunction, L is |5,7,-3,2,6|, so nops(L) is 5.

(8.58)
With a diIIerent list as input, nops(L) is 7. Since at least one oI
the elements oI the list was a limited precision number (5. has
8.5 User-deIined Iunctions whose inputs and outputs aren't numbers 137
A function that takes a list as its input
a decimal point), the limited precision arithmetic is perIormed
with it and subsequent steps oI the sum.
In analyzing mathematical models as we have been doing, it is also useIul to produce abbreviations Ior common combinations oI plot options by
creating a Iunction that produces a plot as its result.
Table 8.7: A function that returns a plot as its output, rather than a number
A function that returns a plot as its output, rather than a number
Problem

We are given two lists oI data, pData, and tData. Plot pData as a Iunction oI tData, and vice versa.

Solution

Build a Iunction that becomes an abbreviation Ior the operations in the plot. Provide a third argument that is the string Ior the
color.

(8.59)
(8.60)
(8.61)

Plot pressure versus temperature, in red. Note: there seems to be a bug in Maple that suppresses the printing of the
horizontal axis label.


138 8 Chapter 8 Programming with Iunctions
A function that returns a plot as its output, rather than a number
Plot temperature versus pressure, in red. Copying and pasting the invocation of the PlotIt function we defined is easier
than changing the insides of the original plot operation.

8.5 User-deIined Iunctions whose inputs and outputs aren't numbers 139
A function that returns a plot as its output, rather than a number

Because we have parameterized the data and the axis labels, it's easy to plot pressure versus temperature instead. We
don't have to rewrite the directions in PlotIt, we just invoke it with the values and labels in reverse order.

140 8 Chapter 8 Programming with Iunctions
A function that returns a plot as its output, rather than a number


It is possible to return a list or sequence as a result oI a Iunction. Such a Iunction can be put in a chain.
Table 8.8: A problem solved with a function that outputs a sequence of two numbers
A problem solved with a function that outputs a sequence of two numbers
Problem A

A right triangle has a hypotenuse of length 5 feet. The angle between it and one of its sides is 10 degrees. What is the
length of the perimeter?

We Iirst build a Iunction that computes the two sides oI the right triangle and returns the two values as a sequence. We have to
convert degrees into radians in order to do this because the Maple trig Iunctions all use radians..

8.5 User-deIined Iunctions whose inputs and outputs aren't numbers 141
A problem solved with a function that outputs a sequence of two numbers
(8.62)

Let's test the sideSide Iunction.


(8.63)

Now, develop a function that takes a sequence of three numbers and adds them together.

(8.64)
By chaining together the output of sideSide and making it part of the input of sumSides, we can get the whole
computation done in one function.

(8.65)
(8.66)
(8.67)

Problem B

A right triangle has a hypotenuse of length 10 feet. The angle between it and one of its sides is 42 degrees. What is the
length of the perimeter?
142 8 Chapter 8 Programming with Iunctions
A problem solved with a function that outputs a sequence of two numbers

(8.68)

Once we have done the work to design and test the Iunctions out on a problem, we can present a script that can solve several diIIerent versions oI
the problem:
Table 8.9: several versions of a function with function definitions
Solving several versions of a function with function definitions
Begin function definitions

A Iunction that computes the two sides oI a right triangle given the angle and the length oI the hypotenuse

(8.69)


A Iunction that takes a sequence oI three numbers and adds them together.

(8.70)
Compute the perimeter by summing the three sides.

(8.71)

End function definitions


8.6 Chapter Summary 143
Solving several versions of a function with function definitions
Problem A Solution

A right triangle has a hypotenuse of length 5 feet. The angle between it and one of its sides is 10 degrees. Approximately,
what is the length of the perimeter in feet?

(8.72)

Problem B Solution

A right triangle has a hypotenuse of length 10 feet. The angle between it and one of its sides is 42 degrees.
Approximately, what is the length of the perimeter in feet?

(8.73)
8.6Chapter Summary
Function design
Designing Iunctions Irom context a) What will the inputs be?
b) What will the output be?
c) How do we calculate the output Irom the inputs?
Function composition
(8.74)
(8.75)
(8.76)
144 8 Chapter 8 Programming with Iunctions
Function design

Unit conversion
Units can be converted directly into compatible units.
(8.77)

Compound units expressed in ratio Iorm can also be
converted into compatible compound units.
(8.78)

Converting between incompatible units will generate
an error message.
Error, (in convert/units) unable to convert `d` to `mi`

II we create a variable with the same name as a unit,
trying to convert using the variable name will throw
an error.




To avoid this, we can use single quotes around the
unit names to speciIy using the unit string, as opposed
to the variable value.
(8.79)
Error, (in convert/units) unable to convert `d` to `5`

(8.80)

We can use Maple's built in unit converter to convert
units using drop-down menus.
This tool is located in Tools~Assistants~Unit Calculator
Non-number inputs and outputs of a function
Using a list oI numbers as an
input
(8.81)
8.6 Chapter Summary 145
Non-number inputs and outputs of a function
(8.82)

Returning a plot instead oI a
number
(8.83)

Returning a sequence oI
numbers instead oI a single
number
(8.84)
(8.85)


146 8 Chapter 8 Programming with Iunctions
147
9Chapter 9 Visualization, modeling, and
simulation
9.1Chapter Overview
1. Plot structures are the result oI the plot Iunction. They can be assigned to variables through : just as numbers, Iormulas, lists, and Iunction
deIinitions, can be.
2. The displav operation oI the plots package allows you to combine together plots. OIten times a single picture can be more enlightening or
easier to understand than looking at multiple pictures separately.
3. Simulation is the art oI predicting the behavior oI system entities as they change over time, through the use oI mathematical models. It can
be as simple as using Iunctions that, given the time t as input, calculate the position, size, weight, or other changing properties oI a situation.
With the appropriate mathematics, personal computer or supercomputer-class calculations can be used to come up with reasonably accurate
descriptions oI phenomena. Computational simulations have become a mainstay oI modern engineering because "build it and see" can be much
slower and more expensive to do Ior serious projects.
4. The animate operation oI the plots package is explained. Its use is illustrated with a session oI question-answering using a mathematical models
oI moving bodies. Computer-generated animations are another useIul tool besides solve, and plot.
9.2plot structures
Like solve, Maple plot is a Iunction: it has inputs and produces outputs. What kind oI output does the plot Iunction produce? In Maple, the result
oI plot is a special type oI result called a plot structure. When you evaluate an expression in Maple that invokes the plot Iunction, a plot structure
is created.
As with any expression in Maple, you have the choice oI just typing in the expression in and seeing the result oI evaluating it, or assigning the
value to a variable through :. Just typing in the plot expression displays the plot picture.When the plot structure is then assigned to a variable ,
then an ellipsis oI the plot structure is displayed rather than the plot picture.
148 9 Chapter 9 Visualization, modeling, and simulation
Table 9.1: Plot results, displayed and not displayed
Example Commentary

The plot structure result oI this operation is not being assigned
to a variable through :. So Maple displays the structure in a
pretty way -- by drawing a picture oI all the points and labels
established by the plot operation and put into the plot structure.

(9.1)
Evaluating an expression and then assigning it to the name p.
This does not display the plot. We just see PLOT(...) which is a
sign that the value oI p is a plot structure.
9.3 plots|display | and combining plots 149
Example Commentary
Typing in p causes Maple to evaluate it. Since p's value is the
plot structure assigned to it previously, Maple will display the
plot.
9.3plots]display | and combining plots
The display Iunction Irom the plots package takes as its Iirst argument a list oI plot structures. It will produce a plot structure that combines all the
plots together. displav is the way to get a multi-plot in a script without doing cutting and pasting oI plots.
Table 9.2: Display combines plots
display combines plots
(9.2)
(9.3)

Someone we know has used a variant on LeastSquares
Curve Fitting, to derive a Iormula with that Iits the
data. We wish to plot both the data points and the
Iormula on a single graph.
150 9 Chapter 9 Visualization, modeling, and simulation
display combines plots

This is the point plot we get Irom this data.
(9.4)
p1 is assigned the plot structure oI the point plot.
Because the result is assigned to the variable p1, only
an ellipsis oI the plot structure is displayed rather than
the picture oI the plot.
(9.5)

Let's assume that we have a Iriend who has done
curve Iitting to timeData and tempData and gotten this
Iormula. In ENGR 101 Fall 2010, Matlab was used
to do exponential curve Iitting like this. We could
replication that calculation in Maple, but don't show
it here. II you're interested in seeing how to do it, ask
your instructor about it.
(9.6)
p2 is assigned another plot structure, the smooth
plot oI the Iormula. We skip the step where we show
this plot as a separate picture because we are more
interested in seeing it combined with the point plot.
We load the plots package. Because we ended the line
with a colon (:), the list oI Iunctions in the package
is suppressed. In general, ending a line with a colon
suppresses the normal output.
9.4 plottools, lines, and other shapes 151
display combines plots
displav's argument must be a list oI plot structures.
This Iorm oI the operation will take all the plots and
superimpose them together. This is a good way to
have a multi-plot in a script.
9.4plottools, lines, and other shapes
The plottools package has a number oI Iunctions that are useIul Ior inclusion in visualizations (plots). For example, you can create a line segment
oI any desired color and line thickness with the line function.
One can then use the displav Iunction to merge together lines, plots, and other shapes.
The on-line documentation on plottools contains links to Iurther description.
Table 9.3: BROKEN - MISSING TITLE!
We create a bit oI "modern art" by drawing a circle and a point plot.
(9.7)
As the on-line documentation indicates, the Iirst two arguments to line are lists indicating the coordinates oI the starting point
and ending point oI the line segment. There are optional arguments that indicate color and line thickness, etc.

152 9 Chapter 9 Visualization, modeling, and simulation
(9.8)
(9.9)
We suppress printing oI the plot structure CURVES(....) Ior c3 with a colon because it's long and we don't want to see it, we
want to see the picture it describes. Note that Maple exposes its Canadian roots by using "colour".
One oI the options to plot (and display) is to not show the axes. Another option is to indicate that the scaling should be
constrained to be equivalent in both horizontal and vertical directions (to make the circle look like a circle). That gives us an
unIramed work oI art!

9.4 plottools, lines, and other shapes 153
We can use lines and circles Ior less Irivulous purposes, too.
Table 9.4: Combining a line with other plots using display
(9.10)

(9.11)

We create a plot structure that is a green line segment
running between the points (12,25), and (12, 58).



154 9 Chapter 9 Visualization, modeling, and simulation

We combine the line with the data and curve p1 and
p2 we created in Combining Plots (page 149). This
highlights the value oI the curve at t12 minutes,
which would be good graphic design iI we were to
include this in a report on a situaiton where that time
was oI special interest.


9.5Mathematical models and simulation
Mathematical models try to describe a "real" situation in terms oI equations and Iormulae. The point oI modeling is to try, through mathematical
or computational means, to determine what will happen without having to run experiments in the "real" situation. As Dr. Jay Brockman oI the
University oI Notre Dame says in his book Introduction to Engineering.
Some engineering students have been Iortunate enough to participate in pre-engineering programs such as the Iirst LEGO(TM)
League robotics design competition or American Society oI Civil Engineering bridge-building contests. In addition to Iostering
creative problem-solving skills, such projects also introduce students to the important notion that seemingly good ideas don't
always work out in practice. OIten in such programs, students have ample opportunity to test and modiIy their designs beIore
they Iormally evaluate them. II the design doesn't work, then like a sculptor working with clay, the designer adds something here
or removes something there until the design is acceptable.

This cut-and-try methodology is also sometimes used in industry, particularly in circumstances where the design is simple,
or where, the risk or cost oI Iailure is low. In many situations, however, there is no second chance in the event oI Iailure. For
engineering systems such as buildings, bridges, or airplanes -- top name just a Iew -- Iailure to meet speciIications could mean
a loss oI liIe. For others -- such as the integrated circuit chip -- the cost oI Iabrication is so high that a company may not be
able to aIIord a second chance. In these situations, it's critical Ior the engineering team to be highly conIident that a design
will be acceptable before it's built. To do this, engineers use models to predict the behavior oI their designs. A model is an
approximation to a real system, such that when actions are perIormed on the model, it will respond in a manner similar to the
9.5 Mathematical models and simulation 155
real system. Models can have many diIIerent Iorms, ranging Irom physical prototypes such as a crash-test dummy to complex
computer simulations.

We can think oI a mathematical model as a kind oI virtual system... whose input is a set oI variables that represent either aspects
oI the design or aspects oI the environment, and whose output is a set oI variables that represent the behavior oI the system.
Inside is a set oI mathematical relationships that describe the operation oI the system.

Jay B. Brockman, Introduction to Engineering, John Wiley and Sons, 2009. pp 151-152.
The point oI expressing a situation mathematically is to use mathematics and computation to better understand the situation. Usually we are
given or derive a Iormulas that allow us to calculate key properties oI the system. For models involving only a Iew variables, this can involve the
Iollowing kinds oI actions:
1. Get a single number, by evaluating a Iormula or Iunction.
2. Get a single number, by solving an equation.
3. Gain an understanding oI a relationship between one or more entities oI interest and the"input variables" by producing a Iormula.
4. Gain a visual understanding oI the relationship by plotting a Iunction, or possibly several plots merged together.
5. Gain an understanding oI how a a system changes over time. Rather than computing the value oI a variable once, we repeatedly compute the
value oI the variable at several diIIerent points at time. This is called computational simulation oI the system. We can view plots changing over
time by producing an animation.
Examples of the first four types of computation
Third type: producing a formula
In the temperature-pressure data fitting example (page 118) the mathematical model is the Iormula that expresses the
relationship between temperature and pressure. We had only data and no Iormula to begin with, but we developed the Iormula
using the CurveFitting|LeastSquares| operation. This was a computation oI the third type mentioned above..
First type: Evaluating a formula or function
Once we had the Iomrula, we used the relationship to calculate pressure at several given temperatures. This was a computation
oI the Iirst type mentioned above.
Second type: solving an equation to get a desired value
We also Iound a temperature corresponding to a speciIied pressure by using solve -- a computation oI the second kind
mentioned above.
Fourth type: visualization (plotting)
In the example with an exponential curve fit (page 149), we got a visual impression oI how the Iormula Iit the data by
combining the point plot oI the data and the plot oI the Iormula together -- the Iourth kind oI computation mentioned above.
We haven't explained how to do the IiIth type oI computation -- animation -- yet. This will be discussed in upcoming section Animations (movies)
using animate (page 158)
156 9 Chapter 9 Visualization, modeling, and simulation
9.6Drawing x-y position as a function of time through
parameterized plots
The mathematical models oIten describe the position oI a system entity as a Iunction(s) oI time. II the entity's position is two dimensional, then
we have two Iunctions, oIten called x(t) and v(t). We can generate a plot oI position Ior various values oI t with a special Iorm oI plot.
plot( [ x-position expression, v-position expression, var low..high], plot options)
will draw a two dimensional graph connecing the (x,y) points traced out Ior the values oI the expression as the variable var takes on values
between low and high.
Table 9.5: Examples of plots where the x and y positions are parameterized by t
Examples of plots where the x and y positions are expressions parameterized by t
Example 1
In this example, we describe the x and v positions as periodic Iunctions oI time. Every time units, the positions return back
to where they were originally.

(9.12)
(9.13)
We plot the position using parameterized plotting. By deIault the scaling used Ior the horizontal and vertical axes are diIIerent,
but in this case the deIault makes the graph look misleadingly like a circle when it should look more like an ellipse. To
compensate, we use the scalingconstrained option to plot.

9.6 Drawing x-y position as a Iunction oI time through parameterized plots 157
Examples of plots where the x and y positions are expressions parameterized by t
Example 2

In this example, we have parameterized expressions Ior an object shot out oI a cannon with horizontal velocity 10 Ieet/second
and vertical velocity 10 Ieet/second minus the acceleration due to gravity.

(9.14)
(9.15)
158 9 Chapter 9 Visualization, modeling, and simulation
Examples of plots where the x and y positions are expressions parameterized by t
9.7A first animation example using animate, animation
controls
Let's look an animation. First we have to create it. We can use the animate Iunction oI the plots package. For the time being, let's not worry the
details oI why the operation is entered the way it is. Rather we Iocus on what the animation is trying to do, and how to view it in Maple once it
has been created.
Table 9.6: First animation example
This produces a movie oI a point moving through the points (0,0), (.1, .01), (.2, .04), etc. up to (10, 100).

9.7 A Iirst animation example using animate, animation controls 159

II we click on the plot, the Maple tool bar changes and shows us animation controls.
Table 9.7: Animation controls

These controls are highly similar to video playback controls Iound in many applications (e.g. You Tube), so we won't discuss at length here. Note
that using them you can:
1. Start playing the animation.
2. Stop the animation.
3. Display only a particular Irame, "Irozen".
160 9 Chapter 9 Visualization, modeling, and simulation
4. Control the number oI Irames per second it plays.
5. Set it to play once or continually repeat in a loop.
See Graphics-~Animation-~Animation Toolbar under the Table oI Contents oI the on-line Maple help.
Right-clicking on the animation will also produce a menu oI operations that provide an alternative Ior controlling the animation.
Table 9.8: Animation pop-up menu
Animation pop-up menu
II we play the animation, we will see a point move upwards in a parabolic path:
Table 9.9: Frames 1, 5, 10, 20, and 25 of the animation
Frames 1, 5, 10, 20, and 25 of the animation

9.8 The Iirst animate example, part 2 161
Frames 1, 5, 10, 20, and 25 of the animation

9.8The first animate example, part 2
Now that we've gotten the general idea oI what animates results are like, let's look again at what was entered and look at the details oI what was
computed. The general Iorm oI the operation is:
General form of the animate operation
with(plottools):
animate(plotting function, | parameters to plotting Iunction|, time variable time range)
What this does is to apply the plotting Iunction to the parameters Ior various values oI the time variable. Each value oI the time variable used is a
separate Irame oI the movie.
In the previous section we used:
In the Iirst Irame oI the movie, it used the value t0 Ior the time variable t. Thus it plotted
plot(|0|,|0|,stylepoint, color"Purple");
plots a small circle at the coordinate (0,0).
The next time value used is t0.4667. The second Irame oI the movie is the result oI
plot([.41667], [.416672], stylepoint, color"Purple");
which plots a small circle at the coordinate (.41667, ).
9.9Designing a function to use in animate
In this section, we explore the design oI a Iunction that we use with animate to create an animation.
162 9 Chapter 9 Visualization, modeling, and simulation
Problem
Create an animation oI a circle whose radius expands Irom r1 at time t0 to r5 at time t10.
Solution
The actions we create are a short script that deIines a Iunction drawCircle(t) . Then it uses it in animate. While we there are other ways to use
animate other than this style, we choose to do things this way because the drawCircle Iunction allows us to test the plotting one Irame at a time
beIore we try to create the animation.
Solution, part 1

First, we Iind Irom the on-line documentation Ior plottools that
the Iunction circle(|a,b|,r, color"red") draws a circle centered
at the point (a,b) with radius r:
While this seems to do the job that we wish Ior the starting Irame, we have to consider how this will vary as t0 to t10. We want the radius
oI the circle to be 1 when t0 (as in the Iigure above), and 5 when t10. We can do some line Iitting to come up with the Iormula iI we can't
remember enough high school algebra to Iigure it out ourselves:
Figuring out the formula for the radius as a function of time

(9.16)


We plot the Iormula we get Irom the data Iitting with the data
points to convince ourselves that the line agrees exactlv with
9.9 Designing a Iunction to use in animate 163

the two points. We'd get a similar result Ior curve Iitting a line
any time we Iit a line with only two data points.
(9.17)
Testing the solution function
(9.18)

We can create a Iunction that creates a circle oI the appropriate
size given a value Ior t;
164 9 Chapter 9 Visualization, modeling, and simulation

We try this Iunction out at a t0.

The Iunction evaluated at t5 looks the same, but we see Irom
the axis labels that it is actually a bigger circle
9.9 Designing a Iunction to use in animate 165

The Iunction evaluated at t10 draws a circle oI radius 5, as it
should.
Once we are convinced that the Iunction drawCircle works Ior the range oI values oI t that we need it to, we can use it in animate. The way we
invoke drawCircle within animate is diIIerent than when we were using it one Irame at a time. The name oI the Iunction is kept separate Irom the
parameters, and the range. This delays the creation oI any plot structures until animate starts computing them.
Complete script solving the "expanding circle animation" problem
(9.19)
We see the Iirst Irame oI the animation
below. When we click on the
animation, the animation toolbar
appears and we can then play it. We
see the circle expand over time. Unlike
the individual plots we created during
our testing, all the circles use the same
scaling.
166 9 Chapter 9 Visualization, modeling, and simulation

Some frames from the expanding circle animation

9.10 Designing a more elaborate animation 167
9.10Designing a more elaborate animation
Problem
A satellite is in a circular orbit around the earth, at a distance oI Iive earth radii. Create an animation oI it circling.
Solution -- discussion
We will create a Iunction drawPlanetAndSat (t) that Ior any time t draws both the Earth and the satellite at time t. We browse through the on-line
plottools package and discover the Iunction disk(c,r, color....) that draws a solid disk whose center is at the point c, has radius r, and speciIied
color, in a Iashion similar to the circle Iunction we used in the Designing a function to use in animate (page 161). Browsing through on-line
help Ior the color names known to plot reveals that one color is "DarkKhaki". We decide to draw the Earth as a disk oI radius 1 centered at (0,0).
Drawing a disk
168 9 Chapter 9 Visualization, modeling, and simulation

We decide to represent the satellite as a point. Recall that (cos(t), sin(t)) describes circular motion moving around a circle oI radius r1. To
parameterize orbital motion at radius 5, we use (5*cos(t), 5*sin(t)).
Drawing a point at (0,5)
9.10 Designing a more elaborate animation 169

We can combine the two together using displav.
Drawing a disk, and a point at (0,5)
We invoke displav with a list oI two plot structures: the disk,
and the point plot.
AIter we draw this we see the two plots together, but it isn't
exactly what we want, because the horizontal and vertical
scaling is not equal.
170 9 Chapter 9 Visualization, modeling, and simulation


We include the plot option scalingconstrained as an extra
argument to displav.
9.10 Designing a more elaborate animation 171
We're almost done. As with the animation design oI the previous section, we write a Iunction oI t that describes the disk and the position oI the
point at time t.
A function describing a frame to draw at time t
(9.20)



We test this Iunction out at a Iew values oI t. The idea is that
the satellite makes one orbit during the period t0.. .
172 9 Chapter 9 Visualization, modeling, and simulation




We now can use this Iunction with animate to draw the orbiting satellite.
Complete script to solve the "orbiting satellite animation problem"
9.10 Designing a more elaborate animation 173
(9.21)



We have the animation have the satellite circling three times.
Some oI the Irames oI the animation can be seen in the table of
frames from the orbiting satellite animation (page 173).
Table of selected frames from the orbiting satellite animation

174 9 Chapter 9 Visualization, modeling, and simulation
9.11Solving a problem with the help of animation
Problem
A boy throws a ball straight up in the air with an initial velocity oI 15 miles per hour. Once released, the ball's position is described by the
Iunction , where is the initial velocity, and g is the Iorce oI gravity, .
(a) How many seconds does the ball stay in the air?
(b) Generate an animation that shows the ball's motion. t should be measured in seconds, and position in Ieet Irom ground level. Assume that the
ball starts at 0 Ieet altitude even though this would be unrealistic Ior a boy to do unless he was standing in a pit!
(c) Use the animation to determine roughly when the maximum altitude is, and what that altitude is.
(d) Find how Iast the initial velocity should be so that the ball goes up over 30 Ieet.
Solution
First convert 15 miles per hour into a velocity in Ieet per second using the convert Iunction Iirst described in the section on problem solving
functions in Chapter 7 (page 121).
(9.22)
DeIine the gravitational constant
(9.23)
Next, deIine the Iunction Ior position.
9.11 Solving a problem with the help oI animation 175
(9.24)
Doing a rough plot oI position versus time shows us roughly when the ball will hit the ground. We guess that it might take three seconds.
Table 9.10: A rough-draft plot of ball versus time.

(9.25)
That's too much. The answer seems to be about 1.4 seconds, but we can use solve to come up with an exact value.
(9.26)
There are actually two solutions. The obvious one is when t0 and the ball hasn't yet been thrown.
The non-obvious one is the larger one. To get that one, we daisy-chain the max Iunction with solve.
(9.27)
This answers part (a) oI the problem.
We veriIy our computation by evaluating x at that time and Iinding that the position oI the ball really is at altitude 0.
176 9 Chapter 9 Visualization, modeling, and simulation
(9.28)
Now we need to make the movie. We need to create a Iunction which Ior time t, draws a point at the coordinate (0, x(t)). To illustrate what we
mean, let's compute the position oI the ball at t1 second.
(9.29)
We want the ball to be at position (0,6). A plot command that would do this would be:
Table 9.11: A plot that draws a ball as a circle at (0,6).
We've drawn the circle larger (symbolsize30) to make it stand out.
9.11 Solving a problem with the help oI animation 177
Similarly, at t.5, the ball's position would be at x(0.5):
Table 9.12: A plot that draws a ball as a time t0.5.

Evidently x(.5) is 7.
We create a user-deIined Iunction that creates a plot structure as a result.
(9.30)
Let's try this out Ior t.5 and see iI we get the same result.
178 9 Chapter 9 Visualization, modeling, and simulation
Table 9.13: Trying out a function that draws the ball at a particular time.
We can now use this Iunction with animate. Note that we needed to compute flightTime in order to make the movie run exactly as long as it takes
Ior the ball to come back down to the ground.
Table 9.14: Animating the ballFrame function to produce a movie of the ball in flight.
9.11 Solving a problem with the help oI animation 179

This answers part (b) oI the problem.
By playing the movie, we see that the maximum altitude is about 7.5 Ieet, at time t0.687 seconds.
This answers part (c) oI the problem.
To answer part (d), we need to do more programming. We Iirst modiIy the plot so that it draws a line segment at 30 Ieet as well as plotting the
position oI the ball. We use the line Iunction oI the plottools package discussed in Plottools section (page 151) to draw a line segment at (-5,30)
to (5,30), and to color it green:
(9.31)
(9.32)
The display Iunction can be used to combine this line with a Irame oI the movie. Here is an example oI this:
Table 9.15: A ball with a vertical target line.

180 9 Chapter 9 Visualization, modeling, and simulation
The automatic scaling oI plot chops oII the vertical distance between 6 Ieet and 0 because there is nothing in this Irame that needs that. In the
animation, the scale is adjusted so that all Irames operate in the same axes.
Now we can create a new user-deIined Iunction that plots both the line and the ball.
(9.33)
To look at the behavior oI the ball at a particular velocity, we can now execute a two line script, consisting oI assigning v0 to the desired initial
velocity, and then the operation that draws the movie.
Table 9.16: A first try at shooting the ball up 30 feet: 15 miles per hour
(9.34)


9.11 Solving a problem with the help oI animation 181

As we already have seen, is not Iast enough. We set it higher and recalculate the movie:
Table 9.17: Second try: shoot upwards at 22 feet per second
(9.35)


We can continue with trial-and-error. All we need to do is vary the velocity parameter. The reusability oI the ballWithLine Iunction is heavily
exploited.
182 9 Chapter 9 Visualization, modeling, and simulation
Third and Iourth tries: shoot upwards at 40 and 45 Ieet per second
(9.36)

(9.37)
(9.38)
According to the movies, 45 Ieet per second seems to be about right. We could get a more precise determination through movie-watching, but
Ior high accuracy we should use more mathematics. In a subsequent chapter, we will introduce additional Maple operations that can calculate
the velocity exactly (or a close approximation) without the trial-and-error oI movie watching. However, using the movies did give us a better
understanding oI the phenomenon, and allowed us to do trial-and-error Iairly easily.
9.12 Exporting animations and non-animated plots 183
9.12Exporting animations and non-animated plots
One operation available in the popup menu is Export. Right-click (or control-click) -~ Export -~ Graphics Interchange Format will produce an
animation Iile in .giI Iormat. As the animation Iile is being created, a dialog box will appear asking you to speciIy the directory where the .giI Iile
should be written. Once created, the Iile can be included on web pages or other documents.
This Ieature is also available Ior ordinary (non-animated) plots. Right-clicking (control-click Ior Macintosh) will create a Iile oI the plot
in .giI. .jpeg, or .ps Iormat. However, .giI Iile is the only Iormat oI the three that is supported by web browsers Ior animations.
Table 9.18: Exporting animations through the pop-up menu
9.13Summary of Chapter 9
Combining plots and shapes using display
Supplying the displav Iunction with a list oI
two or more plots will cause those plots to be
plotted on top oI one another.
We can save a plot in a variable to display
later.


Marking a speciIic value on the plot can
be accomplished using the line or circle
Iunction.
(9.39)
(9.40)

184 9 Chapter 9 Visualization, modeling, and simulation
Parameterized plots
For plots that may have values corresponding to 2d positions,
we use multiple Iunctions to deIine both and
(9.41)
(9.42)

We put these Iunctions in a list, along with the range oI
the independent variable , and plot.
Animating plots using animate
Creating an animation
9.13 Summary oI Chapter 9 185

Controling the animation using the animation tool bar or the
animation popup menu
The animation tool bar is located above the workspace window,
while the popup menu can be displayed by right-clicking the
animation.
Exporting an animation to a graphics Iile Right click animation~Export~Graphics Interchange Format
186 9 Chapter 9 Visualization, modeling, and simulation
187
10Moving into programming
10.1Chapter Overview
The interactive document interIace that we've worked with so Iar is good Ior quick development oI calculations involving a Iew steps. Because
Ieedback occurs aIter each step entered, this style oI working is oIten the Iastest way oI getting such short calculations done. We've also seen how
the eIIort to develop a script can be made to yield greater payback by Iinding and exploiting situations where reuse with minor editing and re-
execution.
With longer scripts or extensive re-execution, editing and execution (other than oI the entire worksheet) becomes tedious even iI we are saving
time by having the computer do the calculations. In this chapter, we introduce a new way oI entering and executing scripts: code edit regions.
While such regions make it more convenient to work and execute blocks oI Maple instructions, the programmer must enter things using the
textual version oI Maple expressions and operations. Individual lines must be separated by semi-colons (;) so that the Maple language processor
can more easily tell where one instruction ends and the next begins.
With code edit regions, the programmer must also work harder to see how the state oI the computation is changing during the computation, rather
than only looking at the Iinal output.. This is because iI the Iinal result is wrong, the programmer must Iind at what point in the script mistakes
(called program bugs) occur. Work with code edit regions oIten requires adding print statements into the script to better see changes in variables
and intermediate results.
As our Iirst extended experience with code edit regions, we look at the CarSimulator package used Ior the Iirst two labs oI CS 122. To use it, two
new programming elements are needed: procedures (blocks oI code that are delimited by proc()... end proc) and conditional loops (blocks oI code
that are delimited by while condition do ... end do). Procedures (multi-line Iunctions) and loops are used in most programming languages and in
many oI the programs that we will write Irom this point onwards.
10.2Code edit regions: entering a block of code,
executing it all at once
We have seen that even some oI the easiest technical calculations break down into a series oI operations, chained together. Realizing that we will
typically need the computer to perIorm a "series oI operations" is the motivation Ior making the transition to programming, where execution oI
many operations in a block is typical.
The Maple document interIace that we have been using so Iar easily supports calculations where the user prompts to computer to do a series
oI steps by positioning the cursor at the Iirst operation and then hitting return (or enter) repeatedly. We now introduce a second way to enter
instructions and have them executed as a block. This alternative is oIten easier to work with when you have a Iew dozen instructions and will
want to execute them all in a chain.
One can open a text Iield in the Maple document where one can enter a series oI instructions. To do this, position your cursor where you want the
text Iield to appear in the document, and with the mouse perIorm Insert-~Code Edit Region
Insert->Code Edit Region menu
188 10 Moving into programming
Once the Iield has been created, you can type the textual version oI the Maple instructions you want executed. Each instruction (commonly
reIerred to as a statement) must be separated by either a semi-colon, or a colon. II the statement ends in a semi-colon, then its value will be
printed during execution oI the region. II the statement ends in a colon, then printing oI its value will be suppressed just as it is in operation oI
documents.
Once all the instructions have been entered, you can run them all in a series by typing control-e (command-e on Macintosh), or by entering right-
click-~Execute Code (on Macintosh, control-click-~Execute Code). Instead oI typing control-e, you can as an alternative click on the "!" icon
on the Maple toolbar. (The "!!!" icon executes the entire worksheet, including all code edit regions.) The results Ior each statement will appear
below the region in blue, except Ior the statements whose printing is suppressed by a colon. The output in blue is reIerred to as the execution trace
produced when the code region's actions are carried out.
Execute Group icon ("!" icon) on Maple Toolbar

10.2 Code edit regions: entering a block oI code, executing it all at once 189
Any portion oI a line that begins with a "#" is a program comment. The rest oI the line is regarded by Maple as something that people will read,
not an instruction to be perIormed by the computer. Typically what appears aIter the # is commentary written in English (or whatever language
is convenient Ior communication with the intended audience) that helps explain/remind human readers what a segment oI code is about. As you
read programs with comments, you will see that some programmers are "chattier" than others. This is because they Ieel that their intended reading
audience needs more explanation.
In proIessionally written code, you will oIten see comments at the beginning oI the region that give an overview oI what the code region does,
the name oI the author(s), and the date oI creation/modiIication. ProIessional organizations oIten mandate a particular style and content Ior such
"header" comments, so that it will be easy to Iind inIormation about any code written by several programmers working on a single project.
An example from Chapter 7 as a code edit region
Code region Commentary

#We reset Maple so previous assignments are ditched.
restart;
#We assign two variables with pressure and #temperature
data.
pData := [134.2, 142.5, 155.0, 159.8, 171.1, 184.2];
tData := [0, 20.1, 39.8, 60.0, 79.9, 100.3];
#Load in the package
with(CurveFitting);
#Develop a linear formula for temperature
#as a function of pressure.
pressureFormula := LeastSquares(tData, pData, t);
#We can also fit a line to pressure as a function
# of temperature.
Segments oI lines that begin with # are
regarded as program comments (Ior the
program reader's eyes), not operations Ior
Maple to carry out.

Separate actions are separated by semi-colons.


The results in blue are displayed aIter we
position the cursor in the code edit region and
type control-e (command-e on Macintosh), or
enter Execute Code Region via the clickable
menu.


190 10 Moving into programming
tempFormula := LeastSquares(pData, tData, p);
(10.1)

The code region will develop a scroll bar iI the amount oI text entered exceeds the size oI the window. The size oI the Iield can be adjusted by
right click-~Component Properties, and then modiIying the integers listed Ior the width and height.
Enlarging a code Region window by changing component properties via the clickable menu
10.2 Code edit regions: entering a block oI code, executing it all at once 191
192 10 Moving into programming
The "Collapse Code Edit Region" menu item oI the clickable menu will reduce the entire window to an icon. II the Iirst line oI the region is a
program comment, it will be listed to the right oI the icon. Clicking on the code icon will execute the code within.
A collapsed code edit region
Code region Commentary

# Figure 11.1.1
#initialize variables
i := 1;
val := .3; #evaluation point
term := (val^i)/i!; # a term to compute
print("term is", term); #message
s := term; #s has a copy of the term
tol := 10e-7; #A small value.
We collapsed the region by positioning the cursor in the code
region window, and then entering right click-~ Collapse Code
Edit Region. Note that the comment on the Iirst line becomes
displayed next to the code edit icon.
10.3 Troubleshooting common errors in entry into code edit regions 193
(10.2)
10.3Troubleshooting common errors in entry into code
edit regions
New kinds oI errors and warnings can appear in code regions, due primarily to the new requirement that statements
be separated by semi-colons and colons. Trying to enter too many lines at once beIore testing what you have so Iar can make older errors appear
more mysterious.
Table 10.1: BROKEN - MISSING TITLE!
Code regions with entry errors
Example Commentary

a := sin(3.5 + cos( Pi/2) ;
Error, `;` unexpected

(Above region with Iixes)
a := sin(3.5) + cos( Pi/2) ;
(10.3)

The error message indicates that a semi-colon was not expected
where Maple scanned it. This is typically semi-colons can
appear only at the end oI a complete Maple statement. This
message typically indicates that something went wrong in what
was entered beIore the semi-colon. Edit the code region to
make the statement correct, and re-execute the region.

The problem here is a missing parentheses ). To Iix this, you'd
put a closing ) aIter the "3.5", and then re-execute the region
with a control-e(or command-e).

a := sin(3.5) + cos( Pi/2)
Warning, inserted missing semicolon at
end of statement
(10.4)

This is only a warning, and executing this region does perIorm
the computation correctly.

a := sin(3.5) + cos( Pi/2)
b := cos(3.5) + sin( Pi/2)
Error, missing operator or `;`
II we add another line without a semi-colon between, we get
just an error. Both lines are messed up because Maple needs an
explicit separator between the statements. Just putting the next
instruction on another line is not enough.
194 10 Moving into programming


(Above region edited with Iixes)
a := sin(3.5) + cos( Pi/2) ;
b := cos(3.5) + sin( Pi/2);
(10.5)


a := sin(3.5 + cos( Pi/2) ;
b := cos(3.5) + sin( Pi/2);
Error, `;` unexpected


II you try to enter several lines at once beIore entering, you
have the additional problem oI Iiguring out which line the
mistake occurred on. In this case, you have to decide which oI
the two semi-colons was unexpected.

Fortunately, you can sometimes get some help iI you are
observant -- the cursor will oIten Ilash on the line where the
problem is Iirst noticed (the Iirst line in this case).

a := sin(3.5 + cos( Pi/2)
Warning, premature end of input,
use <Shift> + <Enter> to avoid this
message.

There's both a missing closing parentheses and a missing semi-
colon here. The symptom is that there is a warning message
(which isn't necessarily a sign oI trouble, but is here), plus
the Iact that you are expecting a value to be printed Irom the
computation but just see a red "~" with no numeric output.

Fixing both problems and re-executing the region will produce
the correct result.

a := sin(3.5 + cos( Pi/2)
b := cos(3.5) + sin( Pi/2);
Error, missing operator or `;`
Sometimes when you enter a line, you make more than one
error. Even though the same mistake has been made as in
the previous example, there is a diIIerent error message than
beIore. Executing this region causes the message indicated,
plus the cursor will be observed to Ilash at the start oI the
second line.

Rather than being in an "I was expecting more" state which
produced the Iirst warning message, Maple is in a "I was
expecting more, but wait what I've seen isn't what I was
expecting". Since there's nothing beIore the cursor started
Ilashing beIore any oI the second line was parsed, it means that
the error must have happened beIore that on the previous line.
10.4 print and annotating an execution trace 195

There are still two things to Iix. Fixing them will produce two
lines oI output.
10.4print and annotating an execution trace
While executing a code region is convenient, a disadvantage is that the output oI a result is separated Irom the line oI code that produced it.
Furthermore, many lines oI output are not very useIul other than as conIirmation that an intermediate action didn't Iail.
As is true with interactive operation oI Maple, ending a line with a colon (:) instead oI a (;) will suppress the output. print(....) will cause a line
oI output which will list the value oI each item between the parentheses. The programmer can tailor the use oI each to produce more intelligible
inIormation.
An example from Chapter 7 as a code edit region
Code region Commentary

#We reset Maple so previous assignments are lost.
restart;
#We assign two variables with pressure and #temperature
data.
pData := [134.2, 142.5, 155.0, 159.8, 171.1, 184.2]:
print("Pressure data is:", pData);
tData := [0, 20.1, 39.8, 60.0, 79.9, 100.3]:
print("Temperature data is:", tData);
#Load in the package
with(CurveFitting):
#Develop a linear formula for temperature
#as a function of pressure.
pressureFormula := LeastSquares(tData, pData, t):
print("Fit of temperature to pressure is:",
p=pressureFormula);
#We can also fit a line to pressure as a function
# of temperature.
tempFormula := LeastSquares(pData, tData, p):
print("Fit of pressure to temperature is:",
t=pressureFormula);
(10.6)


This is the same computation as the previous
example, but the programmer has tailored
the output to make the execution trace more
intelligible. The execution trace output oI
most oI the assignments is suppressed with a
colon, as is the listing oI the operations Irom
the with. print statements have been added
to make the results more intelligible. Since
the result oI Least Squares does not name the
dependent variable, we tweak the output to
include that as well as the result oI the least
squares calculation.


196 10 Moving into programming
10.5The CarSimulator package: a first view
Let's see an example oI working with a package that requires us to write Maple programs in code regions. We are given the problem and a place
to work out its solution as a Maple worksheet with multiple code edit regions.
Executing the Iirst region does Simulator Set Up -- initializing some oI the variables and data structures needed Ior the simulator to work. It also
loads in the CarSimulator package (in the Iile CarSimulator.mla, which must be in the same directory as the worksheet we opened up).
Table 10.2: First code edit region for CarSimulator work
Problem: run the simulator in a 20 x 20 empty arena,
so that the car moves Iorward one square, turns leIt, and then moves Iorward two squares.


#Code region #1
#Reset all assignments in Maple so that they are forgotten
restart;
#Look for packages in the current directory (".") as well as
#built-in Maple library.
libname := ".", libname;
#Load CarSimulator package
with(CarSimulator);
(10.7)

AIter loading the package, Maple on-line help Ior it is also available. Clicking on the CarSimulator Iolder in the table oI contents Ior on-line
help indicates that there are several levels oI help. For example, under Setup we some oI the operations mentioned in the Iirst code region. Under
Car Operations we see move and turn. As with the rest oI on-line help, clicking on any oI the entries will bring up a help page with a Iurther
description. We realize that using a typical system's package, with dozens oI Ieatures, will require doing more reading oI on-line help to get the
details oI Ieatures we are likely to need.
10.5 The CarSimulator package: a Iirst view 197
Table 10.3: On-line help for CarSimulator after package is loaded into worksheet
Executing the second region sets up and shows the arena. The arena has a rectangular coordinate system. The
car typically starts at coordinate (1,1) Iacing rightwards. The orientation oI the car is given in radians, and the initial orientation is 0 radians. The
arena is shown by calling the package operation carMovie with argument stateTable (a Ieature oI the simulator that we could read about iI we
looked Iurther in the Data Iolder oI the package).
Table 10.4: Second code edit region with Car Simulator

198 10 Moving into programming
#Code region #2
#We don't put a restart here because we want to
#remember what we did in the previous code region.
#To start afresh, rerun all of the regions.
#Initialize car simulator
initialize();
establishBoundaries(20,20): #draw a 20 x 20 sandbox for the car to roam
around in. Coordinates range from [0,0] to [20,20].
drawBackground(): #Create a plot of the wall and the target for later
use, but don't display it yet
carMovie(stateTable); #Show initial configuration

10.5 The CarSimulator package: a Iirst view 199
We want to have the car move Iorward one square and turn leIt,and then move Iorward two more squares, like this:
Table 10.5: First control program: move, turn left, move twice.
At start AIter a one square move
(partial image zoom in)
AIter turn in place AIter another move AIter last move
To do this, we write a control program named run1a. The program consists oI a script using operations Irom the CarSimulator package. The
arguments we give to turn we derive Irom the inIormation Iound in the on-line help page: a leIt turn requires an argument oI a/2 radians. We see
some new syntax in the control program: the control script is bracketed by proc() and the phrase end proc. This deIines the script as a procedure.
A procedure is a way oI encapsulating a script so that it has a clearly deIined beginning and end, along with named parameters. A procedure is
expected to name parameters and optionally have a result, like the Iunctions deIined by the arrow notation described in chapter 7. run1a needs
no parameters, which is why it has ( ) rather than a parenthesized parameter sequence ( x) or (tData, pData). For the next chapter all oI our
procedures will have no parameters, but we will eventually visit situations where we need to write procedures that need one or more parameters.
Table 10.6: Second code edit region with Car Simulator

#Code Region #3
run1a := proc()
move(1); #move forward one square
turn(Pi/2); #turn Pi/2 radians (left 90 degrees)
move(1); #move forward one square
move(1); #move again.
end proc:
run(run1a);
(10.8)

AIter deIining the control procedure Ior the car, we submit the procedure (by its name, run1a) to the run operation oI the CarSimulator package.
The execution trace indicates that the simulation has taken 4 moves. However, there is no indication oI what happened in the trace.
To see the simulation at work, we again perIorm the operation carMovie(stateTable). Because the simulation had multiple moves, this now causes
an animation window to appear which when played displays the movement oI the car Irame-by-Irame. The animation is created by the Maple
programming in the package -- we will learn how to make movies in that Iashion in a Iew chapters.
Table 10.7: Fourth code edit region with Car Simulator animation result

#Code Region #4
#Play an animation of the results of the simulation.
#After the "run", the stateTable has all the information #needed to
make the animation.
200 10 Moving into programming
carMovie(stateTable);

The worksheet with the Iour code regions thus sets up the arena Ior the problem, describes the control procedure that is the solution to the stated
problem, and shows the results oI running the simulator with the procedure, which demonstrates that it solves the problem.
10.6A second task with the CarSimulator
We set up a more elaborate arena (set up code not shown):
#Establish an arena
restart;
libname := ".", libname;
with(CarSimulator):
#Set up the problem. Note that by parameterizing the problem we can use random
#numbers to set up sizes and locations to vary the problem a little.
size := 30: #width and length of the sandbox.
k := 15;
L := 12;
h := 10:
y := 10:
distanceBetweenWalls := 10:
10.6 A second task with the CarSimulator 201
wallHeight := 5: #parameters for location and sizes
halfDistance := floor(distanceBetweenWalls/2):
halfHeight := floor(wallHeight/2):
initialize();
pickAStart := 3:

#Set the starting location by resetting the state table and CURRENTLOCATION
stateTable[0] := [ k+pickAStart,y,0,CARNORMAL]:
CURRENTLOCATION := [ k+pickAStart,y,0,CARNORMAL]:
#draw walls
setBackground([seq('[k-halfDistance, y+i],[k+halfDistance, y+i]',i=-halfHeight..halfHeight)],
BARRIER):
establishBoundaries(size,size): #draw a sandbox for the car to roam around in. Coordinates
drawBackground():
print("Starting configuration is:");
carMovie(stateTable);
202 10 Moving into programming


The goal is to move Iorward until the barrier is reached, then turn around and move Iorward until the opposite barrier is reached.
The control program, illustrated in the Iollowing code region, uses two new programming elements, while and isTouching. A description oI
the isTouching Iunction can be Iound in the on-line help Ior the CarSimulator package. isTouching is a Iunction that returns true or false --
truth or Boolean values, rather than a number, Iormula, or plot structure as we have seen with other Iunctions. while is part oI the programming
language notation Maple uses to describe a loop. Programming loops typically contain a block oI actions that are to be perIormed a number oI
times, as well as a continuation condition that allows the computer to decide whether to continue to perIorm the block. The decision to repeat
the loop is done typically only at the start oI the would-be repetition. Once the decision is made to do it, the whole block is perIormed beIore the
continuation condition is considered again.
barrierBounce := proc()
while not(isTouching(0, 'bt')) do
move(1);
In the loop in the control program, the block consists oI the
single action "move(1)". Thus the loop tells the simulated car
10.6 A second task with the CarSimulator 203
end do;

turn(Pi); #turn around

while not(isTouching(Pi, 'bt')) do
move (1);
end do;
end:
run(barrierBounce);
print("Simulation result is:");
carMovie(stateTable);
to check iI the Iront oI the car is touching a barrier. II not, it
moves Iorward one square. Repeating the loop means to check
the continuation condition again and then perIorm the block's
actions. Thus this loop perIorms check-move-check-move-
check-move.... stopping when the next check determined that
the Iront oI the car is touching a barrier.

Once the loop is over, the program will perIorm the next action
listed, which is the turn. AIter that, there is another loop which
does the check-move-check.... repetition until the other barrier
occurs.



At the start (partial image
zoom in)
Taking several more moves
to the wall, and and turning.
(zooming in some more)
Taking several moves towards
the other wall
At the Iinish -- touching the
other wall.
204 10 Moving into programming
10.7Programming to solve a class of problems
At this point, those oI you not be Iamiliar with while may be wondering why we go through the trouble oI writing a loop with a conditional
expression when we can just build something that has exactly the right number oI moves and turns:

#Something that solves one kind of arena
lessRobustSoln := proc()
move(1);
turn(Pi);
move(1);
move(1):
move(1):
move(1):
move(1):
move(1);
move(1);
move(1);
end proc:
However, consider the Iollowing similar arena problem (details oI set up code hidden in code edit region):
#We're not resetting everything, but we will initialize with a new arena
#Set up the problem. Note that by parameterizing the problem we can use random
#numbers to set up sizes and locations to vary the problem a little.
size := 40: #width and length of the sandbox.
k := 15;
L := 12;
h := 10:
y := 10:
distanceBetweenWalls := 20:
wallHeight := 5: #parameters for location and sizes
halfDistance := floor(distanceBetweenWalls/2):
halfHeight := floor(wallHeight/2):
initialize();
pickAStart := 7:

#Set the starting location by resetting the state table and CURRENTLOCATION
stateTable[0] := [ k+pickAStart,y,0,CARNORMAL]:
CURRENTLOCATION := [ k+pickAStart,y,0,CARNORMAL]:
#draw walls
setBackground([seq('[k-halfDistance, y+i],[k+halfDistance, y+i]',i=-halfHeight..halfHeight)],
BARRIER):
establishBoundaries(size,size): #draw a sandbox for the car to roam around in. Coordinates
drawBackground():
print("Starting configuration is:");
carMovie(stateTable);
10.7 Programming to solve a class oI problems 205


Our "less robust" solution won't work here because the number oI moves is diIIerent. II we tried to write a similar program that would solve
this new arena, it would get a bit tedious repeating the "move(1)" instruction the necessary number oI times. However, the original solution
barrierBounce will work Iine:
run(barrierBounce);
print("Simulation result is:");
carMovie(stateTable);
206 10 Moving into programming
Building a program that Iits exactly one situation means that you will have to reprogram it iI the situation changes. II you're going to go through
the eIIort oI building something, it will pay to design Ior a reasonable amount oI Ilexibility. For that reason, appropriate testing oI a program will
put it through its paces in a number oI diIIerent situations, to check that the desired breadth is really there.
10.8Reviewing old and introducing new programming
concepts
The Iirst two labs oI CS 122 uses a package called the CarSimulator. It allows you to write simple programs that control a simulated car moving
around an obstacle course. The simulator is useIul Ior rapidly prototyping ideas Ior controlling an actual robot car.
Using the CarSimulator on a problem requires the use oI elements oI Maple programming that we have already seen:
Element First introduced in Chapter
Assignment 4
Lists 5
Scripts 5
Library packages 7
On-line help 6
10.9 Summary oI Chapter 10 207
Textual entry oI actions 6
Functions 7,8
Animations 9
Code edit regions 10
We will also get a Iirst look at some new elements:
Element Discussed further in
Procedures 11
While loops 12
Methodical testing and
troubleshooting code blocks
13
Tables 14
For loops 15
If -then-else statements 16
10.9Summary of Chapter 10
Example Commentary
Use of print

(10.13)

(10.14)
We set up a list oI values. L has all Iloating-point
values because the "10.0" in the denominator
causes all computations to be done using limited-
precision arithmetic. The display oI the value oI L is
suppressed because we ended the assignment with a
colon.


print prints out the sequence oI values given to it
as arguments. You can give any number oI items
in the sequence. The commas separating the string
and the number are Irom the sequence. Note that
as Iirst mentioned in the section in Chapter 7 on
including external packages (page 117), the result is
not displayed because the line ends with a colon.
Code edit regions
Insert -~ Code Edit Region to create a region.
control-e (command-e in Macintosh) or
right-click -~ Execute Code in a code edit region to execute all the code in it.
208 10 Moving into programming
right-click-~Component Properties to open a dialog box with the dimensions oI the code region (in pixels). Change these
dimensions by entering diIIerent numbers to enlarge or shrink the box. Sorry, dragging will not resize the code region.
right-click-~Collapse Region to turn the code region into an icon.

Clicking on the icon will then execute the code inside.

The Iirst line oI code appears as the label Ior the icon.

right-click-~Expand Region to turn a code region icon to display the region again.
Semi-colons and colons
All Maple statements must be separated or terminated with
either a semi-colon (;) or a colon (:). Usually each statement
executed generates a line oI output which appears in blue aIter
the region. This sequence oI output is called the execution trace
oI the code. II a colon is used, then the output that normally
results Irom evaluating the statement is suppressed Irom the
execution trace.

Comments
Any portion oI a line oI code starting Irom a number sign (#) is
treated as commentary and not as an operation to be perIormed.

Example Commentary

# Figure 11.1.1
#initialize variables
i := 1;
val := .3; #evaluation point
term := (val^i)/i!; # a term to compute
print("term is", term); #message
s := term; #s has a copy of the term
tol := 10e-7; #A small value.
(10.15)
Segments oI lines that begin with # are regarded as program
comments (Ior the program reader's eyes), not operations Ior
Maple to carry out.



The results in blue are displayed aIter we position the cursor
in the code edit region and type control-e (command-e on
Macintosh), or enter Execute Code Region via the clickable
menu.




Result oI Iirst assignment (to i)
10.9 Summary oI Chapter 10 209
Result oI assignment to val.
Result oI assignment to term.
Result oI calling the print Iunction.
Result oI assignment to s.
Result oI assignment to tol.
CarSimulator package loading

#Code region #1
#Reset all assignments in Maple so that they are forgotten
restart;
#Look for packages in the current directory (".") as well as
#built-in Maple library.
libname := ".", libname;
#Load CarSimulator package
with(CarSimulator);
(10.16)

To successIully use the CarSimulator,
the CarSimulator.mla and
CarSimulator.hdb library Iiles need
to be located in a directory speciIied
by the variable libname. Assigning
libname to include ".", and then
opening and working on a worksheet
located in the same directory as the
library Iiles will work.


You can tell that when the loading
works iI the with (ending with a
semi-colon rather than a colon)
produces a list oI all the items
included in the simulator package.

Only do a restart iI you want all
previous assignments to be erased.
CarSimulator initialization

#Code region #2
#We don't put a restart here because we want to
#remember what we did in the previous code region.
#To start afresh, rerun all of the regions.
#Initialize car simulator
Some actions are necessary to
initialize the simulator beIore
it is used. Typically they are
CarSimulator set up actions such as
initiali:e and establishBoundaries.
210 10 Moving into programming
initialize();
establishBoundaries(20,20): #draw a 20 x 20 sandbox for the car to
roam around in. Coordinates range from [0,0] to [20,20].
drawBackground(): #Create a plot of the wall and the target for
later use, but don't display it yet
carMovie(stateTable); #Show initial configuration

Control procedure for car control
barrierBounce := proc()
while not(isTouching(0, 'bt')) do
move(1);
end do;

turn(Pi); #turn around

while not(isTouching(Pi, 'bt')) do
move (1);
end do;
end:
A procedure begins with proc( ) and
ends with proc end. The procedure
contains a number oI actions such
as move and turn, although it can
contain any other Maple operation
such as assignments, print, or
calculations.

A control loop within a procedure is
oI the Iorm
while condition do
action,
action,
....
end do;
10.9 Summary oI Chapter 10 211

II the condition is true, then the
actions are perIormed. Then the
condition is evaluated again. II
it is still true, the actions will be
repeated. This continues until the
condition is Iound to be Ialse when
it is evaluated. The conditionis
evaluated only once per repetition oI
the actions, at the start.



CarSimulator simulation and visualization of results

run(barrierBounce);
print("Simulation result is:");
carMovie(stateTable);
Use the run operation oI the
CarSimulator, giving it the name
oI a control procedure that has
been previously created. AIter the
simulation is over, see the results
by applying the carMovie operation
oI the package to the stateTable
generated by the simulation.
212 10 Moving into programming
213
11More on proc and while
11.1Chapter Overview
We continue our discussion oI programming with two new elements, procedures and loops under the control oI while.
Procedures are an alternative way (compared to the -~ notation) oI creating a Iunction. As with all Iunctions, they take arguments and returns
results. Unlike arrow notation Iunctions, their deIinition can span more than one line oI code, including control constructs such as while.
Procedures can have local variables, whose use is encapsulated. This allows a procedure programmer to use a variable Ior internal use during a
procedure execution and not have to worry about interIering with the use oI that variable outside oI the procedure.
11.2Review of function terminology
Although they don't use the same syntax, procedures are a kind oI Iunction. The technical terminology oI procedures shares a lot with the
terminology we have already used Ior Iunctions. Let's review what we know Iunction terminology in Maple.
Review of function terminology
1. Functions process input values and produce an output value.
The Iunction definition describes how the values oI those
parameters are used to compute a Iunction result. When we
deIined Iunctions using the -~ (arrow) notation, the result was
described as an expression involving arithmetic, structure
creators such as | |, and built-in (e.g. sin, sqrt, or seq) Iunctions
or Iunctions previously deIined by the user.
The names given to the symbols used Ior the input values are
called the formal parameters to the Iunction. Functions can
have zero or more parameters.

The function result is typically the expression computed by the
Iunction.

There are two Iormal parameters to the Iunction deIined above.
a, and b are their name.


2. Functions are associated by a name by making the Iunction
deIinition the right hand side oI an assignment statement.
The Iunction is named f through the
assignment.
3. DeIining a Iunction does not perIorm the computation, it
just stores the directions Ior how to do the computation within
the computer. AIter a Iunction is deIined, you can get the
computer to perIorm the computation by invoking the function.
Invocation involves giving the name oI the Iunction, and values
Ior the Iunction's arguments, enclosed in parentheses. The
values supplied to the Iunction when it is invoked are called the
actual parameters or arguments to the Iunction.


# The actual parameters (arguments)
# are 3.1 and
4. We can combine the actions oI several Iunctions together by
composing (also known as daisv-chaining ) them. Rather than
the "one aIter another" sequencing oI actions that we might
get Irom using Iunctions on several lines oI a script, Iunction
composition uses "inside out" sequencing.
This is the same as the sequence

214 11 More on proc and while

Programming languages Ieature Iunctions (and procedures) prominently because oI their advantages in the construction oI programs. Writing
a computation as a Iunction encapsulates it: there is a well-deIined starting point given by the values supplied Ior the parameters, there is a
speciIic computation described within the Iunction deIinition, and there is a Iinal result to the computation when the Iunction delivers its output.
Only when the Iunction is later invoked are those actions perIormed. The Iunction may be invoked multiple times (as would occur when the
users invokes it when doing several problems, or when map is used with a Iunction) aIter it is deIined. The expectation oI reuse is one oI the
justiIications Ior the extra work needed to deIine a Iunction, since re-invocation is typically easier than even copying and pasting the actions.
11.3Creating a short function with proc... end
One line Iunctions created by -~ have the limitation that we can't create Iunctions that have code which includes assignment, loops, or other
actions that require more than one line oI code. In the previous chapter, we saw the use oI proc ()... end proc
Maple procedure definition -- the simplest form
function name : proc (formal parameters )
statement,
statement,
... (one or more)

end proc;
The procedural result is by deIault the last thing computed in the sequence oI actions computed by the Iunction when it is invoked.
Example: The simple function defined with -> and then
again with proc


This is a Iunction named f deIined using the arrow notation.
f2 has the same behavior as f. Note that there are no local
variables here -- all the symbols that appear in the Iunction
come Irom the parameters.

Note that you can end a procedure with either "end" or "end
proc".
11.4 Creating a Iunction with encapsulated (local) variables 215
11.4Creating a function with encapsulated (local)
variables
General form of a Maple procedure definition with local variables
function name : proc ( formal parameters )
local variables used within procedure;
statements,
end proc;
Notes
1. There is no semi-colon between the proc ( ) containing the Iormal parameters and the word local. II there are no local
variables Ior the procedure, then you need to put in the semi-colon: proc ( formal parameters ) ; statements, ....
2. The opposite oI a "local variable" is a "global variable". We don't use them much in the Maple procedures we will be
working with. You can read more about them iI you look up "global" in Maple on-line help.
3. You can't assign a value to a Iormal parameter in the statements oI the procedure. parameters are used only as a way oI
providing input inIormation to the Iunction.
Usually local variables arise when the Iunction needs to compute and store intermediate results that are used Ior the Iinal answer but are not part
oI them.
An important Ieature oI local variables is that use oI their use inside the Iunction is encapsulated . When a variable is declared to be a local
variable oI a Iunction, the value that the symbol outside oI the Iunction invocation may have is temporarily set aside. The local symbol within the
Iunction starts oII with no value, but may be assigned values within the programming oI the Iunction. Once the Iunction invocation is over, the
old value that the symbol had is restored to it.
Table 11.1: Demoonstrate of assignment of local variables
Demonstration that local variables do not pay attention to assignments made outside the scope of the procedure
definition.
#This function demonstrates the use of local variables.
testFunction := proc()
local i;
print("Line 1: i's value is",i);
i := 1;
print("Line 2: i's value is",i);
end proc;
We deIine a testFunction, whose
purpose is to assign a value to i, and
to print out it's value beIore and aIter
assignment.

#This is a "non-local" use of i;
i := 0;
print("Line 0: i's value is",i);
testFunction();
print("Line 3: i's value is",i);

The execution trace we get Irom
executing this code region aIter the
previous one is explained by:

a) the Iirst zero is Irom the assignment
to i.
216 11 More on proc and while
b) The "Line 0" output is the assigned
value oI i.
c) Once testFunction begins execution,
the declaration oI i as a local variable
means that previous assigned values
are temporarily set aside. This is why
the Line 1 output is just i.
d) The Line 2 output reIlects the
assignment within testFunction.
e) The Line 3 output comes aIter
the invocation oI testFunction
is completed. By that time, the
previously assigned value oI i is
restored.
The behavior oI local variables is seems more complicated, but the reality is that it Irees the programmer Irom having to worry about the whether
their use oI a variable in a Iunction will break other code outside oI the Iunction iI they don't restore the previously assigned value. One could
avoid local variables iI one never used Iunctions or procedures, but this isn't practical Ior serious projects -- the complexity oI coding will deIeat
the attempt to complete the project successIully without the ability to break the task into the smaller pieces that coding with procedures allows.
11.5Conditional repetition through while... do... end do
The key understanding oI loops written with while is: "keep on repeating this while a condition is true".
The general form of the while statement
while condition do
loop bodv;
end do;
The condition is something that is either true or false. An error results iI it isn't. The condition is evaluated only once per
repetition, at the start beIore any oI the code in the loop bodv is executed. II the condition is true, then the entire loop bodv is
executed. II not, the repetition is Iinished and execution moves onto the next instruction aIter the end do.

The condition is evaluated at the start oI the loop iteration. It does not matter iI the condition changes in the middle oI the loop
bodv; this won't terminate execution oI the loop until the execution reaches the end oI the loop bodv and thecondition looked at
again beIore the start oI the next repetition oI the loop.

The continuation condition is can be a Iunction invocation that returns true or false, such as the isTouching or onTopOfSign
Iunctions oI the CarSimulator package. It can also be an inequality such as x.1 or abs(y-z)~ tol. It always must be an
expression that can be evaluated to be true or false. II the inequality is numeric such as x .1, variables in the condition such as
x must have been assigned a numeric value earlier in the Maple session.

11.6 Example: car control procedures with conditional expression built Irom Boolean operations and multiple loops 217
It is also possible to combine several conditions together using the operations and, or, and not(....).

true and false are reIerred to as Boolean values (aIter the British 19th century, George Boole -- http://www.gap-system.org/
~history/Biographies/Boole.html Ior more inIormation). Functions that return Boolean values are reIerred to as Boolean
functions, while and, or and not are reIerred to as Boolean operations.
11.6Example: car control procedures with conditional
expression built from Boolean operations and multiple
loops
The problem here is to go to the Iirst Iour way intersection, and turn leIt. AIter the turn, go Iorwards to the target.
Here is a sample scenario. We want to write a control program that will work this scneioar as well as with similar ones. The similar scenario will
have a Iour way intersection and a target to the
leIt, but it may have a number oI other (three way) intersections beIore that. The distance to the Iour way intersection may vary between
scenarios.

#Generate test problem (code not shown)
#Turn left at the first 4 way intersection, go to target.
#Reset all assignments in Maple so that they are forgotten
restart;
#Look for packages in the current directory (".") as well as
#built-in Maple library.
libname := ".", libname;
#Load CarSimulator package
with(CarSimulator):
StringTools[Randomize]();
randomize():
#Pick a starting location for the car
xStart := 5:
yStart := 15:
#Pick typical distance to intersection
reach := 20:
#Pick distance to target intersection
distance := rand(reach-4..reach+4)():
#Distance to target
yTarget := rand(5..10)():
#Initialize car simulator
initialize():
#Size of arena
218 11 More on proc and while
arenaSize := 40:
#pick two intersection locations
L1, R1:= op(combinat[randperm]([seq('2*i',i=1..floor(distance/2)-1)])[1..2]):
establishBoundaries(arenaSize,arenaSize):
#draw a sandbox for the car to roam around in. Coordinates range from [0,0] to [arenaSize,
arenaSize].
#Set the starting location by resetting the state table and CURRENTLOCATION
stateTable[0] := [ xStart,yStart,0,CARNORMAL]:
CURRENTLOCATION := [xStart,yStart,0,CARNORMAL]:
minLR := min(L1,R1):
maxLR := max(L1,R1):
#draw wall to a first intersection
setBackground([seq('[xStart+i, yStart-1],[xStart+i, yStart+1]',i=0..minLR-1)],
BARRIER):
#draw wall to a second intersection
setBackground([seq('[xStart+i, yStart-1],[xStart+i, yStart+1]',i=minLR+1..maxLR-1)],
BARRIER):

#draw wall to a third intersection
setBackground([seq('[xStart+i, yStart-1],[xStart+i, yStart+1]',i=maxLR+1..distance-1)],
BARRIER):

#Draw wall to the end
setBackground([seq('[xStart+i, yStart-1],[xStart+i, yStart+1]',i=distance+1..min(arenaSize-
xStart-2,2*distance))],
BARRIER):

#Draw wall at intersection
setBackground([seq('[xStart+distance-1, yStart+i],[xStart+distance+1, yStart
+i]',i=1..min(arenaSize-yStart-2, distance))],
BARRIER):

setBackground([seq('[xStart+distance-1, yStart-i],[xStart+distance+1, yStart-
i]',i=1..min(yStart-2, distance))],
BARRIER):

#set target location
setBackground([seq('[xStart+distance, yStart+yTarget]',i=0..distance)],
TARGET):

#Draw left wall at intersection
setBackground([seq('[xStart+L1-1, yStart+i],[xStart+L1+1, yStart+i]',i=1..min(arenaSize-
yStart-2, distance))],
BARRIER):

setBackground([[xStart+L1, yStart-1]],
BARRIER):
11.6 Example: car control procedures with conditional expression built Irom Boolean operations and multiple loops 219

#Draw right wall at intersection
setBackground([seq('[xStart+R1-1, yStart-i],[xStart+R1+1, yStart-i]',i=1..min(yStart-2,
distance))],
BARRIER):

setBackground([[xStart+R1, yStart+1]],
BARRIER):
setBackground([seq('[xStart+distance-1, yStart-i],[xStart+distance+1, yStart-
i]',i=1..min(yStart-2, distance))],
BARRIER):
drawBackground(): #Create a plot of the wall and the target for later use, but don't display it
yet
carMovie(stateTable); #Show initial configuration

220 11 More on proc and while
The car will automatically stop when it is pointing at and directly next to a blue target, so the control program should just run the car Iorward
while it hasn't seen the 4 way intersection, then turn and run Iorward until it is touching the target. We can express "not at a Iour way intersection"
by using Boolean operations to construct the condition that the car is touching something in one oI the Iour directions -- (istouching(-Pi,'bit')) or
(isTouching(Pi/2,'bt')) or (isTouching(0,'bit')) or (isTouching(-Pi/2,'bt'));
goToFourWay := proc()
while isTouching(-Pi,'bt') or(isTouching(Pi/2,'bt')) or (isTouching(-Pi/2,'bt')) or
(isTouching(0)) do
move(1);
end do;
turn(Pi/2);
while not(isTouching(Pi/2,'bt')) do
move(1);
end do;
end proc:
run(goToFourWay);
carMovie(stateTable);
11.7Example: car control procedures with a local variable
Our scenario here is to go around Iour corners in a counterclockwise direction, then stop.
11.7 Example: car control procedures with a local variable 221

#Generate test problem
#Turn left at the first 4 way intersection, go to target.
#Reset all assignments in Maple so that they are forgotten
restart;
#Look for packages in the current directory (".") as well as
#built-in Maple library.
libname := ".", libname;
#Load CarSimulator package
with(CarSimulator):
StringTools[Randomize]();
randomize():
#Pick a starting location for the car
xStart := 5:
yStart := 15:
#Pick lengths of rectangle
bottomTop := rand(4..10)();
leftRight := rand(4..10)();
#Initialize car simulator
initialize():
#Size of arena
arenaSize := 40:
establishBoundaries(arenaSize,arenaSize):
#draw a sandbox for the car to roam around in. Coordinates range from [0,0] to [arenaSize,
arenaSize].
#Set the starting location by resetting the state table and CURRENTLOCATION
stateTable[0] := [ xStart,yStart,0,CARNORMAL]:
CURRENTLOCATION := [xStart,yStart,0,CARNORMAL]:
minLR := min(L1,R1):
maxLR := max(L1,R1):
#draw bottom and top walls
setBackground([seq('[xStart+i, yStart+1],[xStart+i, yStart+leftRight]',i=0..bottomTop)],
BARRIER):
#draw left and right walls
setBackground([seq('[xStart, yStart+i],[xStart+bottomTop, yStart+i]',i=1..leftRight)],
BARRIER):
drawBackground(): #Create a plot of the wall and the target for later use, but don't display it
yet
222 11 More on proc and while
carMovie(stateTable); #Show initial configuration

We present two solutions to this. The Iirst solution repeats the code that says "move Iorward while touching the wall to the leIt, then turn leIt"
Iour times. Since the isTouching operation uses absolute rather than relative directions, we must change the angle with each turn Irom to
to to 2 . We do this by introducing a local variable "angle" that we increment by Pi/2 each time. The reason why we do the incrementing
angle : angle Pi/2; rather than just angle : Pi/2; angle : Pi; angle : 3*Pi/2, etc. will become clear in the sequel.
turnFourTimes := proc()
local bt,angle;
#Move forward while touching a wall, then
#turn left
#Initial scan angle is Pi/2
angle := Pi/2;
11.7 Example: car control procedures with a local variable 223
while isTouching(angle,'bt') do
move(1);
end do;
turn(Pi/2);
move(1);
angle := angle+Pi/2; #angle is now Pi
#Do that again.
while isTouching(angle,'bt') do
move(1);
end do;
turn(Pi/2);
move(1);
angle := angle + Pi/2; #angle now 3*Pi/2
#Do that again.
while isTouching(angle,'bt') do
move(1);
end do;
turn(Pi/2);
move(1);
angle := angle+Pi/2; #2*Pi is same as 0.
#Do that one more time
while isTouching(angle,'bt') do
move(1);
end do;
turn(Pi/2);
move(1);
end proc:
run(turnFourTimes);
224 11 More on proc and while
carMovie(stateTable);
11.8Example: the four corner problem with a shorter
program using a nested loop
In the previous example, our solution involved describing the actions needed to traverse one side oI our Iour sided problem, then repeating that
code Iour times. We write a shorter program by putting all oI this inside a loop, whose purpose is to perIorm everything inside it Iour times.
#Generate test problem
#Turn left at the first 4 way intersection, go to target.
#Reset all assignments in Maple so that they are forgotten
restart;
#Look for packages in the current directory (".") as well as
#built-in Maple library.
libname := ".", libname:
#Load CarSimulator package
with(CarSimulator):
StringTools[Randomize]():
randomize():
11.8 Example: the Iour corner problem with a shorter program using a nested loop 225
#Pick a starting location for the car
xStart := 5:
yStart := 15:
#Pick lengths of rectangle
bottomTop := rand(4..10)():
leftRight := rand(4..10)():
#Initialize car simulator
initialize():
#Size of arena
arenaSize := 40:
establishBoundaries(arenaSize,arenaSize):
#draw a sandbox for the car to roam around in. Coordinates range from [0,0] to [arenaSize,
arenaSize].
#Set the starting location by resetting the state table and CURRENTLOCATION
stateTable[0] := [ xStart,yStart,0,CARNORMAL]:
CURRENTLOCATION := [xStart,yStart,0,CARNORMAL]:
minLR := min(L1,R1):
maxLR := max(L1,R1):
#draw bottom and top walls
setBackground([seq('[xStart+i, yStart+1],[xStart+i, yStart+leftRight]',i=0..bottomTop)],
BARRIER):
#draw left and right walls
setBackground([seq('[xStart, yStart+i],[xStart+bottomTop, yStart+i]',i=1..leftRight)],
BARRIER):
drawBackground(): #Create a plot of the wall and the target for later use, but don't display it
yet
carMovie(stateTable); #Show initial configuration
226 11 More on proc and while
turnFourTimesA := proc()
local bt,angle,count;
#Move forward while touching a wall, then
#turn left
#Initial scan angle is Pi/2
angle := Pi/2;
#Initial count is zero.
count := 0;
while(count <4) do
while isTouching(angle,'bt') do
move(1);
end do;
turn(Pi/2);
move(1);
angle := angle+Pi/2;
count := count +1;
end do;
end proc:
run(turnFourTimesA);
carMovie(stateTable);
11.9 Example: a procedure that does a calculation instead oI controlling a car. 227
The programming Ior this solution is dramatically more succinct. II you can handle in stride the greater conceptual complexity oI putting one loop
inside another loop, you will probably get your solution working Iaster, since there are Iewer characters to type, thereIore decreased chances oI
making an entry error.
11.9Example: a procedure that does a calculation instead
of controlling a car.
Much oI technical programming has to do with the calculation oI mathematical Iormulae, visualizations, or numbers rather than control oI
devices. Procedures can be used Ior such tasks. II you have a calculational script, in many cases it is easy to turn it into a procedure:
1. Rather than initializing script parameters at the start, remove those assignments and put the parameters into the sequence oI parameters oI the
procedure.
2. The last thing calculated should be the result returned. In the example below, we return a list oI numbers, which allows us to communicate two
results that the procedure has computed. One oI the advantages oI using a data structure such as a list is that it can be used as a "data container"
Ior several results.
#Given two sides of a right triangle, compute
#the hypotenuse and the angle formed by it.
triangleInfo := proc (side1, side2)
local hypotenuse, angle, dAngle;
#Use whatever values are provided when the
#procedure is invoked
228 11 More on proc and while
angle := arctan(side1,side2);
dAngle := convert(angle,units,radians,degrees);
hypotenuse := sqrt(side1^2+side2^2);
#Return a list of the hypotenuse and the angle
#(in degrees).
[hypotenuse, dAngle];
end proc:
Once we've deIined the Iunction, we can use it Ior several triangles:
(11.6)
(11.7)
II we want one value or the other, we can use part extraction as explained in chapter 5.
(11.8)
(11.9)
We note that while the procedure will calculate either with exact or Iloating point numbers, the Iloating point results are only approximations and
may diIIer slightly Irom the correct answer.
(11.10)
(11.11)
11.10Troubleshooting small function definitions
Function deIinitions with proc... end proc will exhibit typical "imbalance" error messages such as errors iI you Iorget to balance the typical things
that occur in computer syntax: ( and ), iI and end iI, | and |, proc and end proc. II you do it in a code edit region, then in addition to printing out an
error message, the cursor will pause over the Iirst place where the computer saw the imbalance.
Troubleshooting problems with small function definitions

findPower := proc (x,tol)
local i;
i := 0;
This Iunction loops to Iind last power oI x that is greater than
tol.
11.10 Troubleshooting small Iunction deIinitions 229
term := x^i;
while (term>tol) do
term := x^i;
Warning, premature end of input,
use <Shift> + <Enter> to avoid this
message.
We note that there's no "end proc" to end the procedure
deIinition. We edit the code region and execute it again.
findPower := proc (x,tol)
local i;
i := 0;
term := x^i;
while (term>tol) do
term := x^i;
i := i+1;
end proc;
Error, reserved word `proc` unexpected
AIter making that correction, we get a message that indicates
that the proc wasn't expected. The cursor is positioned in the
midst oI the "end proc" so we know that it's the last proc where
things ran into problems.

We know that procedures must end with end proc, so removing
it doesn't seem to be an option. But another cause oI end proc
being unexpected is that something else was expected instead.

Looking harder, we realize, that we Iorgot the "end iI" to end
the iI statement, too.
findPower := proc (x,tol)
local i;
i := 0;
term := x^i;
while (term>tol) do
term := x^i;
i := i+1;
end do;
end proc;
Warning, `term` is implicitly declared
local to procedure `findPower`
(11.12)
(11.13)
We get a warning message that term is not a local variable.
Since we want it to be (it doesn't have a value outside oI the
Iunction), we change the declaration.
230 11 More on proc and while
findPower := proc (x,tol)
local i,term;
i := 0;
term := x^i;
while (term>tol) do
term := x^i;
i := i+1;
end do;
end proc;
(11.14)

(11.15)
This doesn't have any errors or warnings, but is it right? We
need to test the Iunction.
(11.16)
(11.17)

The Iunction returns 23. But isn't greater than .1, so that
can't be the last power greater than .1.
findPower := proc (x,tol)
local i,term;
i := 0;
term := x^i;
while (term>tol) do
term := x^i;
i := i+1;
end do;
i -1;
end proc;
When we analyze what is going on, the value oI i computed in
the loop is one too Iar. So we think we should really make the
last value oI the Iunction one less.
11.10 Troubleshooting small Iunction deIinitions 231
(11.18)

(11.19)
(11.20)
(11.21)

Evidently our analysis is still oII. Looking harder, we see that
we should change the order oI operations inside the loop, or we
will go one too Iar.
findPower := proc (x,tol)
local i,term;
i := 0;
term := x^i;
while (term>tol) do
i := i+1;
term := x^i;
end do;
i -1;
end proc;
(11.22)

(11.23)
This works. But it took some analysis even beyond the point
where we stopped getting error messages and warnings.
232 11 More on proc and while
(11.24)

Troubleshooting problems with use of variables in procedures
g := proc(a,n);
local i, count;
count := 0;
while count<n do
a := sqrt(a); #take multiple square roots
count := count +1;
end do;
return a;
end;
Error, unexpected `local` declaration in procedure
body
This is a Iunction that's supposed to take n
square roots oI a number: .
There are several problems with it. The Iirst
problem is that in order to declare a local
variable, Maple insists that there be no extra
semi-colon between the parameters (a,n) and
the word "local". It should be proc(a,n) local
i; ....
g := proc(a,n)
local i, count;
count := 0;
while count<n do
a := sqrt(a); #take multiple square roots
count := count +1;
end do;
return a;
end;
(11.25)

With this corrected, we get Maple to accept
the deIinition, but it still doesn't work. The
problem is that Iormal parameters can't be
assigned to in Maple procedures. Formal
parameters do not work the same as local
variables.

The way to get the desired result is to set up a
local variable and to initialize it with the value
oI a beIore the loop starts.
g := proc(a,n)
local i, result,count;
result := a;
for i from 1 to n do
result := sqrt(result); #take multiple square roots
count := count +1;
end do;
return result;
end;
The Iunction works both Ior exact numbers
and Iloating point numbers.
11.11 Troubleshooting local variable declarations 233
(11.26)
(11.27)
(11.28)

11.11Troubleshooting local variable declarations
Warning messages from undeclared local variables

eApprox2 := proc(x, tol)
local term,i;
s := 1;
term := 1;
i := 1;
while abs(term)>=tol do
term := term*x/i;
s := s+term;
i := i+1;
end do;
return s;
end proc:
Warning, `s` is implicitly declared local to
procedure `eApprox2`
II you Iorget to declare a variable within a Iunction
deIinition as a local variable, then Maple will do it Ior
you but print out a warning. While this deIault action is
almost always what is wanted, the warning is get you to
think about whether you really wanted s to be a global
variable (the same values used within the procedure and
outside oI it) rather than local.

x := 0;
while (x<=1.0) do
print("Computed:",eApprox2(x, 10^(-5)),"Built-
in", exp(x));
x := x + .1;
end do:
printf("s's value is: %a", s);
We see that s's value within this script is not aIIected to
what happened to s within eApprox2. s was automatically
declared local to the eApprox2 procedure when it was
deIined, even iI we didn't include it in the code.

By putting a colon at the end oI the "end do", we
suppress the execution trace oI what's happening inside
the loop. The print actions do cause output though.
234 11 More on proc and while
s's value is: s


eApprox2 := proc(x, tol)
local term,i,s;
s := 1;
term := 1;
i := 1;
while abs(term)>=tol do
term := term*x/i;
s := s+term;
i := i+1;
end do;
return s;
end proc:
(11.30)
(11.31)
We can redeIine eApprox2. This doesn't change the
meaning but it does get rid oI the warning messages.


Note that the computed result is based on x (the script
version, not the local variable used inside eApprox2)'s
last value Irom the previous computation in the
worksheet. Since a the last computed value oI x Irom
the while loop is 1.1, this is the value used in the
computation here.
11.12 Troubleshooting while loops: loops that don't stop 235

11.12Troubleshooting while loops: loops that don't stop
We've already reviewed some oI the basic problems with loops that come Irom Iorgetting to put in a end do or do, or putting too many in. These
typically give some kind oI Error, message and were discussed here.
More pernicious are loops which do not give error messages but either stop with the wrong answer, or do not stop. Such programming is called an
infinite loop and they are usually something to avoid because the program will never come to a conclusion and let you regain interactive control
oI the computer.
You can tell that this is happening because you are either Ilooded with a lot oI output, or the status window in the lower leIt hand corner oI the
worksheet says Evaluating.... but never seems to return to Readv.
A runaway "infinite" loop

x := 3.0;
while (x<10) do
print("sin(x)", sin(x));
end do;
3.0
"sin(x)", 0.1411200081

"sin(x)", 0.1411200081

"sin(x)", 0.1411200081

"sin(x)", 0.1411200081

etc., etc.
The problem with this loop is that it will never stop. The continuation condition is x10, which is true initially. Because the loop body never
changes the value oI x, it will continue to be true Ior every successive repetition. Presumably the programmer made a mistake and Iorgot to put in
lines oI code into the loop body that changed the value oI x and (with enough repetition) made it eventually grow to 10 or larger.
In theory, you can stop long-running computations such as inIinite loops with the "red stop hand" Iound on the Maple tool bar. Clicking on the
red stop hand will eventually stop the output, but not until aIter a long, long wait while all the output that resulted beIore you told the computer to
stop looping, is printed. It typically takes a long time Ior it to print all the output Irom the repetitions, and you won't be able shut oII the printing
with the red hand.
So in practice, iI you have a runaway loop you oIten will have to Iorce-quit your Maple session and recover the most recent version oI your
worksheet through File -~ Recent Backups-~ Recover Backup, as described in chapter 2. II you absolutely must execute a runaway loop again,
Iirst change it to print very little so that it will stop beIore you get old waiting Ior it.
Interrupting an evaluation with the "red stop hand"
236 11 More on proc and while
AIter entering an equation
to solve, we have realized
that we really don't want to
see millions oI roots listed,
assuming that Maple can
Iind them all. So we click
on the red hand (circled in
green).
AIter a signiIicant pause,
the message "computation
interrupted" appears and
we can enter another
computation Ior Maple
to perIorm. There is no
result Irom the interrupted
computation, so no result
is displayed. The hand
becomes gray instead oI red
when execution is over.

This technique does not
work well iI the long
running computation prints
a lot, as might occur in a
runaway loop with a verbose
execution trace.
The CarSimulator package won't have the problem oI having the car take inIinite numbers oI moves by accident, because it has a built-in limit
to the number oI steps that the car can take. However, you may discover that you have a "runaway car" who was only prevented Irom an inIinite
loop by the built-in limits. It's still possible to write a car control program that doesn't end iI you include a loop that doesn't move the car (such as
one that only prints, or assigns a variable a value).
11.13Summary of Chapter 11
General form of a Maple procedure definition with local variables
function name : proc ( formal parameters )
local variables used within procedure;
statements,
11.13 Summary oI Chapter 11 237
end proc;
Notes
1. There is no semi-colon between the proc ( ) containing the Iormal parameters and the word local. II there are no local
variables Ior the procedure, then you need to put in the semi-colon: proc ( formal parameters ) ; statements, ....
2. The opposite oI a "local variable" is a "global variable". We don't use them much in the Maple procedures we will be
working with. You can read more about them iI you look up "global" in Maple on-line help.
3. You can't assign a value to a Iormal parameter in the statements oI the procedure. parameters are used only as a way oI
providing input inIormation to the Iunction.
The general form of the while statement
while condition do
loop bodv;
end do;
The condition is something that is either true or false. An error results iI it isn't. The condition is evaluated only once per
repetition, at the start beIore any oI the code in the loop bodv is executed. II the condition is true, then the entire loop bodv is
executed. II not, the repetition is Iinished and execution moves onto the next instruction aIter the end do.

The condition is evaluated at the start oI the loop iteration. It does not matter iI the condition changes in the middle oI the loop
bodv; this won't terminate execution oI the loop until the execution reaches the end oI the loop bodv and thecondition looked at
again beIore the start oI the next repetition oI the loop.

The continuation condition is can be a Iunction invocation that returns true or false, such as the isTouching or onTopOfSign
Iunctions oI the CarSimulator package. It can also be an inequality such as x.1 or abs(y-z)~ tol. It always must be an
expression that can be evaluated to be true or false. II the inequality is numeric such as x .1, variables in the condition such as
x must have been assigned a numeric value earlier in the Maple session.

It is also possible to combine several conditions together using the operations and, or, and not(....).

true and false are reIerred to as Boolean values (aIter the British 19th century, George Boole -- http://www.gap-system.org/
~history/Biographies/Boole.html Ior more inIormation). Functions that return Boolean values are reIerred to as Boolean
functions, while and, or and not are reIerred to as Boolean operations.
Stopping a never-ending (infinite) loop
Stop the "red hand" icon on the Maple toolbar. II that doesn't work, then shut down the Maple session by killing the process (in
Windows) or doing a Force Quit (in Macintosh). UnIortunately the latter will cause you to lose all the work you've done since
last saving the worksheet, so it should be done only as a last resort.
238 11 More on proc and while
239
12More control: for and if
12.1Chapter Overview
We introduce two more kinds oI control actions. for is a way oI doing looping. It doesn't do anything that we couldn't do with while, but it
provides convenience -- less typing to get the same thing to happen. if allows us to tell the computer to decide between alternative courses oI
action.
12.2for and indexed repetition for... from ... to ... do...
end do
In Chapter 11, we saw a car control problem where the car was supposed to turn Iour times. The way the solution was handled was by setting up
a loop programmed with a while that used a variable called count. The variable was initialized to zero, and incremented by one each time through
the loop. The condition oI the while was set to look at the value oI count and to stop the repetition when the count reached 4.
This pattern -- using a counter variable to do things a set number oI times -- occurs so oIten in programming situations that most languages have
notation to make setting up such loops convenient.
General Iorm:
Ior stepperJariable Irom initialJalue to finalJalue do
loop bodv,
end do;
BeIore the loop starts, the stepperJariable is initialized to initialJalue. Then the loop runs as iI it were:
while(stepperVariable IinalValue) do
loop bodv,
stepperJariable . stepperJariable 1,
end do;
II this is used within a proc....end, then the stepperJariable should be declared as a local variable.
Here is shorter solution oI the "Iour turn" problem using for. The stepperJariable is i.

#Generate test problem
#Turn left at the first 4 way intersection, go to target.
#Reset all assignments in Maple so that they are forgotten
restart;
#Look for packages in the current directory (".") as well as
#built-in Maple library.
libname := ".", libname;
#Load CarSimulator package
with(CarSimulator):
StringTools[Randomize]();
randomize():
#Pick a starting location for the car
xStart := 5:
yStart := 15:
240 12 More control: Ior and iI
#Pick lengths of rectangle
bottomTop := rand(4..10)();
leftRight := rand(4..10)();
#Initialize car simulator
initialize():
#Size of arena
arenaSize := 40:
establishBoundaries(arenaSize,arenaSize):
#draw a sandbox for the car to roam around in. Coordinates range from [0,0] to [arenaSize,
arenaSize].
#Set the starting location by resetting the state table and CURRENTLOCATION
stateTable[0] := [ xStart,yStart,0,CARNORMAL]:
CURRENTLOCATION := [xStart,yStart,0,CARNORMAL]:
minLR := min(L1,R1):
maxLR := max(L1,R1):
#draw bottom and top walls
setBackground([seq('[xStart+i, yStart+1],[xStart+i, yStart+leftRight]',i=0..bottomTop)],
BARRIER):
#draw left and right walls
setBackground([seq('[xStart, yStart+i],[xStart+bottomTop, yStart+i]',i=1..leftRight)],
BARRIER):
drawBackground(): #Create a plot of the wall and the target for later use, but don't display it
yet
carMovie(stateTable); #Show initial configuration
12.2 Ior and indexed repetition Ior... Irom ... to ... do... end do 241

turnFourTimesWithFor := proc()
local bt,angle,i;
#Move forward while touching a wall, then
#turn left
#Initial scan angle is Pi/2
angle := Pi/2;
for i from 1 to 4 do
while isTouching(angle,'bt') do
move(1);
end do;
turn(Pi/2);
move(1);
angle := angle+Pi/2; #angle is now Pi
end do;
242 12 More control: Ior and iI
end proc:
run(turnFourTimesWithFor);
carMovie(stateTable);
With a little analysis, we note that rather than incrementing angle each time, we can come up with the right angle by multiplying the current value
oI the stepperJariable by , producing an even more succinct solution.

#Generate test problem
#Turn left at the first 4 way intersection, go to target.
#Reset all assignments in Maple so that they are forgotten
restart;
#Look for packages in the current directory (".") as well as
#built-in Maple library.
libname := ".", libname;
#Load CarSimulator package
with(CarSimulator):
12.2 Ior and indexed repetition Ior... Irom ... to ... do... end do 243
StringTools[Randomize]();
randomize():
#Pick a starting location for the car
xStart := 5:
yStart := 15:
#Pick lengths of rectangle
bottomTop := rand(4..10)():
leftRight := rand(4..10)():
#Initialize car simulator
initialize():
#Size of arena
arenaSize := 40:
establishBoundaries(arenaSize,arenaSize):
#draw a sandbox for the car to roam around in. Coordinates range from [0,0] to [arenaSize,
arenaSize].
#Set the starting location by resetting the state table and CURRENTLOCATION
stateTable[0] := [ xStart,yStart,0,CARNORMAL]:
CURRENTLOCATION := [xStart,yStart,0,CARNORMAL]:
minLR := min(L1,R1):
maxLR := max(L1,R1):
#draw bottom and top walls
setBackground([seq('[xStart+i, yStart+1],[xStart+i, yStart+leftRight]',i=0..bottomTop)],
BARRIER):
#draw left and right walls
setBackground([seq('[xStart, yStart+i],[xStart+bottomTop, yStart+i]',i=1..leftRight)],
BARRIER):
drawBackground(): #Create a plot of the wall and the target for later use, but don't display it
yet
carMovie(stateTable); #Show initial configuration
244 12 More control: Ior and iI

turnFourTimesWithAngleFor := proc()
local bt,i;
#Do the following for each of four angles:
for i from 1 to 4 do
#Move forward while touching a wall, then
while isTouching(Pi/2*i,'bt') do
move(1);
end do;
#turn left and move forward one square
turn(Pi/2);
move(1);
end do;
end proc:
run(turnFourTimesWithAngleFor);
carMovie(stateTable);
12.3 variations on Ior: to ... do ... end do, Ior... Irom ... by ... to ... do ... end do 245
12.3variations on for: to ... do ... end do, for... from ...
by ... to ... do ... end do
II you just care about repetition without an index variable, you can use:
to finalJalue do
loop bodv,
end do;
This is not as Ilexible as the indexed repetition given in the previous section, but it will cause the loop body to be repeated finalJalue oI times.
Programming by Bart Simpson
to 5 do
print("I will not burp in class.");
246 12 More control: Ior and iI
end do;
(12.5)
You may have noticed in this example that the loop is not encapsulated in a procedure. Just like all other actions, you can include loops in
scripts built without procedures. The advantage is that you don't have to go through the extra work oI including the procedure "header" or the
declaration oI local variables. The downside oI procedure-less scripting is that it's harder to identiIy the scope oI the script, it is not given an easy-
to-remember name, multiple invocation oI it requires more tedious editing oI the script, and lack oI encapsulation means that it is more diIIicult to
make sure that you can move and successIully execute the script in another worksheet.
II you want a stepper variable but you want to use steps other than 1, you can use:
Ior stepperJariable Irom initialJalue to finalJalue by stepJalue do
loop bodv,
end do;
For loop with a decrementing rather than incrementing steps
print("x and sin(x)");
for x from 1 by -.3 to -1 do
print( x, evalf(sqrt(x)));
end do;
(12.6)

Since the step size is -.3, then the finalJalue should be less than initialJalue oI 1. With a negative step value, the condition
because while(x~-1) do... rather than while(x-1) do... which would cause the loop to stop immediately.
12.4 Choosing alternatives: iI...then...end iI, iI... then... else...end iI 247
12.4 Choosing alternatives: if...then...end if, if... then...
else...end if
The if statement allows a programmer to selectively execute code only iI a condition is true.
if condition then
code to executed iI the condition is true
end if
If-then-else-end if : one possible outcome

grade := 3.7: #This value will cause execution of the "then" code
if (grade >= 3.5)
then
print("An excellent grade:", grade);
end if;
(12.7)
II an "else" is added , then the computer will chose which oI two segments oI code it will execute, depending on whether the condition is true or
Ialse.
if condition then
code to executed iI the condition is true
else
code to be executed iI the condition is Ialse
end if
If-then-else-end if : another possible outcome

grade := 3.2: #This value will cause execution of the "else" code
if (grade >= 3.5)
then
print("An excellent grade", grade);
else
print("A grade:", grade);
end if;
(12.8)

Obviously, iI you are planning to do something once, you can probably just look at what's being computed and decide yourselI what to do. The
value oI if is when you plan to use a script many times, or iI you need to make many conditional decisions within a loop in a single execution oI
a script. If is helpIul because you won't have to stop to think about what happens each time. You can just set things up ahead oI time so that the
right actions occur in either case.
248 12 More control: Ior and iI
If-then-else-end if in a loop
#Initialize a list with some numbers, some grade point
# averages
L := [3.6, 3.9, 2.1, 3.9, 2.1, 0.7, 1.3, 2.8, 3.6, 3.7, 2.9];
#Initialize some accumulation variables. We will use them
#to count how many grades are 3.5 or above, and
#how many are below.
countHigh := 0;
countLow := 0;
#This goes first, before a loop, as a table header
printf("Grades at least 3.5:\n");
#i steps through the possible index values for L.
#Recall that nops(L) is the length of the list.
for i from 1 to nops(L) do
if (L[i] >= 3.5)
then
print( L[i]);
countHigh := countHigh+1; #add one to the "high" counter
else
countLow := countLow+1; #add one to the "low" counter
end if;
end do:
print("There were",countHigh,"grades at or above, and ", countLow, "grades below 3.5.");
Grades at least 3.5:
(12.9)
A Iairly common occurrence is when the else alternative would speciIy to do nothing at all. Instead oI having an empty else branch, you should
write the code without the else as if...then...end if.
Example of If-then-end if

12.5 Coding situations with three or more cases using iI-then-eliI-eliI...else-end iI 249
In University oI West Dakota at Hoople, the Dean's list is determined by computing the average oI all student's grade point
averages, and selecting those whose gpa is at least .5 point higher than the average.

restart; #unassign any previous values
L := [3.6, 3.9, 2.1, 3.9, 2.1, 0.7, 1.3, 2.8, 3.6, 3.7, 3.8, 2.0]:
deanCount := 0:
#Compute the average for all students by
#using sum to add up all the grades, and dividing by the
#number of grades
avg := sum(L[i],i=1..nops(L))/nops(L):
#Print out table header.
printf("Grade average of %d students is: %f.\n", nops(L), avg);
printf("Grades at least one point above average (Dean's List):\n");
for i from 1 to nops(L) do
if (L[i] >= (avg+1) )
then
printf("%.2f\n", L[i]);
deanCount := deanCount+1;

end if;
end do:
printf("\nThere were %d students making the Dean's List.",deanCount);
Grade average of 12 students is: 2.791667.
Grades at least one point above average (Dean's List):
3.90
3.90
3.80
There were 3 students making the Dean's List.
12.5 Coding situations with three or more cases using if-
then-elif-elif...else-end if
The if statements we have seen so Iar can be described as being useIul when:
1. The condition was used to select between one series, and doing nothing as the alternative. (if.. then...end if).
2. The condition was used to select between two alternative series oI actions (if...then..else..end if).
Another situation that arises in practice is to select between three or more alternatives. This can be thought oI as "case by case" situation.
if-then-elif-else-end if
General form Notes
if (condition 1) then
statements separated bv , or .
II condition 1 is true when the statement is executed, then the
Iirst series oI statements is executed.

250 12 More control: Ior and iI
elif (condition 2) then
statements separated bv , or .

(more alternatives specified bv more elifs)


else
statements separated bv , or .

end if
Otherwise, iI condition 1 is not true but condition 2 is true,
then the second series as executed.

Additional conditions will be checked iI none oI the previous
ones turn out to be true. The sequence oI actions corresponding
to the Iirst true condition will be executed.

II none oI the conditions are true, then last series oI statements
are done.

The else section can be omitted. This is equivalent to doing
nothing iI none oI the other conditions apply.
The else portion oI the statement speciIies what is to be done iI none oI the previous cases apply.
Example of If-then-elif-elif-...-else-end if
#Initialize L to be a list of grades
L := [3.6, 3.9, 0.9, 2.1, 3.9, 2.1, 0.7, 1.3, 2.8, 3.6, 3.7, 2.9, 1.3];
#Initialize counters foer the number of As, Bs, etc. (silently)
countA := 0:
countB := 0:
countC := 0:
countD := 0:
countF := 0:
#Process each element of the list, with the following
#cases:
#As (3.3 and above)
#Bs (at or above 2.7 and below an A)
#Cs (at or above 1.7 and
#Ds (at or above 1.0 and below a C)
#Fs (below 1.0).
for i from 1 to nops(L) do
x := L[i]; #Use "x" instead of typing L[i] several times
if (x >= 3.3)
then
countA := countA +1;
elif (x >=2.7) #this is tried only if first condition false
then
countB := countB +1;
elif (x >=1.7)
then
countC := countC +1;
elif (x>=1.0)
then
12.6 A CarSimulator scenario solved with iI 251
countD := countD +1;
else #we will be here only if everything else hasn't worked
# so we know x<1.0.
countF := countF +1;
end if;
end do:
print(countA,"As", countB, "Bs", countC, "Cs", countD, "Ds", countF, "Fs");
(12.10)
It is possible to include iI statements as one oI the actions inside another iI statement. This is useIul where a case may have subcases. We will
discussion this situation in more detail in later chapters.
12.6A CarSimulator scenario solved with if
The car passes over a numbered oI colored signs. When it detects a red sign, it should turn leIt. When it detects a yellow sign, it should turn right.
Eventually it will encounter a purple sign and stop on it.
Here is a sample scenario:

#Generate test problem
#Turn left at red signs, #Turn right at yellow signs
#Stop at target
restart;
#Look for packages in the current directory (".") as well as
#built-in Maple library.
libname := ".", libname;
#Load CarSimulator package
with(CarSimulator):
StringTools[Randomize]();
randomize():
#Starting location of the car is the default
#Initialize car simulator
initialize():
#Size of arena
arenaSize := 40:
establishBoundaries(arenaSize,arenaSize):
#draw a sandbox for the car to roam around in. Coordinates range from [0,0] to [arenaSize,
arenaSize].
numTurns := rand(4..14)():
lastLocX := 0:
252 12 More control: Ior and iI
lastLocY := 0:
for i from 1 to numTurns do
distance := rand(2..5)();
if i mod 2 = 1
then #move horiz
nextLocX := lastLocX +distance;
nextLocY := lastLocY;
setBackground([nextLocX, nextLocY], REDSIGN);
else #move vertical
nextLocX := lastLocX;
nextLocY := lastLocY + distance;
setBackground([nextLocX, nextLocY], YELLOWSIGN);
end if;
lastLocX := nextLocX;
lastLocY := nextLocY;
end do:
distance := rand(2..5)();
if numTurns+1 mod 2 = 1
then #move horiz
nextLocX := lastLocX +distance;
nextLocY := lastLocY;
setBackground([nextLocX, nextLocY], PURPLESIGN);
else #move vertical
nextLocX := lastLocX;
nextLocY := lastLocY + distance;
setBackground([nextLocX, nextLocY], PURPLESIGN);
end if;
drawBackground(): #Create a plot of the wall and the target for later use, but don't display it
yet
carMovie(stateTable); #Show initial configuration
12.6 A CarSimulator scenario solved with iI 253

We use the onTopOfSign operation oI the CarSimulator (described in the on-line documentation, which you should look at iI you want to
understand what the diIIerence is between PURPLESIGN and PURPLESIGNHIT, Ior example).
turnRedYellow := proc();
#No local variables needed in this procedure
#Move while you're not on a purple sign
while not(onTopOfSign(PURPLESIGNHIT)) do
#turn right if you're on a red sign
if onTopOfSign(REDSIGNHIT)
then turn(Pi/2);
254 12 More control: Ior and iI
#turn left if you're on a yellow sign
elif onTopOfSign(YELLOWSIGNHIT)
then turn(-Pi/2);
end if;
move(1); #move forward regardless of whether
#you've turned.
end do;
end proc:

run(turnRedYellow);
carMovie(stateTable);
12.7 For the curious: comparing control statements in
conventional programming languages
Almost all conventional programming languages (C, Mathematica, C#, Java, Matlab, Maple, C, NXT, Python, Perl, Visual Basic, etc.) have the
"loop" idea and the "selection/case" ideas in them. The main diIIerence in how they speciIy them through the use oI diIIerent words or symbols.
Java Ior example, uses and } rather than "then" or "end iI" to delimit the begin and end oI segments oI actions.
Portion of a 1ava program with a conditional
int countThree, countSubThree;
12.7 For the curious: comparing control statements in conventional programming languages 255
int [] L= {1,5,9,2,6,7,-2,1,7};
// Set up counters
countThree=0;
countSubThree=0;
// Loop through L counting how many are above three
for (int i=0; i<L.length; i++)
{
if (L[i]>3)
\011{ countThree++;}
else
\011{countSubThree++;}
};
System.out.println("three and above:"+countThree+
\011 " below three: "+ countSubThree);

While this portion oI the Java program is quite intelligible to someone who is more experienced in Maple, a Iull program to
do this would require additional statements that we don't should here. Note that the printing statement println uses "" not Ior
numerical addition but to combine the output into a single string that println can handle. Getting used to these language-speciIic
idiosyncrasies are part oI the overhead oI working with more than one tool on a computer.
The Matlab program Iollow with a loop and a conditional may look similar to a Maple program, although it uses diIIerent Iunctions to extract size
("size" rather than "nops") and there is diIIerent syntax (no do, just "end", "" instead oI from and ":" instead oI to.
Portion of a Matlab M-file with a loop and a conditional
%Program to count how many elements of a vector/list are greater than and less %than three
countThree=0;
countSubThree=0;
L=[1,5,9,2,6,7,-2,1,7];
[r,c]=size(L);
%Loop for i from 1 to c, where c is the number of items in L (length of L)
for i=1:c
if L(i)>3
countThree=countThree+1;
else
countSubThree=countSubThree+1;
end;
256 12 More control: Ior and iI
end;
%Use formatted printing to print out things.
sprintf('Three and Above: %d, Below three: %d', [countThree, countSubThree])


There may be other ways oI doing the same task in a language that may be easier in that particular language. For example, summing together
numbers in Java involve writing a loop that gathers the sum oI the numbers, whereas in Maple one could use sum and get the job done in one line.
C, Java and C have a "?" operator which one could use to write the if oI the above example more succinctly even though it would work the
same as the kind we have worked about. No language will ever be "the most convenient" Ior all tasks, which is one oI the reasons why so many
languages endure.
Regardless oI what you think about the way other languages handle loops and conditional execution, learning how to do it in Maple transIers to
most other languages you may choose to work with in the Iuture.
12.8 Troubleshooting if statements
While, for, and if statements require you have to have words such as "while" "do", "iI", "else", etc. written in exactly the right sequence or
else the language processor will reject them. Sixty years oI advances in programming language technology haven't made the situation Ior
programmers much easier. It's still necessary to learn how to do it correctly, and how to troubleshoot your way out oI problems when you make a
mistake.
You can use the Iact that the computer will let you experiment as much as you want to help you learn. Now that you have learned how to enter
code into worksheets and code edit regions, you can start oII by learning new syntax by exactly reproducing an example that contains output
as well as input, and seeing whether you can get the same results. Next, you can try modiIying the examples to do slightly diIIerent things, and
seeing whether you are successIul at the modiIications. Finally, you can try to enter something oI your own devising, Irom scratch.
Getting expert help can be a useIul way oI quickly seeing how to notice and Iix the basics. However, the point Irom getting such help is not to
only to get your mistake Iixed, but also to learn Irom the experience so that you can avoid and/or recognize how to Iix similar things in the Iuture
on your own.
Examples of Errors with if statements
Example Commentary

x := 3;
countThree := 0;
countSubThree := 0;
if x>3
then
countThree := countThree +1;
else
countSubThree := countSubThree +1;
end if;
printf("three and above:%d, below three: %d",
No missing pieces here.
12.8 Troubleshooting iI statements 257
countThree, countSubThree);
three and above:0, below three: 1


x := 3;
countThree := 0;
countSubThree := 0;
if x>3
countThree := countThree +1;
else
countSubThree := countSubThree +1;
end if;
printf("three and above:%d, below three: %d",
countThree, countSubThree);
Error, missing operator or `;`


In this example, we Iorgot to put in
the word "then". AIter the error, the
cursor is positioned in the midst oI
the line aIter the "iI", which is the
Iirst time that Maple's input processor
notices that there's something wrong.
To handle this error, one looks
backwards Irom the position oI
the cursor to see iI there's anything
missing. Admittedly there's not a lot
oI inIormation given that the cause
is a missing then, but this is Iairly
typical in working with programming
languages. You'll get an indication
that there's something wrong, and
the point at which the wrongness
was Iirst noticed, but you won't be
able to depend on the error message
to tell you exact what was wrong.
II computer technology could do
that reliably, systems would just
Iix the errors automatically. But
doing that well seems to require
artiIicial intelligence beyond present
conventional technology.



x := 3;
countThree := 0;
countSubThree := 0;

We Iorgot to end the statement with
end iI (or Ii).
258 12 More control: Ior and iI
if x>3
then
countThree := countThree +1;
else
countSubThree := countSubThree +1;
printf("three and above:%d, below three: %d",
countThree, countSubThree);
Warning, premature end of input, use <Shift> + <Enter> to
avoid this message.




x := 3;
countThree := 0;
countSubThree := 0;
if x>3
then
countThree := countThree +1;
else
countSubThree := countSubThree +1;
end do;
printf("three and above:%d, below three: %d",
countThree, countSubThree);
Error, reserved word `do` unexpected


We said "end do" rather than "end iI".


x := 3;
countThree := 0;
countSubThree := 0;
if x>3
then
countThree := countThree +1;
Actually, this is not a mistake.
Evidently "end iI" and "end" are
acceptable here. "Ii" is another way oI
ending an if statement.
12.8 Troubleshooting iI statements 259
else
countSubThree := countSubThree +1;
end;
printf("three and above:%d, below three: %d",
countThree, countSubThree);
three and above:0, below three: 1


x := 3;
countThree := 0;
countSubThree := 0;
if x>3
then
countThree := countThree +1;
else
countSubThree := countSubThree +1;
end if
printf("three and above:%d, below three: %d",
countThree, countSubThree);
Error, missing operator or `;`

We get the same error message as
in the second item in this example.
However, this time the problem is
that we leIt oII the semi-colon aIter
the end iI. We need it iI there's an
instruction aIter the iI statement in the
script.

Programmers need to the consider
that there could be more than one
kind oI mistake that would provoke
an error message. This kind oI
situation can be encountered with
almost any sort oI programming
system (e.g. C, Matlab, Java, etc.)
due the limitations oI conventional
language-processing technology.

x := 3;
countThree := 0;
countSubThree := 0;
IF x>3
THEN
countThree := countThree +1;
ELSE
countSubThree := countSubThree +1;
END IF;

The mistake here is that "IF" does
not mean the same thing as "iI" to
Maple -- case matters. Executing
this script gives an error message
with the cursor positioned right
aIter the IF. The error message is
saying that Maple doesn't know
what to do with the IF. It is not
particularly inIormative about the
260 12 More control: Ior and iI
printf("three and above:%d, below three: %d",
countThree, countSubThree);
Error, missing operator or `;`

cause. You could discover the cause
by comparing what is done with
examples that are known to work
which all use "iI" rather than "IF".
12.9Summary of Chapter 12
General form of for statement Example
Ior stepperJariable Irom initialJalue to finalJalue do
loop bodv,
end do;


for angle from 5.0 to 10.0 by .5 do
x := convert(angle, units, degrees,
radians);
print("Cosecant of ", angle, " degrees
is", csc(x));
end do:
(12.20)
General forms of if...then..else...end if Example


if condition then
#print out a message if a sin of a multiple of
Pi/4 is 1 or -1.
for i from 1 to 10 do
if abs(sin(i*Pi/4)) = 1
then
12.9 Summary oI Chapter 12 261
code to executed iI the condition is true
end if


print(i,"sin(",i*Pi/4," ) is ",
sin(i*Pi/4));
#i*Pi/4 automatically reduced to simplest
terms in output
end if;
end do;
(12.21)
if condition then
code to executed iI the condition is true
else
code to be executed iI the condition is Ialse
end if




grade := 3.2: #This value will cause execution
of the "else" code
if (grade >= 3.5)
then
print("An excellent grade", grade);
else
print("A grade:", grade);
end if;
(12.22)

if (condition 1) then
statements separated bv , or .
elif (condition 2) then
statements separated bv , or .

(more alternatives specified bv more elifs)

else
statements separated bv , or .

end if


for i from 1 to 10 do
if i=1
then
print("i is one");
elif isprime(i) #built-in primality
testing function
#Returns true if its argument is a
prime number
#See on-line documentation for
isprime.
then
print (i, "is prime");
elif i mod 2 = 0 #test for divisibility by
two
#See on-line documention for "mod"
then
print(i, "is even");
else
262 12 More control: Ior and iI
print(i, "is neither prime nor even
nor one");
end if;
end do;
(12.23)
263
13Looking inside execution of a procedure or
script
13.1Chapter Overview
In this chapter, we show ways oI providing more printed output while a script or procedure is running. printf and sprintf provide formatted
printing -- Iine control oI exactly how numbers, Iormulas, and other results appear when they are printed. We revisit the use oI colons as a way oI
suppressing execution trace output Irom loops. Finally, we introduce the trace operation as a way oI causing deeper amounts oI execution tracing
to occur, allowing programmers to view the internal operation oI user-deIined procedures.
13.2More on printing: print, printf and sprintf
In Chapter 10, we introduced print that prints the value oI things out during execution oI a script. A synopsis oI print is that it is a Iunction
that takes a sequence oI values as arguments, and returns NULL as a result. Its use is in its side effect -- it causes the sequence oI values to be
displayed in the worksheet in two dimensional Iormat.
Use of print
Example Commentary

(13.1)

(13.2)
We set up a list oI values. L has all Iloating-point
values because the "10.0" in the denominator
causes all computations to be done using limited-
precision arithmetic. The display oI the value oI L is
suppressed because we ended the assignment with a
colon.


print prints out the sequence oI values given to it as
arguments. You can give any number oI items in the
sequence. The commas separating the string and the
number are Irom the sequence. Note that the result is
not displayed because the line ends with a colon.
printf is a similar Iunction, but it provides the programmer more precise control over the Iormat oI the output. The Iirst item in the sequence
oI parameters to printf is a format string: a string that contains ordinary words plus special Iormat codes that begin with . The table below
summarizes common Iormat codes.
Common formatting codes for printf and sprintf
Format code Example Explanation
a Algebraic format. Works Ior any type oI
value. The value is printed out in a Iormat
suitable Ior textual input.
264 13 Looking inside execution oI a procedure or script
The answer is: x = x^2+1

f
w.df The answer is : 0.009095
0.00910





2432902008176640000.00000
0.750000

The answer is:
Error, (in IprintI) number expected Ior Iloating point Iormat

Fixed format Ior Iloating point numbers.
w and d are optional. II given, w
describes the number oI columns Ior
the entire number, and d the number oI
columns Ior the number aIter the decimal
point.


Values will be converted to Iloats where
possible.
e
w.de

9.095277e-03
9.09528e-03
3.628800e+06
ScientiIic Iormat Ior Iloating point
numbers. wand d are optional. II given,
w describes the number oI columns Ior
the entire number, and d the number oI
columns Ior the number aIter the decimal
point. II w is wider than the number oI
digits required, then it is Iilled out with
leading white space.

Values will be converted to Iloats where
possible.
13.2 More on printing: print, printI and sprintI 265
7.500000e-01
(13.4)
2.43290e+18
7.500000e-01

The answer is:
Error, (in IprintI) number expected Ior Iloating point Iormat

d
2432902008176640000

Error, (in IprintI) integer expected Ior integer Iormat

Error, (in IprintI) integer expected Ior integer Iormat

(13.6)
3

Format Ior integers.

This Iormat rejects any value that is not
oI integer type. The built-in whattvpe
Iunction can provide guidance about what
is and what isn't going to work, iI you
aren't sure.

266 13 Looking inside execution oI a procedure or script
(13.7)

s
The color is DodgerBlue.

Format Ior strings.
\n
One two,
buckle my shoe.
3, 4,
shut the door.
Causes a line break between preceding
and succeeding parts oI the output.
\t
1,\0112,\011buckle my shoe.
3\0114,\011shut the door.

Causes next output to appear starting at
the next columnar (tab) position.
The on-line documentation Ior printf describes all Iormat codes comprehensively. Some oI them allow Ior quite intricate eIIects such as right or
leIt justiIication, padding with leading or trailing zeroes, etc.
printf can allow computed inIormation to be printed quite expressively. However, the Iormat codes will give error messages iI you supply a value
that can't Iit into the Iormat.
Use of printf
Example Commentary
We set up a list oI values as in the previous
example.
The last element of the list is 0.800000.

The Iirst argument to printI is a string
describing how to Iormat the rest oI the
values. The string contains Iormat codes
describing where to insert the values oI the
later arguments into the message described in
the string. In this example, there is one Iormat
code I, Ior printing numbers in conventional
decimal point notation. Since printI joins
together the message and the value together
into a single message without a sequence
into a single result, what is printed is without
commas.
13.2 More on printing: print, printI and sprintI 267
There are 4 elements in this list.
The largest is 9.048374e-01.

Additional printI codes are "d" (Ior integer
values), and "e" (Ior Iloating point numbers
in scientiIic notation). The Iormat code \n
causes the next output to occur on the next
line.
L has 4 elements. The largest is 9.048374e-01.

Note that a printI statement does not
automatically cause the output on a new line.
Thus, several printI statements executed
together as these are will have all the output
on a single line.
There are 4.000000 elements in this list. The
largest is
Error, (in IprintI) integer expected Ior integer Iormat

II you attempt to use I Iormat on an integer,
or d Iormat on a number that is not an
integer, you will get incorrect values (with no
warning that they are incorrect), or possibly
an error message. Let the user beware; the
computer isn't going to necessary save you
Irom making a mistake.
Error, (in IprintI) number expected Ior Iloating point Iormat

You can't print a list oI Iloating point numbers
using just I Iormat.
[.9048374180, .8187307531, .7408182207, 4/5]
The a Iormat will handle arbitrary numerical
or non-numeric values. The value being
printed out is a list, which is a "non-numeric
value" because it isn't a single number.
However, the output uses the textual Iormat
rather than the "pretty" Iormat used by print.
sprintf is like printf in that its Iirst argument is a string with Iormat codes, and the rest oI its arguments are values to be Iormatted. However,
unlike printf it does not print (display) any inIormation. It returns a string which is the Iormatted inIormation instead. This is useIul in situations
where Iunctions need Iormatted inIormation, such as titles or labels oI plots.
Use of sprintf
Commentary
(13.9)
(13.10)


We set up a list oI data points. Using sprintI, we
create a string. The d Iormat is used to insert the
number oI points into the message.
268 13 Looking inside execution oI a procedure or script

We need the string to put a title onto our pointplot.
You can read more about the title argument to
pointplot by entering plot,options in Maple's on-
line help.

The advantage oI using sprintf is in re-use. II
we want to plot a diIIerent set oI points, we can
change the line assigning L (and the plot line oI x
coordinates), but we don't have to alter the sprintI
since it will automatically use the right value
corresponding to the length oI the list.
print and printf are oIten included in procedures being debugged by programmers. This is to make the execution trace more intelligible which
oIten allows the programmer to more quickly Iind where a computation is going awry.
"Commenting out": The idea with diagnostic print/printI output is that it is put in to Iind the cause oI a problem, and then taken out aIter a
programmer thinks that they've Iound and Iixed the problem. However, it is oIten the case that there is more than one problem in the program.
OIten times, aIter Iurther work Iixing multiple problems, the need Ior the same diagnostic output as beIore returns. For this reason, rather
than deleting printing statements (and then needing to type them in again), the programmer puts a comments symbol # in Iront oI the printing
operation. This deactivates the printing until the # is removed. It's a lot easier to delete and add a comment sign than it is to retype a printing
statement.
13.3For the curious: why printf is so cryptic
The Iormat codes Ior printf are a "programming language within a programming language" in that they describe in abbreviated Iorm how the
computer should output values. The style used Ior Iormat codes is very much unlike the rest oI Maple, in that brevity rather than readability
seems to be the criterion used to design this portion oI the language. A language Ior science and engineering popular starting in the middle oI the
twentieth century was Fortran, which had Iormatting that was similar in approach but was much diIIerent in syntax. Given your knowledge oI
Maple (and English), the intent oI the program is Iairly obvious: it will print There were 3 values, and the answer was 4.7.
A Fortran program that demonstrates formatted printing
n = 3
x = 4.7
print (*,35) n,x
13.4 Suppressing execution trace output Irom loops 269
35 format('There were',i2,' values, and the answer was',f5.1, '.')
stop
end
The design oI printf can be explained by the Iact that Maple's printf is a close imitation oI the printf Iound in the programming language C
(popular with many programmers at the time Maple was invented in the '80s, and still popular among computer engineers and scientists doing
device programming). Matlab, being invented in the same era, also a sprintf Iunction oI highly similar design. Imitation makes it easier to transIer
programming skill Irom one language to another. Inventors oI programming languages will tend to imitate unless they have a goal or insight that
justiIies a diIIerence. Programming languages designed by skillIul designers do not introduce arbitrary diIIerences without a good cause. When
you notice similarities between languages, take it as a recognition that many people liked that style oI doing things, across many diIIerent kinds oI
programmers and kinds oI problems. When you notice a diIIerence between languages, you should explore the reasons why the designers made
diIIerent choices. Usually there is a problem or issue that the designer is trying to handle better by being diIIerent Irom the rest oI the pack.
Matlab script for plotting a few points, in blue

This probably isn't the only way, or the shortest way, of producing the title in Matlab, but we use it to illustrate that
knowledge of formatting codes is one of the things that transfers between languages. Matlab also has a sprintf function.
About the only formatting features that it (as of February, 2012) lacks is a a format code to make conversion of
formulas convenient.
Programming language Ieatures Ieatures that work across tens oI thousands oI programmers and programs are usually motivated by signiIicant
design rationales. The design oI most programming languages is nowadays inIormed by what worked well or didn't work well in the past.
13.4Suppressing execution trace output from loops
We mentioned it in Chapter 12 beIore, but do so again as our use oI loops grows: putting a colon (:) rather than a semi-colon (;) aIter the end do
oI a while or for loop suppresses the execution trace output normally produced by the loop. II you want to see particular results Irom within the
loop, use print or printf, which are not suppressed by the use oI colons.
Accumulation of two results in the same loop; suppression of output
n := 5;
result := 0;
result2 := 1;
for i to n do
What the execution trace oI this loop looks like without output
suppression. Since the loop computes two things per repetition,
270 13 Looking inside execution oI a procedure or script
result := result+i;
result2 := result2*i^2;
end do;
printf("The sum of the numbers from 1 to %d
is: %d.\n", n, result);
printf("The product of squares i^2 from i=1 to
%d is %d.\n",n,result2);
The sum of the numbers from 1 to 5 is:
15.
The product of squares i^2 from i=1 to 5
is 14400.

and repeats 10 times, there are 10 lines oI output to look at.
Imagine what would happen iI there n were 100.
n := 10;
result := 0;
result2 := 1;
for i to n do
result := result+i;
result2 := result2*i^2;
end do:
printf("The sum of the numbers from 1 to %d
is: %d.\n", n, result);
We suppressed output Irom the execution trace by placing a
colon at the end do, which suppresses the execution trace oI
everything in the loop body. This is something to do aIter you
have gotten the loop to work correctly.
13.5 Looking inside oI procedure invocations (Iunction calls) with trace 271
printf("The product of squares i^2 from i=1 to
%d is %d.\n",n,result2);
The sum of the numbers from 1 to 10 is:
55.
The product of squares i^2 from i=1 to
10 is 13168189440000.
n := 10;
result := 0;
result2 := 1;
for i to n do
result := result+i;
result2 := result2*i^2;
printf("i is: %2d, result2 is: %4.2e
\n",i,result2);
end do:
printf("The sum of the numbers from 1 to %d
is: %d.\n", n, result);
printf("The product of squares i^2 from i=1 to
%d is %d.\n",n,result2);
i is: 1, result2 is: 1.00e+00
i is: 2, result2 is: 4.00e+00
i is: 3, result2 is: 3.60e+01
i is: 4, result2 is: 5.76e+02
i is: 5, result2 is: 1.44e+04
i is: 6, result2 is: 5.18e+05
i is: 7, result2 is: 2.54e+07
i is: 8, result2 is: 1.63e+09
i is: 9, result2 is: 1.32e+11
i is: 10, result2 is: 1.32e+13
The sum of the numbers from 1 to 10 is:
55.
The product of squares i^2 from i=1 to
10 is 13168189440000.
We can see a trace oI just the result2 values by including a
printf statement. This gives us the opportunity to include other
inIormation that might help understand the execution trace.
Note that we are using a Iormat code Ior i that makes its output
always appear in two columns, which lines up the output that
is easier to read. You can read about this in the on-line help Ior
printI.
13.5Looking inside of procedure invocations (function
calls) with trace
II we have a script where a procedure is invoked, then the execution trace normally only displays the result oI the procedure. However iI there is a
programming error in a procedure we've written, we won't see the execution trace oI the actions within the procedure.
272 13 Looking inside execution oI a procedure or script
Maple provides another operation, trace, which will turn on the execution trace within a procedure. untrace is a way to turn oII tracing once we
no longer need it. Or write a code edit region that begins with restart, which will also cause Maple to stop tracing (as well as Iorget about the state
oI everything else that had happened in the session.
By tracing the procedure we see the execution trace oI each action during the procedure. It is easier to see which line is going wrong with the
trace on. AIter you are Iinished tracing, you can
stop the tracing by doing a untrace('procedure name),. II your code edit region includes both the procedure deIinition and its invocation, then you
can also turn oII tracing by putting a restart as the Iirst line oI the region.
Tracing and untracing a function
Problem: We give a team oI two students a series oI 4 individual tests. We wish to award them a point total that's the sum oI
the best score Irom each test. For example, iI one student had scores oI 1,3,5, and 10, and the other student had scores oI 0, 4, 6
and 7, then the sum would be max(1,0) max(3,4) max(5,6) max(10,7) 21. Write a procedure that takes two lists oI scores
produces the desired sum as a result.

maxSum := proc(xdata, ydata)
local s,i, n;
#Loop over the 1st, 2nd, etc. value of the
lists
for i from 1 to n do
#Compare them and add the larger to s.
if xdata[i]>ydata[i]
then s := s + xdata[i];
else s := s + ydata[i];
end if;
end do;
return s;
end proc;

(13.14)
We deIine the procedure.
(13.15)
(13.16)
We try it out. We get error messages. Bummer. But what is
causing the problem?
13.5 Looking inside oI procedure invocations (Iunction calls) with trace 273
Error, (in maxSum) Iinal value in Ior loop must be numeric
or character

(13.17)
--~ enter maxSum, args |1, 3, 5, 10|, |0, 4, 6, 7|
-- ERROR in maxSum (now at top level) Iinal value in
Ior loop must be numeric or character}
Error, (in maxSum) Iinal value in Ior loop must be numeric
or character

Tracing the procedure and running the test again shows the
execution trace oI the procedure. It seems to being with
the inIormation we expect Irom L1 and L2. However, it
immediately gives an error, which means that the Iirst line oI
action -- "Ior i Irom 1 to n do..." must be causing the error.
The trace doesn't tell us how to diagnose the cause or the cure
-- we have to apply our knowledge oI what Ior loops want.
We consult examples oI loops -- we seem to using the right
syntax. But the message says that "Iinal value in Ior loop must
be numeric or character". The Iinal value in this case is n --
and that causes us to realize that we never assigned n a value. n
does not have a numeric or character value. The way to Iix this
mistake is to decide how we are going to give n the right value,
which is the number oI items in the list.
maxSum := proc(xdata, ydata)
local s,i, n;
n := nops(xdata);
#Loop over the 1st, 2nd, etc. value of the
lists
for i from 1 to n do
#Compare them and add the larger to s.
if xdata[i]>ydata[i]
then s := s + xdata[i];
else s := s + ydata[i];
end if;
end do;
return s;
end proc;

274 13 Looking inside execution oI a procedure or script
(13.19)
(13.20)
(13.21)
--~ enter maxSum, args |1, 3, 5, 10|, |0, 4, 6, 7|
-- exit maxSum (now at top level) s21}
(13.22)

RedeIining maxSum again has removed the trace. We need to
invoke trace again.

Once we do so, we see the execution trace. We see that the Iirst
operation "n: ..." results in 4, which is correct since that's the
length oI L1 (what was given as the Iirst argument xdata).

The next operation would be one oI the s : s xdata|i| or
s : s ydata|i|. We expect this to be an accumulation oI the
maximum values. However we are getting "s1" rather than
just "1".

This causes us to realize that we have Iorgotten to initialize s.
maxSum := proc(xdata, ydata)
local s,i, n;
n := nops(xdata);
s := 0; #Initialize accumulation variable
to zero.
#Loop over the 1st, 2nd, etc. value of the
lists
for i from 1 to n do
#Compare them and add the larger to s.
if xdata[i]>ydata[i]
We edit the code edit region and execute it again, redeIining the
procedure.
13.5 Looking inside oI procedure invocations (Iunction calls) with trace 275
then s := s + xdata[i];
else s := s + ydata[i];
end if;
end do;
return s; #Return accumulated sum as the
result.
end proc;
(13.23)
(13.24)
--~ enter maxSum, args |1, 3, 5, 10|, |0, 4, 6, 7|
-- exit maxSum (now at top level) 21}
(13.25)

We have to trace the procedure again since the redeIinition oI
maxSum wipes out Maple's notes
that the procedure was traced. It seems to work better now.
(13.26)
To untrace, we have to put the name oI the procedure in single
quotes.

276 13 Looking inside execution oI a procedure or script
(13.27)

Once we untrace, we can use the Iunction without seeing the
execution trace.
Recall that you can also look at the value oI variables after execution is ended by looking at the Variables panel oI the Maple window, as
described in Chapter 4.
13.6 Summary of Chapter 13
printf, sprintf
L has 4 elements.
The largest is 9.048374e-01.

The Iirst argument to printI is a string containing Iormat codes
speciIying how the values oI the subsequent arguments should
appear. Common Iormats are d, I, e, s. There are large
number oI modiIiers. \n is the Iormat code to cause next output
to appear on a diIIerent line.
(13.28)
(13.29)
sprintI is good Ior manuIacturing plot titles, among other
things.
13.6 Summary oI Chapter 13 277

Suppressing output in loops
Ending a while or for loop with a end do: rather than end do; will suppress the execution trace oI all the activity within that
loop. There is no need to end each line within with a colon.
Tracing procedures
trace(maxSum); Turns on execution tracing oI the named procedure.
untrace('maxSum'); Turns oII execution tracing oI the procedure. Name oI
procedure must be enclosed in apostrophes (single quotes).
278 13 Looking inside execution oI a procedure or script
279
14Tables and other data structures
14.1Chapter Overview
This brieI chapter introduces Maple tables. Tables are another way in Maple to store multiple values in a single "data container". Unlike Maple
lists, Maple tables are mutable, which means that you can add or change data in them without having to copy anything. We revisit another data
structure, the sequence and show how it can be used to do multiple simultaneous assignment, and to return multiple results from a procedure.
14.2A review of Maple data structures and properties
The value oI computers in technical work comes not only Irom doing long calculations conveniently, but being able to do thousands or millions oI
shorter computations quickly. Loops (for and while) are one way to harness this power without making the programming very tedious. Another
need is a way to store lots oI data. We've already seen lists, sequences, and sets as general-purpose containers. Maple also has specialized
containers: ranges, equations, and expressions.
Let's review the data containers we've seen and used in Maple so Iar. Each kind oI container has a diIIerent syntax to create it, and to access its
parts. Variables can be assigned as their value their whole container. Some Maple Iunctions use the whole container, while in other cases you use
"parts access" to use particular items within the container. One common aspect oI these data containers is that they are immutable: once you've
created them, you can't change them, you can only access (Iind out the value oI) their value. In order to "change" a list or expression, Ior example,
what you have to do is to create a new list/expression with the changes in it, and assign it to the same variable as had the old list.
Type of
container
Examples How to create it How to access a
part of it
Operations that can
be performed with
container
Ways of
changing data
container
after it's been
created
Math
expression
a b c
exp(-b*x/3)
sqrt(32)- y`2
Type it in using
an arithmetic
operation or
mathematical
Iunction
op(i, expr) gets the
i-th piece oI the
expression
Arithmetic and math
Iunctions: , -, *, `, /,
sqrt, !, etc.,
nops(expr) tells you the
number oI terms in the
expression.
Can't do it. Can
only create a
new container
with edited
content.
Range 1..10 as in
plot(x`2, x
1..10)
Type it in using
a ".."
op(1, r) gets the low
end oI the range.
op(2,r) gets the high
end
None other than access
Iunctions, but many
Iunctions use ranges as
one oI their parameters.
Can't do it.
Equation 2*x-35 as in
solve(2*x-35,x)
Type it in using
a ""
rhs(eqn), lhs(eqn)
gets the right and
leIt hand sides oI the
equation
solve, Isolve, plot, etc.
although usually other
arguments are necessary
Can't do it.
List |1.1, 2.3, 5.5|

| | #empty list
Type in items
separated
by commas,
enclosed by
square brackets.
L|i| gets the i-th
item Irom the list.
II i is negative,
counting starts Irom
the end.

nops tells you the number
oI items in the list
Can't do it.
280 14 Tables and other data structures
Type of
container
Examples How to create it How to access a
part of it
Operations that can
be performed with
container
Ways of
changing data
container
after it's been
created
L|i..j| gets a list oI
the ith through fth
items.
Plot
structures
Result oI
plot(x`2,
x1..10)
As a result oI a
plot or display
operation.
Typically, you don't
want to, you just
want to print it.
op does work on
plot structures (see
structure, plot in
Maple on-line help)
printing plots is the
main thing that you
want to do. You can also
operate on them with the
plots|display| operation.
Can't do it.
Sequences
NULL #empty
sequence
Type in items,
separated by
commas.
L|i| gets the i-th
item Irom the list.
II i is negative,
counting starts Irom
the end.

L|i..j| gets a list oI
the ith through fth
items.
All Maple Iunctions
accept sequences,
although not all oI them
work with sequences that
are the wrong length or
provide the wrong kinds
oI values.
Can't do it.
Set 1.1, 2.3, 5.5}

} #empty set
Type it in:
items separated
by commas,
enclosed by
curly brackets.
S|i| gets the i-th item
Irom the set.

S|i..j| gets a set oI
the ith through fth
items.
nops tells you the number
oI items in the list

union, intersect, minus,
perIorm set operations
Can't do it.
14.3 Tables
In this section, we introduce another kind oI data structure, tables. A table is a structure where each element is accessed using a value called a
kev or index. Individual elements oI the table can be assigned, in any order. Individual elements can be accessed using the index notation, e.g t|2|
to get the element oI the table t whose access key is 2. This is similar to how lists worked, where L|2| Ior a list L was an expression whose value
was the 2nd element oI the list L.
The kev can be any valid Maple value, not necessarily an integer (although they oIten are). Tables can add elements over time with additional
assignments.
Tables are useIul in situations where we will be adding items to the data structure Irequently, but we don't know in advance how many entries
we will ultimately have. They will be more eIIicient to use than lists or sets in that kind oI situation. In Iuture chapters we will also see ways to
exploit the other useIul property oI tables, where it is more convenient to have non-integer values Ior access keys.
Example
Demonstration of tables
14.3 Tables 281
Example
(14.1)

This creates a table named T. Initially there are no items in the
table. Whatever previous value T may have had is Iorgotten.
(14.2)

Through assignment we create a table entry Ior T|1|.
(14.3)

Through assignment we create a table entry Ior T|2|.
(14.4)

Through assignment we create a table entry Ior T|4|. The value
oI this entry depends on the values oI T|1| and T|2|.
(14.5)

We reassign the value oI T|2|, overwriting the previous value.
(14.6)

Expressions mentioning assigned elements oI the table will use
the assigned value. II that element oI the table does not have an
assigned value, it remains just as a symbol.
(14.7)

The indices Iunction will return a sequence oI the access keys
Ior the table. Each access key is enclosed in brackets so that
you can tell where one key stops and the next begins.
(14.8)

The entries Iunction will return a sequence oI values stored in
the table. The sequence is generally listed in the same order
that the indices Iunction lists access keys.
(14.9)

This is the way to tell how many items are in a table at any
time. Note that nops(indices(T)) does not work because indices
produces a sequence oI items, so that nops would typically be
given several arguments instead oI the one that it expects.
282 14 Tables and other data structures
Example
(14.10)

The convert command can be used to put all the entries oI a
table into one list.

(14.11)
(14.12)
(14.13)


The indices oI a table can also be letters. Here we assign a
value to the index T|x|.
(14.14)

This is the way to remove an entry Irom the table -- putting
quotes (', not backquotes `) around the item being removed.
This is also the way unassignment oI values were done Ior
ordinary variables in section 4.2.
(14.15)
(14.16)

AIter removing the item, we see Iewer items in the table.
We can create a list oI all the entries oI T using the convert
operation. This does not change T, it just creates an additional
list structure that has the same data as T's entries which can be
assigned to another variable.
(14.17)

(14.18)
(14.19)


II we want to Iill up a table by giving a multiple entries at once,
we can give the table Iunction a list oI equations indicating
indices and the associated value.

14.4 Troubleshooting with tables 283
There are three major diIIerences between tables in Maple and the data containers we've seen previously: a) there is a separate creation step
(creating a table), b) you can change the contents oI the table be creating, changing or deleting an item/entry c) items within a table are stored
under a kev or index value. The latter works similar to the indexing used Ior lists, but the index does not have to be an integer. A table can store
any kind oI item under any kind oI key. For the time being, we will stick with integer keys, but eventually we will take greater advantage oI this
"key" Ieature.
14.4Troubleshooting with tables
1. Table creation tbl . table( );
should always appear at the beginning oI a script, which assigns tbl a Iresh, empty table. II you don't do this, you may see items leIt over in tbl
Irom the last time you executed a script in the same worksheet with tbl in it. Alternatively, you can do a restart; beIore running a script that uses a
table.
2. Use entries and indices to see what's in your table. Check the results to make sure that there are no unexpected indices and that the values are
computed correctly.
3. Be able to recall the circumstances where you should use square braces | | and when to use parentheses ( ) in working with tables.
Troubleshooting with tables: mistaking ] and ( )
(14.20)

Error, invalid leIt hand side in assignment



Accessing a table value requires the use oI square brackets | |
to enclose the index. Use oI parentheses makes Maple think
you are talking about Iunctions rather than tables.


(14.21)

We place an entry into the table -- the value 4.7 accessed by the
key 3.
(14.22)
(14.23)

This does not access values in the table. It uses ( )s rather than
the necessary | | s.
Troubleshooting tables: checking for extra keys
(14.24)
This a demonstration oI how the Iact that "a table key can be
any value" can create conIusion. We create a table and place
three entries into it.
284 14 Tables and other data structures
Troubleshooting tables: checking for extra keys
(14.25)
(14.26)
(14.27)

(14.28)

(14.29)
(14.30)
(14.31)
(14.32)
(14.33)
(14.34)

We assign i, f, and k the values oI 2, 3.0 and 4.






We expect that this will result in creating or changing certain
entries oI the table.
(14.35)

Where did the 5.0 go? We thought we did T|3.0| : 5.0 with
the use oI k
(14.36)
We use the indices and entries Iunctions to tell us what is going
on in T. Evidently the keys used Ior T include both 3 and 3.0.
14.5 Using sequences in multiple simultaneous assignments 285
Troubleshooting tables: checking for extra keys
(14.37)

The light bulb in our heads comes on where we recall that
3 and 3.0 are regarded as diIIerent in Maple. "3" is an exact
integer and "3.0" is a Iloating point number.
(14.38)
(14.39)
(14.40)
(14.41)
(14.42)


We see that there are two distinct entries in the table.
Reassigning one oI them does not erase the other entry,
although the value oI the Iirst one does change.
(14.43)
(14.44)
(14.45)

As indicated in the demonstration oI tables, unassigning a table
entry will erase it Irom the table.
14.5Using sequences in multiple simultaneous
assignments
Maple provides a way oI doing several assignments at once. Rather than variable . expression; we can give sequences on both sides:
variable1 , variable2 : expression1, expression2;
286 14 Tables and other data structures
will assign variable1 the value oI expression1, and variable2 the value oI expression2. 3 or more assignments can be perIormed in a similar
Iashion.
Use sequences to do multiple simultaneous assignments.
(14.46)
a=1, b is 3.141593, c is
arctan(1/3*Pi)
a=1.000000, b is 3.141593, c is
0.808449

Multiple assignment at work. This is also a demonstration that
the I Iormat will Iorce an evalI approximation as long as the
value is some kind oI real number.
Error, ambiguous multiple assignment
Error, ambiguous multiple assignment

We get error messages iI the number oI variables assigned don't
match the number oI values being supplied on the right hand
side. Maple regards this as "ambiguous" because we could
be thinking oI assigning a the value 1, and b the sequence
3.1415..., arctan( ) or perhaps assigning a the sequence 1,
3.1415..., and b the value arctan( ) . Yet another possibility
is to assign a the value NULL, and b the sequence 1, 3.1415...,
arctan( ) . Maple doesn't try to decide and signals an error
instead.
(14.47)
(14.48)
(14.49)

II there's only variable on the leIt hand side, then it is assigned
the entire sequence as a result. We can use a sequence in a
Iashion similar to a list.
14.6 Using multiple assignment with procedure results: return sequences oI results 287
The one time that you run into problems with this approach is when some oI the values you want to assign are sequences themselves. Maple
automatically combines sequences separated together by commas into a single sequence. This is a situation where it is better to use lists as the
values, since two lists don't automatically combine with a comma.
A tricky situation: sequences of sequences
(14.50)
(14.51)
(14.52)

We see that sequences blend together into a single sequence
when combined together with a comma.
Error, ambiguous multiple assignment

Maple can't tell what part oI 1,2,x,y is supposed to go to var1,
and what part is supposed to go to var2.

(14.53)
(14.54)
(14.55)
(14.56)

A sequence oI lists doesn't automatically combine the
lists together. This allows us to do multiple assignment
unambiguously.
14.6Using multiple assignment with procedure results:
return sequences of results
OIten the programmer wants a procedure to calculate several things at once. We can Iinesse the limitation that Maple procedures can return only
one result by making that result a sequence. We can then do multiple assignment to separate out the pieces.
#This procedure takes two lists of scores from n tests.
#It finds the average of the best scores, and the average
#of the worst scores
maxMinAvg := proc(list1, list2)
local i, n, smax, smin;
288 14 Tables and other data structures
#Initialize the accumulator variables
smax, smin:= 0,0;
n:= nops(list1);
#Add together the maximum and minimums.
for i from 1 to n do
smax := smax + max(list1[i], list2[i]);
smin := smin + min(list1[i], list2[i]);
end do;
return smin/n, smax/n;
end proc;
L1 := [ 95, 75, 74, 83];
L2 := [ 72, 98, 0, 85];
worst, best := maxMinAvg(L1, L2);
printf("Avg of best scores is: %4.2f\nAvg of worst scores is: %4.2f", best,worst);
L1 := [ 10, 9, 7.5, 6.4, 7.4];
L2 := [ 8, 8, 8, 8, 9];
worst, best := maxMinAvg(L1, L2):#suppress output with a colon
printf("Avg of best scores is: %4.2f\nAvg of worst scores is: %4.2f", best,worst);
Avg of best scores is: 88.00
Avg of worst scores is: 57.50
Avg of best scores is: 8.80
Avg of worst scores is: 7.46
14.7 A pattern oI looping: accumulation oI results 289
14.7A pattern of looping: accumulation of results
Many for and while loops accumulate results: they use variables whose value is updated each loop repetition to be incrementally more than they
were beIore. For example, the smax and smin variables oI the previous section are are initialized beIore the loop as zero. Each trip through the
loop they accumulate one more term oI the sums. When the loop ends, their values will be the sum oI all the maxes and mins, respectively. Thus
the role oI smax and smin are as accumulators; the loop oI the procedure might be reIerred to as an accumulation loop.
Tables can be used as accumulators as well. Rather than adding together data, they accumulate all the results oI a particular variety as entries in
the table. When the loop is Iinished, the table will be a data container holding all oI the computed results oI that variety. The example in the next
section shows a way as well.
14.8Application of tables in accumulative looping
Problem: a little old lady deposits some money into a savings account, which pays an interest rate of r paid yearly. She leaves it in the
account for numYears years. Calculate how much money she will have at the end of this time. Also, plot the growth of her money over
time.
Solution: To handle this problem, we couple two ideas:
1. We use a loop to calculate the amount in the account. II A is the amount on deposit, then
newA : A r*A;
expresses the relationship that the amount in the next year will be the current amount plus the interest. (We're assuming that r is a decimal Iraction
representing the interest, e.g. r is .03 iI we are talking about 3 interest.
AIter doing the computation , we can do A : newA; to set up the loop Ior the next iteration.
It's not diIIicult to write a procedure that does this. We parameterize it by choosing initialAmt to be the amount Iirst deposited, rate the interest
rate, numYears the number oI years. We could use i, r, and N as the parameter names instead but choose the longer ones because it helps remind
us oI the intended purpose oI each parameter.


interestCalculator := proc (initialAmt, rate,
numYears)
local i, A, newA;
#initialize A
A := initialAmt;
#Calculate new amount based on present amt
#and interest rate
for i from 1 to numYears do
newA := A + rate*A;
A := newA; #Get ready for next loop trip
end do;
return A;

290 14 Tables and other data structures
end proc;
(14.58)
(14.59)

We get a dollar amount by clicking on the result Irom the
computation and doing right-click -~ Numerical Formatting -~
Currency.
AIter we get this working, we add in the second idea:
2. We use tables to accumulate the year-values and the amount-values so that we can generate a plot. Once we have accumulated the values,
we convert the tables into lists because the plot Iunction wants lists oI values.


interestCalculator2 := proc (initialAmt, rate,
numYears)
local i, A, newA, yearTab, amtTab;

#initialize A
A := initialAmt;

#initialize tables
yearTab := table();
amtTab := table();

yearTab[0] := 0;
amtTab[0] := initialAmt;

#Calculate new amount based on present amt
#and interest rate
for i from 1 to numYears do

newA := A + rate*A;
A := newA;
#Remember this year's info.
yearTab[i] := i;
amtTab[i] := A;
end do;
return convert(yearTab,list),
convert(amtTab,list);

14.8 Application oI tables in accumulative looping 291
end proc;
(14.60)
(14.61)


interestCalculator3 := proc (initialAmt, rate,
numYears)
local i, A, newA, yearTab, amtTab;

#initialize A
A := initialAmt;

#initialize tables
yearTab := table();
amtTab := table();

yearTab[0] := 0;
amtTab[0] := initialAmt;

#Calculate new amount based on present amt
#and interest rate
for i from 1 to numYears do

newA := A + rate*A;
A := newA;
#Remember this year's info.
We develop a version oI the procedure so that it returns a plot
structure. Note the use oI sprintI to create a title that changes
when the parameters change.
292 14 Tables and other data structures
yearTab[i] := i;
amtTab[i] := A;
end do;
return plot(convert(yearTab,list),
convert(amtTab,list),
style=POINT, color="Red", symbol=diamond,
labels=["years", "dollars"],
title=sprintf("Growth of a $%5.2f deposit
at %4.2f percent interest",initialAmt,
rate*100));
end proc:

Looking at the compound interest Ior this period oI time, we
can see that the growth is beginning to diverge Irom being a
straight line.
Having developed the idea Ior this procedure in pieces, we create a single procedure that returns both results as a sequence.
interestCalculator4 := proc (initialAmt, rate,
numYears)
local i, A, newA, yearTab, amtTab;

#initialize A
A := initialAmt;

#initialize tables
yearTab := table();
amtTab := table();

yearTab[0] := 0;
amtTab[0] := initialAmt;

#Calculate new amount based on present amt
#and interest rate
We develop a version oI the procedure so that it returns a plot
structure. Note the use oI sprintI to create a title that changes
when the parameters change.
14.8 Application oI tables in accumulative looping 293
for i from 1 to numYears do

newA := A + rate*A;
A := newA;
#Remember this year's info.
yearTab[i] := i;
amtTab[i] := A;
end do;
return
amtTab[numYears],plot(convert(yearTab,list),
convert(amtTab,list),
style=POINT, color="Red",
labels=["years", "dollars"],
title=sprintf("Growth of a $%5.2f deposit
at %4.2f percent interest",initialAmt,
rate*100));
end proc:
(14.62)
(14.63)

Our little old lady might be really, really old aIter 40 years,
but it is an instructive demonstration oI compound interest
that someone younger might use. The graph shows why doing
retirement savings in your twenties has much more impact than
equivalent savings in later years.
Note that the graph does not get drawn when it is assigned as
part oI a sequence oI two values. In order to see it, we have to
ask Maple to display the value oI the variable that was assigned
the plot structure.

294 14 Tables and other data structures
interestCalculator5 := proc (initialAmt, rate,
numYears)
local i, A, newA, yearTab, amtTab;

#initialize A
A := initialAmt;

#initialize tables
yearTab := table();
amtTab := table();

yearTab[0] := 0;
amtTab[0] := initialAmt;

#Calculate new amount based on present amt
#and interest rate
for i from 1 to numYears do

amtTab[i] := amtTab[i-1] + rate*amtTab[i-1];
#Remember this year's info.
yearTab[i] := i;
end do;
return
amtTab[numYears],plot(convert(yearTab,list),
convert(amtTab,list),
style=POINT, color="Red",
labels=["years", "dollars"],
title=sprintf("Growth of a $%5.2f deposit
at %4.2f percent interest",initialAmt,
rate*100));
end proc:
We notice that the use oI A/newA and amtTab are somewhat
redundant. Using the ability to reIerence previously stored
values in the table, we rewrite the loop to use amtTab|i| Ior
newA and amtTab|i-1| Ior the previously calculated value oI A.
As the loop runs, it reIers to amtTab|1| and amtTab|0| the Iirst
time through the loop, amtTab|2| and amtTab|1| the second
time, etc. Everything is placed in the table in the correct place
to make this kind oI indexing work.
(14.64)
(14.65)
This version oI the procedure produces the same result as the
last, but it is a Iew lines shorter.
For those who are comIortable with equations such as
, this is the way it's done in the syntax oI a
typical programming language.
14.9 Summary oI Chapter 14 295

14.9Summary of Chapter 14
Type of
container
Examples How to create it How to access a
part of it
Operations that can
be performed with
container
Ways of
changing data
container
after it's been
created
Table T : table();
T|1| : 32.5;
T|2| : 32.6;
tblName :
table();

This creates an
empty table and
assigns it to the
variable tblName
replacing
whatever
previous value it
had.
T|kev| retrieves the
value stored under
the key Irom within
the table. Thus
iI you have done
T|1| : 32.5; then
subsequently doing
T|1| will retrieve the
32.5.

The kev can be
any value Maple
expression. Thus x,
or t3..5 can be kevs
Ior a Table
provides a
sequence oI the values
stored in the table
(including duplicate
values)

indices(T) provides a
sequence oI the kevs used
by the table.

convert(T,list) creates a
list oI the entries oI T.
The entries will be listed
in the order Maple uses
Ior the indices. When the
Key-value
assignment:
T|1| : 32.7
296 14 Tables and other data structures
Type of
container
Examples How to create it How to access a
part of it
Operations that can
be performed with
container
Ways of
changing data
container
after it's been
created
indices are non-negative
integers, this will be the
expected order. The order
will be less predictable
when non-integer indices
are used.
Use sequences to do multiple simultaneous assignments.
(14.66)
a=1, b is 3.141593, c is arctan(1/3*Pi)
a=1.000000, b is 3.141593, c is 0.808449

297
15More on looping
15.1Chapter Overview
We Iirst look at another way oI creating an animation in Maple, using a variant oI the plots|display| operation. This way oI creating animations is
useIul in situations where it is inconvenient to describe the animation through a single mathematical Iunction oI a time variable t.
Many activities in programming use repeated actions to accomplish a result. We've already seen this in Chapters 11-14 in the application oI while
and for. In this chapter, we look at patterns oI looping driven by their application. We also look at a number oI operations Irom the Maple library
that cause repetition to happen even without explicit loop programming. Knowing and using these short-cut operations that use looping implicitly
can save you time coding and troubleshooting.
15.2Animations, revisited
We've seen how the displav operation oI the plots package can be used Ior multiplotting -- creating a single plot that combines together separately
created plots. II insequencetrue is given as the second parameter to displav,, then an animation will be created with each element oI the list oI
plots given displayed, one per Irame oI the animation.
General Iorm:
plots|display|( list of plot structures, insequencetrue, other plot options)
creates an animation Irom a list oI plots or plot structures.
Problem: Create an animation with two parts: a square 1 unit wide and long moves from (0,0) to (0,1). Then it moves sidesways to (1,1)
while smoothly growing to be a square 1.5 units wide and long.
Solution: write a loop that generates the frames for the first part of the animation. Then add on more code that generates the frames for
the second part of the animation. Use display(..., insequencetrue) with the list of plots.
restart;
#Use drawBoxB function to draw a box
drawBoxB := proc(x0, y0, width, height, c, plotTitle)
local L1, L2, L3, L4, ptlf;
#short hand abbreviation of line
ptlf := plottools[line];
#this avoids having to remember to do "with(plottools)" in the session.
L1 := ptlf([x0, y0], [x0+width, y0], color=c);
L2 := ptlf([x0, y0], [x0, y0 + height], color=c);
L3 := ptlf([x0+width, y0], [x0+width, y0+height], color=c);
L4 := ptlf([x0, y0+height], [x0+width, y0+height], color=c);
plots[display]([L1, L2, L3, L4], title=plotTitle);
end proc;
movie1 := proc(u1, u2, numsteps1, numsteps2, boxColor)
local indexTab, frames, frames2, i,
h, uh, s1, s2, s3, L1;
#initialize tables
frames, frames2:= table(), table();

#part 1 -- draw square u1 units wide moving from (0,0) to (0,1) taking numsteps1 frames.
#h is the step distance
298 15 More on looping
h := (1-0)/numsteps1; #step size of movement
for i from 0 to numsteps1 do
frames[i] := drawBoxB(0,0+i*h,u1, u1, boxColor, "Part 1");
end do;
#part 2 -- draw square growing from u1 units wide
#to u2 units wide in numsteps2 steps,
#moving from (0,1) to (1,1) in numsteps2 frames.
#h is the distance step size
#uh is the size step size
h := 1/numsteps2;
uh := (u2-u1)/numsteps2;
frames2 := table();
for i from 0 to numsteps2 do
frames2[i] := drawBoxB(0+h*(i),1,u1+uh*(i), u1+uh*(i), boxColor, "Part 2");
end do;

#Convert tables into lists, then the two lists into sequences
#Section 6.8 (p. 85) talked about this conversion.
s1 := op(convert(frames,list));
s2 := op(convert(frames2,list));
#Merge the two sequences into a single sequence
s3 := s1, s2;
#convert the sequence into a list
L1 := [s3];
#Create a movie of all the frames
plots[display](L1, insequence=true);
end proc;
#Test 0 Draw a box
printf("Test 0 -- draw a box\n");
drawBoxB(0, 0, 1, 1,"Red", "Box Test");
#Test 1
#draw red square 1 unit in size changing into
#a square 1.5 units in size.
#Part 1 is 11 frames long, Part 2 is 16 frames long.
printf("Test 1 -- do a two part movie\n");
movie1(1, 1.5, 10, 15,"Red");
15.2 Animations, revisited 299
Test 0 -- draw a box
Test 1 -- do a two part movie
300 15 More on looping
The table below shows several Irames oI the animation.
Table 15.1: Several frames of movie1 animation

Frame 1 (Iirst Irame) oI
animation
Frame 12 Frame 20 Frame 27 (last Irame)
It is sometimes easier to code an animation one Irame at a time, separately, rather than try to Iigure out how to represent all the dynamics in a
single Iormula involving x,v, and t as the animate operation discussed in Chapter 9 required.
Programmers used to Iunctional notation will daisy chain together some operations and will avoid as many variables. This might be what a
more experienced programmer would do iI they thought that their intended readership could keep up. A programmer might use variables Ior
intermediate results s1, s2, s3, L1, L2, iI the same result was to be used several times, or iI they wanted to print out the value oI the intermediate
result Ior debugging purposes along the way. But here they are used only as a temporary way points to the Iinal result, so we can do without them.
The code region below illustrates the same program without use oI those intermediate variables. It does the same thing, but it is shorter.
restart;
#Use drawBoxB function to draw a box
drawBoxB := proc(x0, y0, width, height, c, plotTitle)
local ptlf;
#short hand abbreviation of line
ptlf := plottools[line];
#this avoids having to remember to do "with(plottools)" in the session.
plots[display]([ ptlf([x0, y0], [x0+width, y0], color=c),
ptlf([x0, y0], [x0, y0 + height], color=c),
ptlf([x0+width, y0], [x0+width, y0+height], color=c),
ptlf([x0, y0+height], [x0+width, y0+height], color=c)], title=plotTitle);
end proc;
movie1 := proc(u1, u2, numsteps1, numsteps2, boxColor)
local indexTab, frames, frames2, i, h, uh;
#initialize tables
frames, frames2:= table(), table();

#part 1 -- draw square u1 units wide moving from (0,0) to (0,1) taking numsteps1 frames.
#h is the step distance
15.3 Combining Ior and while 301
h := (1-0)/numsteps1; #step size of movement
for i from 0 to numsteps1 do
frames[i] := drawBoxB(0,0+i*h,u1, u1, boxColor, "Part 1");
end do;
#part 2 -- draw square growing from u1 units wide
#to u2 units wide in numsteps2 steps,
#moving from (0,1) to (1,1) in numsteps2 frames.
#h is the distance step size
#uh is the size step size
h := 1/numsteps2;
uh := (u2-u1)/numsteps2;
frames2 := table();
for i from 0 to numsteps2 do
frames2[i] := drawBoxB(0+h*(i),1,u1+uh*(i), u1+uh*(i), boxColor, "Part 2");
end do;

#Convert tables into lists, then the two lists into sequences
plots[display]([op(convert(frames,list)), op(convert(frames2,list))],
insequence=true);
end proc: #Suppress listing the procedure
#Test 0 Draw a box
printf("Test 0 -- draw a red box\n");
drawBoxB(0,0,1,1,"Red", "Box Test");
#Test 1
#draw red square 1 unit in size changing into
#a square 1.5 units in size.
#Part 1 is 11 frames long, Part 2 is 16 frames long.
printf("Test 1 -- do a two part movie\n");
movie1(1, 1.5, 10, 15, "Green");
15.3Combining for and while
Maple's syntax allows the use oI an stepper variable and a continuation condition (other than that oI the stepper variable reaching a Iinal value.
for stepperJariable from initialJalue to finalJalue by stepJalue while condition do .... end do
promotes a repetition which stops when either the stepperVariable reaches the finalJalue, or the condition becomes Ialse. II the to initialJalue is
omitted, then the while condition is the sole criterion used to decide whether to continue the repetition. This can abbreviate the coding in many
standard cases.
Example of for and while repetition: accumulate terms of a sum until next term is very small.
The code region below computes the sum

302 15 More on looping
until it Iinds a term that is tiny (less than the value oI the parameter tol). Compared to the previous
example, the main diIIerence is that we can skip the initialization and increment oI the stepper variable i since the for loop
handles it Ior us automatically. Since there is no finalJalue, the loop will not stop because the stepper variable reached a
Iinal value. However, at the start oI each repetition, the condition abs(term)~tol will also be checked. This is what stops the
repetition aIter a Iew trips through the loop.
restart;
sumIt := proc(x,tol)
local term, s, i; #Variables that will be used within the procedure.
# Initialize script parameters
#tol is the size of the stopping criterion.
#x is the value of sin(x) that we are trying to approximate
#Initialize accumulation variables to first term of sum.
term := x;
s := x; #s accumulates the sum, we initialize it to the first term.
#add on successive terms
for i from 3 by 2 while abs(term)>=tol do
term := (-1)^((i-1)/2)*(x^i)/i!;
printf("term is %e\n", term);
s := s+term;
end do;
end proc;
#Test 1
x := .3;
tol := 10^(-2); #want at least two decimal digits to be stable.
#Print out the final sum, and compare it to what it's supposed to be.
printf("Test 1, sum is: %e compared to %e.",sumIt(x,tol), sin(x));
#Test 2 -- same argument for x, smaller tolerance.
x := .3;
tol := 10^(-5); #want at least five decimal digits to be stable.
#Print out the final sum, and compare it to what it's supposed to be.
printf("Test 2, sum is: %e compared to %e.",sumIt(x,tol), sin(x));
15.4 Patterns oI looping 303
term is -4.500000e-03
Test 1, sum is: 2.955000e-01 compared to 2.955202e-01.
term is -4.500000e-03
term is 2.025000e-05
term is -4.339286e-08
Test 2, sum is: 2.955202e-01 compared to 2.955202e-01.
We can get more decimal places oI agreement between our approximation and the reIerence result by changing the value oI tol.
This will cause more repetitions oI the loop. We would get an inIinite loop (a bug) iI term doesn't become tiny eventually in the
repetition, but we know that it will, since .

One surprising Iact about this approximation is that it works very well Ior small values oI x. Only a Iew multiplications and
sums gives us an approximation to sin(x)) that's accurate to seven decimal digits.

While the built-in sin procedure is a perIectly good substitute Ior writing a loop, there are situations where this approach will
work at approximating a Iunction that there is no built-in library procedure Ior. You'd want to ask a numerical analvst to provide
assurances that it's going to work well.
15.4Patterns of looping
We have introduced quite a lot oI machinery. It can be easy Ior beginners to be uncertain about the connection between intent (what you want to
achieve) and svntax (how you say it in the programming language). Here as a short table summarizing the connections we have introduced so Iar
Patterns of looping
304 15 More on looping
Intent Example
Repeat exactly the same thing n times
to 5 do
print("I will not teach others to fly");
end do;
(15.3)
The action within the loop depends on the
value oI variable stepping Irom an initial to
a Iinal value, adding a Iixed amount to the
stepper variable between each step
for t from 0 by .1 to 1 do
printf("arccos(%f) = %f\n", t, arccos(t))
end do;
arccos(0.000000) = 1.570796
arccos(0.100000) = 1.470629
arccos(0.200000) = 1.369438
arccos(0.300000) = 1.266104
arccos(0.400000) = 1.159279
arccos(0.500000) = 1.047198
arccos(0.600000) = 0.927295
arccos(0.700000) = 0.795399
arccos(0.800000) = 0.643501
arccos(0.900000) = 0.451027
arccos(1.000000) = 0.000000
Accumulation: Variables within the
loop accumulate results. Each repetition
incrementally changes onto the partial result.
When the repetition is Iinished, the entire
result has been computed.
x := .3: #Parameter
prod := 1: #Accumulation variable
#"sum" and "product" are reserved Maple names
#So we should use something else like "s" or "prod"
#for variables.
#Compute the product of (x+2)*(x+3)*(x+4)*...(x+15)
for i by 2 to 15 do
prod := (x+i)*prod;
end do: #Suppress execution trace inside the loop

print(prod);
(15.4)
The loop is repeated while a continuation
condition is true.
(The loop stops when the continuation
condition is no longer true.)
This computes an approximation to

15.5 Constructing iterations and the roles oI variables 305
.

#script parameters
x := .2:
tol := 10^(-6):
#accumulation variables, initialized
i := 1:
s := 1: #First term of sum
term := 1: #Next term of sum
while (abs(term)>=tol) do
term := (x^(i))/(i!);
s := s + term;
i := i+1;
end do:
printf("compare %e to %e\n", s, exp(x));
compare 1.221403e+00 to 1.221403e+00
15.5Constructing iterations and the roles of variables
Setting up iterations is one oI the most prized skills Ior a programmer, and one oI the most diIIicult Ior beginners to learn. To construct an
iteration Irom scratch, we recommend the Iollowing process:
Without trying to create an iteration, just write down the series oI directions Ior the computational process you want to program. II you keep on
writing down steps, eventually you will notice similarities between actions in the series. OIten times the similarity allows you to discover that
rather than giving a new variable name to every result you compute, you can reuse certain variables because you no longer need them beyond a
certain point in the series.
Finding the pattern is key. Once you have it, then the repeating pattern becomes the loop body. Steps beIore the repetition kicks in become
initiali:ation beIore the loop. Usually there are summary actions (e.g. making a plot or animation, oI all the results computed by the loop) which
become part oI the post-loop actions.
Only through experience do you get to the point where you can construct loops without explicitly Iinding the repetition pattern is. Don't expect it
to come without practice and thoughtIul reIlection on experience.
In trying to construct a code Ior a successIul loop, it can be good practice to decide upon the role oI each variable you are using. II you are
unclear about a variable's role or purpose, it may mean that the variable is superIluous. It could also mean that you have skipped over a key
computational action which you need to develop code Ior. Here are typical roles variables can play in a loop:
Roles
Jariables that are fixed values. They are assigned during the initialization but don't change aIterwards.
Jariables that are steppers. Stepper variables change value methodically with each iteration oI the loop.
Jariables that gather the final result. Each repetition oI the loop leaves the accumulation variable one step closer to its Iinal
objective (e.g. adding together all the terms oI a sum).
Jariables that hold the most recent value going through a succession. Usually (but not always) the most recent value is
computed Irom previous values.
306 15 More on looping
Jariables that hold the previous value. Sometimes in order to compute the next value, it is necessary to keep not only the present
value, but the one before that. Since a variable can hold only one value at a time, an extra variable needs to be used to play that
role. The examples given so Iar do not have any variables with this role, but we will soon see situations where this is needed.
Consider Ior example this starter code Irom Lab 2, that computed chemical concentrations
Lab 3 CS 122 Chemical problem starter code
#Lab 2 Problem 2.1
restart; #just in case there are unwanted assignments hanging around from
#previous runs.
# Chemical reaction modelled by Lotka-Volterra Model
#Discrete approximation
numTimeSteps := 50;
#Initial concentrations
A0 := 1000;
B0 := 0;
X0 := 10;
Y0 := 50;
#Chemical constants
k1 := .001;
k2 := .001;
k3 := .1;
Atab := table();
indexTab := table();
A := A0; B := B0; X := X0; Y := Y0;
for i from 1 to numTimeSteps do
#Calculate next values of chemical concentrations
newA := A - k1*A*X;
# newX :=
# newY :=
# newB :=
#Update all four chemicals
A := newA; #B, X, Y updated similarly.
indexTab[i] := i;
Atab[i] := A; #Each entry is a (time, concentration)
end do:
with(plots): with(plottools):
#Given a table of x coordinates and a table of y coordinates,
#Create a point plot.
Fplot := (itab,vtab,COLOR) ->
plot(convert(itab,list),convert(vtab,list), color=COLOR, symbol=diamond, style=point);
#Display this point plot, along with a title and labels.
display([Fplot(indexTab,Atab,"Green")], title="graph of A", labels=["time", "concentration"]);
15.5 Constructing iterations and the roles oI variables 307
The analysis oI the roles oI variables in this code is as Iollows:
Roles Example in CS 122 Lab 3 Problem 2.1
Jariables that are fixed values. They are procedure parameters
or are assigned during the initialization but don't change
aIterwards.
k1, k2, k3 are Iixed values. So are A0, B0, X0, and Y0. So is
numTimeSteps.
Jariables that are steppers. Stepper variables change value
methodically with each iteration oI the loop.
i is a stepper variable. It begins at 1. One is added onto its
value each trip through the loop, so it steps through the values
1, 2, 3, ....
Jariables that gather the final result.
Each repetition oI the loop leaves the accumulation variable
one step closer to its Iinal objective (e.g. adding together all the
terms oI a sum).
While there are no variables that sum or count in this example,
indexTab and Atab are tables that store one value per iteration.
The tables gather the whole collection oI values generated
during the simulation, which are needed to produce the plot
which is the Iinal result.
Jariables that hold the most recent value going through a
succession. Usually (but not always) the most recent value is
computed Irom previous values.
newA is holds the most recent value oI the concentration oI
chemical A. newB, newX, and newY also take on this kind oI
role.
Jariables that hold the previous value. Sometimes in order to
compute the next value, it is necessary to keep not only the
present value, but the one before that. Since a variable can hold
only one value at a time, an extra variable needs to be used
to play that role. The examples given so Iar do not have any
variables with this role, but we will soon see situations where
this is needed.
At the beginning oI the loop, A (and as the code was
developed, B, X, and Y) took on the role oI the previous value.
It was used to compute newA, newB, etc. At the end oI the loop
body, A is set to the most recent value in preparation Ior the
next iteration oI the loop when those values will once again
become "previous".
Here's a classic example oI a task accomplished through a loop: Iinding the smallest value oI a list oI numbers
An example using a variable that holds the previous value
#We want to find the smallest value in a list of numbers
L := [3, 6, -4, 2, 1, 9];
#minSoFar's role is an accumulator variable:
# it keeps track of the "smallest number so far". We initialize it to infinity
# so that whatever we see first will automatically become the smallest so far.
minSoFar := infinity;
#Step through all the positions of L: 1, 2, 3... 6
for i from 1 to nops(L) do
printf("Comparing %a to %d\n", minSoFar, L[i]);

#If the previously seen minimum is larger than L[i]....
if minSoFar>L[i]
#Make it the new previously computed minimum
then minSoFar := L[i];
end if;
end do;
308 15 More on looping
printf("The minimum is: %a.", minSoFar);
Comparing infinity to 3
Comparing 3 to 6
Comparing 3 to -4
Comparing -4 to 2
Comparing -4 to 1
Comparing -4 to 9
The minimum is: -4.
The analysis oI the roles oI variables in this code is as Iollows:
Roles Example in CS 122 Lab 2 Problem 2.1
Jariables that are fixed values. They are assigned during the
initialization but don't change aIterwards.
L is the only Iixed variable.
Jariables that are steppers. Stepper variables change value
methodically with each iteration oI the loop.
i is a stepper variable. It begins at 1. One is added onto its
value each trip through the loop, so it steps through the values
1, 2, 3, ....
Jariables that gather the final result.
Each repetition oI the loop leaves the accumulation variable
one step closer to its Iinal objective (e.g. adding together all the
terms oI a sum).
There aren't any "gatherers" in this short bit oI code.
Jariables that hold the most recent value going through a
succession. Usually (but not always) the most recent value is
computed Irom previous values.
minSoFar hold the most recent value oI the minimum value
seen in L so Iar. BeIore the loop starts, it is initialized at
infinity, which guarantees that aIter the Iirst element oI L
is inspected it will become the value oI minSoFar. Note that
infinity will be printed as the minimum at the end oI the
code iI L : | | (the list with nothing in it).
Jariables that hold the previous value. Sometimes in order to
compute the next value, it is necessary to keep not only the
present value, but the one before that. Since a variable can hold
only one value at a time, an extra variable needs to be used
to play that role. The examples given so Iar do not have any
variables with this role, but we will soon see situations where
this is needed.
minSoFar also plays this role too. As the i-th execution oI the
loop body begins, the value oI minSoFar is the minimum oI
L|1| through L|i-1| -- the minimum computed Irom all the
previous repetitions oI the loop. As the i-th execution oI the
loop body ends, minSoFar has changed to be the minimum oI
L|1| through L|i|. The code works because iI you know the
minimum oI a collection oI numbers, and add a new number to
the collection, the overall minimum is the smaller oI the new
number and the minimum oI all the other numbers.
15.6lmplicit looping: seq, list addition
In chapter 12, we learned about for and its ability to loop with an index (stepping) variable. While this is an abbreviation over the pattern
as implemented with while that we saw in chapter 11, it still takes work to set up. Certain repetitious actions occur so commonly that many
programming languages provided even shorter ways oI perIorming looping idioms.
For example, suppose we wanted to build a list consisting oI the numbers 1,4,9, 25, ... 169. We could type such a list in, but Irom what we know
oI looping we suspect that there's an easier way. Using the ideas we've seen so Iar, we could use a table valueTab that has an accumulation
15.6 Implicit looping: seq, list addition 309
variable L. With i as the stepping variable going Irom 1 to 100, we would Iill up the table with values. AIter the loop Iinishes, we could
convert the table into a list to get our result:
Making a list through a loop

#initialize table
valueTab := table();
#fill up the table
for i from 1 to 13 do
valueTab[i] := i^2;
end do;
#Produce list as result
L := convert(valueTab,list):
printf("L is: %a", L);
L is: [1, 4, 9, 16, 25, 36, 49, 64, 81, 100, 121, 144, 169]
The nice thing about this loop is that it is no more work Ior us to create a list with 100 items 1, 4, 9, ... 10,000. All we have to do is change the
"13".
However, there is an even shorter way to create this list. Maple has a Iunction seq whose syntax borrows heavily Irom for. The seq Iunction
creates a sequence as indicated by the values oI the stepping variable. To get a list, we convert the sequence into a list, as explained in Chapter 12.
Making a list with seq
310 15 More on looping

#initialize table
#Create a sequence
s := seq(i^2,i=1..13);
#Convert it to a list
L := [s];
printf("L is: %a", L);
L is: [1, 4, 9, 16, 25, 36, 49, 64, 81, 100, 121, 144, 169]

With this kind oI Iacility, it is easy to quickly solve many problems that are solved with Maple Iunctions that take lists.
Creating a list of values with seq
Create a list oI the approximations to , , ... . We chain together the sequence creation and the conversion to a list
in one line oI code.

(15.8)

The general Iorms oI seq are
seq( expression involving steppingJariable, steppingJariablerange) and
seq( expression involving steppingJariable, steppingJariablerange, stepJalue).
The second Iorm behaves like the for... from... to ... by ...do...end do described in the previous chapter.
Giving lists to Maple functions using seq with a step value
Create a list oI the approximations to , , ... .

(15.9)
Adding together the elements oI two lists is another common operation that has a built-in shortcut.
Adding together two lists with +
15.6 Implicit looping: seq, list addition 311
Create a list oI values oI (approximately).
(15.10)
Create a similar list oI values Ior
(15.11)
Add them together using "".
(15.12)


We don't get component-wise "automatic multiplication" with "*". Maple doesn't give an error, but it doesn't do the
multiplication, either.
(15.13)

We get this done without writing a loop to do it with the use oI seq.

(15.14)
"Automatic addition" works with numbers, expressions, lists, sequences, ranges, and equations, but does not work with tables, plots, or sets.
"Automatic multiplication" only works with numbers and expressions. Maple does have additional data containers with use with linear algebra,
vectors, and matrices Ior which both addition and multiplication work -- but that should wait until you need to do linear algebra.
312 15 More on looping
15.7An example that uses for...while... looping and
display(..., insequence=true) for animation
Problem: a little old lady deposits some money into a savings account, which pays an interest rate of r paid yearly. She leaves it in
the account for numYears years. However, she has a targetAmount in savings in mind, so she will stop as sooner if the target amount is
reached. Show the growth of her money over time for a variety of interest rates.
interestCalculator6 := proc (initialAmt, loRate, hiRate, numYears,
targetAmt)
local i, A, newA, yearTab, amtTab, rate,
frame, frameIndex;

#Initialize frame (a table of plots) and frameIndex (a counter)
frame := table(); #A table of graphs
frameIndex := 0;

#Try for a variety of rates

for rate from loRate by .001 to hiRate do
#initialize A, and tables.
A, yearTab, amtTab := initialAmt, table(), table();
#initialize first entry of table before looping
yearTab[0] := 0;
amtTab[0] := initialAmt;

#Calculate new amount based on present amt
#and interest rate.
for i from 1 to numYears while amtTab[i-1] <=targetAmt do
amtTab[i] := amtTab[i-1] + rate*amtTab[i-1];
#Remember this year's info.
yearTab[i] := i;
end do; #loop with years

#Store a plot into the frame table.
frame[frameIndex] := plot(convert(yearTab,list),
convert(amtTab,list),
style=POINT, color="Red",
labels=["years", "dollars"],
title=sprintf("Growth of a $%5.2f deposit at %4.2f percent
interest",initialAmt, rate*100));
frameIndex := frameIndex +1; #update frameIndex (counter)
end do; #loop with rates

#Create an animation with all the graphs together
plots[display](convert(frame,list),insequence=true);
end proc:
The program is written as a
nested loop. For each value oI
the interest rate, we go through
the calculation described
in interestCalculator5 oI
the previous chapter. The
inner loop rather than being
a straight repetition with i
taking on all the values Irom
1 to numYears, will stop early
iI the target amount is ever
reached early.

Recall Irom Chapter XX
that the op operation is a
way oI selecting parts oI
an expression. II rateRange
is the range (.5)..(.6).
op(1,rateRange) is .5 and
op(2,rateRange) is .6.
15.7 An example that uses Ior...while... looping and display(..., insequencetrue) Ior animation 313


This produces an animation
Ior various rates oI interest.
We show the Iirst Irame oI the
animation here. As described
in Chapter 6, the animation
player can be used to play the
animation.




We show a Iew Irames Irom
the animation here. For the
last Iew Irames, the amount
on deposit eventually exceeds
targetAmt, so no data points
are displayed above that level.
This is because the Ior...while
loop in the simulation stopped
repeating without reaching
numYears because the while
condition became true beIore
the limit oI the for repetition
was reached.

The animation shows a series
oI plots. The idea is "the
314 15 More on looping
Frame 1 Frame 8 Frame 29

interest rate is higher as time
elapses in the animation".
It is not literally showing a
process that changes over time
since a Irame oI the animation
shows how the deposit amount
increases over time.


15.8Summary of Chapter 15
Animations through plots[displav](..., insequencetrue)

To create an animation Irom a list oI plot structures, do

with(plots):
display(list of plots, insequencetrue);

or

plots|display|(list of plots, insequencetrue);

Either will create an animation where each plot in the list is a single Irame oI the animation.

Plot options given as additional arguments will aIIect all the Irames. For example,

plots|display|(list of plots, insequencetrue, axesnone);
15.8 Summary oI Chapter 15 315

will remove the horizontal and vertical axes Irom all Irames oI the animation.


Creating a sequence through seq
seq( expr, varlo..hi) will create a sequence using the index variable var looping through the values lo, lo1, lo 2, ... hi.

Example:

(15.15)

Adding all the elements oI two lists together, element-wise
(15.16)
(15.17)
(15.18)
Note: list subtraction works in the same way, but list multiplication does not.
A single loop with both for and while control
Maple's syntax allows the use oI an stepper variable and a continuation condition (other than that oI the stepper variable
reaching a Iinal value.

for stepperJariable from initialJalue to finalJalue by stepJalue while condition do .... end do
promotes a repetition which stops when either the stepperVariable reaches the finalJalue, or the condition becomes Ialse. II the
to initialJalue is omitted, then the while condition is the sole criterion used to decide whether to continue the repetition. This
can abbreviate the coding in many standard cases.

Example
316 15 More on looping

restart;
sumIt := proc(x,tol)
local term, s, i; #Variables that will be used within the procedure.
# Initialize script parameters
#tol is the size of the stopping criterion.
#x is the value of sin(x) that we are trying to approximate
#Initialize accumulation variables to first term of sum.
term := x;
s := x; #s accumulates the sum, we initialize it to the first term.
#add on successive terms
for i from 3 by 2 while abs(term)>=tol do
term := (-1)^((i-1)/2)*(x^i)/i!;
printf("term is %e\n", term);
s := s+term;
end do;
end proc;
#Test 1
x := .3;
tol := 10^(-2); #want at least two decimal digits to be stable.
#Print out the final sum, and compare it to what it's supposed to be.
printf("Test 1, sum is: %e compared to %e.",sumIt(x,tol), sin(x));
term is -4.500000e-03
Test 1, sum is: 2.955000e-01 compared to 2.955202e-01.

15.8 Summary oI Chapter 15 317
Variable types
Fixed value These variables are set beIore the loop begins, and do not
change while the loop is being executed.
Steppers These variables change some Iixed amount every time a loop
iterates. They hold the current step a loop is on. For example,
the index variable in a Ior loop is a stepper variable.
Variables that gather the Iinal result These variables accumulate a result that is returned at the end
oI a loop.
Variables that hold the most recent value in a loop These variables hold some value that was computed during
the current iteration oI a loop. They are oIten compared to a
previous value to compute a change.
Variables that hold a previous value in a loop These variables hold some value computed in the previous
iteration oI a loop.
318 15 More on looping
319
16Calculus and optimization
16.1Chapter Overview
One oI the most distinctive Ieatures oI Maple is its ability to do mixed computations where the results are a mixture oI Iormulas and numbers.
We introduce some oI the built-in Iunctions Ior doing calculus: diIIerentiation, limits, and Iinding minima and maxima. Like solve, the minima
and maxima-Iinders come in two styles -- one that tries to calculate Iormulae Ior answers (exact calculation) and one that uses approximation
techniques that work only when the answer is numeric.
16.2Learning about additional kinds of mathematical
computations
We have become accustomed to looking at problems that can be attacked by using some combination oI solve, plot (and other visualization
tools), and numerical evaluation/assignment. Contrasting this with the kind oI calculation available Irom today's calculators, we observe that it is
diIIerent qualitatively Irom the calculation provided by a calculator because
a) Programming (e.g. looping) allows us to do much larger quantities oI calculation than we could through operations only through keystroke.
b) Programming (procedures and scripts) allows us to more easily reuse calculations on similar but diIIerent problems.
c) The document interIace allows us to record the results and commentary, Ior Iuture reIerence and/or use in presentations and reports.
d) PowerIul operations such as solve make solving certain types oI problems much more convenient, even with the work to needed to enter the
expression into the computer.
e) PowerIul visualization and animation Ieatures available on a computer enhance understanding oI the phenomena being studied.
In addition to programmability and a more extensive user interIace, computers in 2012 have another edge over calculators -- they can
conveniently access much more library programming because they have access to more Iiles and more interIace Ilexibility. Systems such as
Maple, Mathematica and Matlab use this room to provide thousands of extra mathematical features. Some oI it is immediately useIul to Iirst
year university/college students in the Iorm oI svmbolic computation -- results Irom calculus. Some oI it involves operations and Iunctions that
you will hear about aIter the Iirst year -- linear algebra, multivariate calculus, ordinary and partial diIIerential equations, Laplace transIorms, etc.
Learning how to operate a system such as Maple on the basic Ieatures gives you the understanding to pick up many oI the more advanced Ieatures
on your own, aIter you learn the mathematics. Right now, we exploring some oI operations that are comprehensible with Iirst year mathematics.
However, no course could possible teach all the Ieatures needed in a career. Taking advantage oI systems with thousands oI Ieatures means
users must be able to learn on their own aIter their initial period oI training. The process is still the same -- using a combination oI written
documentation and interaction with more experienced users. However, no matter how good the support system is, there will always be gaps that
the user will have to Iill in, through thought and experimentation, on their own. It's work to do this, but the payoII is that you are Ireed Irom only
using the Ieatures you saw in a course.
16.3Differentiation, simplification
As it is taught in traditional Iirst-year calculus, diIIerentiation is an operation on Iunctions. Maple knows how to diIIerentiate all oI the common
Iunctions Iound in calculus. It is particularly useIul Ior perIorming diIIerentiation when it would take a lot oI algebraic manipulation to do the
operations by hand.
diff as a Iunction takes two or more arguments. The Iirst argument must be (or evaluate to) an expression such as or . The
second argument must be (or be an expression that evaluates to) the variable oI diIIerentiation. The there are third, Iourth, etc. arguments
provided, they are used as variables oI higher derivatives.
The result oI diff is an expression or a number iI the derivative is a numerical constant. Evaluation oI a derivative expression can occur through
eval as with other expressions.
320 16 Calculus and optimization
Table 16.1: Symbolic differentiation using the expression palette

Enter the expression, and then right-click (control-click
on Macintosh) to bring up the calculation options. Select
diIIerentiate-~x



We can calculate the second derivative with respect to t oI this
expression by perIorming diIIerentiation twice. We simpliIy
the expression a bit by perIorming the operation simpliIy-
~symbolic.

II we want the document to display the mathematical notation
Ior the derivative, we can select Irom the expression
palette and then Iill in the slots f and x. We can then get Maple
to calculate the derivative by typing control-.
Table 16.2: Symbolic differentiation and evaluation of derivatives, textually
Example Commentary
(16.1)
(16.2)

Find the Iirst derivative oI the expression with respect to x.





(16.3)
(16.4)

Find the Iirst derivative oI the expression with respect to t.

(16.5)

Find the second derivative oI the expression with respect to t.
(16.6)
diII's second argument must be the variable oI diIIerentiation.
Note that iI the variable doesn't occur in the expression, the
derivative (according to the mathematical deIinition) is zero. II
16.3 DiIIerentiation, simpliIication 321
you are surprised by getting a zero derivative, check that you
are using the correct variable.
(16.7)

simpliIy can reduce the size oI the expression, although
Iactor or expand sometimes work better. Sometimes
additional trigonometric identities need to be applied, through
simpliIy(...., trig).
(16.8)

This is a way to compute . Through
daisy-chaining, we Iirst calculate the derivative, then do the
evaluation.
(16.9)

This is a way to compute The result
oI evaluation here is a Iormula, not a number, since there were
two symbols, t and in the second derivative, but we only
evaluated one oI them, t.
Table 16.3: Plotting a function and its derivative together
Example Commentary
(16.10)




We're already used to Iunctions that take one or two arguments.

(16.11)
In this example, we plot a "strange" Iunction built out oI
trigonometric parts, and plot it and its derivative. Since g is a
Iunction g(t) will evaluate to the expression . Thus the plotting
variable should be t rather than or some other variable.
Recall Irom Chapter 3 that we can get the character u either
by using the Greek Palette. Another way oI getting u is to type
alpha; Maple will auto-transIorm it typographically.

322 16 Calculus and optimization
(16.12)

The value oI Ideriv is an expression involving alpha. In
programming, unlike mathematics, variable names oIten have
names that are more than one letter long. This is to increase
intelligibility to readers looking at the programming. Although
it seems minor, ease oI comprehension can play an signiIicant
role in the cost oI developing and using soItware, so is an
important engineering concern.
We plot two expressions together involving the variable on
the same plot. The use oI a set }as a Iirst argument to plot to
do multiple plots was Iirst explained in section XX.

We need to use as the plotting variable here since the value
oI fderiv is an expression involving .

16.3 DiIIerentiation, simpliIication 323
Table 16.4: Plotting a function and its derivative together
Example Commentary
(16.13)




We can use proc to deIine a Iunction that is mathematically
identical to f oI the previous example. x, and v are used as the
names oI the Iormal parameters here instead oI a and b, but
it still deIines the same Iunction. For "small mathematical
Iunctions" such as this one, we generally preIer the use oI
since it's shorter. The arrow notation can handle this because
the Iunction doesn't involve any use oI for, while, or if.
(16.14)

g is the same Iunction as beIore, since I and I2 deIine
equivalent Iunctions. We use s in the same way that was used
in the previous deIinition, and we've already mentioned that I2
is the same as I. So g2 is identical to g.
324 16 Calculus and optimization

We using t as the plotting variable since g2(t) and diII(g2(t),t)
are both expressions involving t.


16.4Limits
You can use the clickable interIace to compute limits by selecting the appropriate item Irom the Expression palette and then Iilling in the template
as needed. Maple uses calculus techniques (e.g. l'Hpital's Rule) to compute limits symbolically.
Table 16.5: Clickable interface version of limits
Example Commentary
(16.15)
The result oI limit can be an expression (possibly involving
positive or negative inIinity) as well as the symbol undefined.



(16.16)
(16.17)
Even though the limit exists as t approaches Irom the right or
leIt, they do not agree, so there is no "two-sided" limit.
16.5 Finding minima and maxima by using Maple as a calculus calculator 325
(16.18)

To take a one-sided limit, add a "" or "-" superscript to the
limit point.
(16.19)
Note that the value returned as the limit Ior xa, while true Ior
most values oI a, is not really valid Ior a0.
The textual version oI taking limits involves the limit Iunction. limit takes at least two arguments. The Iirst argument is the expression that you
wish to take the limit oI. The second argument is an equation indicating the limit variable and the limiting value. II you supply a third argument, it
indicates whether a "right sided", "leIt sided" limit is desired instead oI a two-sided limit.
Table 16.6: Textual version of limits
Example Commentary

(16.20)



The Iirst argument is the expression that you wish to take the
limit oI. The second argument is an equation indicating the
limit variable and the limiting value.
(16.21)
You can use "inIinity" or "-inIinity" as the way oI speciIying
or - textually without use oI the palette.
(16.22)
The third optional third argument to limit can speciIy a one
sided limit. This is the textual way oI speciIying .
There is no simple way oI speciIying this
(16.23)
This is the textual way oI speciIying .
It would be nice iI the second argument Ior the textual Iorm oI limit was var -~ limitJalue instead oI var limitJalue, since arrows are the
more conventional notation. But because oI the limitations oI Maple's processing capabilities Ior its programming language, it would be more
expensive to support arrows Ior both this meaning in limits and the use oI -~ in Iunction deIinitions. So users must get used to using equations
rather than the standard math symbol Ior "approaches".
16.5Finding minima and maxima by using Maple as a
calculus calculator
The minimum or maximum oI a continuous Iunction in an interval is sometimes reIerred to as the local extrema of the function. The Iunction
whose extremum you are Iinding is reIerred to as the obfective function. In elementary calculus, you learn how to Iind extrema through the use oI
derivatives and a little algebra.
326 16 Calculus and optimization
Maple can be used to do such calculations using diff to Iind the derivative, and then Iinding where the derivative is zero through solve or fsolve.
The advantages oI doing it with a system such as Maple are the same as with other calculations: a) it is easier to do iI the Iormula is involved or iI
you don't remember all the math, b) you can easily organize the inIormation about the problem and its solution using Maple word processing so
that the work can be easier to present or read, c) the computer's speed oI re-execution makes it easy to handle a bundle oI similar problems.
Table 16.7: Finding the minimum and maximum using textually-specified commands
From Anton, Calculus 8th ed. p. 307 problem 13 (modiIied)

Find the absolute maximum and minimum values oI

Ior x in


Solution

We deIine a continuous Iunction
(16.24)
and compute its derivative symbolically.
We generate a plot to get an idea oI where to look Ior minima
and maxima. Evidently the minimum is in the interior oI the
interval middle and the maximum is at the leIt.
16.5 Finding minima and maxima by using Maple as a calculus calculator 327
(16.25)
Finding where the derivative is zero should produce the
minimum.
(16.26)
The minimum is conIirmed to be zero -- the plot seems to
indicate that but the picture doesn't let us know whether it's
really zero or just really close to zero.
(16.27)

Standard calculus procedure is to check out the value oI the
end points, since the minimum might be there instead oI in the
interior oI the interval. However, the values oI f at both oI the
end points are larger than the value at .We conclude that f
attains a minimum at x .
Table 16.8: Writing a procedure to calculate minimum of a function and produce a plot that will help understanding of it
Writing a procedure to calculate minimum oI a Iunction and produce a plot that will help understanding oI it
Anton, Calculus, 8th ed. Problem 307 (modiIied)

Find the absolute minimum oI
Ior x in |-2..3|







We enter the expression, and plot it on -2..3 using the 2-D plot
builder.







328 16 Calculus and optimization





To develop the script, Iirst we do it interactively.
We do a chain oI calculations to Iind what is obviously the
minimum point.

We had to copy the expression again, and evaluate it at
to calculate the minimum value. Once we have a
number, we calculate what it is, approximately.

We omit calculating the value oI the objective expression at the
end points because the plot tells us that minimum is not located
there.

(16.28)
(16.29)
The next step is to develop the textual version oI this.

We turn the sequence oI actions into a textual script. The
parameters are the expression to be minimized, the variable
in the expression and the region. The plot is just to help us
have a visual check that the numeric answer result is probably
correct. Since it's something we decide we'd want to do with
16.5 Finding minima and maxima by using Maple as a calculus calculator 329
(16.30)
(16.31)
(16.32)
(16.33)
(16.34)

any version oI a minimization problem, we decide to include it
in the script.

findMin := proc(expr,var, interval)
local deriv, soln, valAtPoint;
print(plot(expr,var=interval));
We turn this into a procedure that does the same thing,
returning the minimum value and its location in a list. Unless
we print the plot, it won't appear during execution oI the
procedure because by deIault Maple only displays the result oI
330 16 Calculus and optimization
deriv := diff(expr,var);
soln := solve(deriv,var);
valAtPoint := eval(expr, var=soln);
return [evalf(valAtPoint), var=soln];
end;
(16.35)
the procedure. print causes something to be printed in addition
to the Iinal result.


(16.36)

We solve the minimization problem.
16.6 One-step extrema-Iinding using maximize and minimize 331
(16.37)

Now that we have a procedure, we can solve a similar problem:
minimize Ior v in the interval -1..4. We see that
the plot supports the numerical calculation that the value oI the
expression at the minimum is approximately 1.65 at v .
16.6One-step extrema-finding using maximize and
minimize
Maple's library provides both exact and approximate numerical Iunctions that do the entire sequence oI steps needed to Iind extrema. In this
section, we discuss the Maple Iunctions that provide exact solutions to such problems. The exact version automates the work that the user would
have to do with diff and solve. The approximate version will provide usable inIormation in situations where exact solution methods don't work.
The Maple Iunction Ior Iinding maxima is, appropriately enough, called maximi:e. II expr is an expression involving a variable x, then
maximi:e(expr,xrange) will produce maximum value oI the expression Ior x in that range. II there is no maximum value, or iI Maple can't Iind
it, NULL is returned. Thus iI maximi:e returns an answer, it should be "believable", but a result oI NULL doesn't meant that there is no minimum.
It may just mean that Maple was unable to Iind it.. infinitv or -infinitv may be an answer iI the Iunction is unbounded within the range.
We have already seen max and min, which are also built-in Maple Iunctions. However, they only work on lists, sets oI values. They don't work at
Iinding minima or maxima oI expressions.
It is oIten a good idea when working on extrema problems to do a plot oI the expression in question, so that you can some notion oI where the
maxima will be and what their values will be like.
Table 16.9: minimize and maximize
minimize and maximize
From Anton, Calculus 8th ed. p. 307 problem 13 (modiIied)

Find the absolute maximum and minimum values oI



332 16 Calculus and optimization
minimize and maximize

Ior x in








Solution

(16.38)
We deIine a Iunction.
We generate a plot to get an idea oI where to look Ior minima
and maxima. Evidently the minimum is in the middle and the
maximum is at the leIt.
(16.39)
The minimum is conIirmed to be zero -- the plot seems to
indicate that but the picture doesn't let us know whether it's
really zero or just really small. Note again that this is the
minimum value, it is not the location oI the minimum value.
16.6 One-step extrema-Iinding using maximize and minimize 333
minimize and maximize
(16.40)
Finding the maximum is a copy/edit job once we've worked out
how to do the minimum.
Sometimes, you want not only what the maximum or minimum value is, but also where it is. II you give maximi:e or minimi:e an extra third
parameter, location, then it will return a sequence Ior a result. This sequence has a logical but somewhat intricate structure. You can use a chain oI
selections to extract the location var location point Irom the answer given by Maple.
Table 16.10: Solving a problem with minimize and maximize
Solving a problem with minimize and maximize
From Anton, Calculus 8th ed. p. 307 problem 13 (modiIied)

Find the absolute maximum and minimum values oI

Ior x in



Solution

(16.41)
We deIine a Iunction.
334 16 Calculus and optimization
Solving a problem with minimize and maximize
We generate a plot to get an idea oI where to look Ior minima
and maxima. Evidently the minimum is in the middle and the
maximum is at the leIt.
(16.42)
We get the smallest value oI I in the interval.
(16.43)
(16.44)
(16.45)
(16.46)
This is what minimize does with the extra parameter location.
The result returned by minimi:e with this extra parameter
contains not only the value oI the minimum, but all the values
oI x where the minimum occurs. It takes a lengthy chain oI
part-extraction operations to get the value out.
16.6 One-step extrema-Iinding using maximize and minimize 335
Solving a problem with minimize and maximize
(16.47)
(16.48)


(16.49)
(16.50)


However, once we have Iigured it out, we can get the same
result by daisy-chaining the operations together.

(16.51)

A copy/edit job will produce the similar sequence oI operations
Ior the maximum.
(16.52)
(16.53)

Finding the maximum is a copy/edit job once we've worked out
how to do the minimum.
Table 16.11: Using optimization in a chemical production problem with minimize and maximize
From Anton, Calculus 8th ed. p. 316-317 Example 7.

A liquid Iorm oI penicillin manuIactured by a pharmaceutical
Iirm is sold in bulk at a price oI $200 per unit. II the total
production cost (in dollars) Ior x units is


336 16 Calculus and optimization


and iI the production capacity oI the Iirm is at most 30,000
units in a speciIied time, how many units oI penicillin must
be manuIactured and sold in that time to maximize the proIit?
What will the revenue be at the maximal proIit?
(16.54)

Solution

(16.55)
(16.56)
(16.57)
We deIine a Cost, Revenue, and ProIit Iunction, using the
inIormation given in the problem about cost, and the standard
deIinitions Irom economics Ior revenue and proIit. We
could use the proc.. end proc way oI deIining a Iunction to
equal eIIect but the arrow notation was designed to make it
convenient to enter these short deIinitions that don't need if or
for.



We plot the proIit Iunction to scope out its behavior. Evidently
the Iunction does hit a maximum, roughly around x20000. II
we got x35 as the answer Irom our maximum calculation in
Maple, we'd wonder iI we did things correctly.


16.7 One-step approximations to extrema using Optimization|Maximize| and Optimization|Minimize| 337

(16.58)

(16.59)


We are getting Iloating point answers Irom Maple because
some oI the numbers in C are Iloating point. II we wanted an
exact calculation, we would change the .003 into 3/1000.

We use the daisy-chaining oI operations we saw in the previous
example to extract the value oI the x where the maximum is
obtained. We use the floor Iunction which rounds down to
the nearest integer. This is the way to guarantee that we have
a result that is a whole integer, without Iractional part. ,The
chemical process can't produce a Iraction oI a unit.
(16.60)
(16.61)
(16.62)
We used the clickable menu item "Numeric Formatting" (see
Ior example the "A script that uses Iunctional composition"
example in Chapter 8) to reIormat the result in terms oI
currency.

16.7One-step approximations to extrema using
Optimization]Maximize| and Optimization]Minimize|
Maple's solve and fsolve provide exact and approximate solutions. In a similar Iashion, maximi:e and Optimi:ation[Maximi:e], and minimi:e and
Optimi:ation[Minimi:e] provide exact and approximate solution to extrema problems.. Optimi:ation[Maximi:e] and Optimi:ation[Minimi:e]
only work when there is a numerical answer. They use only numerical, not analytical knowledge about Iunctions. They produce a "best eIIort"
approximation.
The clickable interIace's "Optimization" operation invokes these approximate optimizers.
Optimi:ation[Maximi:e] and Optimi:ation[Minimi:e] always return a list. The Iirst item oI the list is the extreme value, and the second item
in the list is a list oI equations describing the location oI the extreme value. As with the exact optimizing Iunctions maximi:e and minimi:e, the
second item can be used as a parameter to eval to evaluate an expression at the location oI an extreme value.
While the inventiveness oI numerical analysts make such approximation techniques usually reliable, they can be Iooled and without warning
produce a bad approximation. For that reason, the results oI fsolve, Maximi:e and Minimi:e should be checked against what else is known about
the properties oI the actual solution. For example, by plugging the location oI the approximate maximum into the objective Iunction, does it really
produce a value that is larger than any other value oI the objective Iunction that you can think oI? OIten times a plot makes this easy to tell.
Table 16.12: Finding an approximation to the minimum with OptimizationMinimize]
Anton, Calculus 8th ed. problem 47, p. 321 (modiIied)

Two particles, A and B, are in motion in the xv-plane. Their
coordinates at each instant oI time t are given by




338 16 Calculus and optimization
, , , . Find the minimum distance
between A and B.

(16.63)
(16.64)
(16.65)
(16.66)










(16.67)
(16.68)
(16.69)
(16.70)


p1 and p2 are lists oI the x and y positions Ior each point.

dist is the objective Iunction. This is the Euclidean distance
between the two points. We can compute the distance between
p1 and p2 using it.

16.7 One-step approximations to extrema using Optimization|Maximize| and Optimization|Minimize| 339
It's always good to have an independent way oI evaluating the
answer you get Irom a "solver". In this case, we can plot the
Iunction and Irom that see approximately where the answer
ought to be. Our Iirst plot does not show the minimum that
clearly, so we do another plot that emphasizes the region oI
interest.
(16.71)
Using with, we can then say Minimize rather than
Optimization|Minimize|.

340 16 Calculus and optimization
(16.72)

The exact location is t 0.4, but the rounding error
made during the operation oI Minimize leads to only an
approximation oI the exact answer. This is typical.
(16.73)
(16.74)
(16.75)
According to the structure returned by this Iunction, the
location inIormation is in the second element oI the result.

We can Iind the location oI particles A and B at t0.4 by
evaluating the expression p1 and p2 at that value oI t. Note that
minLocation is a list consisting oI an equation, which eval can
use as its second argument.
(16.76)
(16.77)

This is a problem that Maple's exact solver can solve
exactly. The numerical approximation produced by
Optimization|Minimize| is in close agreement with the exact
result.
(16.78)
This expression has a complicated expression Ior the exact
minimum. We could try to approximate it using the right-click-
~approximate menu, or we could give it to the approximate
minimizer instead.
16.7 One-step approximations to extrema using Optimization|Maximize| and Optimization|Minimize| 341
(16.79)
(16.80)
(16.81)


Table 16.13: An example where the approximate minimizer gets fooled
(16.82)

While the approximate solvers will probably Iare well on
most Iunctions that have only a Iew extrema, they can be
Iooled iI the Iunction has a lot oI extrema close to each other.
In that case, it may return something that is an extremum in
a small neighborhood around the point, but not iI a larger
interval is considered. This "tricky" example is Irom Cheney
and Kincaid's textbook on numerical analysis, Numerical
342 16 Calculus and optimization
mathematics and computing, 6th edition, 2008 Thomson
Brooks/Cole, p.626.
(16.83)

We get an answer Irom approximation technique. There are no
error or warning messages. To check this answer, let's do a plot
and see what the Iunction looks like.

We see why an approximate solver might have problems --
there are lots oI "local minima". x.56 does not seem to be a
true minimum Ior in the interval. We can draw a zoomed in
plot and look closer:
16.7 One-step approximations to extrema using Optimization|Maximize| and Optimization|Minimize| 343

It's clear now that x0.56 is not the location oI the minimum.
x's value should be about x-0.02 and the minimum value is
about -1. These approximate coordinates were Iound using the
cursor positioning Ieatures described in Table XX.
(16.84)
(16.85)


Applying the exact minimizer to this problem produces a hard
to understand result that takes much more time to compute
than the approximate technique. Nevertheless, it is the true
minimum in the interval. We ask evalf to approximate it. That
produces something that, while still approximate, agrees more
with what the plot leads us to expect Ior the right answer.

The moral: numerical optimization typically produces results
quickly but even the best can be Iooled. You need to veriIy
the results iI the cost oI making an error is high. We might be
tempted to see iI we can change the circumstances so that the
objective Iunction we work with is not so "squirrelly".
344 16 Calculus and optimization
16.8Multivariate operations in Maple
diff , minimi:e, solve, and the rest can handle problems and situations where there are multiple variables involved. Approximation methods (e.g.
fsolve or Optimization|Maximize| oIten take more prominence with multivariate problems because oI the diIIiculty and/or expense oI Iinding
oI succinct Iormulas that express the solution. We won't explore Maple's abilities to handle multivariate problems here, but they are there to be
explored and used when you're ready Ior it.
Table 16.14: Multivariate operations in Maple
(16.86)


Compute the second partial derivative with respect to x.

(16.87)

The textual way oI entering partial derivatives.
(16.88)
(16.89)

We are interested in Iinding the points oI intersection with the
unit circle with center at (0,0), and the line deIined by xy5.
Doing it exactly gives an answer, but it's not easy to understand
-- x and y can be either the positive or negative solution to a
quadratic expression. Introducing a Iloating point number into
the equation (1.25 instead 10/8) tells solve that it's all right to
approximate the results.
16.8 Multivariate operations in Maple 345
(16.90)
(16.91)

(16.92)
Warning, computation interrupted
(16.93)


We want to Iind the minimum value oI an expression in two
variables. minimi:e takes a long time to work and doesn't seem
to come up with much. However, the approximate minimizer
Iinds a value oI 5.
Multivariate plotting is handled by the plot3d Iunction. By
clicking on the plot window, the point oI view can be moved or
rotated. Animations, etc. can be built in a Iashion similar to two
dimensional plots.
346 16 Calculus and optimization

16.9Chapter summary
Symbolic differentiation using the expression palette, right-
click, control-.


Enter the expression, and then right-click (control-click
on Macintosh) to bring up the calculation options. Select
diIIerentiate-~x



We can calculate the second derivative with respect to t oI this
expression by perIorming diIIerentiation twice. We simpliIy
the expression a bit by perIorming the operation simpliIy-
~symbolic.

II we want the document to display the mathematical notation
Ior the derivative, we can select Irom the expression
palette and then Iill in the slots f and x. We can then get Maple
to calculate the derivative by typing control-.
Symbolic differentiation and evaluation of derivatives,
textually

Example Commentary
(16.94)
(16.95)
(16.96)
(16.97)
(16.98)




Find the Iirst derivative oI the expression with respect to x.




Find the Iirst derivative oI the expression with respect to t.

Find the second derivative oI the expression with respect to t.
The result is the same as iI we had taken the derivative oI 1.2.4.

diffs second argument must be the variable oI diIIerentiation.
Note that iI the variable doesn't occur in the expression, the
16.9 Chapter summary 347
(16.99)









(16.100)



(16.101)


(16.102)
derivative (according to the mathematical deIinition) is zero. II
you are surprised by getting a zero derivative, check that you
are using the correct variable to diIIerentiate with respect to.



simplifv can reduce the size oI the expression, although factor
or expand sometimes work better. Sometimes additional
trigonometric identities need to be applied, through simplifv(....,
trig).


This is a way to compute .

This is a way to compute The
result oI evaluation does not have to be a number even iI a
numeric value is being supplied Ior one oI the variables in the
expression being evaluated.
Textual version of limits
Example Commentary

(16.103)


The Iirst argument is the expression that you wish to take the
limit oI. The second argument is an equation indicating the
limit variable and the limiting value.

You can use "inIinity" or "-inIinity" as the way oI speciIying
or - textually without use oI the palette.
348 16 Calculus and optimization

(16.104)


(16.105)





(16.106)




The third optional third argument to limit can speciIy a one
sided limit. This is the textual way oI speciIying .
There is no simple way oI speciIying this

This is the textual way oI speciIying .
Minima and maxima
Operation Example
Exact minimum or maximum value
Ior x in


Solution

(16.107)

It's usually a good idea to plot the objective Iunction to
get some sense oI where the extrema are. Remember that the Iunction to be
16.9 Chapter summary 349
investigated should be continuous iI you expect good results Irom calculus
techniques.

(16.108)
(16.109)

Exact minimum or maximum value location
(16.110)
(16.111)

Evaluation oI an Iunction at an extremum point
Evaluate g at the location of the smallest value of f.
(16.112)
(16.113)
(16.114)

Evaluation oI an expression at an extremum
point
(16.115)
350 16 Calculus and optimization
(16.116)
(16.117)

Approximate minimum or maximum value and
location
(16.118)
(16.119)
(16.120)
(16.121)
Remember that numerical methods can be fooled into returning answers
that are seriously incorrect. Verify your computed results using other
techniques. Plots can often indicate that something is weird, although
even that is not foolproof.
More mathematical modeling: area-finding, piecewise
expressions, and splines
Chapter Overview
We introduce three more computational features from Maple's built-in library:
Computing definite and indefinite integrals
Piecewise expressions
Curve fitting using splines
Many scientific or engineering problems benefit from the use of these mathematical concepts.
Recalling basic facts about integration
We review some of the basic facts and terminology about integration from your calculus courses.
Definite integration can be related to finding the area under a curve described by a function. The
definite integral
a
b
f x dx
is notation for this. If an answer for a definite integration problem exists, then it is a number which
could be described exactly or as an approximation.
Indefinite integration is sometimes called the problem of antidifferentiation: given a function f,
find another function F whose derivative is f:
d
dx
F x = f x .
Another way of writing this relationship is:
f x dx = F x .
In most cases, there is no single function F(x) that solves the antiderivative problem for f(x). There
is a whole family of functions that differ from each other by a constant. Sometimes the
antiderivative F(x) is written with an extra symbolic constant: f x dx = F x CC, particularly if
antidifferentiation is going to be applied as only one step of a problem-solving process. Since the
hard part of doing indefinite integration is finding F(x) without the constant and since the name of
the constant can be freely chosen, systems that can compute antiderivatives often omit the constant
in their answers, letting the user add in the constant of their choice later if they want to.
The answer F(x) is expected to be a formula involving expressions and functions that the user
already knows about. F(x), when it exists, is referred to as a closed form solution to the integration
problem or the antiderivative of f(x). For calculus freshmen, these are expressions that involve
numbers, polynomials, roots, rational functions, trig, logarithm and exponential functions. For
more advanced mathematics students, it might include other functions: erf (error function), Bessel,
G, etc.
The Fundamental Theorem of Calculus (FTC) states that the definite integral can computed if
the antiderivative is known:
a
b
f x dx = F b KF a ,
where F is any function such that
d
dx
F x = f x .
Programs for doing indefinite integration/antidifferentiation have been around from the early '60s.
They can be found in a number of symbolic systems: Maple, Mathematica, Macsyma (now found
in SAGE), and MuPad (now part of Matlab). It is not an easy program to write. Even today in the
era of "open source", the quality and power of such programs varies significantly.
The ideas for how to program an indefinite integrator comes from seminal work done by Robert
Risch in the late 1960s, who was in turn continuing work done previously by other mathematicians
such as Joseph Liouville (1809-1882). Earlier approaches in the early '60s treated symbolic anti-
differentiation as a problem in artificial intelligence. Those AI programs (such as SAINT by James
Slagle (1934-1994) and SIN by Joel Moses (1941-) tried to simulate the expertise of a college
calculus student. However, they would sometimes give up on problems that could be solved
through a clever trick of substitution or other reformulation. Risch's work used advanced
mathematics to come up with programs that would guarantee that they could find a solution if there
was one to be found, for certain classes of integrands. However, while programs using Risch's
techniques are more powerful, they don't necessarily give the answer in simplest terms as a
mathematics textbook would. That may require more symbolic manipulation, under human
guidance.
Indefinite integration
Doing anti-differentiation with
(1.3.1.4) (1.3.1.4)
(1.3.1.1) (1.3.1.1)
(1.3.1.2) (1.3.1.2)
(1.3.1.3) (1.3.1.3)
(1.3.1.5) (1.3.1.5)
Indefinite integration with int
The clickable interface can be used to input an indefinite integration problem for Maple to
compute. Use the f dx item in the Expression Palette, then edit the expression f and the
variable of integration x. This computes the antiderivative of the expression, with respect to the
variable of integration.
The general form of the textual version of the integration command is
int( expression , var )
Indefinite integration with int
Example! Commentary
1
x
dx
ln x
cos x
2
dx
1
2
cos x sin x C
1
2
x
cos
2
x dx
1
2
cos x sin x C
1
2
x
The clickable interface can be used to enter the
integration problem.
In the clickable interface there are two ways
of entering what is "cos x times cos x ".
int sqrt x , x
2
3
x
3 / 2
int cos a $sin
a
2
, a
K
1
3
cos
3
2
a Ccos
1
2
a
The textual form of integration works in the
same way.
This uses the textual form of the command in
a small script. The script computes the
symbolic integral and stores it as the value of
the variable i1. Then it takes the derivative of
i1 and assigns that to d1.
According to the Fundamental Theorem of
(1.3.1.4) (1.3.1.4)
(1.3.1.1) (1.3.1.1)
(1.3.1.2) (1.3.1.2)
(1.3.1.3) (1.3.1.3)
(1.3.1.5) (1.3.1.5)
expr := tan(t) + x^2;
i1 := int(expr,t);
d1 := diff(i1,t);
difference := d1-expr;
printf("difference: %a\n",
difference);
printf("simplified
difference: %a\n",
simplify(difference));
tan t Cx
2
Kln cos t Cx
2
t
sin t
cos t
Cx
2
sin t
cos t
Ktan t
difference: sin(t)/cos(t)-tan
(t)
simplified difference: 0
Calculus, d1 should be the same as expr.
However, we have to get Maple to simplify
the difference in order to have it actually be
reduced to zero, using the built-in library
procedure simplify.
Understanding the answers from int
If Maple does not know how to find an answer it will return an expression that is the display
version of the input expression. It does this instead of giving an error message or NULL so that
a script or program using integration can continue even if the attempt to integrate fails. For
example, even if you don't have the anti-derivative of the integrand, you can still try to do an
approximate numerical definite integration. We discuss that in section 20.4.
Sometimes the answer given by the symbolic integrator will be given in terms of functions that
beginning students haven't heard of. This can be confusing, but unfortunately most systems
written for professionals to use do not have a "beginners mode" that avoids using advanced
math. If you see a function you don't know of, you can get a brief explanation of it through on-
line help.
Whether or not the answer computed for you is satisfactory depends on the situation. A
strange-looking answer might be a sign that you may have mis-entered the problem ("garbage in,
garbage out"). It might mean that there's a way to simplify the strange-looking answer into
something you recognize. If the answer involves functions you haven't heard of, it may mean
that it's time to learn more math to deal with the situation.
To summarize, the responsibilities that users of symbolic integration programs have are a) to be
prepared to learn about more advanced functions if answer is of that form, b) understand when
the computer system is saying that it can't find a closed form solution.
Integration answers using advanced functions
e
x
2
dx
K
1
2
I p erf I x (1.3.2.1)
You wouldn't see this integration problem in
an elementary calculus textbook, because the
answer involves the "error function". The
term "error function" is not indication that
there is a mistake, It's the official
terminology for the function, in the same way
that "sine", "cotangent" or "exponential" are
names of functions. The name arose from
the function's original use in probability and
statistics. You can read more about the error
function by typing "erf" or "error function"
into Maple's on-line help.
Integration answers when Maple can't find an expression for the anti-derivative
answer d int f x , x
f x dx (1.3.2.2)
answer2 := int(g(t),t);
printf("The answer was:
%a",answer2);
g t dt
The answer was: int(g(t),t)
Maple can't find the antiderivative because we
haven't told it what the function f is.
It can't find the symbolic integral for g either,
for the same reason. We see that printf says
that the value of answer2 is, in textual form,
the same as the input was.
int 1 / sqrt log sin x C1 Cexp x , x
1
ln sin x C1 Ce
x
dx (1.3.2.3)
Here's another example of an integral that
Maple can't do even though the expression
being integrated is something that it
recognizes.
1. 1.
(1.3.3.5) (1.3.3.5)
3. 3.
(1.3.3.9) (1.3.3.9)
(1.3.3.11) (1.3.3.11)
(1.3.3.7) (1.3.3.7)
(1.3.3.10) (1.3.3.10)
(1.3.3.12) (1.3.3.12)
(1.3.3.2) (1.3.3.2)
(1.3.3.3) (1.3.3.3)
(1.3.3.8) (1.3.3.8)
2. 2.
(1.3.3.13) (1.3.3.13)
(1.3.3.1) (1.3.3.1)
(1.3.3.6) (1.3.3.6)
(1.3.3.4) (1.3.3.4)
Checking that your integration answer is correct
Like everything else produced by a computer, it's still your responsibility to verify that the
computed answer is correct. After all, even if you think it unlikely that there's a programming
error in int's programming, you might have mis-entered the function you want to integrate.
However, it's easy to use Maple to check integration answers it provides:
Is the derivative of your answer the same or equivalent to what you started with?
Because of the relationship between integration and differentiation, it is possible to check
integration answers without having to do the integration by hand:
Use int to calculate the answer
Apply diff to the answer. The result of differentiation should be the function you were
originally tried to integrate with int.
If the derivative does not look the same as the integrand, try simplify on the difference
between the result of step 2 and the original expression. If you get zero, the two are
equivalent.
Checking answers
Example Commentary
expr d 1 C x C
cos x
2
1 Cx C
1
2
cos x
answer d int expr, x
x C
1
2
x
2
C
1
2
sin x
dAnswer d diff answer, x
1 Cx C
1
2
cos x
expr KdAnswer
0
The derivative of the symbolic integral is the
same as the original expression. This is
obvious just by "eyeballing" it, but we can
also confirm that the difference between expr
and the derivative is 0.
expr d sin x
2
sin x
2
answer d int expr, x
K
1
2
cos x sin x C
1
2
x
dAnswer d diff answer, x
1
2
sin x
2
K
1
2
cos x
2
C
1
2
The derivative of the symbolic integral does
not appear at first glance to be the same as the
original expression. However, the difference
between them does simplify to zero, so they
are equivalent. Maple does not automatically
employ trig identities such as
sin
2
a Ccos
2
a = 1 because it would often
introduce tedious delay to try to apply all
possible identities all the time.
(1.3.3.5) (1.3.3.5)
(1.3.3.10) (1.3.3.10)
(1.3.4.5) (1.3.4.5)
(1.3.3.3) (1.3.3.3)
(1.3.4.10) (1.3.4.10)
(1.3.3.8) (1.3.3.8)
(1.3.3.13) (1.3.3.13)
(1.3.4.12) (1.3.4.12)
(1.3.4.3) (1.3.4.3)
(1.3.4.9) (1.3.4.9)
(1.3.4.6) (1.3.4.6)
(1.3.4.15) (1.3.4.15)
(1.3.4.4) (1.3.4.4)
(1.3.3.9) (1.3.3.9)
(1.3.3.11) (1.3.3.11)
(1.3.3.7) (1.3.3.7)
(1.3.4.7) (1.3.4.7)
(1.3.4.11) (1.3.4.11)
(1.3.4.1) (1.3.4.1)
(1.3.3.12) (1.3.3.12)
(1.3.3.2) (1.3.3.2)
(1.3.4.2) (1.3.4.2)
(1.3.3.1) (1.3.3.1)
(1.3.3.6) (1.3.3.6)
(1.3.4.14) (1.3.4.14)
(1.3.3.4) (1.3.3.4)
(1.3.4.8) (1.3.4.8)
(1.3.4.13) (1.3.4.13)
d dexpr KdAnswer
1
2
sin x
2
C
1
2
cos x
2
K
1
2
simplify d
0
expr d x^2 C1 / xK1
x
2
C1
x K1
answer dint expr, x
1
2
x
2
Cx C2 ln x K1
dAnswer d diff answer, x
x C1 C
2
x K1
simplify expr KdAnswer
0
Here is another example of a symbolic
integration answer whose derivative does not
appear to be the same as the starting
expression. However, simplification of the
difference between the two indicates that the
two are equivalent because their difference
simplifies to zero.
Sometimes the "simplify the difference to zero" trick is not needed to check work because it is
obvious that the two are the same. Maple has other simplification operations such as normal,
factor, convert, combine, and expand that will attempt to put results into alternative forms. See
on-line help about simplify and the other operations for more information.
Troubleshooting int for indefinite integration problems
int is intended to work with expressions involving the variable of integration, not names of
functions. If an expression does not seem to involve the variable of integration, it is treated as a
symbolic constant, regardless of whether it is also the name of a function. This can lead to
misleading "answers"
Variables assigned expressions work with int, but not variables assigned function
definitions
badResult d cos dt
cos t
Since the integrand is just cos, and does not
involve the variable of integration, Maple
treats it as a symbolic constant. Since for a
name a, a dt = a t the result of this integral
is the name "cos", times t. It sort of looks like
cos(t) but it isn't.
(1.3.4.5) (1.3.4.5)
(1.3.4.10) (1.3.4.10)
(1.3.4.12) (1.3.4.12)
(1.3.4.3) (1.3.4.3)
(1.3.4.9) (1.3.4.9)
(1.3.4.6) (1.3.4.6)
(1.3.4.15) (1.3.4.15)
(1.3.4.4) (1.3.4.4)
(1.3.4.7) (1.3.4.7)
(1.3.4.11) (1.3.4.11)
(1.3.4.1) (1.3.4.1)
(1.3.4.2) (1.3.4.2)
(1.3.4.14) (1.3.4.14)
(1.3.4.8) (1.3.4.8)
(1.3.4.13) (1.3.4.13)
eval badResult, t = 0
0
eval badResult, t = .5
0.5 cos
One way to demonstrate that this answer isn't
cos(t) is to evaluate it at t=0. 0$cos = 0. That
certainly isn't the value of cos(0).
Evaluating this expression for another value
indicates that Maple's integration result is
actually "t*cos": the variable t times the name
of a function.
goodResult d cos t dt
sin t
eval goodResult, t = 0
0
eval goodResult, t = .5
0.4794255386
Giving the symbolic integrator an expression
involving the variable t gives the desired
result.
g d x / x^2
x/x
2
int g, x
g x
Another mistaken attempt to integrate a
function name instead of an expression
involving the integration variable. Like the
example above, it produces the answer "g
times x". Note that the value of g is not the
expression x^2, it's the function definition x
-> x^2.
int g x , x
1
3
x
3
int g t , t
1
3
t
3
Evaluating g(x) does produce x
2
, which int
can integrate properly.
Evaluate g(t) produces an expression in t,
which int can integrate properly if it is given t
as the variable of integration.
int g y , x
y
2
x
This is the integral of y
2
with respect to x,
which is x times y
2
.
int abs b , b
K
1
2
b
2
b %0
1
2
b
2
0 !b
Integrating the absolute value function returns
a piecewise expression. We talk more about
piecewise expressions in Section 20.6.
(1.3.4.5) (1.3.4.5)
(1.3.4.10) (1.3.4.10)
(1.3.4.12) (1.3.4.12)
(1.3.4.3) (1.3.4.3)
(1.3.4.9) (1.3.4.9)
(1.3.4.6) (1.3.4.6)
(1.3.4.15) (1.3.4.15)
(1.3.4.4) (1.3.4.4)
(1.3.4.7) (1.3.4.7)
(1.3.4.11) (1.3.4.11)
(1.3.4.1) (1.3.4.1)
(1.3.4.2) (1.3.4.2)
(1.3.4.14) (1.3.4.14)
(1.3.4.8) (1.3.4.8)
(1.3.4.13) (1.3.4.13)
procAbs d proc x
if x R0
then
return x
else
return Kx
end if;
end
proc x
if 0 != x then
return x
else
return Kx
end if
end proc
int procAbs b , b
Error, (in procAbs) cannot
determine if this
expression is true or
false: 0 <= b
We write a Maple procedure to emulate the
absolute value function...
We can't integrate this procedure because the
if statement gives an error when it tries to
decide whether x is greater than zero when
there is no numerical value for x.
procNew d proc x
eval exp x $a, a = 2$x
end
proc x
eval exp x * a, a = 2 * x
end proc
int procNew y C1 , y
2 e
y C1
y
Maple can integrate the results of procedures,
as long as the procedure can execute with
symbolic-valued parameters.
This computes the integral of exp(y+1)*(y+1)
. Note that we could use it to calculate the
integral of exp(z)*2*z, exp(w)*2*w, etc. in a
similar fashion.
Definite integration
Definite integration with int
(1.4.9) (1.4.9)
(1.4.8) (1.4.8)
(1.4.10) (1.4.10)
(1.4.6) (1.4.6)
(1.4.1) (1.4.1)
(1.4.11) (1.4.11)
(1.4.2) (1.4.2)
(1.4.12) (1.4.12)
(1.4.5) (1.4.5)
(1.4.4) (1.4.4)
(1.4.3) (1.4.3)
(1.4.7) (1.4.7)
To use the clickable interface to calculate a definite integral, use the
a
b
f dx item in the Expression
Palette, then edit the expressions f, a, and b, and the variable of integration x to suit your
problem.
If the answer to a definite integration problem returns as the symbolic answer, then doing evalf( ...
) of it will produce an approximation to the area,. These approximation techniques do not require
the use of symbolic integration to find the answer. Approximation techniques will be used
automatically, without evalf, if the integrand or the limits of integration involve floating point
numbers rather than exact fractions.
The general form of the textual version of the command is
int( expression , var = a..b )
Definite Integration Examples
Example Commentary
Anton, Calculus 8th ed. problem 6..5.14.
Calculate
K10
5
6 dt ,
K
p
3
p
3
sin x dx,
0
3
x K2 dx ,
0
2
4 Kx
2
dx .
K10
5
6 dt
90
Using the clickable interface.
int sin x , x =K
p
3
..
p
3
0
Using the document interface but with
the textual form of the operation.
0
3
x K2 dx
5
2
Using the clickable interface again.
Entry of the operation within a code
(1.4.9) (1.4.9)
(1.4.8) (1.4.8)
(1.4.10) (1.4.10)
(1.4.6) (1.4.6)
(1.4.1) (1.4.1)
(1.4.11) (1.4.11)
(1.4.2) (1.4.2)
(1.4.12) (1.4.12)
(1.4.5) (1.4.5)
(1.4.4) (1.4.4)
(1.4.3) (1.4.3)
(1.4.7) (1.4.7)
result := int(sqrt(4-x^2),x=0..2);
p
edit region, followed by its execution
with control (command)-E
Anton problem 32 (a). Calculate
K3
3
x
2
K1 K
15
x
2
C1
dx . Answer should be 30*
arctan(13/9) + 28/3.
int abs x^2 K1 K15 / x^2 C1 , x =K3 ..3
28
3
K30 arctan 3 C60 arctan 2
simplify (1.4.4)
30 arctan
13
9
C
28
3
The original form of the answer given
by Maple does not match the answer
key, but fortunately the simplify
function reduces the result to the
desired form.
Anton problem 32 (b). Calculate
0
3
2
1
1 Kx
2
K 2 dx using Theorem 6.5.5,
which states that for a !b !c,
a
c
f x dx =
a
b
f x dx C
b
c
f x dx .
You will find it necessary to choose a good splitting
point -- try plotting the function and finding where it is
zero.
expr dabs 1 / sqrt 1Kx^2 Ksqrt 2
1
1 Kx
2
K 2
int expr, x = 0 ..sqrt 3 / 2
Integration is a domain where even the
best programs may not be able to find
the answer. It usually doesn't hurt to
see if Maple's software can find the
answer by itself. But we see here that
since the result is the same as the input,
that Maple can't do the problem without
help.
(1.4.9) (1.4.9)
(1.4.8) (1.4.8)
(1.4.10) (1.4.10)
(1.4.6) (1.4.6)
(1.4.1) (1.4.1)
(1.4.11) (1.4.11)
(1.4.2) (1.4.2)
(1.4.12) (1.4.12)
(1.4.5) (1.4.5)
(1.4.4) (1.4.4)
(1.4.3) (1.4.3)
(1.4.7) (1.4.7)
0
1
2
3
1
1 Kx
2
K 2 dx
plot expr, x = 0 ..sqrt 3 / 2
x
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
0.1
0.2
0.3
0.4
0.5
Plotting the function suggests that
there's quite a change occurring around
0.7. This point is probably the splitting
point that the problem is suggesting we
find. But what is it, exactly?
zeroes dsolve expr = 0 , x
x =
1
2
2 , x = K
1
2
2
splitPt d zeroes 2
x = K
1
2
2
bpt d eval x, (1.4.9)
K
1
2
2
Giving solve the problem of where the
function is zero gives us two solutions.
One is negative and we don't care about
it since our definite integral is looking
at the function only between 0 and
3
2
.
We use the splitting idea to ask Maple
to evaluate two definite integrals. It can
(1.4.9) (1.4.9)
(1.4.19) (1.4.19)
(1.4.16) (1.4.16)
(1.4.14) (1.4.14)
(1.4.8) (1.4.8)
(1.4.10) (1.4.10)
(1.4.6) (1.4.6)
(1.4.1) (1.4.1)
(1.4.11) (1.4.11)
(1.4.2) (1.4.2)
(1.4.12) (1.4.12)
(1.4.5) (1.4.5)
(1.4.18) (1.4.18)
(1.4.17) (1.4.17)
(1.4.13) (1.4.13)
(1.4.4) (1.4.4)
(1.4.3) (1.4.3)
(1.4.7) (1.4.7)
(1.4.15) (1.4.15)
0
bpt
expr dx C
bpt
3
2
expr dx
1
4
p K1 C
K
1
2
2
1
2
3
1
1 Kx
2
K 2 dx
do each piece.
We predict that within 20 years Maple
or similar systems will be able to do
this problem completely on their own
without this kind of manual analysis
and manipulation.
Anton Example 4, ch. 8 Evaluate
KN
N
1
1 Cx
2
dx
KN
N
1
1 Cx
2
dx
p
Maple can do improper definite
integrals since it can do symbolic limits
(e.g. the functionality available through
limit).
You can input infinity by through the
Common Symbols Palette that can be
found on the left hand side of the
Maple application window.
Anton Example 6 (c) Ch 8. Evaluate
0
N
1
x $ x C1
dx
answer := int(1/(sqrt(x)*(x+1)), x
= 0 .. infinity);
p
Infinity is can also be input textually as
the symbol .
If the coefficients or limits of integration are floating point numbers, then Maple may use alternative
techniques to approximate the integral. These approximations are often good and sometimes work
when the exact techniques do not.
Numerical approximation of definite integration
Maple is unable to calculate the exact solution
(1.4.19) (1.4.19)
(1.4.16) (1.4.16)
(1.4.14) (1.4.14)
(1.4.18) (1.4.18)
(1.4.17) (1.4.17)
(1.4.13) (1.4.13)
(1.4.15) (1.4.15)
result dint
exp x
2
log x
2
, x = 2 ..3
2
3
e
x
2
ln x
2
dx
to this definite integration problem.
evalf result
694.1492846
evalf of the expression for the integral will
invoke a numerical approximation technique. It
says that the area is approximately 694.15.
int
exp x
2
log x
2
, x = 2 ..3.0
694.1492846
The difference between this and the original
input is that the upper limit of integration has
changed from "3" to "3.0". Because this is a
floating point number, Maple will use
numerical approximation automatically.
int
exp x
2
log x
2
, x = 1 ..3
1
3
e
x
2
ln x
2
dx
evalf (1.4.16)
Float N
This is the same integrand, but the lower limit
of integration has been changed to a point
where the integrand is infinite.
Not surprisingly, the numerical approximation
method doesn't give us a number for the
answer. Since the approximation method
doesn't guarantee that its answer is close to the
actual answer, this result doesn't prove that the
area under the curve is infinite. Further analysis
by a mathematician would be needed to
establish this conclusively.
int
x
6
C2$x C1
x
3
C2$x
2
C1
, x = 0 ..1
K
>
_R= RootOf _Z
3
C2 _Z
2
C1
1
_R 3 _RC4
2 5 K_R
C10 _R
2
ln K_R K
89
12
int finds a symbolic solution to the integral, but
it's not that enlightening. A mathematician
might find this answer useful, but the typical
user will prefer a numerical approximation. We
can get that either by applying evalf to the
result, or by using int with a floating point
number as one of the limits of integration.
(1.4.19) (1.4.19)
(1.4.16) (1.4.16)
(1.4.14) (1.4.14)
(1.5.1) (1.5.1)
(1.5.2) (1.5.2)
(1.5.4) (1.5.4)
(1.5.5) (1.5.5)
(1.4.18) (1.4.18)
(1.5.3) (1.5.3)
(1.4.17) (1.4.17)
(1.4.13) (1.4.13)
(1.4.15) (1.4.15)
C
>
_R= RootOf _Z
3
C2 _Z
2
C1
1
_R 3 _RC4
2 5 K_R
C10 _R
2
ln 1 K_R
int
x
6
C2$x C1
x
3
C2$x
2
C1
, x = 0 ..1.0
1.161103706
Arc-length integration
Recall the following information about the arc length of parametric curves (Anton, ch 7.4, theorem
7.4.3).
If no segment of the curve represented by the parametric equations
x = x t , y = y t , a %t %b
is traced more than once as t increases from a to b, and if
dx
dt
and
dy
dt
are continuous functions for
a %t %b, then the arc length L of the curve is given by
L =
a
b
dx
dt
2
C
dy
dt
2
dt . We can use this formula and Maple's definite integration
operation to calculate arc length easily.
Maple has an arclength operation built into its "Student1" package:
Calculating arclength
Student[Calculus1][Arclength]([ x expression, y expression], parameter = a..b)
calculates the arclength of a curve described by the parametric expressions in the two coordinates.
The package uses the definite integration feature of Maple. Therefore, it will use approximate
methods if a, b, or any number in the expressions is a floating point number.
Student[Calculus1][Arclength](f(x), x = a..b); calculates the arc of the curve given by the
parameterization x=t, y=f(t)
(1.5.1) (1.5.1)
(1.5.4) (1.5.4)
(1.5.5) (1.5.5)
(1.5.3) (1.5.3)
(1.5.2) (1.5.2)
Solving a problem involving arclength
Problem 11, section 7.4 Anton p. 489:
Let the parameterization be x = cos 2$t , y =sin 2$t , 0 %t %
p
2
. Find the arclength.
plot cos 2$t , sin 2$t , t = 0 ..
p
2
, scaling = constrained
K1 K0.5 0 0.5 1
0.2
0.4
0.6
0.8
1
Student Calculus1 ArcLength cos 2 * t , sin 2 * t , t = 0 ..p / 2
p
This is a parametric plot (see this section). We say scaling =constrained so that the x
and y axes are given the same scaling.
Student Calculus1 ArcLength cos 2 * t , sin 2 * t , t = 0 ..p / 2.0
3.141592654
The use of "2.0" makes Maple use approximation techniques in the arclength calculation.
(1.5.1) (1.5.1)
(1.5.4) (1.5.4)
(1.5.5) (1.5.5)
(1.5.3) (1.5.3)
(1.5.2) (1.5.2)
npExpr d sqrt 1 Kx
2
1 Kx
2
plot npExpr, x =K1 ..1, scaling = constrained
x
K1 K0.5 0 0.5 1
0.2
0.4
0.6
0.8
1
This is an alternative formulation for the same curve, as the plot indicates.
We load in the Student Calculus1 package using with. Then we can just use ArcLength instead
of the full name of the library procedure.
with Student Calculus1
AntiderivativePlot, AntiderivativeTutor, ApproximateInt, ApproximateIntTutor,
ArcLength, ArcLengthTutor, Asymptotes, Clear, CriticalPoints, CurveAnalysisTutor,
DerivativePlot, DerivativeTutor, DiffTutor, ExtremePoints, FunctionAverage,
FunctionAverageTutor, FunctionChart, FunctionPlot, GetMessage,
GetNumProblems, GetProblem, Hint, InflectionPoints, IntTutor, Integrand,
InversePlot, InverseTutor, LimitTutor, MeanValueTheorem,
MeanValueTheoremTutor, NewtonQuotient, NewtonsMethod, NewtonsMethodTutor,
(1.5.4) (1.5.4)
(1.6.3) (1.6.3)
(1.5.5) (1.5.5)
(1.5.3) (1.5.3)
(1.6.2) (1.6.2)
(1.5.1) (1.5.1)
(1.6.4) (1.6.4)
(1.6.1) (1.6.1)
(1.5.2) (1.5.2)
PointInterpolation, RiemannSum, RollesTheorem, Roots, Rule, Show,
ShowIncomplete, ShowSolution, ShowSteps, Summand, SurfaceOfRevolution,
SurfaceOfRevolutionTutor, Tangent, TangentSecantTutor, TangentTutor,
TaylorApproximation, TaylorApproximationTutor, Understand, Undo,
VolumeOfRevolution, VolumeOfRevolutionTutor, WhatProblem
ArcLength npExpr, x =K1 ..1
p
Piecewise expressions and piecewise functions
Mathematical situations sometimes need functions that have more than one part to their definition.
Consider the following expression:
a d
- x x !0
x x R0

This defines a piecewise expression. It means "-x if x is less than 0, x if x is R 0."
Piecewise expressions
Example Commentary
a d
- x x !0
x x R0

Kx x !0
x 0 %x
Define a piecewise expression. It means "-x if
x is less than 0, x if x is R 0."
plot a, x =K5 ..5 Plot the piecewise expression for x in the range
-5..5.
(1.6.6) (1.6.6)
(1.6.3) (1.6.3)
(1.6.2) (1.6.2)
(1.6.4) (1.6.4)
(1.6.5) (1.6.5)
(1.6.1) (1.6.1)
x
K4 K2 0 2 4
1
2
3
4
5
b d piecewise x !0, Kx, x R0, x
Kx x !0
x 0 %x
b is assigned the same piecewise expression,
using the textual way of entering the
information.
bMinus1 d simplify 2$ b K1
K1 K2 x x !0
K1 C2 x 0 %x
Maple can do limited algebra on piecewise
expressions.
eval bMinus1, x = 3
5
Since x is greater than or equal to 0, the
bottommost piece of the expression applies, and
the result of evaluation is K1 C2$3 = 5.
We can provide more branches to a piecewise expression. In the Palette version, you can add an
extra row by typing control-shift-R (on Macintosh, command-shift-R) in the bottom row. In the
textual version, you can add extra items onto the sequence of inputs to the function.
More extensive piecewise expressions with the clickable and textual interfaces
Clickable interface Commentary
c d
1 x !K1
2 1 O x RK1
5 x R1
/
To get this we typed c :=, then selected the
- x x !a
x x Ra
item from the palette. We edited
what was there. Then we typed control-shift-R
and filled in the bottom row. After completing
the piecewise expression, we right-clicked
(control-clicked in Mac) the expression and
(1.6.6) (1.6.6)
(1.6.5) (1.6.5)
x
K10 K5 0 5 10
2
3
4
5

selected the 2D plot item from the popup menu.
Textual interface
step2 d piecewise x !K1, 1,K1 %x !2,
x
2
, x R2, x
3
1 x !K1
x
2
K1 %x and x !2
x
3
2 %x
plot step2, x =K2 ..4
x
K2 K1 0 1 2 3 4
10
20
30
40
50
60
eval step2, x = 1.5
2.25
This is an expression with three pieces that fit
together at -1, and 2. To get the %, we typed
the characters <= consecutively. Maple
automatically reset them as the mathematical
symbol %.
This uses the plot facility using the textual
interface.
(1.6.6) (1.6.6)
(1.6.8) (1.6.8)
> >
(1.6.5) (1.6.5)
(1.6.7) (1.6.7)
> >
We can evaluate the expression at x=1.5. Since
the input is limited-precision and the expression
can be completely evaluated using arithmetic,
we get a limited-precision result.
It is possible to use otherwise as one of the clauses of a expression. Usually there are
an even number of inputs to , but if there is an odd number, then the last input is the
"otherwise" value.
More extensive piecewise expressions with the clickable and textual interfaces
Clickable interface Commentary
c d
1 K2 ! x !K1
2 1 O x RK1
5 otherwise
1 K2 !x and x !K1
2 x !1 and K1 %x
5 otherwise
/
smartplot (1.6.7)
x
K10 K5 0 5 10
2
3
4
5

To get this we typed c :=, then selected the
- x x !a
x x Ra
item from the palette. We edited
what was there. Then we typed control-shift-
R and filled in the bottom row. After
completing the piecewise expression, we right-
clicked (control-clicked in Mac) the expression
and selected the 2D plot item from the popup
menu.
Textual interface Commentary
(1.6.10) (1.6.10)
(1.6.8) (1.6.8)
(1.6.9) (1.6.9)
> >
(1.6.11) (1.6.11)
(1.6.12) (1.6.12)
(1.6.7) (1.6.7)
> >
(1.6.14) (1.6.14)
(1.6.13) (1.6.13)
step3 d piecewise K2$p %x %p, sin x , p
!x !5, 0,K2
sin x K2 p %x and x %p
0 p !x and x !5
K2 otherwise
plot step3, x =K10 ..10
x
K10 K5 0 5 10
K2
K1
1
We define an expression that is sin(x) for x
between K2$p and p, 0 between p and 5, and
-2 everywhere else. If the number of terms in
the sequence of inputs to piecewise is odd, then
the last term is the "otherwise" clause.
Since a piecewise expression is still an expression, we can use it as part of a function definition.
Within a function definition, a piecewise expression can be used instead of an if then ... end if.
More piecewise function definitions and uses
Clickable interface Commentary
h dx/
x x !1.5
x$log x x R1.5
x/piecewise x !1.5, x, 1.5 %x,
x log x

h 1.5
0.6081976622
solve h x = 3.2
2.954165523
plot h x , x = 0 ..4
To get this we selected the f := a /y, from the
Expression Palette, and then filled in for f,
and for a. Positioning the cursor at y, we
then selected the
- x x !a
x x Ra
item from the
palette. We edited what was there.
Once defined, h is available for evaluation
within expressions.
The solve operation tries to find (numerically,
using limited-precision arithmetic) a value of x
such that h(x) is approximately equal to 3.23.2
is greater than 1.5, and 2.954165523*log
(2.954165523) is approximately 3.2.
(1.7.5) (1.7.5)
(1.7.2) (1.7.2)
(1.6.11) (1.6.11)
(1.7.4) (1.7.4)
(1.6.13) (1.6.13)
(1.7.1) (1.7.1)
(1.6.10) (1.6.10)
(1.6.9) (1.6.9)
(1.7.3) (1.7.3)
(1.6.12) (1.6.12)
(1.6.14) (1.6.14)
x
1 2 3 4
0
1
2
3
4
5

Textual interface Commentary
h2 d x / piecewise x !1.5, x, 1.5 % x, x
$log x
x/piecewise x !1.5, x, 1.5 %x,
x log x
solve h2 x = 3.3
3.001983443
h x Kh2 x
0
Maple formatted the as a right-arrow after
we entered it. it also formatted a that we
typed as as %.
Calculations with piecewise expressions
Many scientific and engineering situations can be modeled with piecewise expressions. Maple can
apply solve, diff , and int on piecewise expressions, This can often save you time over trying to
perform the operations on the pieces separately.
Calculus on piecewise expressions and functions
Example Commentary
Problem: An ion-powered space probe is
moving through a gaseous nebula. For the
first ten hours the ion engine is on, and, it
(1.7.1) (1.7.1)
(1.7.5) (1.7.5)
(1.7.2) (1.7.2)
(1.7.3) (1.7.3)
(1.7.4) (1.7.4)
travels at a rate of .05$t
2
meters per second.
Then the motor shuts off an the gas in the
nebula gradually slows the space probe down.
The velocity of the space probe can be
described as:

0 t !0
5
100
$t
2
0 %t !10
5
100
$10
2
$exp Kt C10 t R10
What is the position of the space probe at t=20
hours?
Solution
velocity d
0 t !0
5
100
$t
2
0 %t !10
5
100
$10
2
$exp Kt C10 t R10
0 t !0
1
20
t
2
0 %t and t !10
5 e
Kt C10
10 %t
plot velocity, t = 0 ..20
We assign the piecewise expression to the
variable velocity and plot it to better understand
it.
The plot reveals that there are not any
discontinuities in velocity, although there is an
abrupt transition at t=10.
(1.7.5) (1.7.5)
(1.7.2) (1.7.2)
(1.7.9) (1.7.9)
(1.7.8) (1.7.8)
(1.7.7) (1.7.7)
(1.7.4) (1.7.4)
(1.7.1) (1.7.1)
(1.7.3) (1.7.3)
(1.7.6) (1.7.6)
t
5 10 15 20
0
1
2
3
4
position d int velocity, t
0 t %0
1
60
t
3
t %10
K5 e
Kt C10
C
65
3
10 !t
eval position, t = 20.0
21.66643967
Since position is the integral of velocity, we can
get an expression for the position by
symbolically integrating velocity with respect to
time.
To find the position at t=20, we evaluate the
expression position at that value. By giving the
floating point number 20.0 instead of the exact
number 20, we ensure that the result is given as
a floating point number.
distance d int velocity, t = 0 ..20.0
21.66643967
We should get the same result if we integrate
velocity for t=0 to t=20
Piecewise integration
Problem 4(c) Anton, problem 6.7.42. A sprinter in a 100 m race explodes out of the starting block
with an acceleration of 4.0
m
s
2
, which she sustains for 2.0s. Her acceleration then drops to zero
for the rest of the race. (a) What is her time for the race? Make a graph of her distance from the
starting block versus time.
Solution
Create a piecewise expression modeling the acceleration.
aexpr d piecewise 0 %t and t %2, 4, 0
(1.7.9) (1.7.9)
(1.7.8) (1.7.8)
(1.7.7) (1.7.7)
(1.7.6) (1.7.6)
4 0 %t and t %2
0 otherwise
We integrate the acceleration expression to get a velocity expression.
vexpr d int aexpr, t
0 t %0
4 t t %2
8 2 !t
We integrate the velocity expression to get a position expression.
dexpr d int vexpr, t
0 t %0
2 t
2
t %2
8 t K8 2 !t
We use fsolve to find approximately the time when the runner reaches 100 meters.
finishTime d fsolve dexpr = 100, t
13.50000000
We plot distance versus time to confirm the answer from fsolve.
plot dexpr, t = 0 ..finishTime, labels = "time (in seconds)", "position (in meters)"
(1.7.9) (1.7.9)
(1.7.8) (1.7.8)
(1.7.7) (1.7.7)
(1.7.6) (1.7.6)
time (in seconds)
0 2 4 6 8 10 12
position (in meters)
0
20
40
60
80
100
Piecewise integration, continued
Problem
Two racers are in the race. One accelerates at 2m/sec^2 for 4 seconds, as described previously.
The other accelerates at 2.1 m/sec^2 for 3.8 seconds. Who gets to the finish line first?
Solution
Create a procedure that turns the previous script into something that can handle other
accelerations. The procedure should take two parameters, the initial acceleration rate aInit, and the
period of time tInit that this acceleration lasts. After the initial acceleration, the racer is assumed to
run at the final velocity attained for the rest of the race. We want the procedure to return the time
to complete the race. It should also, as a side-effect, print out the plot but it should not return the
plot structure as a result.
racer := proc( aInit, tInit)
local aexpr, vexpr, dexpr, finishTime, plotStructure;
#Define acceleration
aexpr := piecewise(0 <= t and t <= aInit, tInit, 0);
#Calculate velocity and distance travelled.
vexpr := int(aexpr, t);
dexpr := int(vexpr, t);
#Calculate approximate time to finish.
finishTime := fsolve(dexpr = 100, t);
#print out postion vs. time.
print(plot(dexpr, t = 0 .. finishTime, labels = ["time (in
seconds)", "position (in meters)"]));
#return finishing time.
return finishTime;
end proc:
#Test the procedure on the original problem
r1 :=racer(2.0, 4.0);
time (in seconds)
0 2 4 6 8 10 12
position (in meters)
0
20
40
60
80
100
13.50000000
r1 :=racer(2.0, 4.0);
r2 := racer(2.1, 3.8);
#while the if statement is not really worthwhile if
#we only wanted to compare two racers, we are considering
#the possibility that we might want to do this
#frequently and so write for easy reuse.
if r1< r2 then print("Racer 1 won!");
elif r1>r2 then print("Racer 2 won!");
else print("They tied!");
end if;
time (in seconds)
0 2 4 6 8 10 12
position (in meters)
0
20
40
60
80
100
13.50000000
(1.7.10) (1.7.10)
time (in seconds)
0 2 4 6 8 10 12
position (in meters)
0
20
40
60
80
100
13.58132832
"Racer 1 won!"
Because piecewise expressions are similar to if then .. end if, it is tempting to think that they are
interchangable. One thing to keep in mind is that operations such as int evaluate all their arguments
before they start to try to figure out their results. If one of the arguments involves a procedure, then
the procedure needs to work with symbolic rather than numeric arguments.
The difference between piecewise and a procedure with if then else.
int abs b , b
Error, (in int) integration
range or variable must be
specified in the second
argument, got piecewise(x <
0, -x, 0 <= x, x)
Integrating the absolute value function returns a
piecewise expression.
(1.7.10) (1.7.10)
procAbs d proc x
if x R0
then
return x
else
return Kx
end if;
end
proc x
if 0 != x then
return x
else
return Kx
end if
end proc
int procAbs b , b
Error, (in procAbs) cannot
determine if this
expression is true or
false: 0 <= piecewise(x <
0, -x, 0 <= x, x)
We write a Maple procedure to emulate the
absolute value function...We find that Maple
can't integrate it because it tries to get a formula
for procAbs(b) and then integrate that. But the
procedure can't run with b only a symbol.
Curve fitting with splines
In scientific and engineering work, we often start with numerical measurements of a phenomenon.
but need ways of knowing something about the mathematical behavior of the phenomenon for
values other than those measured -- to make predictions, or to do mathematical modeling operations
such as equation-solving, integration, or optimization that require formulae. A typical way of
handling this is to construct a model (a formula) from the data.
Linear least squares data fitting, described in this section, is a way of doing this. However, it can
only find formulas that are straight lines. . While there are techniques for non-linear least squares
for certain kinds of models-- linear, quadratic, exponential, logarithmic, etc -- they all require that the
modeler know ahead of time that whether the relationship should be linear, quadratic, or whatever.
Picking the wrong kind of relationship may produce a procedure that doesn't very closely model the
data.
There is a different kind of formula-fitting called spline interpolation. The advantage of the data
fitting using splines is that the modeler does not have to know ahead of time whether the
relationship is linear, quadratic, etc. A spline interpolant builds a piecewise expression that is
guaranteed to agree exactly at every data point, instead of hopefully being close by as with least
squares fitting.
(1.8.2) (1.8.2)
(1.8.8) (1.8.8)
(1.8.7) (1.8.7)
(1.8.5) (1.8.5)
(1.8.3) (1.8.3)
(1.8.9) (1.8.9)
(1.8.4) (1.8.4)
(1.8.1) (1.8.1)
(1.8.6) (1.8.6)
Spline interpolants differ in the kind of function used for each piece of the expression. A linear
spline connects the data points together with straight lines. That is, it is a piecewise expression
where a linear expression describes each piece. A piece corresponds to the interval between two
consecutive data points.
A quadratic spline is a piecewise expression where each piece connects two consecutive data points
together with a degree 2 polynomials. cSimilarly there are cubic splines, quartic splines, etc.
Typically quadratic or cubic splines are used for data fitting since they produce smooth curves that
can be computed quickly.
Linear splines
We enter a collection of data values
dataValues d 1, 3 , 1.5, 2.8 , 2.1, 4.6 , 3.0, 4.4
dataValues d 1, 3 , 1.5, 2.8 , 2.1, 4.6 , 3.0, 4.4
... and plot them.
dataPlot d plot dataValues, style = point, view = 1 ..4, 0 ..5 , symbol = diamond, symbolsize
= 20 , color = "Purple"
dataPlot dPLOT ...
print dataPlot
(1.8.2) (1.8.2)
(1.8.8) (1.8.8)
(1.8.7) (1.8.7)
(1.8.5) (1.8.5)
(1.8.3) (1.8.3)
(1.8.9) (1.8.9)
(1.8.4) (1.8.4)
(1.8.1) (1.8.1)
(1.8.6) (1.8.6)
1 2 3 4
0
1
2
3
4
5
We load in the CurveFitting package, and construct a linear spline which we call s. Without
the with, we could just do CurveFitting[Spline].
with CurveFitting
ArrayInterpolation, BSpline, BSplineCurve, Interactive, LeastSquares,
PolynomialInterpolation, RationalInterpolation, Spline, ThieleInterpolation
s dSpline dataValues, t, degree = 1
s d
3.400000000 K0.4000000000 t t !1.5
K1.700000000 C3.000000000 t t !2.1
5.066666667 K0.2222222222 t otherwise
We display a plot of the spline, with the data point plot. This shows that the spline s does agree
with all the data points.
(1.8.2) (1.8.2)
(1.8.8) (1.8.8)
(1.8.7) (1.8.7)
(1.8.5) (1.8.5)
(1.8.3) (1.8.3)
(1.8.9) (1.8.9)
(1.8.4) (1.8.4)
(1.8.1) (1.8.1)
(1.8.6) (1.8.6)
plots display plot s, t = 0 ..4 , dataPlot
t
0 1 2 3 4
0
1
2
3
4
5
To get a prediction of what to expect for say t=2.3, we just evaluate s at that point.
eval s, t = 2.3
4.555555556
(2.3, 4.556) is on the line connecting the data points 2.1, 4.6 and 3.0, 4.4 .
Maple's standard math operations know how to work with the piecewise expression that the
spline generator produces.
Optimization Maximize s, t = 1 ..3.1
4.59999999843952, t = 2.10000000873216
int s, t = 1 ..3.1
8.158888889
fsolve will find one solution, but if we want the solution to be inside the range of values that we
measured, we should specify the interval
fsolve s = 3.1
(1.8.13) (1.8.13)
(1.8.15) (1.8.15)
(1.8.8) (1.8.8)
(1.8.5) (1.8.5)
(1.8.3) (1.8.3)
(1.8.4) (1.8.4)
(1.8.6) (1.8.6)
(1.8.2) (1.8.2)
(1.8.12) (1.8.12)
(1.8.7) (1.8.7)
(1.8.9) (1.8.9)
(1.8.1) (1.8.1)
(1.8.11) (1.8.11)
(1.8.14) (1.8.14)
0.7500000000
fsolve s = 3.1, t = 1 ..4
1.600000000
One problem with linear splines is that they can have sharp corners, as demonstrated by the example
just given. It is possible, with sufficient data points, to construct splines where the pieces are
quadratic, cubic, or of even higher degree. This has the advantage that the bends in the spline will
be smoother. How smooth depends on the degree. However, higher degree splines may do
undesirable things for values outside the ranges being measured. For that reason, it's better to use
splines for problems where you need results for values inside the ranges being measured.
Quadratic splines and splines of higher degrees
The alternative form of spline construct takes two lists, one for first coordinate, and one for the
second.
tValues d 1, 1.7, 2.1, 3.7, 4.1
tValues d 1, 1.7, 2.1, 3.7, 4.1
mValues d 3, 2.7, 4.8, 4.4, 3.1
mValues d 3, 2.7, 4.8, 4.4, 3.1
We can plot them.
dataPlot2 d plot tValues, mValues, style = point, view = 1 ..8, 0 ..8 , symbol = diamond,
symbolsize = 20 , color = "Purple"
dataPlot2 dPLOT ...
print dataPlot2
(1.8.13) (1.8.13)
(1.8.15) (1.8.15)
(1.8.12) (1.8.12)
(1.8.11) (1.8.11)
(1.8.14) (1.8.14)
1 2 3 4 5 6 7 8
0
1
2
3
4
5
6
7
8
With load in the CurveFitting package, and construct a quadratic spline which we call s2.
Without the with, we could just do CurveFitting[Spline].
with CurveFitting
ArrayInterpolation, BSpline, BSplineCurve, Interactive, LeastSquares,
PolynomialInterpolation, RationalInterpolation, Spline, ThieleInterpolation
s2 dSpline tValues, mValues, t, degree = 2
s2 d
3.00000000000000 C7.93016446160826 10
K16
t K3.50975329890993 t K1
2
t !1.35000000000000
K2.73892713711991 C3.19936890418818 t C8.08028030489304 t K1.7
2
t !1.90000000000000
K5.56910499139415 C4.93766904352102 t K3.73452995656094 K2.1 Ct
2
t !2.90000000000000
18.6912916974019 K3.86251126956807 t K1.76558273911974 t K3.7
2
t !3.90000000000000
3.10000000000000 C11.4218609130399 t K4.1
2
otherwise
(1.8.13) (1.8.13)
(1.8.15) (1.8.15)
(1.8.12) (1.8.12)
(1.8.11) (1.8.11)
(1.8.14) (1.8.14)
We display a plot of the spline, with the data point plot. This shows that the spline s does agree
with all the data points.
plots display plot s2, t = 0 ..5 , dataPlot2
t
0 1 2 3 4 5 6 7 8
0
1
2
3
4
5
6
7
8
To get a prediction of what to expect for say t=2.3, we just evaluate s at that point.
eval s, t = 2.3
4.555555556
The formula does have the drawback that it starts increasing after t=4.1, even though the data
doesn't. This is a property of higher degree polynomials. We might not want to use this model if
we need to extrapolate very far beyond the values that were measured.
Higher degree spline interpolation produces piecewise expressions that all agree fairly closely for
values within the t values measured within the region defined by the data points. However, they
may differ significantly outside the region of the data points.
s1 d Spline tValues, mValues, t, degree = 1 :
s3 d Spline tValues, mValues, t, degree = 3 :
s4 d Spline tValues, mValues, t, degree = 4 :
(1.8.13) (1.8.13)
(1.8.15) (1.8.15)
(1.8.12) (1.8.12)
(1.8.11) (1.8.11)
(1.8.14) (1.8.14)
plots display plot s1, t = 0 ..5, color = black , plot s2, t = 0 ..5, color = red , plot s3, t = 0
..5, color = green , plot s4, t = 0 ..5, color = blue , dataPlot2
t
0 1 2 3 4 5 6 7 8
0
1
2
3
4
5
6
7
8
Chapter summary
Indefinite integration with int
The clickable interface can be used to input an indefinite integration problem for Maple to
compute. Use the f dx item in the Expression Palette, then edit the expression f and the variable
(1.9.5) (1.9.5)
(1.9.2) (1.9.2)
(1.9.3) (1.9.3)
(1.9.4) (1.9.4)
(1.9.1) (1.9.1)
of integration x to suit your problem.
The general form of the textual version of the command is
int( expression , var )
This computes the anti-derivative of the expression, with respect to the variable var. If the answer
is just a formula involving expressions that the user already knows about (e.g. ones that involve
numbers, polynomials, rational functions, trig, logarithm and exponential functions), when Maple
can find one, is sometimes referred to as a closed form solution to the integration problem.
Indefinite integration with int
Example! Commentary
1
x
dx
ln x
cos x
2
dx
1
2
cos x sin x C
1
2
x
cos
2
x dx
1
2
cos x sin x C
1
2
x
The clickable interface can be used to enter the
integration problem.
Note that in the clickable interface there are two
ways of entering what is "cos x $cos x ".
int sqrt x , x
2
3
x
3 / 2
int cos a $sin
a
2
, a
K
1
3
cos
3
2
a Ccos
1
2
a
The textual form of integration works in the
same way.
This is use of the textual form of the command
in a small script. The script computes the
symbolic integral and stores it in i1. Then it
takes the derivative of i1 and assigns that to d1.
(1.9.11) (1.9.11)
(1.9.6) (1.9.6)
(1.9.2) (1.9.2)
(1.9.10) (1.9.10)
(1.9.4) (1.9.4)
(1.9.7) (1.9.7)
(1.9.1) (1.9.1)
(1.9.9) (1.9.9)
(1.9.12) (1.9.12)
(1.9.5) (1.9.5)
(1.9.8) (1.9.8)
(1.9.3) (1.9.3)
expr := tan(t) + x^2;
i1 := int(expr,t);
d1 := diff(i1,t);
difference := d1-expr;
print("difference:".
difference);
print("simplified
difference:",
simplify(difference));
tan t Cx
2
Kln cos t Cx
2
t
sin t
cos t
Cx
2
sin t
cos t
Ktan t
"difference:".
sin t
cos t
Ktan t
"simplified difference:", 0
According to the Fundamental Theorem of
Calculus, d1 should be the same as expr.
However, we have to get Maple to simplify the
difference in order to have it actually be reduced
to zero.
Definite Integration examples
Example Commentary
Anton, Calculus 8th ed. problem 6..5.14.
Calculate
K10
5
6 dt ,
K
p
3
p
3
sin x dx,
0
3
x K2 dx ,
0
2
4 Kx
2
dx .
K10
5
6 dt
Using the clickable interface.
(1.9.11) (1.9.11)
(1.9.6) (1.9.6)
(1.9.10) (1.9.10)
(1.9.7) (1.9.7)
(1.9.9) (1.9.9)
(1.9.12) (1.9.12)
(1.9.8) (1.9.8)
90
int sin x , x =K
p
3
..
p
3
0
0
3
x K2 dx
5
2
int(sqrt(4-x^2),x=0..2);
p
Using the document interface but with
the textual form of the operation.
Using the clickable interface again.
Entry of the operation within a code
edit region.
Anton problem 6.6.25. Calculate
1
4
3
t
K5 t Kt
K
3
2
dt
int 3 / sqrt t K5 * sqrt t Kt^ K3 / 2 , t = 1 ..4
K
55
3
Anton problem 32 (a). Calculate
K3
3
x
2
K1 K
15
x
2
C1
dx . Answer should be 30*
arctan(13/9) + 28/3.
int abs x^2 K1 K15 / x^2 C1 , x =K3 ..3
28
3
K30 arctan 3 C60 arctan 2
simplify (1.9.10)
30 arctan
13
9
C
28
3
The original form of the answer given
by Maple does not match the answer
key, but fortunately the simplify
function reduces the result to the
desired form. This does not always
happen for definite integration results.
Sometimes further manual operation
may be necessary.
Maple can do improper definite
integrals since it can do symbolic limits
(1.9.17) (1.9.17)
(1.9.11) (1.9.11)
(1.9.6) (1.9.6)
(1.9.10) (1.9.10)
(1.9.7) (1.9.7)
(1.9.9) (1.9.9)
(1.9.16) (1.9.16)
(1.9.14) (1.9.14)
(1.9.18) (1.9.18)
(1.9.12) (1.9.12)
(1.9.15) (1.9.15)
(1.9.8) (1.9.8)
(1.9.19) (1.9.19)
(1.9.13) (1.9.13)
Anton Example 4, ch. 8 Evaluate
KN
N
1
1 Cx
2
dx
KN
N
1
1 Cx
2
dx
p
Anton Example 6 (c) Ch 8. Evaluate
0
N
1
x $ x C1
dx
answer := int(1/(sqrt(x)*(x+1)), x
= 0 .. infinity);
printf("sin evaluated at the
answer is: %a", sin(answer));
p
sin evaluated at the answer is: 0
(e.g. the functionality available through
limit).
You can input infinity by through the
"Common Symbols" palette that can be
found on the left hand side of the
Maple application window.
Infinity is can also be input textually as
the symbol .
Approximate definite integration examples
int
exp x
2
log x
2
, x = 2 ..3
2
3
e
x
2
ln x
2
dx
evalf (1.9.13)
694.1492846
int
exp x
2
log x
2
, x = 2 ..3.0
694.1492846
int
exp x
2
log x
2
, x = 1 ..3
Maple is unable to calculate the exact solution
to this definite integration problem.
evalf says that the area is approximately 694.15.
Note that the difference between this and the
original input was that we have changed the
upper limit of integration from "3" to "3.0".
Because this is a floating point number, Maple
will use evalf automatically.
(1.9.16) (1.9.16)
(1.9.14) (1.9.14)
(1.9.18) (1.9.18)
(1.9.17) (1.9.17)
(1.9.15) (1.9.15)
(1.9.19) (1.9.19)
(1.9.13) (1.9.13)
1
3
e
x
2
ln x
2
dx
evalf (1.9.16)
Float N
This is the same integrand. Note that the lower
limit of integration is where the integrand is
infinite.
Not surprisingly, the approximation method
doesn't give us a number for the answer. Since
the approximation method doesn't guarantee
that its answer is close to the actual answer, this
result doesn't prove that the area under the
curve is infinite. Further analysis by a
mathematician might establish this conclusively.
int
x
6
C2$x C1
x
3
C2$x
2
C1
, x = 0 ..1
K
>
_R= RootOf _Z
3
C2 _Z
2
C1
1
_R 3 _RC4
2 5 K_R
C10 _R
2
ln K_R K
89
12
C
>
_R= RootOf _Z
3
C2 _Z
2
C1
1
_R 3 _RC4
2 5 K_R
C10 _R
2
ln 1 K_R
int
x
6
C2$x C1
x
3
C2$x
2
C1
, x = 0 ..1.0
1.161103706
There is a symbolic solution to the integral, but
it's messy. An explanation of "RootOf" can be
found in Example 17.1.3.2.
As with the integrand above, making the upper
limit of integration into a floating point number
invokes the approximation technique that evalf
uses.
Calculating arclength
Student[Calculus1][Arclength]([ x expression, y expression], parameter = a..b)
calculates the arclength of a curve described by the parametric expressions in the two coordinates.
The package uses the definite integration feature of Maple. Therefore, it will use approximate
methods if a, b, or any number in the expressions is a floating point number.
(1.9.21) (1.9.21)
(1.9.23) (1.9.23)
(1.9.20) (1.9.20)
(1.9.22) (1.9.22)
Student[Calculus1][Arclength](f(x), x = a..b); calculates the arc of the curve given by the
parameterization x=t, y=f(t)
Student[Calculus1][Arclength](f(x), x = a..b, showfunction=true); calculates the arc of the curve
given by the parameterization x=t, y=f(t)
Student[Calculus1][Arclength]([ x expression, y expression], parameter = a..b, output=plot,
showfunction=true, showintegral=true, showintegrand=true);
Piecewise expressions
A piecewise expression uses different expressions to describe the value for different ranges of the
controlling variable. It can be entered either with the clickable interface or textually.
To add more branches to a piecewise expression. In the clickable version, you can add an extra
row by typing control-shift-R (on Macintosh, command-shift-R) in the bottom row. In the textual
version, you can add extra items onto the sequence of inputs to the function.
If the word otherwise is used for a range in the clickable interface, then the expression
associated with that branch will be used for all other values of the controlling variable. The
otherwise branch is specified in the textual version of piecewise as the odd last argument to
piecewise .
A difference between piecewise expressions and if then ... end if is that Maple can do
mathematical operations such as +, solve, diff, int, or plot on piecewise expressions, while it can't
do so as easily with if then ... end if.
h dx/
x x !1.5
x$log x x R1.5
x/piecewise x !1.5, x, 1.5 %x, x log x

h 1.5
0.6081976622
solve h x = 3.2
2.954165523
step3 d piecewise K2$p %x %p, sin x , p !x !5, 0,K2
(1.9.21) (1.9.21)
(1.9.22) (1.9.22)
(1.9.23) (1.9.23)
(1.9.20) (1.9.20)
sin x K2 p %x and x %p
0 p !x and x !5
K2 otherwise
plot step3, x =K10 ..10
x
K10 K5
0
5 10
K2
K1
1
351
lndex
Symbols
! (exclaimation point
Iactorial), 12
! icon (execute group), 188
!!! (execute worksheet), 66
" (double-quote), 77
# (program comment), 189
' (apostrophe), 84
(, 11
), 11
- (minus or dash), 11
-inIinity, 325, 331
: (colon), 188
:, 45
; (semi-colon), 188
(equation), 23
` (caret), 12
(underscore), 45
` (backquote), 84
a
symbolic constant, 29, 33, 34
, 111
A
abs (absolute value), 107
accumulation loop, 289
actual parameter, 77
acute accent marks (`), 84
addition (), 10
alpha (Greek letter), 321
altitude, 174
and (Boolean operation), 217, 237
animate (plots package), 158, 180
animation controls, 159
animations
exporting to GIF Iormat, 183
apostrophe, 132
apostrophes, 84
approximation
Iloating point, 228
arccos (inverse oI cosine), 29
arcsin (inverse oI sine), 29, 107
arctan (inverse oI tangent), 29
arguments
oI Iunction, 77, 105
arithmetic, 10
approximate, 29
exact, 13
operations, 10
arrow, 111
arrow key
right, 13
assignment, 24
mental model oI, 48
multiple, 285
oI a plot structure to a variable, 147
undoing, 49
assistant
Unit Converter, 136
asterisk, 11
autosaving, 22
average
oI numbers in a list, 93
Axiom, 4
B
backquotes, 84
backspace key, 17
backups
retrieving, 22, 22
ball, 174
base oI natural logarithm (e), 33
blue, 151
Boole
George (1815-1864), 213, 215, 217, 236, 237
Boolean values, 202
brackets, 62
by, 245, 247, 247, 247, 247, 247, 260, 261, 261, 261, 261
C
C (programming language), 4, 6, 55
C (programming language), 5
Canadian, 92
Canadian roots, 152
Canvas (Maple word processor Ieature), 43
caret (power), 12
circle, 152, 156, 176
Iunction, 152
circle-X, 7
clickable interIace, 80
code edit region, 187
Code Edit Region
collapse, 192
execute, 192
resize, 190
code icon
clicking on, 192
Collapse Code Edit Region, 192
colon, 188, 195, 269
Colorado, 63
colornames (plot), 167
352 Index
colors
plotting, 91
comma, 62
command-, 41
command-C (copy operation), 17, 66
command-e (execute region), 188
command-R, 42
switch to math mode, 40
command-T
switch to text entry mode, 40
command-V (paste operation), 17, 66
command-X (cut operation), 17
comment
program (#), 189
Commenting out
printing statements, 268
comments
header, 189
completion
command, 109
Component Properties, 190
composition
oI Iunctions, 125, 128, 128
computing
algebraic, 1
graphical, 1
high perIormance, 2
logical, 1
numerical, 1
constant
symbolic, 57
control-, 41, 42, 320
control-C (copy operation), 17, 66
control-click, 23
control-e (execute region), 188
control-R, 42
switch to math mode, 40
control-T, 42
switch to text entry mode, 40
control-V (paste operation), 17, 66
control-X (cut operation), 17
convert, 174, 174, 282
a list into a sequence, 93
a list into a string, 93
between units, 121, 131
converting
table into list, 282
copy (editing operation), 17
copy and paste, 66
crash
Maple, 22
curly braces, 62
currency, 130
current state oI a Maple session, 48
cut (editing operation), 17
D
daisy chaining, 125
Iunctions, 128
dashed rectangle, 8
days, 132
degrees, 121
delimiters in Maple: ( ) | | } , 15
derivative, 320
diII, 319, 322, 344
diIIerentiate
right-click menu item, 328
diIIerentiate menu item, 320
diIIerentiation, 319
oI expression palette, 320
digits, 30
disk (plottools Iunction), 167
display, 149, 179
Iunction, 149, 152
divide and conquer, 68
division (/), 11
documentation
need Ior, 2
dot
centered, 13
double-quotes, 84
E
e
Irom keyboard not a symbolic constant, 33
natural logarithm and symbolic constant, 29, 33, 34, 42
Edit-~Execute-~Selection, 66
Edit~Execute~Worksheet., 66
ellipse, 156
ENGR101, 68
enter, 41, 49, 66
entries, 281, 285
equation, 23, 61, 279
error
attempting to assign to Pi which is protected, 57
attempting to assign to `solve` which is protected , 57
controlling variable oI Ior loop must be a name, 57
illegal use oI an object as a name, 56
invalid Iraction, 16
invalid leIt hand side in assignment, 283
invalid matrix/vector product, 14
invalid operator parameter name, 116
invalid power, 78
invalid sequence, 15
invalid sum/diIIerence, 14
logic, 20
missing operation, 14
Index 353
numeric exception: division by zero, 14
semantic, 19
unable to match delimiters, 15, 15, 78
unexpected option, 84
vocabulary, 19
Error
(in IprintI) integer expected Ior integer Iormat, 265, 268
(in IprintI) number expected Ior Iloating point Iormat, 264
ambiguous multiple assignment, 286
missing operator or `;`, 257
reserved word `do` unexpected, 258
unexpected `local` declaration, 232
error message, 13
escape key (ESC), 42, 109
Euro, 128
eval, 120, 319
evalI, 121
evaluating
preventing, 132
state, 22
evaluation, 46
at a point, 66
oI a Iormula, 155
oI a Iunction, 77
oI expressions, 10
evaluation oI a derivative expression, 319
execute region, 188
execution trace, 269, 272
exp (exponential Iunction), 108
exponentiation, 18
Exporting animations, 183
expression, 279
expressions
algebraic, 23
Irom palette, 28
extrema , 325
F
Iactorial (!), 12
Ialse (Boolean operation), 213, 214, 214, 216, 237
Ialse (Boolean value), 202
Ience post error, 32
Iind something similar, 68
Format -~ Create Document Block, 43
Formatting
output
as currency, 130
Fortran, 4
Ireeze Irame
animation, 159
Iruit Ilies, 111
Isolve, 119, 279
It, 131
Iunction
Boolean, 213, 214, 217, 237
idea oI, 105
Iunction deIinition, 116
Iunction deIinition alternative, 117
Iunction evaluation, 77
Iunctions
deIine, 111
designing Irom word problems, 125
thousands oI built-in, 108
Iuzzy notions, 67
G
gallons, 121, 132
Imperial, 92
gigaIlop, 3
graphical user interIace (GUI), 4
Graphics Exchange Format (GIF), 183
Greek letters, 321
green, 151
H
header comments, 189
help
on-line, 88, 101, 109
table oI contents Ior, 196
quick, 7
through command completion, 109
using on-line, 110
Help (Maple menu), 43
horsepower, 131
hour, 121, 121, 131
hypotenuse, 140
l
I (square root oI -1), 57, 63
Ideal Gas Law, 125
Image (Maple word processor Ieature), 43
immutable, 279
inch, 121, 131
index, 61, 152, 280
indices, 281, 284
inequalities, 61
inIin (shortcut Ior inIinity in math mode), 42
inIinite loop, 235
inIinity, 324, 325, 331
Insert (Maple menu item), 43
Insert-~2-D Math, 40
Insert-~Code Edit, 187
Insert-~Text, 40
interactive, 5
inverse trig Iunctions, 29
354 Index
isprime, 246, 262
]
Java, 4, 5, 55
K
Kessel Run, 132
key, 280
kilometer, 131, 133
kilopascals, 126
kilowatt, 131
L
label, 10
labels Ior axes
plotting, 91
lady
little old, 289
LeastSquares, 117
leIt arrow key, 17
leIt hand side, 61
leIt hand side (right-click menu), 31, 35
lhs, 120, 279
limit
leIt sided, 325
one sided, 325
right sided, 325
line
Iunction, 151, 152
linear algebra, 14
list, 62, 81, 279
add together items Irom, 93
count number oI items in, 93
empty, 279
one item Irom, 93
operations on, 92
part oI, 93
liters, 121
log10 (base 10 logarithm), 107
logarithm
base e, 29
logic error, 20, 32
loop
accumulation, 289
inIinite, 235
suppressing output Irom, 269
Los Alamos National Laboratory, 3
M
Macsyma, 4
Maple, 4, 4
as a word processor, 39
document mode, 7
math entry mode, 39
Mathematica, 4, 6
mathematical model, 1
mathematical models, 154
Matlab, 4, 55
Symbolic Toolkit, 6
max (maximum oI arguments), 107, 118
maxdigits, 19
maximize, 331, 335
meter, 121, 131
micrometer, 131
mile, 131
millimeter, 131
min (minimum oI arguments), 107, 118
minimize, 331, 334, 340, 345
minus, 11
minute, 132
mistake
typographical
how to Iix, 17
mod, 262
multiple assignment, 285
multiple systems, 6
multiplication (*), 11
multivariate, 344
MuPad, 4, 55
N
name
assignment to, 45
National Forest Service, 65
negation (-), 11
negative inIinity, 325
nops, 279, 280
not (Boolean operation), 213, 215, 217, 237
NULL, 331
O
Oak Ridge National Laboratory, 3
objective Iunction, 325
Octave, 4
old
really
really, 293
omega (Greek letter), 42
Omega (upper case Greek letter), 42
on-line documentation, 87
op, 279
operation
Boolean (and
or, not), 213, 214, 217, 237
Index 355
Optimization, 339
Optimization|Maximize|, 337
Optimization|Minimize|, 337, 345
options
plot, 120
or (Boolean operation), 213, 215, 217, 237
orange, 151
oscilloscope, 68
P
packages
library, 117
palette, 28
Expression, 80, 105, 111, 111, 118, 320
square root, 28
Palette
Greek, 321
palettes, 42
parameter, 66
parameterized plots, 156
parentheses, 11, 77
parsec, 132
paste (editing operation), 17
petaIlop, 3
Pi, 57
Pi (symbolic constant), 57
pick names, 67
pitIalls
common
in Maple, 105
playing
animation, 159
plot, 119, 279
a Iormula, 25
builder, 87
colors, 84
Ior insight into a mathematical relationship, 155
labels, 84
options, 91
structure, 147
symbol Ior point plot, 176
symbol size Ior point plotting, 176
plot options, 120
Plot structures, 280
plot3d, 345
PlotBuilder, 26
plots
package, 150
plottools, 151
plus, 10
pop-up, 117
power (`), 12
pressure, 137
preventing
evaluation, 132
print, 195
procedure
with a numerical result, 227
procedures
returning multiple results, 287
Python, 6

Quick ReIerence (help menu item), 43

quotation, 132, 132
R
radians, 121
range, 279
ready (status oI Maple), 22
rectangle
dashed (entry), 8
red, 151
reIactor
code, 129
replay
animation, 160
restart, 57, 57, 68, 272
reuse
oI soItware, 2
rhs, 120, 279
right arrow key, 13
right hand side, 61
right hand side (oI equation), 30
right hand side (right-click menu), 31, 35
right-click, 23, 320
optimization menu item, 337
S
SAGE, 4
scalingconstrained (option to plot), 156
scripting, 63
search Iield
Ior help, 88
select
entry, 61
item Irom a set, 30
select entry (right-click menu), 31, 35
semantic error, 19
semi-colon, 188
sequence, 15, 62, 140, 280, 285, 287
empty, 280
set, 62, 280
empty, 74, 280
sheep, 63
356 Index
shortcuts
keyboard, 42
Show Al Palettes, 42
simpliIy, 320, 321
simpliIy -~ assuming real (clickable operation), 108
simpliIy(....
trig) , 321
simulation, 1
single-quotes, 84
slash, 11
small steps, 67
Solo
Han, 132
solve, 61, 77, 101, 119, 279
a system oI equations, 93
by right-clicking or control-clicking, 24
numerical, 30
won't work with an assigned variable, 57
solving an equation, 155
speed
animation, 160
sqrt (square root), 107
square root
Irom palette, 28
Star Wars, 132
state
current oI a Maple session, 49
current oI Maple session, 48
state display
oI Maple, 22
statement
program, 188
stop playing
animation, 159
sublist, 93
subscript, 41
subtraction (-), 11
sum, 256
supercomputer, 2
suppress output, 195
suppressing output Irom a loop, 269
symbol
Ior point plotting, 176
symbolsize, 176, 176
T
table, 280
in an accumulative loop, 289
Table (Maple word processor Ieature), 43
temperature, 137
template
textual, 109
teraIlop, 3
test, 68
testing, 126
a program, 117, 117
Iunctions, 127
text entry mode, 40
text mode to math mode
can't switch back, 43
textual entry, 77
to (part oI Ior), 245
Tools-~Spellcheck, 43
trace, 272
triangle, 140
troubleshooting, 67
assignment, 55
Iunction deIinitions, 116
tables, 283
textual input, 101
unit conversion, 132
word processing, 43
true (Boolean value), 202, 213, 216, 237
truth values, 202
ttrace
execution, 269
U
unassign, 47, 47, 49, 57
undeIined, 324
underscore, 41
underscore (), 45
Unit Converter assistant, 136
units, 174
Units Calculator, 92
untrace, 272
V
value
oI an unassigned name, 46
values
Boolean, 213, 214, 217, 237
variable
programming, 45
variable oI diIIerentiation, 320
vertical range, 80
View (Maple menu), 42
Visual Basic, 24
visualization, 1, 1
W
Warning
computation interrupted, 236
premature end oI input, 258
variable is implicitly declared local, 228, 233
Index 357
what you see is what you get, 39
WYSIWYG, 39
Y
yard, 132, 132
358 Index