软件测试用例书籍,有效用例模式-软件测试用书

Preface

Use cases are a popular requirements modeling technique, yet people often struggle when

writing them. They understand the basic concepts of use cases, but find that actually writing

useful ones turns out to be harder than one would expect. One factor contributing to this difficulty

is that we lack objective criteria to help judge their quality. Many people find it difficult to

articulate the qualities of an effective use case.

This book examines the problems people encounter while writing use cases. It describes simple,

elegant and proven solutions to the specific problems of writing use cases on real projects. We

have identified approximately three-dozen patterns that people can use to evaluate their use

cases. We have based these patterns on the observable signs of quality that successful projects

tend to exhibit. Our goals are to provide a vocabulary for discussing and sharing these properties

with other people, provide advice for writing and organizing use cases effectively, and provide

some “diagnostics” for evaluating use cases. We want these patterns to become second nature.

Our hope is to find people saying, “Do we have a

SharedClearVision(95)?”

or “Does this use

case have

CompleteSingleGoal(132)?”

when discussing their use cases.

Audience

This book is intended for anyone who deals with use cases and wishes to learn more about them.

It assumes that you have a working knowledge of use cases, and have some experience writing

them. It is the follow-up to Alistair Cockburn’s Writing Effective Use Cases. If you are unfamiliar or

inexperienced with use cases, then we recommend that you read Alistair’s book first.

Use Cases are helpful for designing business processes, software based or not. We do not

intend for this book to be “yet another software development book”, written in “geek-speak” that is

indiscernible to all but the most technically gifted, and readable only by the most determined.

Yet, we understand that use cases are predominately a software development tool, and we, being

software developers, cannot help but focus on this area. We hope that this book helps all the

participants of a project understand the software development community, and we have tried to

illustrate this topic in a usable manner for all who work with use cases.

Copyright © 2001 Adolph, Bramble All Rights Reserved

1

Organization

This book is organized as a catalog of approximately three dozen patterns offering criteria for

evaluating the quality of use cases. Each pattern describes a specific guideline or “sign of

quality” that you can use to judge the caliber of a use case in a particular area. Because each

organization has its own culture and its own way of doing things, use cases that work well for one

organization may completely miss the mark in another. Patterns document generalized solutions

and as such provide a simple and yet effective mechanism for describing the characteristics of

quality use cases. As such, patterns can transcend organizational differences, allowing people to

tailor the solution to their specific needs.

We document our patterns using the style of Christopher Alexander, in which we define the

problem, the context in which it occurs and a solution. We find this format to be very readable,

and is more appropriate for our topic and diverse audience. Pattern names appear in bold type.

For example,

VisibleActorIntent(161)

and

EverUnfoldingStory(117)

are both patterns

described in this book. We include the page number in parenthesis after the pattern name to help

you locate its description.

The patterns are organized into categories. For example, chapter two describes organizational

patterns, chapter three describes process patterns, and chapter four describes the use case set

patterns.

Each pattern has one or more examples demonstrating either the benefit of implementing the

solution recommended by the pattern or the consequences of what happens when you don’t. We

based as many of these examples on live projects as we could; however, we sanitized them, to

protect the companies and people involved (quite rightly), as well as streamlined them for

readability. Moreover, we simplified many of them because real use cases are often long and

can be quite complicated, especially the ones demonstrating bad practices. We hope that you

find these examples useful. You may apply some of our other patterns to these samples, and find

ways to improve them.

We illustrate our patterns with a story that runs throughout the book, following a group of

developers from a national travel agency as they write some use cases for their new product, the

“Wings

Over the World”

travel reservation system. This story portrays an environment in which

many of the problems described in the book can occur and therefore provides a background for

discussing the patterns. The story also helps us provide examples that succinctly demonstrate a

pattern that may be difficult or impossible to illustrate with real examples. While the

Wings Over

the World

examples may be contrived, they are based on our experiences, and many of the

people, conversations, examples, and even the Red Eye flight are composites of real events, real

conversations, and real use cases.

Copyright © 2001 Adolph, Bramble All Rights Reserved

2

How to Use this Book

Reading a pattern catalog cover to cover is usually not most people’s idea of fun. Your first

reading of this book can be either to familiarize yourself with the patterns or serve as a tutorial for

what makes a quality use case. Either way, read chapter one, which gives you the background for

use cases, patterns, and the pattern categories.

If your objective is just to become familiar with the patterns, simply read the introduction of each

subsequent chapter, and then read the problem, context, and solution of each pattern. If you find

a pattern interesting, or particularly germane to your specific situation, then read it more closely.

Don’t feel that you need to fully understand one pattern before you examine another. Once you

have this knowledge, you can use the book as a reference. Store it on your shelf within easy

reach, so that you can look up a pattern later when you are have a problem writing a use case, or

you are reading one, and something just looks wrong.

This book can also be used as a tutorial to understand the signs of quality for a well-written use

case. After reading chapter one, skim through each chapter and read the

Wings Over the World

example to get a feel for the environment that gives rise to many of the problems addressed by

these patterns. Read the chapter introduction to understand the pattern category and get a brief

overview of each of the patterns in that chapter. Skim the patterns themselves, read the problem

statement, and the solution. Take a look at the examples and make sure you understand how the

pattern improves the quality of the use case or the process to create the use case.

A great approach to learning these patterns is to hold “brown bag” or “lunch and learn” sessions,

which many people use to learn patterns. To do this, have someone present one or two patterns

at a lunchtime seminar every week. Then discuss the pattern, its intention and its tradeoffs, then

look at the examples, and discuss how they compare with your organization’s use cases.

What About UML?

The Unified Modeling Language is the software industry’s latest attempt to sign a modeling

language non-proliferation treaty with itself. When most people think of use cases and UML they

simply think of diagramming tools. Others think about UML defining semantics for include and

extends

in UML parlance. Some are UML enthusiasts, and others are not. One thing on which we

all agree is that rigidly following UML semantics does not guarantee quality use cases. Many

people who have never even heard of UML write great use cases. However, UML is an important

influence on use cases and therefore we have included guidelines for the effective use of UML

with the relevant patterns.

Copyright © 2001 Adolph, Bramble All Rights Reserved

3

Why Don’t We Use a Single Use Case Template for the Examples?

With the exception of the

Wings Over the World

use cases, most of the use cases presented

throughout this book are sanitized versions of real production use cases. Many of these use

cases appear different because they are the works of different writers. We could have created a

uniform template to present all the examples, but we decided not to because we did not want to

advocate a specific use case template. There are enough around now, and just as we did not

want to imply there is only one true way to write a use case, we also did not want to imply there is

only one true use case template.

We chose to use different styles for our use case diagrams for the same reason, using a variety

of different drawing and CASE tools to generate them.

The Agile Software Development Series.

This book is one in a collection of books, the Agile Software Development Series that highlights

light, human-powered software development techniques. Some books discuss a single technique,

some a single role on the project, and some discuss team collaboration issues.

The book series is based on two common principles:

Different projects have different needs. Systems have different characteristics, and are

built by teams of differing sizes, containing people having differing values and priorities. It

cannot be possible to describe the one, best way of producing software;

Focusing on skills, communication and community allows the project to be more effective

and more agile than focusing on processes and plans.

Accordingly, the book series runs along three main tracks:

How one person can improve their effectiveness on projects through particular

techniques;

How a group of people can improve their combined effectiveness through various group

techniques; and

Examples of particular, successful methodologies that you can use as models for your

own adaptations.

Agile Software Development elaborates the ideas of software development as a cooperative

game, of methodology as coordination of culture, and of methodology families. It separates the

different aspects of methodologies, techniques from activities, work products and standards.

Copyright © 2001 Adolph, Bramble All Rights Reserved

4