<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8"/>
<title>JUnit FAQ</title>
<style type="text/css">
body {
font-style: normal;
font-weight: normal;
margin-left: 10pt;
margin-right: 10pt;
}
a {
text-decoration: none;
}
a:hover {
text-decoration: underline;
}
.header {
color: black;
font-size: 125%;
font-weight: bold;
background: #33ff33;
padding-top: 2px;
padding-bottom: 2px;
padding-left: 5px;
margin-top: 25px;
}
.code {
background: white;
border-left: 5px solid #33ff33;
}
.code-red {
background: white;
border-left: 5px solid #cc0000;
}
</style>
</head>
<body>
<h1>
<font color="#33ff33">J</font><font color="#cc0000">U</font>nit FAQ
</h1>
<hr size="1"/>
<!--
Summary
-->
<p>
<i>
JUnit is an open source Java testing framework used to write and run
automated test.
<br/>
It is an instance of the xUnit architecture for unit
testing frameworks.
</i>
</p>
<hr size="1"/>
<p>
Editor: <a href="mailto:mike@clarkware.com">Mike Clark</a>,
<a href="http://www.clarkware.com">Clarkware Consulting</a>
</p>
<p>
Last modified on December 28, 2004
</p>
<hr/>
<!--
Table of Contents
-->
<div class="header">
Table of Contents
</div>
<ol>
<li>
<p>
<b><a href="#faqinfo">FAQ Info</a></b>
</p>
<ol>
<li><a href="#faqinfo_1">Who is responsible for this FAQ?</a></li>
<li><a href="#faqinfo_2">How can I contribute to this FAQ?</a></li>
<li><a href="#faqinfo_3">Where do I get the latest version of
this FAQ?</a></li>
</ol>
</li>
<li>
<p>
<b><a href="#overview">Overview</a></b>
</p>
<ol>
<li><a href="#overview_1">What is JUnit?</a></li>
<li><a href="#overview_2">Where is the JUnit home page?</a></li>
<li><a href="#overview_3">Where are the JUnit mailing lists and
forums?</a></li>
<li><a href="#overview_4">Where is the JUnit documentation?</a></li>
<li><a href="#overview_5">Where can I find articles on JUnit?</a></li>
<li><a href="#overview_6">What's the latest news on JUnit?</a></li>
<li><a href="#overview_7">How is JUnit licensed?</a></li>
<li><a href="#overview_8">What awards has JUnit won?</a></li>
</ol>
</li>
<li>
<p>
<b><a href="#started">Getting Started</a></b>
</p>
<ol>
<li><a href="#started_1">Where do I download JUnit?</a></li>
<li><a href="#started_2">How do I install JUnit?</a></li>
<li><a href="#started_3">How do I uninstall JUnit?</a></li>
<li><a href="#started_4">How do I ask questions?</a></li>
<li><a href="#started_5">How do I submit bugs, patches, or
feature requests?</a></li>
</ol>
</li>
<li>
<p>
<b><a href="#tests">Writing Tests</a></b>
</p>
<ol>
<li><a href="#tests_1">How do I write and run a simple test?</a></li>
<li><a href="#tests_2">How do I use a test fixture?</a></li>
<li><a href="#tests_3">Why isn't my test fixture being run?</a></li>
<li><a href="#tests_4">How do I test a method that doesn't
return anything?</a></li>
<li><a href="#tests_5">Under what conditions should I test get()
and set() methods?</a></li>
<li><a href="#tests_6">Under what conditions should I not test
get() and set() methods?</a></li>
<li><a href="#tests_7">How do I write a test that passes when an
expected exception is thrown?</a></li>
<li><a href="#tests_8">How do I write a test that fails when an
unexpected exception is thrown?</a></li>
<li><a href="#tests_9">What's the difference between a failure
and an error?</a></li>
<li><a href="#tests_10">How do I test protected methods?</a></li>
<li><a href="#tests_11">How do I test private methods?</a></li>
<li><a href="#tests_12">Why does JUnit only report the first
failure in a single test?</a></li>
<li><a href="#tests_13">In Java 1.4, 'assert' is a
keyword. Won't this conflict with JUnit's assert()
method?</a></li>
<li><a href="#tests_14">How do I test things that must be run in
a J2EE container (e.g. servlets, EJBs)?</a></li>
<li><a href="#tests_15">Do I need to write a TestCase class for
every class I need to test?</a></li>
<li><a href="#tests_16">Is there a basic template I can use to
create a test?</a></li>
<li><a href="#tests_17">How do I write a test for an abstract
class?</a></li>
<li><a href="#tests_18">When are tests garbage collected?</a></li>
</ol>
</li>
<li>
<p>
<b><a href="#organize">Organizing Tests</a></b>
</p>
<ol>
<li><a href="#organize_1">Where should I put my test files?</a></li>
<li><a href="#organize_2">How do I write a test suite for all of
my tests?</a></li>
<li><a href="#organize_3">How can I run setUp() and tearDown()
code once for all of my tests?</a></li>
</ol>
</li>
<li>
<p>
<b><a href="#running">Running Tests</a></b>
</p>
<ol>
<li><a href="#running_1">What CLASSPATH settings are needed to
run JUnit?</a></li>
<li><a href="#running_2">Why do I get a NoClassDefFoundError
when trying to test JUnit or run the samples?</a>
</li>
<li><a href="#running_3">How do I run the JUnit GUI test runner?</a>
</li>
<li><a href="#running_4">How do I run JUnit from my command window?</a>
</li>
<li><a href="#running_5">How do I run JUnit using Ant?</a>
</li>
<li><a href="#running_6">How do I use Ant to create HTML test reports?</a>
</li>
<li><a href="#running_7">How do I pass command-line arguments to a test execution?</a>
</li>
<li><a href="#running_8">Why do I get an error (ClassCastException or
LinkageError) using the GUI TestRunners?</a>
</li>
<li><a href="#running_9">Why do I get a LinkageError when using
XML interfaces in my TestCase?</a>
</li>
<li><a href="#running_10">Why do I get a ClassCastException when I
use narrow() in an EJB client TestCase?</a>
</li>
<li><a href="#running_11">Why do I get the warning "AssertionFailedError: No
tests found in XXX" when I run my test?</a>
</li>
<li><a href="#running_12">Why do I see "Unknown Source" in the stack trace of
a test failure, rather than the source file's line number?</a>
</li>
<li><a href="#running_13">Why does the "excluded.properties" trick not
work when running JUnit's GUI from inside my favorite IDE?</a>
</li>
<li><a href="#running_14">How do I get one test suite to invoke another?</a>
</li>
<li><a href="#running_15">How do I organize all test classes in a TestSuite
automatically and not use or manage a TestSuite explicitly?</a>
</li>
</ol>
</li>
<li>
<p>
<b><a href="#best">Best Practices</a></b>
</p>
<ol>
<li><a href="#best_1">When should tests be written?</a></li>
<li><a href="#best_2">Do I have to write a test for
everything?</a></li>
<li><a href="#best_3">How simple is 'too simple to break'?</a></li>
<li><a href="#best_4">How often should I run my tests?</a></li>
<li><a href="#best_5">What do I do when a defect is reported?</a></li>
<li><a href="#best_6">Why not just use System.out.println()?</a></li>
<li><a href="#best_7">Why not just use a debugger?</a></li>
</ol>
</li>
<li>
<p>
<b><a href="#extend">Extending JUnit</a></b>
</p>
<ol>
<li><a href="#extend_1">How do I extend JUnit?</a></li>
<li><a href="#extend_2">What kinds of extensions are
available?</a></li>
</ol>
</li>
<li>
<p>
<b><a href="#misc">Miscellaneous</a></b>
</p>
<ol>
<li><a href="#misc_1">How do I integrate JUnit with my IDE?</a></li>
<li><a href="#misc_2">How do I launch a debugger when a test
fails?</a></li>
<li><a href="#misc_3">Where can I find unit testing frameworks
similar to JUnit for other languages?</a></li>
</ol>
</li>
</ol>
<!--
FAQ Info
-->
<div class="header">
<a name="faqinfo">FAQ Info</a>
</div>
<ol>
<li>
<p>
<b><a name="faqinfo_1">Who is responsible for this FAQ?</a></b>
</p>
<p>
The current version of this FAQ is maintained by <a
href="mailto:mike@clarkware.com">Mike Clark</a>.
</p>
<p>
Most of the wisdom contained in this FAQ comes from the
collective insights and hard-won experiences of the many good
folks who participate on the JUnit mailing list and the JUnit
community at large.
</p>
<p>
If you see your genius represented anywhere in this FAQ without
due credit to you, please send me an email and I'll make things
right.
</p>
</li>
<li>
<p>
<b><a name="faqinfo_2">How can I contribute to this FAQ?</a></b>
</p>
<p>
Your contributions to this FAQ are greatly appreciated! The
JUnit community thanks you in advance.
</p>
<p>
To contribute to this FAQ, simply write a JUnit-related question
and answer, then send the unformatted text to <a
href="mailto:mike@clarkware.com">Mike Clark</a>. Corrections to
this FAQ are always appreciated, as well.
</p>
<p>
No reasonable contribution will be denied. Your name will
always appear along with any contribution you make.
</p>
</li>
<li>
<p>
<b><a name="faqinfo_3">Where do I get the latest version of this
FAQ?</a></b>
</p>
<p>
The master copy of this FAQ is available at <a
href="http://junit.sourceforge.net/doc/faq/faq.htm">http://junit.sourceforge.net/doc/faq/faq.htm</a>.
</p>
<p>
The JUnit distribution also includes this FAQ in
the <code>doc</code> directory.
</p>
</li>
</ol>
<!--
Overview
-->
<div class="header">
<a name="overview">Overview</a>
</div>
<ol>
<li>
<p>
<b><a name="overview_1">What is JUnit?</a></b>
</p>
<p>
JUnit is an open source Java testing framework used to write and
run repeatable tests. It is an instance of the xUnit
architecture for unit testing frameworks.
</p>
<p>
JUnit features include:
</p>
<ul>
<li>Assertions for testing expected results</li>
<li>Test fixtures for sharing common test data</li>
<li>Test suites for easily organizing and running tests</li>
<li>Graphical and textual test runners</li>
</ul>
<p>
JUnit was originally written by Erich Gamma and Kent Beck.
</p>
</li>
<li>
<p>
<b><a name="overview_2">Where is the JUnit home page?</a></b>
</p>
<p>
The official JUnit home page is <a
href="http://junit.org">http://junit.org</a>.
</p>
</li>
<li>
<p>
<b><a name="overview_3">Where are the JUnit mailing lists and
forums?</a></b>
</p>
<p>
There are 3 mailing lists dedicated to everything JUnit:
</p>
<ul>
<li>
<a
href="http://lists.sourceforge.net/lists/listinfo/junit-announce">JUnit
announcements</a>
</li>
<li>
<a href="http://groups.yahoo.com/group/junit/">JUnit user
list</a>
</li>
<li>
<a
href="http://lists.sourceforge.net/lists/listinfo/junit-devel">JUnit
developer list</a>
</li>
</ul>
<p>
You can <a
href="http://groups.yahoo.com/group/junit/">search</a> the JUnit
user list archives for answers to frequently asked questions not
included here.
</p>
<p>
There is also a <a
href="http://www.jguru.com/forums/home.jsp?topic=JUnit">jGuru
discussion forum</a> dedicated to everything JUnit.
</p>
</li>
<li>
<p>
<b><a name="overview_4">Where is the JUnit
documentation?</a></b>
</p>
<p>
The following documents are included in the JUnit distribution
in the <code>doc</code> directory:
</p>
<ul>
<li>
<a
href="http://junit.sourceforge.net/doc/testinfected/testing.htm">JUnit
Test Infected: Programmers Love Writing Tests</a>
</li>
<li>
<a href="http://junit.sourceforge.net/doc/cookbook/cookbook.htm">JUnit
Cookbook</a>
</li>
<li>
<a
href="http://junit.sourceforge.net/doc/cookstour/cookstour.htm">JUnit
- A Cook's Tour</a>
</li>
<li>
<a href="http://junit.sourceforge.net/doc/faq/faq.htm">JUnit
FAQ</a>
</li>
</ul>
</li>
<li>
<p>
<b><a name="overview_5">Where can I find articles on
JUnit?</a></b>
</p>
<p>
The JUnit home page maintains a list of <a
href="http://www.junit.org/news/article/index.htm">JUnit
articles</a>.
</p>
</li>
<li>
<p>
<b><a name="overview_6">What's the latest news on JUnit?</a></b>
</p>
<p>
The JUnit home page publishes the <a
href="http://www.junit.org/news/index.htm">latest JUnit
news</a>.
</p>
</li>
<li>
<p>
<b><a name="overview_7">How is JUnit licensed?</a></b>
</p>
<p>
JUnit is <a href="http://www.opensource.org/">Open Source
Software</a>, released under <a
href="http://oss.software.ibm.com/developerworks/oss/license-cpl.html">IBM's
Common Public License Version 0.5</a> and hosted on <a
href="http://sourceforge.net/projects/junit/">SourceForge</a>.
</p>
</li>
<li>
<p>
<b><a name="overview_8">What awards has JUnit won?</a></b>
</p>
<ul>
<li>
<p>
<a href="http://www.javaworld.com/javaworld/jw-03-2002/jw-0326-awards.html">2002 JavaWorld Editors' Choice Awards (ECA)</a>
</p>
<p>
Best Java Performance Monitoring/Testing Tool
</p>
</li>
<li>
<p>
<a
href="http://www.javaworld.com/javaworld/jw-06-2001/j1-01-awards.html">2001
JavaWorld Editors' Choice Awards (ECA)</a>
</p>
<p>
Best Java Performance Monitoring/Testing Tool
</p>
</li>
</ul>
</li>
</ol>
<!--
Getting Started
-->
<div class="header">
<a name="started">Getting Started</a>
</div>
<ol>
<li>
<p>
<b><a name="started_1">Where do I download JUnit?</a></b>
</p>
<p>
The latest version of JUnit is available on <a
href="http://sourceforge.net/project/showfiles.php?group_id=15278">SourceForge</a>.
</p>
</li>
<li>
<p>
<b><a name="started_2">How do I install JUnit?</a></b>
</p>
<ol>
<li>
<p>
First, <a
href="http://sourceforge.net/project/showfiles.php?group_id=15278">download</a>
the
latest version of JUnit, referred to below
as <code>junit.zip</code>.
</p>
</li>
<li>
<p>
Then install JUnit on your platform of choice:
</p>
<p>
<u>Windows</u>
</p>
<p>
To install JUnit on Windows, follow these steps:
</p>
<ol>
<li>
<p>
Unzip the <code>junit.zip</code> distribution file to
a directory referred to as <code>%JUNIT_HOME%</code>.
</p>
</li>
<li>Add JUnit to the classpath:
<p>
<code>set CLASSPATH=%CLASSPATH%;%JUNIT_HOME%\junit.jar</code>
</p>
</li>
</ol>
<p>
<u>Unix (bash)</u>
</p>
<p>
To install JUnit on Unix, follow these steps:
</p>
<ol>
<li>
<p>
Unzip the <code>junit.zip</code> distribution file to
a directory referred to as <code>$JUNIT_HOME</code>.
</p>
</li>
<li>
<p>
Add JUnit to the classpath:
</p>
<p>
<code>export CLASSPATH=$CLASSPATH:$JUNIT_HOME/junit.jar</code>
</p>
</li>
</ol>
</li>
<li>
<p>
<i>(Optional)</i> Unzip
the <code>$JUNIT_HOME/src.jar</code> file.
</p>
</li>
<li>
<p>
Test the installation by using either the textual or
graphical test runner to run the sample tests distributed
with JUnit.
</p>
<p>
<i>Note: The sample tests are not contained in
the <code>junit.jar</code>, but in the installation
directory directly. Therefore, make sure that the JUnit
installation directory is in the CLASSPATH.</i>
</p>
<p>
For the textual TestRunner, type:
</p>
<div>
<blockquote><code>
java junit.textui.TestRunner junit.samples.AllTests
</code></blockquote>
</div>
<p>
For the graphical TestRunner, type:
</p>
<div>
<blockquote><code>
java junit.swingui.TestRunner junit.samples.AllTests
</code></blockquote>
</div>
<p>
All the tests should pass with an "OK" (textual) or a
green bar (graphical).
</p>
<p>
If the tests don't pass, verify
that <code>junit.jar</code> is in the CLASSPATH.
</p>
</li>
<li>
<p>
Finally, <a href="#overview_4">read</a> the documentation.
</p>
</li>
</ol>
</li>
<li>
<p>
<b><a name="started_3">How do I uninstall JUnit?</a></b>
</p>
<ol>
<li>
<p>
Delete the directory structure where you unzipped the JUnit
distribution.
</p>
</li>
<li>
<p>
Remove <code>junit.jar</code> from the classpath
</p>
</li>
</ol>
<p>
JUnit does not modify the registry so simply removing all the
files will fully uninstall it.
</p>
</li>
<li>
<p>
<b><a name="started_4">How do I ask questions?</a></b>
</p>
<p>
Questions that are not answered in the <a
href="http://junit.sourceforge.net/doc/faq/faq.htm">FAQ</a> or
in the <a href="#overview_4">documentation</a> should be posted
to the <a
href="http://www.jguru.com/forums/home.jsp?topic=JUnit">jGuru
discussion forum</a> or the <a
href="http://groups.yahoo.com/group/junit/">JUnit user mailing
list</a>.
</p>
<p>
Please stick to technical issues on the discussion forum and
mailing lists. Keep in mind that these are public, so
do <b>not</b> include any confidental information in your
questions!
</p>
<p>
You should also read <a
href="http://www.catb.org/~esr/faqs/smart-questions.html">"How
to ask questions the smart way"</a> by Eric Raymond before
participating in the discussion forum and mailing lists.
</p>
<p>
<i> NOTE:
<br/> Please do NOT submit bugs, patches, or feature requests
to the discussion forum or mailing lists. <br/> Refer instead
to <a href="#started_5">"How do I submit bugs, patches, or
feature requests?"</a>.
</i>
</p>
</li>
<li>
<p>
<b><a name="started_5">How do I submit bugs, patches, or
feature requests?</a></b>
</p>
<p>
JUnit celebrates programmers testing their own software. In this
spirit, bugs, patches, and feature requests that include JUnit
tests have a better chance of being addressed than those
without.
</p>
<p>
JUnit is forged on <a
href="http://sourceforge.net/projects/junit">SourceForge</a>.
Please use the tools provided by SourceForge for your
submissions.
</p>
</li>
</ol>
<!--
Writing Tests
-->
<div class="header">
<a name="tests">Writing Tests</a>
</div>
<ol>
<li>
<p>
<b><a name="tests_1"></a>How do I write and run a simple test?</b>
</p>
<ol>
<li>
<p>
Create a subclass of <code>TestCase</code>:
</p>
<div class="code">
<pre><code>
package junitfaq;
import java.util.*;
import junit.framework.*;
public class SimpleTest extends TestCase {
</code></pre>
</div>
</li>
<li>
<p>
Write a test method to assert expected results on the
object under test:
</p>
<div class="code">
<pre><code>
public void testEmptyCollection() {
Collection collection = new ArrayList();
assertTrue(collection.isEmpty());
}
</code></pre>
</div>
</li>
<li>
<p>
Write a <code>suite()</code> method that uses reflection
to dynamically create a test suite containing all
the <code>testXXX()</code> methods:
</p>
<div class="code">
<pre><code>
public static Test suite() {
return new TestSuite(SimpleTest.class);
}
</code></pre>
</div>
</li>
<li>
<p>
Write a <code>main()</code> method to conveniently run the
test with the textual test runner:
</p>
<div class="code">
<pre><code>
public static void main(String args[]) {
junit.textui.TestRunner.run(suite());
}
}
</code></pre>
</div>
</li>
<li>
<p>
Run the test:
</p>
<ul>
<li>
<p>
To run the test with the textual test runner used
in <code>main()</code>, type:
</p>
<div>
<blockquote><code>
java junitfaq.SimpleTest
</code></blockquote>
</div>
<p>
The passing test results in the following textual output:
</p>
<div>
<blockquote>
<pre><code>
.
Time: 0
OK (1 tests)
</code></pre>
</blockquote>
</div>
</li>
<li>
<p>
To run the test with the graphical test runner, type:
</p>
<div>
<blockquote><code>
java junit.swingui.TestRunner junitfaq.SimpleTest
</code></blockquote>
</div>
<p>
The passing test results in a green bar displayed in
the graphical UI.
</p>
</li>
</ul>
</li>
</ol>
</li>
<li>
<p>
<b><a name="tests_2"></a>How do I use a test fixture?</b>
</p>
<p>
A test fixture is useful if you have two or more tests for a
common set of objects. Using a test fixture avoids duplicating
the test code necessary to initialize and cleanup those common
objects for each test.
</p>
<p>
Tests can use the objects (variables) in a test fixture, with
each test invoking different methods on objects in the fixture
and asserting different expected results. Each test runs in its
own test fixture to isolate tests from the changes made by other
tests. That is, tests don't share the state of objects in the
test fixture. Because the tests are isolated, they can be run
in any order.
</p>
<p>
To create a test fixture, define a <code>setUp()</code> method
that initializes common objects and a <code>tearDown()</code>
method to cleanup those objects. The JUnit framework
automatically invokes the <code>setUp()</code> method before
each test is run and the <code>tearDown()</code> method after
each test is run.
</p>
<p>
The following test uses a test fixture to initialize and cleanup
a common <code>Collection</code> object such that both tests are
isolated from changes made by the other:
</p>
<div class="code">
<pre><code>
package junitfaq;
import junit.framework.*;
import java.util.*;
public class SimpleTest extends TestCase {
private Collection collection;
protected void setUp() {
collection = new ArrayList();
}
protected void tearDown() {
collection.clear();
}
public void testEmptyCollection() {
assertTrue(collection.isEmpty());
}
public void testOneItemCollection() {
collection.add("itemA");
assertEquals(1, collection.size());
}
}
</code></pre>
</div>
<p>
Given this test, the methods might execute in the following
order:
</p>
<div>
<blockquote>
<pre><code>
setUp()
testOneItemCollection()
tearDown()
setUp()
testEmptyCollection()
tearDown()
</code></pre>
</blockquote>
</div>
<p>
The ordering of test-method invocations is not guaranteed,
so <code>testEmptyCollection()</code> might be executed
before <code>testOneItemCollection()</code>. This is why the
test methods themselves must be written to be independent of one
another. What is guaranteed is that <code>setUp()</code> will
execute before each test method and <code>tearDown()</code> will
execute after each test method.
</p>
</li>
<li>
<p>
<b><a name="tests_3"></a>Why isn't my test fixture being run?</b>
</p>
<p>
Make sure the test fixture methods are defined as follows,
noting that both method names are case sensitive:
</p>
<div>
<blockquote><pre>
protected void setUp() {
// initialization code
}
protected void tearDown() {
// cleanup code
}
</pre></blockquote>
</div>
</li>
<li>
<p>
<b><a name="tests_4"></a>How do I test a method that doesn't
return anything?</b>
</p>
<p>
<i>(Submitted by: Dave Astels)</i>
</p>
<p>
Often if a method doesn't return a value, it will have some side
effect. Actually, if it doesn't return a value AND doesn't have
a side effect, it isn't doing anything.
</p>
<p>
There may be a way to verify that the side effect actually
occurred as expected. For example, consider
the <code>add()</code> method in the Collection classes. There
are ways of verifying that the side effect happened (i.e. the
object was added). You can check the size and assert that it is
what is expected:
</p>
<div class="code">
<pre><code>
public void testCollectionAdd() {
Collection collection = new ArrayList();
assertEquals(0, collection.size());
collection.add("itemA");
assertEquals(1, collection.size());
collection.add("itemB");
assertEquals(2, collection.size());
}
</code></pre>
</div>
<p>
Another approach is to make use of <a
href="http://www.mockobjects.com">MockObjects</a>.
</p>
<p>
A related issue is to design for testing. For example, if you
have a method that is meant to output to a file, don't pass in a
filename, or even a <code>FileWriter</code>. Instead, pass in
a <code>Writer</code>. That way you can pass in
a <code>StringWriter</code> to capture the output for testing
purposes. Then you can add a method
(e.g. <code>writeToFileNamed(String filename)</code>) to
encapsulate the <code>FileWriter</code> creation.
</p>
</li>
<li>
<p>
<b><a name="tests_5"></a>Under what conditions should I test
get() and set() methods?</b>
</p>
<p>
Unit tests are intended to alleviate fear that something might
break. If you think a <code>get()</code> or <code>set()</code>
method could reasonably break, or has in fact contributed to a
defect, then by all means write a test.
</p>
<p>
In short, test until you're confident. What you choose to test
is subjective, based on your experiences and confidence level.
Remember to be practical and maximize your testing investment.
</p>
<p>
Refer also to <a href="#best_3">"How simple is 'too simple to
break'?"</a>.
</p>
</li>
<li>
<p>
<b><a name="tests_6"></a>Under what conditions should I not test
get() and set() methods?</b>
</p>
<p>
<i>(Submitted by: J. B. Rainsberger)</i>
</p>
<p>
Most of the time, get/set methods just can't break, and if they
can't break, then why test them? While it is usually better to
test more, there is a definite curve of diminishing returns on
test effort versus "code coverage". Remember the maxim: "Test
until fear turns to boredom."
</p>
<p>
Assume that the <code>getX()</code> method only does "return x;"
and that the <code>setX()</code> method only does "this.x =
x;". If you write this test:
</p>
<div>
<blockquote><pre>
testGetSetX() {
setX(23);
assertEquals(23, getX());
}
</pre></blockquote>
</div>
<p>
then you are testing the equivalent of the following:
</p>
<div>
<blockquote><pre>
testGetSetX() {
x = 23;
assertEquals(23, x);
}
</pre></blockquote>
</div>
<p>
or, if you prefer,
</p>
<div>
<blockquote><pre>
testGetSetX() {
assertEquals(23, 23);
}
</pre></blockquote>
</div>
<p>
At this point, you are testing the Java compiler, or possibly
the interpreter, and not your component or application. There is
generally no need for you to do Java's testing for them.
</p>
<p>
If you are concerned about whether a property has already been
set at the point you wish to call <code>getX()</code>, then you
want to test the constructor, and not the <code>getX()</code>
method. This kind of test is especially useful if you have
multiple constructors:
</p>
<div>
<blockquote><pre>
testCreate() {
assertEquals(23, new MyClass(23).getX());
}
</pre></blockquote>
</div>
</li>
<li>
<p>
<b><a name="tests_7"></a>How do I write a test that passes when
an expected exception is thrown?</b>
</p>
<p>
Catch the exception within the test method. If it isn't thrown,
call the <code>fail()</code> method to signal the failure of the
test.
</p>
<p>
The following is an example test that passes when the
expected <code>IndexOutOfBoundsException</code> is raised:
</p>
<div class="code">
<pre><code>
public void testIndexOutOfBoundsException() {
ArrayList emptyList = new ArrayList();
try {
Object o = emptyList.get(0);
fail("Should raise an IndexOutOfBoundsException");
} catch (IndexOutOfBoundsException expected) {
assertTrue(true);
}
}
</code></pre>
</div>
</li>
<li>
<p>
<b><a name="tests_8"></a>How do I write a test that fails when
an unexpected exception is thrown?</b>
</p>
<p>
Declare the exception in the <code>throws</code> clause of the
test method and don't catch the exception within the test
method. Uncaught exceptions will cause the test to fail with an
error.
</p>
<p>
The following is an example test that fails when
the <code>IndexOutOfBoundsException</code> is raised:
</p>
<div class="code-red">
<pre><code>
public void testIndexOutOfBoundsExceptionNotRaised()
throws IndexOutOfBoundsException {
ArrayList emptyList = new ArrayList();
Object o = emptyList.get(0);
}
</code></pre>
</div>
</li>
<li>
<p>
<b><a name="tests_9"></a>What's the difference between a failure
and an error?</b>
</p>
<p>
Assertions are used to check for the possibility of failures,
therefore failures are anticipated. Errors are unanticipated
problems resulting in uncaught exceptions being propagated from
a JUnit test method.
</p>
<p>
In the following example, the <code>FileNotFoundException</code>
is expected and checked with an assertion. If the expected
exception is not raised, then a failure is produced. If any
other unexpected <code>IOException</code> or unchecked exception
(e.g. <code>NullPointerException</code>) is raised, the JUnit
framework catches the exception and signals an error.
</p>
<div class="code">
<pre><code>
public void testNonexistentFileRead() throws IOException {
try {
File file = new File("doesNotExist.txt");
FileReader reader = new FileReader(file);
assertEquals('a', (char)reader.read());
fail("Read from a nonexistent file?!");
} catch (FileNotFoundException expected) {
assertTrue(true);
}
}
</code></pre>
</div>
<p>
In the following example, an <code>IOException</code> is not
expected. The JUnit framework will signal an error if
an <code>IOException</code>
(e.g. <code>FileNotFoundException</code>) or any unchecked
exception (e.g. <code>NullPointerException</code>) is raised.
</p>
<div class="code">
<pre><code>
public void testExistingFileRead() throws IOException {
// exists.txt created in setup(), perhaps
File file = new File("exists.txt");
FileReader reader = new FileReader(file);
assertEquals('a', (char)reader.read());
}
</code></pre>
</div>
<p>
Both failures and errors will cause the test to fail. However,
it is useful to differentiate between failures and errors
because the debugging process is slightly different.
</p>
<p>
In the first example, the use of <code>fail()</code> will not
generate a complete stack trace including the method that raised
the exception. In this case that's sufficient since we
anticipate that the exception will be raised. If it's not
raised, then it's a problem with the test itself.
</p>
<p>
In the second example, the JUnit framework catches the exception
and generates an error with a complete stack trace for the
exception. Since we don't expect this exception to be raised, a
complete stack trace is useful in debugging why it was raised.
</p>
</li>
<li>
<p>
<b><a name="tests_10"></a>How do I test protected methods?</b>
</p>
<p>
Place your tests in the same package as the classes under test.
</p>
<p>
Refer to <a href="#organize_1">"Where should I put my test
files?"</a> for examples of how to organize tests for protected
method access.
</p>
</li>
<li>
<p>
<b><a name="tests_11"></a>How do I test private methods?</b>
</p>
<p>
Testing private methods may be an indication that those methods
should be moved into another class to promote reusability.
</p>
<p>
But if you must...
</p>
<p>
You can use reflection to subvert the access control mechanism.
If you are using JDK 1.3 or higher you can use the <a
href="http://groups.yahoo.com/group/junit/files/src/PrivilegedAccessor.java">PrivilegedAccessor</a>
class. Examples of how to use this class are available in <a
href="http://groups.yahoo.com/group/junit/files/src/PrivilegedAccessorTest.java">PrivilegedAccessorTest</a>.
</p>
</li>
<li>
<p>
<b><a name="tests_12"></a>Why does JUnit only report the first
failure in a single test?</b>
</p>
<p>
<i>(Submitted by: J. B. Rainsberger)</i>
</p>
<p>
Reporting multiple failures in a single test is generally a sign
that the test does too much, compared to what a unit test ought
to do. Usually this means either that the test is really a
functional/acceptance/customer test or, if it is a unit test,
then it is too big a unit test.
</p>
<p>
JUnit is designed to work best with a number of small tests. It
executes each test within a separate instance of the test
class. It reports failure on each test. Shared setup code is
most natural when sharing between tests. This is a design
decision that permeates JUnit, and when you decide to report
multiple failures per test, you begin to fight against
JUnit. This is not recommended.
</p>
<p>
Long tests are a design smell and indicate the likelihood of a
design problem. Kent Beck is fond of saying in this case that
"there is an opportunity to learn something about your design."
We would like to see a pattern language develop around these
problems, but it has not yet been written down.
</p>
<p>
Finally, note that a single test with multiple assertions is
isomorphic to a test case with multiple tests:
</p>
<p>
One test method, three assertions:
</p>
<div>
<blockquote><pre><code>
public class MyTestCase extends TestCase {
public void testSomething() {
// Set up for the test, manipulating local variables
assertTrue(condition1);
assertTrue(condition2);
assertTrue(condition3);
}
}
</code></pre></blockquote>
</div>
<p>
Three test methods, one assertion each:
</p>
<div>
<blockquote><pre><code>
public class MyTestCase extends TestCase {
// Locale variables become instance variables
protected void setUp() {
// Set up for the test, manipulating instance variables
}
public void testCondition1() {
assertTrue(condition1);
}
public void testCondition2() {
assertTrue(condition2);
}
public void testCondition3() {
assertTrue(condition3);
}
}
</code></pre></blockquote>
</div>
<p>
The resulting tests use JUnit's natural execution and reporting
mechanism and, failure in one test does not affect the execution
of the other tests. You generally want exactly one test to fail
for any given bug, if you can manage it.
</p>
</li>
<li>
<p>
<b><a name="tests_13"></a>In Java 1.4, 'assert' is a
keyword. Won't this conflict
with JUnit's <code>assert()</code> method?</b>
</p>
<p>
<i>(Submitted by: David Stagner)</i>
</p>
<p>
JUnit 3.7 deprecated <code>assert()</code> and replaced it
with <code>assertTrue()</code>, which works exactly the same
way.
</p>
<p>
Simply upgrade your JUnit to version 3.7 or higher and change
all <code>assert()</code> calls in your existing tests
to <code>assertTrue()</code>.
</p>
</li>
<li>
<p>
<b><a name="tests_14"></a>How do I test things that must be run
in a J2EE container (e.g. servlets, EJBs)?</b>
</p>
<p>
Refactoring J2EE components to delegate functionality to other
objects that don't have to be run in a J2EE container will
improve the design and testability of the software.
</p>
<p>
<a href="http://jakarta.apache.org/cactus/index.html">Cactus</a>
is an open source JUnit extension that can be used to test J2EE
components in their natural environment.
</p>
</li>
<li>
<p>
<b><a name="tests_15"></a>Do I need to write
a <code>TestCase</code> class for every class I need to
test?</b>
</p>
<p>
<i>(Submitted by: J. B. Rainsberger)</i>
</p>
<p>
No. It is a convention to start with one <code>TestCase</code>
class per class under test, but it is not necessary.
</p>
<p>
<code>TestCase</code> classes only provide a way to organize
tests, nothing more. Generally you will start with
one <code>TestCase</code> class per class under test, but then
you may find that a small group of tests belong together with
their own common test fixture.[1] In this case, you may move
those tests to a new <code>TestCase</code> object. This is a
simple object-oriented refactoring: separating responsibilities
of an object that does too much.
</p>
<p>
Another point to consider is that the <code>TestSuite</code> is
the smallest execution unit in JUnit: you cannot execute
anything smaller than a TestSuite at one time without changing
source code. In this case, you probably do not want to put tests
in the same <code>TestCase</code> class unless they somehow
"belong together". If you have two groups of tests that you
think you'd like to execute separately from one another, it is
wise to place them in separate <code>TestCase</code> classes.
</p>
<p>
<i>
[1] A test fixture is a common set of test data and
collaborating objects shared by many tests. Generally they are
implemented as instance variables in the <code>TestCase</code>
class.
</i>
</p>
</li>
<li>
<p>
<b><a name="tests_16"></a>Is there a basic template I can use to
create a test?</b>
</p>
<p>
<i>(Submitted by: Eric Armstrong)</i>
</p>
<p>
The following templates are a good starting point. Copy/paste
and edit these templates to suit your coding style.
</p>
<p>
SampleTest is a basic test template:
</p>
<div>
<blockquote><pre><code>
import junit.framework.TestCase;
public class SampleTest extends TestCase {
private java.util.List emptyList;
/**
* Sets up the test fixture.
* (Called before every test case method.)
*/
protected void setUp() {
emptyList = new java.util.ArrayList();
}
/**
* Tears down the test fixture.
* (Called after every test case method.)
*/
protected void tearDown() {
emptyList = null;
}
public void testSomeBehavior() {
assertEquals("Empty list should have 0 elements", 0, emptyList.size());
}
public void testForException() {
try {
Object o = emptyList.get(0);
fail("Should raise an IndexOutOfBoundsException");
} catch (IndexOutOfBoundsException success) {
}
}
public static void main(String args[]) {
junit.textui.TestRunner.run(SampleTest.class);
}
}
</code></pre></blockquote>
</div>
<p>
SampleTestSuite is a template for a suite of tests:
</p>
<div>
<blockquote><pre><code>
import junit.framework.Test;
import junit.framework.TestSuite;
public class SampleTestSuite {
public static Test suite() {
TestSuite suite = new TestSuite("Sample Tests");
//
// Add one entry for each test class
// or test suite.
//
suite.addTestSuite(SampleTest.class);
//
// For a master test suite, use this pattern.
// (Note that here, it's recursive!)
//
suite.addTest(AnotherTestSuite.suite());
return suite;
}
public static void main(String args[]) {
junit.textui.TestRunner.run(suite());
}
}
</code></pre></blockquote>
</div>
</li>
<li>
<p>
<b><a name="tests_17"></a>How do I write a test for an abstract
class?</b>
</p>
<p>
Refer to <a
href="http://c2.com/cgi/wiki?AbstractTestCases">http://c2.com/cgi/wiki?AbstractTestCases</a>.
</p>
</li>
<li>
<p>
<b><a name="tests_18"></a>When are tests garbage collected?</b>
</p>
<p>
<i>(Submitted by: Timothy Wall and Kent Beck)</i>
</p>
<p>
By design, the tree of Test instances is built in one pass, then
the tests are executed in a second pass. The test runner holds
strong references to all Test instances for the duration of the
test execution. This means that for a very long test run with
many Test instances, none of the tests may be garbage collected
until the end of the entire test run.
</p>
<p>
Therefore, if you allocate external or limited resources in a
test, you are responsible for freeing those resources.
Explicitly setting an object to <code>null</code> in
the <code>tearDown()</code> method, for example, allows it to be
garbage collected before the end of the entire test run.
</p>
</li>
</ol>
<!--
Organizing Tests
-->
<div class="header">
<a name="organize">Organizing Tests</a>
</div>
<ol>
<li>
<p>
<b><a name="organize_1"></a>Where should I put my test files?</b>
</p>
<p>
You can place your tests in the same package and directory as
the classes under test.
</p>
<p>
For example:
</p>
<div>
<blockquote><pre>
src
com
xyz
SomeClass.java
SomeClassTest.java
</pre></blockquote>
</div>
<p>
While adequate for small projects, many developers feel that
this approach clutters the source directory, and makes it hard
to package up client deliverables without also including
unwanted test code, or writing unnecessarily complex packaging
tasks.
</p>
<p>
An arguably better way is to place the tests in a separate
parallel directory structure with package alignment.
</p>
<p>
For example:
</p>
<div>
<blockquote><pre>
src
com
xyz
SomeClass.java
test
com
xyz
SomeClassTest.java
</pre></blockquote>
</div>
<p>
These approaches allow the tests to access to all the public and
package visible methods of the classes under test.
</p>
<p>
Some developers have argued in favor of putting the tests in a
sub-package of the classes under test (e.g. com.xyz.test). The
author of this FAQ sees no clear advantage to adopting this
approach and believes that said developers also put their curly
braces on the wrong line. :-)
</p>
</li>
<li>
<p>
<b><a name="organize_2"></a>How do I write a test suite for all
of my tests?</b>
</p>
<p>
Write a <code>suite()</code> method that creates
a <code>TestSuite</code> containing all your tests.
</p>
<p>
For example:
</p>
<div class="code">
<pre><code>
import junit.framework.*;
public class AllTests {
public static Test suite() {
TestSuite suite = new TestSuite();
suite.addTest(SomeTest.suite());
suite.addTest(AnotherTest.suite());
return suite;
}
public static void main(String[] args) {
junit.textui.TestRunner.run(suite());
}
}
</code></pre>
</div>
<p>
Running <code>AllTests</code> will automatically run all of its
contained tests in one fell swoop.
</p>
<p>
You can arbitrarily group any tests into test suites as
appropriate by package, logical layers, test type, etc.
</p>
</li>
<li>
<p>
<b><a name="organize_3"></a>How can I run setUp() and tearDown()
code once for all of my tests?</b>
</p>
<p>
The desire to do this is usually a symptom of excessive coupling
in your design. If two or more tests must share the same test
fixture state, then the tests may be trying to tell you that the
classes under test have some undesirable dependencies.
</p>
<p>
Refactoring the design to further decouple the classes under
test and eliminate code duplication is usually a better
investment than setting up a shared test fixture.
</p>
<p>
But if you must...
</p>
<p>
You can wrap the test suite containing all your tests in a
subclass of <code>TestSetup</code> which
invokes <code>setUp()</code> exactly once before all the tests
are run and invokes <code>tearDown()</code> exactly once after
all the tests have been run.
</p>
<p>
The following is an example <code>suite()</code> method that
uses a <code>TestSetup</code> for one-time initialization and
cleanup:
</p>
<div class="code">
<pre><code>
import junit.framework.*;
import junit.extensions.TestSetup;
public class AllTestsOneTimeSetup {
public static Test suite() {
TestSuite suite = new TestSuite();
suite.addTest(SomeTest.suite());
suite.addTest(AnotherTest.suite());
TestSetup wrapper = new TestSetup(suite) {
protected void setUp() {
oneTimeSetUp();
}
protected void tearDown() {
oneTimeTearDown();
}
};
return wrapper;
}
public static void oneTimeSetUp() {
// one-time initialization code
}
public static void oneTimeTearDown() {
// one-time cleanup code
}
}
</code></pre>
</div>
</li>
</ol>
<!--
Running Tests
-->
<div class="header">
<a name="running">Running Tests</a>
</div>
<ol>
<li>
<p>
<b><a name="running_1"></a>What CLASSPATH settings are needed to
run JUnit?</b>
</p>
<p>
<i>(Submitted by: Eric Armstrong)</i>
</p>
<p>
To run your JUnit tests, you'll need the following elemements in
your CLASSPATH:
</p>
<ul>
<li>JUnit class files</li>
<li>Your class files, including your JUnit test classes</li>
<li>Libraries your class files depend on</li>
</ul>
<p>
If attempting to run your tests results in
a <code>NoClassDefFoundError</code>, then something is missing
from your CLASSPATH.
</p>
<p>
<u>Windows Example:</u>
</p>
<p>
<code>set
CLASSPATH=%JUNIT_HOME%\junit.jar;c:\myproject\classes;c:\myproject\lib\something.jar</code>
</p>
<p>
<u>Unix (bash) Example:</u>
</p>
<p>
<code>export CLASSPATH=$JUNIT_HOME/junit.jar:/myproject/classes:/myproject/lib/something.jar</code>
</p>
</li>
<li>
<p>
<b><a name="running_2"></a>Why do I get
a <code>NoClassDefFoundError</code> when trying to test JUnit
or run the samples?</b>
</p>
<p>
<i>(Submitted by: J.B. Rainsberger and Jason Rogers)</i>
</p>
<p>
Most likely your CLASSPATH doesn't include the JUnit
installation directory.
</p>
<p>
Refer to <a href="#running_1">"What CLASSPATH settings are
needed to run JUnit?"</a> for more guidance.
</p>
<p>
Also consider running <a
href="http://www.clarkware.com/software/WhichJUnit.zip">WhichJunit</a>
to print the absolute location of the JUnit class files required
to run and test JUnit and its samples.
</p>
<p>
If the CLASSPATH seems mysterious, read <a
href="http://java.sun.com/j2se/1.4/docs/tooldocs/findingclasses.html">this</a>!
</p>
</li>
<li>
<p>
<b><a name="running_3"></a>How do I run the JUnit GUI test runner?</b>
</p>
<p>
<i>(Submitted by: Eric Armstrong)</i>
</p>
<ol>
<li>
<p>
<a href="#running_1">Set your CLASSPATH</a>
</p>
</li>
<li>
<p>
Invoke the runner:
</p>
<p>
<code>
java junit.swingui.TestRunner <test class name>
</code>
</p>
</li>
</ol>
</li>
<li>
<p>
<b><a name="running_4"></a>How do I run JUnit from my command window?</b>
</p>
<p>
<i>(Submitted by: Eric Armstrong)</i>
</p>
<ol>
<li>
<p>
<a href="#running_1">Set your CLASSPATH</a>
</p>
</li>
<li>
<p>
Invoke the runner:
</p>
<p>
<code>
java junit.textui.TestRunner <test class name>
</code>
</p>
</li>
</ol>
</li>
<li>
<p>
<b><a name="running_5"></a>How do I run JUnit using Ant?</b>
</p>
<p>
<i>(Submitted by: Eric Armstrong)</i>
</p>
<ol>
<li>
<p>
Define any necessary Ant properties:
</p>
<div>
<pre><code>
<property name="src" value="./src" />
<property name="lib" value="./lib" />
<property name="classes" value="./classes" />
<property name="test.class.name" value="com.xyz.MyTestSuite" />
</code></pre>
</div>
</li>
<li>
<p>
Set up the CLASSPATH to be used by JUnit:
</p>
<div>
<pre><code>
<path id="test.classpath">
<pathelement location="${classes}" />
<pathelement location="/path/to/junit.jar" />
<fileset dir="${lib}">
<include name="**/*.jar"/>
</fileset>
</path>
</code></pre>
</div>
</li>
<li>
<p>
Define the Ant task for running JUnit:
</p>
<div>
<pre><code>
<target name="test">
<junit fork="yes" haltonfailure="yes">
<test name="${test.class.name}" />
<formatter type="plain" usefile="false" />
<classpath refid="test.classpath" />
</junit>
</target>
</code></pre>
</div>
</li>
<li>
<p>
Run the test:
</p>
<div>
<code>
ant test
</code>
</div>
</li>
</ol>
<p>
Refer to the <a
href="http://jakarta.apache.org/ant/manual/OptionalTasks/junit.html">JUnit
Ant Task</a> for more information.
</p>
</li>
<li>
<p>
<b><a name="running_6"></a>How do I use Ant to create HTML test
reports?</b>
</p>
<p>
<i>(Submitted by: Eric Armstrong and Steffen Gemkow)</i>
</p>
<ol>
<li>
<p>
Ensure that Ant's <code>optional.jar</code> file is either
in your CLASSPATH or exists in
your <code>$ANT_HOME/lib</code> directory.
</p>
</li>
<li>
<p>
Add an ANT property for the directory containing the HTML reports:
</p>
<div>
<code>
<property name="test.reports" value="./reports" />
</code>
</div>
</li>
<li>
<p>
Define the Ant task for running JUnit and generating reports:
</p>
<div>
<pre><code>
<target name="test-html">
<junit fork="yes" printsummary="no" haltonfailure="no">
<batchtest fork="yes" todir="${test.reports}" >
<fileset dir="${classes}">
<include name="**/*Test.class" />
</fileset>
</batchtest>
<formatter type="xml" />
<classpath refid="test.classpath" />
</junit>
<junitreport todir="${test.reports}">
<fileset dir="${test.reports}">
<include name="TEST-*.xml" />
</fileset>
<report todir="${test.reports}" />
</junitreport>
</target>
</code></pre>
</div>
</li>
<li>
<p>
Run the test:
</p>
<div>
<code>
ant test-html
</code>
</div>
</li>
</ol>
<p>
Refer to the
<a href="http://jakarta.apache.org/ant/manual/OptionalTasks/junit.html">JUnit Ant Task</a>
for more information.
</p>
</li>
<li>
<p>
<b><a name="running_7"></a>How do I pass command-line arguments
to a test execution?</b>
</p>
<p>
Use the <tt>-D</tt> JVM command-line options, as in:
</p>
<div>
<blockquote><code>
-DparameterName=parameterValue
</code></blockquote>
</div>
<p>
If the number of parameters on the command line gets unweildy,
pass in the location of a property file that defines a set of
parameters. Alternatively, the <a
href="http://junit-addons.sf.net">JUnit-addons package</a>
contains the <tt>XMLPropertyManager</tt>
and <tt>PropertyManager</tt> classes that allow you to define a
property file (or XML file) containing test parameters.
</p>
</li>
<li>
<p>
<b><a name="running_8"></a>Why do I get an error
(<code>ClassCastException</code> or
<code>LinkageError</code>) using the GUI TestRunners?</b>
</p>
<p>
<i>(Submitted by: Scott Stirling)</i>
</p>
<p>
JUnit's GUI TestRunners use a custom class loader
(<code>junit.runner.TestCaseClassLoader</code>) to dynamically
reload your code every time you press the "Run" button
so you don't have to restart the GUI to reload your classes if
you recompile them. The default classloaders of the Java Virtual
Machine do not dynamically reload changed classes. But
JUnit's custom class loader finds and loads classes from the
same CLASSPATH used by the JVM's system classloader. So, by
design, it "sits in front of" the system loader and
applies a filter to determine whether it should load a given
class or delegate the loading of a class to the system
classloader. This filter is configured with a list of String
patterns in a properties file
called <code>excluded.properties</code>.
</p>
<p>
The <code>excluded.properties</code> file contains a numbered
list
(<code>excluded.0</code>, <code>excluded.1</code>, <code>excluded.2</code>,
etc.) of properties whose values are patterns for packages. This
file is packaged in <code>junit.jar</code>
as <code>junit/runner/excluded.properties</code>. As of JUnit
3.7 and Java 1.4, its contents are:
</p>
<div>
<blockquote><pre>
#
# The list of excluded package paths for the TestCaseClassLoader
#
excluded.0=sun.*
excluded.1=com.sun.*
excluded.2=org.omg.*
excluded.3=javax.*
excluded.4=sunw.*
excluded.5=java.*
</pre></blockquote>
</div>
<p>
There are some conditions, discussed below, where the default
exclusions are insufficient and you will want to add some more
to this list and then either update the <code>junit.jar</code>
file with your customized version or place your customized
version in the CLASSPATH before <code>junit.jar</code>.
</p>
</li>
<li>
<p>
<b><a name="running_9"></a>Why do I get
a <code>LinkageError</code> when using
XML interfaces in my TestCase?</b>
</p>
<p>
<i>(Submitted by: Scott Stirling)</i>
</p>
<p>
The workaround as of JUnit 3.7 is to
add <code>org.w3c.dom.*</code> and <code>org.xml.sax.*</code> to
your <code>excluded.properties</code>.
</p>
<p>
It's just a matter of time before this fix becomes incorporated
into the released version of
JUnit's <code>excluded.properties</code>, since JAXP is a
standard part of JDK 1.4. It will be just like
excluding <code>org.omg.*</code>. By the way, if you download
the JUnit source from its Sourceforge CVS, you will find that
these patterns have already been added to the default
excluded.properties and so has a pattern for JINI. In fact, here
is the current version in CVS, which demonstrates how to add
exclusions to the list too:
</p>
<div>
<blockquote><pre>
#
# The list of excluded package paths for the TestCaseClassLoader
#
excluded.0=sun.*
excluded.1=com.sun.*
excluded.2=org.omg.*
excluded.3=javax.*
excluded.4=sunw.*
excluded.5=java.*
excluded.6=org.w3c.dom.*
excluded.7=org.xml.sax.*
excluded.8=net.jini.*
</pre></blockquote>
</div>
<p>
This is the most common case where the
default <code>excluded.properties</code> list needs
modification. The cause of the <code>LinkageError</code> is
related to using JAXP in your test cases. By JAXP I mean the
whole set of <code>javax.xml.*</code> classes and the
supporting <code>org.w3c.dom.*</code>
and <code>org.xml.sax.*</code> classes.
</p>
<p>
As stated above, the JUnit GUI TestRunners' classloader relies
on the <code>excluded.properties</code> for classes it should
delegate to the system classloader. JAXP is an unusual case
because it is a standard Java extension library dependent on
classes whose package names (<code>org.w3c.dom.*</code>
and <code>org.xml.sax.*</code>) do not begin with a standard
Java or Sun prefix. This is similar to the relationship
between <code>javax.rmi.*</code> and the <code>org.omg.*</code>
classes, which have been excluded by default in
JUnit'ss <code>excluded.properties</code> for a while.
</p>
<p>
What can happen, and frequently does when using the JUnit Swing
or AWT UI with test cases that reference, use or depend on JAXP
classes, such as Log4J, Apache SOAP, Axis, Cocoon, etc., is that
the JUnit class loader (properly)
delegates <code>javax.xml.*</code> classes it "sees"
to the system loader. But then the system loader, in the process
of initializing and loading that JAXP class, links and loads up
a bunch of <code>org.w3c.dom</code>/<code>org.xml.sax</code>
classes. When it does so, the JUnit custom classloader is not
involved at all because the system classloader never delegates
"down" or checks with custom classloaders to see if a
class is already loaded. At any point after this, if the JUnit
loader is asked to load
an <code>org.w3c.dom</code>/<code>org.xml.sax</code> class that
it's never seen before, it will try to load it because the
class' name doesn't match any of the patterns in the default
exclude list. That's when a <code>LinkageError</code>
occurs. This is really a flaw in the JUnit classloader design,
but there is the workaround given above.
</p>
<p>
Java 2 JVMs keep classes (remember, classes and objects, though
related, are different entities to the JVM - I'm talking
about classes here, not object instances) in namespaces,
identifying them by their fully qualified classname plus the
instance of their defining (not initiating) loader. The JVM will
attempt to assign all unloaded classes referenced by an already
defined and loaded class to that class's defining loader. The
JVM's classresolver routine (implemented as a C function in the
JVM source code) keeps track of all these class loading events
and "sees" if another classloader (such as the JUnit
custom loader) attempts to define a class that has already been
defined by the system loader. According to the rules of Java 2
loader constraints, in case a class has already been defined by
the system loader, any attempts to load a class should first be
delegated to the system loader. A "proper" way for
JUnit to handle this feature would be to load classes from a
repository other than the CLASSPATH that the system classloader
knows nothing about. And then the JUnit custom classloader could
follow the standard Java 2 delegation model, which is to always
delegate class loading to the system loader, and only attempt to
load if that fails. Since they both load from the CLASSPATH in
the current model, if the JUnit loader delegated like it's
supposed to, it would never get to load any classes since the
system loader would always find them.
</p>
<p>
You could try to hack around this in the JUnit source by
catching the <code>LinkageError</code> in
TestCaseClassLoader's <code>loadClass()</code> method and then
making a recovery call to <code>findSystemClass()</code> --
thereby delegating to the system loader after the violation has
been caught. But this hack only works some of the time, because
now you can have the reverse problem where the JUnit loader will
load a host of <code>org.w3c.dom</code>/<code>org.xml.sax</code>
classes, and then the system loader violates the loader
contraints at some point when it tries to do exactly what I
described above with JAXP because it doesn't ever delegate to
its logical child (the JUnit loader). Inevitably, if your test
cases use many JAXP and related XML classes, one or the other
classloader will end up violating the constraints whatever you
do.
</p>
</li>
<li>
<p>
<b><a name="running_10"></a>Why do I get
a <code>ClassCastException</code> when
I use <code>narrow()</code> in an EJB client TestCase?</b>
</p>
<p>
<i>(Submitted by: Scott Stirling)</i>
</p>
<p>
The solution is to prevent your EJB's interface classes from
being loaded by the JUnit custom class loader by adding them
to <code>excluded.properties</code>.
</p>
<p>
This is another problem inherent to JUnit's dynamically
reloading TestCaseClassLoader. Similar to the LinkageErrors
with JAXP and the <code>org.xml.sax</code>
and <code>org.w3c.dom</code> classes, but with a different
result.
</p>
<p>
Here's some example code:
</p>
<div>
<blockquote><pre>
Point point;
PointHome pointHome;
// The next line works in textui, but throws
// ClassCastException in swingui
pointHome = (PointHome)PortableRemoteObject.
narrow(ctx.lookup("base/PointHome"), PointHome.class);
</pre></blockquote>
</div>
<p>
When you call <code>InitialContext.lookup()</code>, it returns
an object that was loaded and defined by the JVM's system
classloader (<code>sun.misc.Launcher$AppClassLoader</code>), but
the <code>PointEJBHome.class</code> type is loaded by JUnit's
TestCaseClassLoader. In the <code>narrow()</code>, the two fully
qualified class names are the same, but the defining
classloaders for the two are different so you get the exception
during the narrow because the JVM doesn't see them as being the
same runtime class type.
</p>
<p>
Recall that in Java 2 an object's class (a.k.a. "runtime type")
is identified in the JVM as the pair of
<fully-qualified-classname;definingClassLoaderInstance> or
(in shorter form) <C;L>. That is, the defining loader's
identity is part of the runtime name identifying that class in
the JVM. Also recall that the JVM will expect a class's defining
loader to load all unloaded classes referenced by the classes it
loads.
</p>
<p>
If interested for debugging purposes, you can find out more
about which loader loaded which class by doing something like
this:
</p>
<div>
<blockquote><pre>
System.out.println(ctx.lookup("base/PointEJBHome").getClass().getClassLoader());
System.out.println(PointEJBHome.class.getClassLoader());
</pre></blockquote>
</div>
<p>
You'll find when using the GUI TestRunners that
the <code>PointEJBHome</code> type is defined by the JUnit
TestCaseClassLoader and the object returned
from <code>InitialContext.lookup()</code> was defined through
the JVM's system class loader. When using the tex-based
TestRunner they'll both have been loaded through the system
loader.
</p>
<p>
If you use Ant's <batchtest> task to run your test cases
and you have this problem, you can work around it by
setting <code>fork="true"</code>
on <code><batchtest></code>, which causes it to run each
test in its own Java Virtual Machine separate from Ant's
launching JVM.
</p>
<p>
For further reading about the principles of Java dynamic
classloading, the best resource is the short paper by Sheng
Liang, the architect of the Java 2 classloader architecture: <a
href="http://java.sun.com/people/sl/"><i>Dynamic Class Loading
in the Java Virtual Machine</i></a>, OOPSLA 1998.
</p>
</li>
<li>
<p>
<b><a name="running_11"></a>Why do I get the warning
"AssertionFailedError: No
tests found in XXX" when I run my test?</b>
</p>
<p>
Make sure the test contains one or more methods with names
beginning with "test".
</p>
<p>
For example:
</p>
<div>
<blockquote><pre>
public void testSomething() {
}
</pre></blockquote>
</div>
<p>
This error also commonly occurs when using
the <code>addTestSuite()</code> in an attempt to create a
hiearchy of test suites. For more information, see <a
href="#running_14">"How do I get one test suite to invoke
another?"</a>.
</p>
</li>
<li>
<p>
<b><a name="running_12"></a>Why do I see "Unknown Source" in the
stack trace of
a test failure, rather than the source file's line number?</b>
</p>
<p>
The debug option for the Java compiler must be enabled in order
to see source file and line number information in a stack trace.
</p>
<p>
When invoking the Java compiler from the command line, use
the <code>-g</code> option to generate all debugging info.
</p>
<p>
When invoking the Java compiler from an
<a href="http://jakarta.apache.org/ant/index.html">Ant</a> task, use the
<code>debug="on"</code> attribute. For example:
</p>
<div>
<blockquote><code>
<javac srcdir="${src}" destdir="${build}" debug="on" />
</code></blockquote>
</div>
<p>
When using older JVMs pre-Hotspot (JDK 1.1 and most/all 1.2),
run JUnit with the <code>-DJAVA_COMPILER=none</code> JMV command
line argument to prevent runtime JIT compilation from obscuring
line number info.
</p>
<p>
Compiling the test source with debug enabled will show the line
where the assertion failed. Compiling the non-test source with
debug enabled will show the line where an exception was raised
in the class under test.
</p>
</li>
<li>
<p>
<b><a name="running_13"></a>Why does the "excluded.properties"
trick not work when running JUnit's GUI from inside my
favorite IDE?</b>
</p>
<p>
<i>(Submitted by: William Pietri)</i>
</p>
<p>
Some IDEs come with a copy of JUnit, so your copy of JUnit in
the project classpath isn't the one being used. Replace
the <code>junit.jar</code> file used by the IDE with
a <code>junit.jar</code> file containing a
custom <code>excluded.properties</code> and your bar will once
again be green.
</p>
</li>
<li>
<p>
<b><a name="running_14"></a>How do I get one test suite to
invoke another?</b>
</p>
<p>
<i>(Submitted by: Eric Armstrong)</i>
</p>
<p>
To add a suite to a suite, use:
</p>
<div>
<blockquote><code>
suite.addTest(someOtherSuite.suite());
</code></blockquote>
</div>
</li>
<li>
<p>
<b><a name="running_15"></a>How do I organize all test classes
in a TestSuite automatically and not use or manage a TestSuite
explicitly?</b>
</p>
<p>
<i>(Submitted by: Bill de hora)</i>
</p>
<p>
There are a number of ways to do this:
</p>
<ol>
<li>
<p>
In Ant, use the <code>junit</code> task and
the <code>batchtest</code> element:
</p>
<div>
<pre><code>
<junit printsummary="yes" haltonfailure="yes">
...
<batchtest fork="yes">
<fileset dir="${src.dir}">
<include name="**/*Test.java" />
<include name="**/Test*.java" />
</fileset>
</batchtest>
</junit>
</code></pre>
</div>
<p>
Idiomatic naming patterns for unit tests
are <code>Test*.java</code> and <code>*Test.java</code>.
Documentation and examples are at <a
href="http://ant.apache.org/manual/OptionalTasks/junit.html">http://ant.apache.org/manual/OptionalTasks/junit.html</a>.
</p>
</li>
<li>
<p>
Use the <code>DirectorySuiteBuilder</code>
and <code>ArchiveSuiteBuilder</code> (for jar/zip files)
classes provided by JUnit-addons project:
</p>
<div>
<blockquote><pre>
DirectorySuiteBuilder builder = new DirectorySuiteBuilder();
builder.setSuffix("Test");
Test suite = builer.suite("/home/project/myproject/tests");
</pre></blockquote>
</div>
<p>
Documentation and examples are at <a
href="http://junit-addons.sourceforge.net/">http://junit-addons.sourceforge.net</a>.
</p>
</li>
<li>
<p>
Write your own custom suite builder.
</p>
<p>
Have your test classes implement an interface and write a
treewalker to load each class in a directory, inspect the
class, and add any classes that implement the interface to a
TestSuite.
</p>
<p>
You might only want to do this if you are <b>very</b>
uncomfortable with using a naming convention for test
classes. Aside from being slow for larger suites, ultimately
it's arguable whether it's more effort to follow a naming
convention that have test classes implement an interface!
</p>
<p>
An example of this approach is at
<a href="http://www.javaworld.com/javaworld/jw-12-2000/jw-1221-junit_p.html">http://www.javaworld.com/javaworld/jw-12-2000/jw-1221-junit_p.html</a>.
</p>
</li>
</ol>
</li>
</ol>
<!--
Best Practices
-->
<div class="header">
<a name="best">Best Practices</a>
</div>
<ol>
<li>
<p>
<b><a name="best_1"></a>When should tests be written?</b>
</p>
<p>
Tests should be written before the code. Test-first programming
is practiced by only writing new code when an automated test is
failing.
</p>
<p>
Good tests tell you how to best design the system for its
intended use. They effectively communicate in an executable
format how to use the software. They also prevent tendencies to
over-build the system based on speculation. When all the tests
pass, you know you're done!
</p>
<p>
Whenever a customer test fails or a bug is reported, first write
the necessary unit test(s) to expose the bug(s), <em>then</em>
fix them. This makes it almost impossible for that particular
bug to resurface later.
</p>
<p>
Test-driven development is a lot more fun than writing tests
after the code seems to be working. Give it a try!
</p>
</li>
<li>
<p>
<b><a name="best_2"></a>Do I have to write a test for
everything?</b>
</p>
<p>
No, just test everything that could reasonably break.
</p>
<p>
Be practical and maximize your testing investment. Remember
that investments in testing are equal investments in design. If
defects aren't being reported and your design responds well to
change, then you're probably testing enough. If you're spending
a lot of time fixing defects and your design is difficult to
grow, you should write more tests.
</p>
<p>
If something is difficult to test, it's usually an opportunity
for a design improvement. Look to improve the design so that
it's easier to test, and by doing so a better design will
usually emerge.
</p>
</li>
<li>
<p>
<b><a name="best_3"></a>How simple is 'too simple to break'?</b>
</p>
<p>
<i>(Submitted by: J. B. Rainsberger)</i>
</p>
<p>
The general philosophy is this: if it can't break <em>on its
own</em>, it's too simple to break.
</p>
<p>
First example is the <code>getX()</code> method. Suppose
the <code>getX()</code> method only answers the value of an
instance variable. In that case, <code>getX()</code> cannot
break unless either the compiler or the interpreter is also
broken. For that reason, don't test <code>getX()</code>; there
is no benefit. The same is true of the <code>setX()</code>
method, although if your <code>setX()</code> method does any
parameter validation or has any side effects, you likely need to
test it.
</p>
<p>
Next example: suppose you have written a method that does
nothing but forward parameters into a method called on another
object. That method is too simple to break.
</p>
<div>
<blockquote><pre>
public void myMethod(final int a, final String b) {
myCollaborator.anotherMethod(a, b);
}
</pre></blockquote>
</div>
<p>
<code>myMethod</code> cannot possibly break because it does nothing: it
forwards its input to another object and that's all.
</p>
<p>
The only precondition for this method is "myCollaborator !=
null", but that is generally the responsibility of the
constructor, and not of myMethod. If you are concerned, add a
test to verify that myCollaborator is always set to something
non-null by every constructor.
</p>
<p>
The only way myMethod could break would be
if <code>myCollaborator.anotherMethod()</code> were broken. In
that case, test <code>myCollaborator</code>, and not the current
class.
</p>
<p>
It is true that adding tests for even these simple methods
guards against the possibility that someone refactors and makes
the methods "not-so-simple" anymore. In that case, though, the
refactorer needs to be aware that the method is now complex
enough to break, and should write tests for it -- and preferably
before the refactoring.
</p>
<p>
Another example: suppose you have a JSP and, like a good
programmer, you have removed all business logic from it. All it
does is provide a layout for a number of JavaBeans and never
does anything that could change the value of any object. That
JSP is too simple to break, and since JSPs are notoriously
annoying to test, you should strive to make all your JSPs too
simple to break.
</p>
<p>
Here's the way testing goes:
</p>
<div>
<blockquote><pre>
becomeTimidAndTestEverything
while writingTheSameThingOverAndOverAgain
becomeMoreAggressive
writeFewerTests
writeTestsForMoreInterestingCases
if getBurnedByStupidDefect
feelStupid
becomeTimidAndTestEverything
end
end
</pre></blockquote>
</div>
<p>
The loop, as you can see, never terminates.
</p>
</li>
<li>
<p>
<b><a name="best_4"></a>How often should I run my tests?</b>
</p>
<p>
Run all your unit tests as often as possible, ideally every time
the code is changed. Make sure all your unit tests always run
at 100%. Frequent testing gives you confidence that your
changes didn't break anything and generally lowers the stress of
programming in the dark.
</p>
<p>
For larger systems, you may just run specific test suites that
are relevant to the code you're working on.
</p>
<p>
Run all your acceptance, integration, stress, and unit tests at
least once per day (or night).
</p>
</li>
<li>
<p>
<b><a name="best_5"></a>What do I do when a defect is reported?</b>
</p>
<p>
Test-driven development generally lowers the defect density of
software. But we're all fallible, so sometimes a defect will
slip through. When this happens, write a failing test that
exposes the defect. When the test passes, you know the defect
is fixed!
</p>
<p>
Don't forget to use this as a learning opportunity. Perhaps the
defect could have been prevented by being more aggressive about
testing everything that could reasonably break.
</p>
</li>
<li>
<p>
<b><a name="best_6"></a>Why not just use <code>System.out.println()</code>?</b>
</p>
<p>
Inserting debug statements into code is a low-tech method for
debugging it. It usually requires that output be scanned
manually every time the program is run to ensure that the code
is doing what's expected.
</p>
<p>
It generally takes less time in the long run to codify
expectations in the form of an automated JUnit test that retains
its value over time. If it's difficult to write a test to
assert expectations, the tests may be telling you that shorter
and more cohesive methods would improve your design.
</p>
</li>
<li>
<p>
<b><a name="best_7"></a>Why not just use a debugger?</b>
</p>
<p>
Debuggers are commonly used to step through code and inspect
that the variables along the way contain the expected values.
But stepping through a program in a debugger is a manual process
that requires tedious visual inspections. In essence, the
debugging session is nothing more than a manual check of
expected vs. actual results. Moreover, every time the program
changes we must manually step back through the program in the
debugger to ensure that nothing broke.
</p>
<p>
It generally takes less time to codify expectations in the form
of an automated JUnit test that retains its value over time. If
it's difficult to write a test to assert expected values, the
tests may be telling you that shorter and more cohesive methods
would improve your design.
</p>
</li>
</ol>
<!--
JUnit Extensions
-->
<div class="header">
<a name="extend">Extending JUnit</a>
</div>
<ol>
<li>
<p>
<b><a name="extend_1"></a>How do I extend JUnit?</b>
</p>
<p>
JUnit is a testing framework intended to be customized for
specialized use. Browsing the JUnit source code is an excellent
way to learn its design and discover how it can be extended.
</p>
<p>
Examples of JUnit extensions can be found in
the <code>junit.extensions</code> package:
</p>
<ul>
<li>
<code>TestDecorator</code>
<p>
A decorator for Tests. You can use it as the base class for
implementing new test decorators that add behavior before or
after a test is run.
</p>
</li>
<li>
<code>ActiveTestSuite</code>
<p>
A <code>TestSuite</code> that runs each test in a separate
thread and waits until all threads have terminated.
</p>
</li>
<li>
<code>TestSetup</code>
<p>
A <code>TestDecorator</code> to initialize and cleanup
test fixture state once before the test is run.
</p>
</li>
<li>
<code>RepeatedTest</code>
<p>
A <code>TestDecorator</code> that runs a test repeatedly.
</p>
</li>
<li>
<code>ExceptionTestCase</code>
<p>
A <code>TestCase</code> that expects a
particular <code>Exception</code> to be thrown.
</p>
<p>
Kent Beck has mentioned that <code>ExceptionTestCase</code>
likely does not provide enough to be useful; it is just as
easy to write the "exception test" yourself. Refer to the <a
href="#tests">"Writing Tests"</a> section for guidance.
</p>
</li>
</ul>
</li>
<li>
<p>
<b><a name="extend_2"></a>What kinds of extensions are available?</b>
</p>
<p>
The JUnit home page has a complete list of available <a
href="http://www.junit.org/news/extension/index.htm">JUnit
extensions</a>.
</p>
</li>
</ol>
<!--
Miscellaneous
-->
<div class="header">
<a name="misc">Miscellaneous</a>
</div>
<ol>
<li>
<p>
<b><a name="misc_1"></a>How do I integrate JUnit with my IDE?</b>
</p>
<p>
The JUnit home page maintains a list of <a
href="http://www.junit.org/news/ide/index.htm">IDE integration
instructions</a>.
</p>
</li>
<li>
<p>
<b><a name="misc_2"></a>How do I launch a debugger when a test fails?</b>
</p>
<p>
Start the <code>TestRunner</code> under the debugger and
configure the debugger so that it catches
the <code>junit.framework.AssertionFailedError</code>.
</p>
<p>
How you configure this depends on the debugger you prefer to
use. Most Java debuggers provide support to stop the program
when a specific exception is raised.
</p>
<p>
Notice that this will only launch the debugger when an expected
failure occurs.
</p>
</li>
<li>
<p>
<b><a name="misc_3"></a>Where can I find unit testing frameworks
similar to JUnit for other languages?</b>
</p>
<p>
XProgramming.com maintains a complete list of available <a
href="http://www.xprogramming.com/software.htm">xUnit testing
frameworks</a>.
</p>
</li>
</ol>
<br/>
<div align="right">
<a href="http://validator.w3.org/check?uri=referer">
<img src="http://www.w3.org/Icons/valid-xhtml10"
alt="Valid XHTML 1.0!" height="31" width="88" /></a>
</div>
</body>
</html>
