/**************************************************************************** ** ** Copyright (C) 2010 Nokia Corporation and/or its subsidiary(-ies). ** All rights reserved. ** Contact: Nokia Corporation (qt-info@nokia.com) ** ** This file is part of the documentation of the Qt Toolkit. ** ** $QT_BEGIN_LICENSE:FDL$ ** No Commercial Usage ** This file contains pre-release code and may not be distributed. ** You may use this file in accordance with the terms and conditions ** contained in the Technology Preview License Agreement accompanying ** this package. ** ** GNU Free Documentation License ** Alternatively, this file may be used under the terms of the GNU Free ** Documentation License version 1.3 as published by the Free Software ** Foundation and appearing in the file included in the packaging of this ** file. ** ** If you have questions regarding the use of this file, please contact ** Nokia at qt-info@nokia.com. ** $QT_END_LICENSE$ ** ****************************************************************************/ /*! \page qtestlib-manual.html \title QTestLib Manual \brief An overview of Qt's unit testing framework. \ingroup frameworks-technologies \ingroup qt-basic-concepts \keyword qtestlib The QTestLib framework, provided by Nokia, is a tool for unit testing Qt based applications and libraries. QTestLib provides all the functionality commonly found in unit testing frameworks as well as extensions for testing graphical user interfaces. \section1 QTestLib Features QTestLib is designed to ease the writing of unit tests for Qt based applications and libraries: \table \header \o Feature \o Details \row \o \bold Lightweight \o QTestLib consists of about 6000 lines of code and 60 exported symbols. \row \o \bold Self-contained \o QTestLib requires only a few symbols from the Qt Core library for non-gui testing. \row \o \bold {Rapid testing} \o QTestLib needs no special test-runners; no special registration for tests. \row \o \bold {Data-driven testing} \o A test can be executed multiple times with different test data. \row \o \bold {Basic GUI testing} \o QTestLib offers functionality for mouse and keyboard simulation. \row \o \bold {Benchmarking} \o QTestLib supports benchmarking and provides several measurement back-ends. \row \o \bold {IDE friendly} \o QTestLib outputs messages that can be interpreted by Visual Studio and KDevelop. \row \o \bold Thread-safety \o The error reporting is thread safe and atomic. \row \o \bold Type-safety \o Extensive use of templates prevent errors introduced by implicit type casting. \row \o \bold {Easily extendable} \o Custom types can easily be added to the test data and test output. \endtable \note For higher-level GUI and application testing needs, please see the \l{Partner Directory} for Qt testing products provided by Nokia partners. \section1 QTestLib API All public methods are in the \l QTest namespace. In addition, the \l QSignalSpy class provides easy introspection for Qt's signals and slots. \section1 Using QTestLib \section2 Creating a Test To create a test, subclass QObject and add one or more private slots to it. Each private slot is a testfunction in your test. QTest::qExec() can be used to execute all testfunctions in the test object. In addition, there are four private slots that are \e not treated as testfunctions. They will be executed by the testing framework and can be used to initialize and clean up either the entire test or the current test function. \list \o \c{initTestCase()} will be called before the first testfunction is executed. \o \c{cleanupTestCase()} will be called after the last testfunction was executed. \o \c{init()} will be called before each testfunction is executed. \o \c{cleanup()} will be called after every testfunction. \endlist If \c{initTestCase()} fails, no testfunction will be executed. If \c{init()} fails, the following testfunction will not be executed, the test will proceed to the next testfunction. Example: \snippet doc/src/snippets/code/doc_src_qtestlib.qdoc 0 For more examples, refer to the \l{QTestLib Tutorial}. \section2 Building a Test If you are using \c qmake as your build tool, just add the following to your project file: \snippet doc/src/snippets/code/doc_src_qtestlib.qdoc 1 If you are using other buildtools, make sure that you add the location of the QTestLib header files to your include path (usually \c{include/QtTest} under your Qt installation directory). If you are using a release build of Qt, link your test to the \c QtTest library. For debug builds, use \c{QtTest_debug}. See \l {Chapter 1: Writing a Unit Test}{Writing a Unit Test} for a step by step explanation. \section2 QTestLib Command Line Arguments \section3 Syntax The syntax to execute an autotest takes the following simple form: \snippet doc/src/snippets/code/doc_src_qtestlib.qdoc 2 Substitute \c testname with the name of your executable. \c testfunctions can contain names of test functions to be executed. If no \c testfunctions are passed, all tests are run. If you append the name of an entry in \c testdata, the test function will be run only with that test data. For example: \snippet doc/src/snippets/code/doc_src_qtestlib.qdoc 3 Runs the test function called \c toUpper with all available test data. \snippet doc/src/snippets/code/doc_src_qtestlib.qdoc 4 Runs the \c toUpper test function with all available test data, and the \c toInt test function with the testdata called \c zero (if the specified test data doesn't exist, the associated test will fail). \snippet doc/src/snippets/code/doc_src_qtestlib.qdoc 5 Runs the testMyWidget function test, outputs every signal emission and waits 500 milliseconds after each simulated mouse/keyboard event. \section3 Options The following command line arguments are understood: \list \o \c -help \BR outputs the possible command line arguments and give some useful help. \o \c -functions \BR outputs all test functions available in the test. \o \c -o \e filename \BR write output to the specified file, rather than to standard output \o \c -silent \BR silent output, only shows warnings, failures and minimal status messages \o \c -v1 \BR verbose output; outputs information on entering and exiting test functions. \o \c -v2 \BR extended verbose output; also outputs each \l QCOMPARE() and \l QVERIFY() \o \c -vs \BR outputs every signal that gets emitted \o \c -xml \BR outputs XML formatted results instead of plain text \o \c -lightxml \BR outputs results as a stream of XML tags \o \c -eventdelay \e ms \BR if no delay is specified for keyboard or mouse simulation (\l QTest::keyClick(), \l QTest::mouseClick() etc.), the value from this parameter (in milliseconds) is substituted. \o \c -keydelay \e ms \BR like -eventdelay, but only influences keyboard simulation and not mouse simulation. \o \c -mousedelay \e ms \BR like -eventdelay, but only influences mouse simulation and not keyboard simulation. \o \c -keyevent-verbose \BR output more verbose output for keyboard simulation \o \c -maxwarnings \e number\BR sets the maximum number of warnings to output. 0 for unlimited, defaults to 2000. \endlist \section2 Creating a Benchmark To create a benchmark, follow the instructions for creating a test and then add a QBENCHMARK macro to the test function that you want to benchmark. \snippet doc/src/snippets/code/doc_src_qtestlib.qdoc 12 The code inside the QBENCHMARK macro will be measured, and possibly also repeated several times in order to get an accurate measurement. This depends on the selected measurement back-end. Several back-ends are available. They can be selected on the command line: \target testlib-benchmarking-measurement \table \header \o Name \o Commmand-line Argument \o Availability \row \o Walltime \o (default) \o All platforms \row \o CPU tick counter \o -tickcounter \o Windows, Mac OS X, Linux, many UNIX-like systems. \row \o Valgrind/Callgrind \o -callgrind \o Linux (if installed) \row \o Event Counter \o -eventcounter \o All platforms \endtable In short, walltime is always available but requires many repetitions to get a useful result. Tick counters are usually available and can provide results with fewer repetitions, but can be susceptible to CPU frequency scaling issues. Valgrind provides exact results, but does not take I/O waits into account, and is only available on a limited number of platforms. Event counting is available on all platforms and it provides the number of events that were received by the event loop before they are sent to their corresponding targets (this might include non-Qt events). \note Depending on the device configuration, Tick counters on the Windows CE platform may not be as fine-grained, compared to other platforms. Devices that do not support high-resolution timers default to one-millisecond granularity. See the chapter 5 in the \l{QTestLib Tutorial} for more benchmarking examples. \section1 Using QTestLib remotely on Windows CE \c cetest is a convenience application which helps the user to launch an application remotely on a Windows CE device or emulator. It needs to be executed after the unit test has been successfully compiled. Prior to launching, the following files are copied to the device: \list \o all Qt libraries the project links to \o \l {QtRemote}{QtRemote.dll} \o the c runtime library specified during installation \o all files specified in the \c .pro file following the \l DEPLOYMENT rules. \endlist \section2 Using \c cetest \section3 Syntax The syntax to execute an autotest takes the following simple form: \snippet doc/src/snippets/code/doc_src_qtestlib.qdoc 6 \section3 Options \c cetest provides the same options as those for unit-testing on non cross-compiled platforms. See \l {QTestLib Command Line Arguments} {Command Line Arguments} for more information. The following commands are also included: \list \o \c -debug \BR Test version compiled in debug mode. \o \c -release \BR Test version compiled in release mode. \o \c -libpath \e path \BR Target path to copy Qt libraries to. \o \c -qt-delete \BR Delete Qt libraries after execution. \o \c -project-delete \BR Delete project files after execution. \o \c -delete \BR Delete project and Qt libraries after execution. \o \c -conf \BR Specifies a qt.conf file to be deployed to remote directory. \endlist \note \c{debug} is the default build option. \section2 QtRemote \c QtRemote is a small library which is build after QTestLib. It allows the host system to create a process on a remote device and waits until its execution has been finished. \section2 Requirements \c cetest uses Microsoft ActiveSync to establish a remote connection between the host computer and the device. Thus header files and libraries are needed to compile cetest and QtRemote successfully. Prior to \l{Installing Qt for Windows CE}{installation} of Qt, you need to set your \c INCLUDE and \c LIB environment variables properly. A default installation of Windows Mobile 5 for Pocket PC can be obtained by: \snippet doc/src/snippets/code/doc_src_qtestlib.qdoc 7 Note that Qt will remember the path, so you do not need to set it again after switching the environments for cross-compilation. \section1 3rd Party Code The CPU tick counters used for benchmarking is licensed under the following license: (from src/testlib/3rdparty/cycle.h) \legalese Copyright (c) 2003, 2006 Matteo Frigo\br Copyright (c) 2003, 2006 Massachusetts Institute of Technology Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. \endlegalese */ /*! \page qtestlib-tutorial.html \brief A short introduction to testing with QTestLib. \contentspage QTestLib Manual \nextpage {Chapter 1: Writing a Unit Test}{Chapter 1} \ingroup best-practices \title QTestLib Tutorial This tutorial gives a short introduction to how to use some of the features of the QTestLib framework. It is divided into four chapters: \list 1 \o \l {Chapter 1: Writing a Unit Test}{Writing a Unit Test} \o \l {Chapter 2: Data Driven Testing}{Data Driven Testing} \o \l {Chapter 3: Simulating GUI Events}{Simulating GUI Events} \o \l {Chapter 4: Replaying GUI Events}{Replaying GUI Events} \o \l {Chapter 5: Writing a Benchmark}{Writing a Benchmark} \endlist */ /*! \example qtestlib/tutorial1 \contentspage {QTestLib Tutorial}{Contents} \nextpage {Chapter 2: Data Driven Testing}{Chapter 2} \title Chapter 1: Writing a Unit Test In this first chapter we will see how to write a simple unit test for a class, and how to execute it. \section1 Writing a Test Let's assume you want to test the behavior of our QString class. First, you need a class that contains your test functions. This class has to inherit from QObject: \snippet examples/qtestlib/tutorial1/testqstring.cpp 0 Note that you need to include the QTest header, and that the test functions have to be declared as private slots so the test framework finds and executes it. Then you need to implement the test function itself. The implementation could look like this: \snippet doc/src/snippets/code/doc_src_qtestlib.qdoc 8 The \l QVERIFY() macro evaluates the expression passed as its argument. If the expression evaluates to true, the execution of the test function continues. Otherwise, a message describing the failure is appended to the test log, and the test function stops executing. But if you want a more verbose output to the test log, you should use the \l QCOMPARE() macro instead: \snippet examples/qtestlib/tutorial1/testqstring.cpp 1 If the strings are not equal, the contents of both strings is appended to the test log, making it immediately visible why the comparison failed. Finally, to make our test case a stand-alone executable, the following two lines are needed: \snippet examples/qtestlib/tutorial1/testqstring.cpp 2 The \l QTEST_MAIN() macro expands to a simple \c main() method that runs all the test functions. Note that if both the declaration and the implementation of our test class are in a \c .cpp file, we also need to include the generated moc file to make Qt's introspection work. \section1 Executing a Test Now that we finished writing our test, we want to execute it. Assuming that our test was saved as \c testqstring.cpp in an empty directory: we build the test using qmake to create a project and generate a makefile. \snippet doc/src/snippets/code/doc_src_qtestlib.qdoc 9 \bold {Note:}If you're using windows, replace \c make with \c nmake or whatever build tool you use. Running the resulting executable should give you the following output: \snippet doc/src/snippets/code/doc_src_qtestlib.qdoc 10 Congratulations! You just wrote and executed your first unit test using the QTestLib framework. */ /*! \example qtestlib/tutorial2 \previouspage {Chapter 1: Writing a Unit Test}{Chapter 1} \contentspage {QTestLib Tutorial}{Contents} \nextpage {Chapter 3: Simulating Gui Events}{Chapter 3} \title Chapter 2: Data Driven Testing In this chapter we will demonstrate how to execute a test multiple times with different test data. So far, we have hard coded the data we wanted to test into our test function. If we add more test data, the function might look like this: \snippet doc/src/snippets/code/doc_src_qtestlib.qdoc 11 To prevent that the function ends up being cluttered by repetitive code, QTestLib supports adding test data to a test function. All we need is to add another private slot to our test class: \snippet examples/qtestlib/tutorial2/testqstring.cpp 0 \section1 Writing the Data Function A test function's associated data function carries the same name, appended by \c{_data}. Our data function looks like this: \snippet examples/qtestlib/tutorial2/testqstring.cpp 1 First, we define the two elements of our test table using the \l QTest::addColumn() function: A test string, and the expected result of applying the QString::toUpper() function to that string. Then we add some data to the table using the \l QTest::newRow() function. Each set of data will become a separate row in the test table. \l QTest::newRow() takes one argument: A name that will be associated with the data set. If the test fails, the name will be used in the test log, referencing the failed data. Then we stream the data set into the new table row: First an arbitrary string, and then the expected result of applying the QString::toUpper() function to that string. You can think of the test data as a two-dimensional table. In our case, it has two columns called \c string and \c result and three rows. In addition a name as well as an index is associated with each row: \table \header \o index \o name \o string \o result \row \o 0 \o all lower \o "hello" \o HELLO \row \o 1 \o mixed \o "Hello" \o HELLO \row \o 2 \o all upper \o "HELLO" \o HELLO \endtable \section1 Rewriting the Test Function Our test function can now be rewritten: \snippet examples/qtestlib/tutorial2/testqstring.cpp 2 The TestQString::toUpper() function will be executed three times, once for each entry in the test table that we created in the associated TestQString::toUpper_data() function. First, we fetch the two elements of the data set using the \l QFETCH() macro. \l QFETCH() takes two arguments: The data type of the element and the element name. Then we perform the test using the \l QCOMPARE() macro. This approach makes it very easy to add new data to the test without modifying the test itself. And again, to make our test case a stand-alone executable, the following two lines are needed: \snippet examples/qtestlib/tutorial2/testqstring.cpp 3 As before, the QTEST_MAIN() macro expands to a simple main() method that runs all the test functions, and since both the declaration and the implementation of our test class are in a .cpp file, we also need to include the generated moc file to make Qt's introspection work. */ /*! \example qtestlib/tutorial3 \previouspage {Chapter 2: Data Driven Testing}{Chapter 2} \contentspage {QTestLib Tutorial}{Contents} \nextpage {Chapter 4: Replaying GUI Events}{Chapter 4} \title Chapter 3: Simulating GUI Events QTestLib features some mechanisms to test graphical user interfaces. Instead of simulating native window system events, QTestLib sends internal Qt events. That means there are no side-effects on the machine the tests are running on. In this chapter we will se how to write a simple GUI test. \section1 Writing a GUI test This time, let's assume you want to test the behavior of our QLineEdit class. As before, you will need a class that contains your test function: \snippet examples/qtestlib/tutorial3/testgui.cpp 0 The only difference is that you need to include the QtGui class definitions in addition to the QTest namespace. \snippet examples/qtestlib/tutorial3/testgui.cpp 1 In the implementation of the test function we first create a QLineEdit. Then we simulate writing "hello world" in the line edit using the \l QTest::keyClicks() function. \note The widget must also be shown in order to correctly test keyboard shortcuts. QTest::keyClicks() simulates clicking a sequence of keys on a widget. Optionally, a keyboard modifier can be specified as well as a delay (in milliseconds) of the test after each key click. In a similar way, you can use the QTest::keyClick(), QTest::keyPress(), QTest::keyRelease(), QTest::mouseClick(), QTest::mouseDClick(), QTest::mouseMove(), QTest::mousePress() and QTest::mouseRelease() functions to simulate the associated GUI events. Finally, we use the \l QCOMPARE() macro to check if the line edit's text is as expected. As before, to make our test case a stand-alone executable, the following two lines are needed: \snippet examples/qtestlib/tutorial3/testgui.cpp 2 The QTEST_MAIN() macro expands to a simple main() method that runs all the test functions, and since both the declaration and the implementation of our test class are in a .cpp file, we also need to include the generated moc file to make Qt's introspection work. */ /*! \example qtestlib/tutorial4 \previouspage {Chapter 3: Simulating GUI Event}{Chapter 3} \contentspage {QTestLib Tutorial}{Contents} \nextpage {Chapter 5: Writing a Benchmark}{Chapter 5} \title Chapter 4: Replaying GUI Events In this chapter, we will show how to simulate a GUI event, and how to store a series of GUI events as well as replay them on a widget. The approach to storing a series of events and replay them, is quite similar to the approach explained in \l {Chapter 2: Data Driven Testing}{chapter 2}; all you need is to add a data function to your test class: \snippet examples/qtestlib/tutorial4/testgui.cpp 0 \section1 Writing the Data Function As before, a test function's associated data function carries the same name, appended by \c{_data}. \snippet examples/qtestlib/tutorial4/testgui.cpp 1 First, we define the elements of the table using the QTest::addColumn() function: A list of GUI events, and the expected result of applying the list of events on a QWidget. Note that the type of the first element is \l QTestEventList. A QTestEventList can be populated with GUI events that can be stored as test data for later usage, or be replayed on any QWidget. In our current data function, we create two \l {QTestEventList}s. The first list consists of a single click to the 'a' key. We add the event to the list using the QTestEventList::addKeyClick() function. Then we use the QTest::newRow() function to give the data set a name, and stream the event list and the expected result into the table. The second list consists of two key clicks: an 'a' with a following 'backspace'. Again we use the QTestEventList::addKeyClick() to add the events to the list, and QTest::newRow() to put the event list and the expected result into the table with an associated name. \section1 Rewriting the Test Function Our test can now be rewritten: \snippet examples/qtestlib/tutorial4/testgui.cpp 2 The TestGui::testGui() function will be executed two times, once for each entry in the test data that we created in the associated TestGui::testGui_data() function. First, we fetch the two elements of the data set using the \l QFETCH() macro. \l QFETCH() takes two arguments: The data type of the element and the element name. Then we create a QLineEdit, and apply the list of events on that widget using the QTestEventList::simulate() function. Finally, we use the QCOMPARE() macro to check if the line edit's text is as expected. As before, to make our test case a stand-alone executable, the following two lines are needed: \snippet examples/qtestlib/tutorial4/testgui.cpp 3 The QTEST_MAIN() macro expands to a simple main() method that runs all the test functions, and since both the declaration and the implementation of our test class are in a .cpp file, we also need to include the generated moc file to make Qt's introspection work. */ /*! \example qtestlib/tutorial5 \previouspage {Chapter 4: Replaying GUI Events}{Chapter 4} \contentspage {QTestLib Tutorial}{Contents} \title Chapter 5: Writing a Benchmark In this final chapter we will demonstrate how to write benchmarks using QTestLib. \section1 Writing a Benchmark To create a benchmark we extend a test function with a QBENCHMARK macro. A benchmark test function will then typically consist of setup code and a QBENCHMARK macro that contains the code to be measured. This test function benchmarks QString::localeAwareCompare(). \snippet examples/qtestlib/tutorial5/benchmarking.cpp 0 Setup can be done at the beginning of the function, the clock is not running at this point. The code inside the QBENCHMARK macro will be measured, and possibly repeated several times in order to get an accurate measurement. Several \l {testlib-benchmarking-measurement}{back-ends} are available and can be selected on the command line. \section1 Data Functions Data functions are useful for creating benchmarks that compare multiple data inputs, for example locale aware compare against standard compare. \snippet examples/qtestlib/tutorial5/benchmarking.cpp 1 The test function then uses the data to determine what to benchmark. \snippet examples/qtestlib/tutorial5/benchmarking.cpp 2 The "if(useLocaleCompare)" switch is placed outside the QBENCHMARK macro to avoid measuring its overhead. Each benchmark test function can have one active QBENCHMARK macro. \section1 External Tools Tools for handling and visualizing test data are available as part of the qtestlib-tools project on the \l{http://labs.qt.nokia.com/}{http://labs.qt.nokia.com/}Qt Labs Web site. These include a tool for comparing performance data obtained from test runs and a utility to generate Web-based graphs of performance data. See the \l{qtestlib-tools Announcement} for more information on these tools and a simple graphing example. */