path: root/doc/src/development/qtestlib.qdoc

/****************************************************************************
**
** Copyright (C) 2011 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.
**
** $QT_BEGIN_LICENSE:FDL$
** 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.
**
** Other Usage
** Alternatively, this file may be used in accordance with the terms
** and conditions contained in a signed written agreement between you
** and Nokia.
**
**
**
**
** $QT_END_LICENSE$
**
****************************************************************************/

/*!
    \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.cpp 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.pro 1

    If you are using other build tools, 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 -datatags \BR
    outputs all data tags available in the test.
    A global data tag is preceded by ' __global__ '.
    \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.cpp 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 five
    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.cpp 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.cpp 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 \l {qtestlib-tools} project in the \l{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}{qtestlib-tools announcement}
    for more information on these tools and a simple graphing example.

*/