// Copyright (C) 2022 The Qt Company Ltd. // Copyright (C) 2016 Intel Corporation. // SPDX-License-Identifier: LicenseRef-Qt-Commercial OR GFDL-1.3-no-invariants-only /*! \page qtest-overview.html \title Qt Test Overview \brief Overview of the Qt unit testing framework. \ingroup frameworks-technologies \ingroup qt-basic-concepts \keyword qtestlib Qt Test is a framework for unit testing Qt based applications and libraries. Qt Test provides all the functionality commonly found in unit testing frameworks as well as extensions for testing graphical user interfaces. Qt Test is designed to ease the writing of unit tests for Qt based applications and libraries: \table \header \li Feature \li Details \row \li \b Lightweight \li Qt Test consists of about 6000 lines of code and 60 exported symbols. \row \li \b Self-contained \li Qt Test requires only a few symbols from the Qt Core module for non-gui testing. \row \li \b {Rapid testing} \li Qt Test needs no special test-runners; no special registration for tests. \row \li \b {Data-driven testing} \li A test can be executed multiple times with different test data. \row \li \b {Basic GUI testing} \li Qt Test offers functionality for mouse and keyboard simulation. \row \li \b {Benchmarking} \li Qt Test supports benchmarking and provides several measurement back-ends. \row \li \b {IDE friendly} \li Qt Test outputs messages that can be interpreted by Qt Creator, Visual Studio, and KDevelop. \row \li \b Thread-safety \li The error reporting is thread safe and atomic. \row \li \b Type-safety \li Extensive use of templates prevent errors introduced by implicit type casting. \row \li \b {Easily extendable} \li Custom types can easily be added to the test data and test output. \endtable You can use a Qt Creator wizard to create a project that contains Qt tests and build and run them directly from Qt Creator. For more information, see \l {Qt Creator: Running Autotests}{Running Autotests}. \section1 Creating a Test To create a test, subclass QObject and add one or more private slots to it. Each private slot is a test function in your test. QTest::qExec() can be used to execute all test functions in the test object. In addition, you can define the following private slots that are \e not treated as test functions. When present, 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 \li \c{initTestCase()} will be called before the first test function is executed. \li \c{initTestCase_data()} will be called to create a global test data table. \li \c{cleanupTestCase()} will be called after the last test function was executed. \li \c{init()} will be called before each test function is executed. \li \c{cleanup()} will be called after every test function. \endlist Use \c initTestCase() for preparing the test. Every test should leave the system in a usable state, so it can be run repeatedly. Cleanup operations should be handled in \c cleanupTestCase(), so they get run even if the test fails. Use \c init() for preparing a test function. Every test function should leave the system in a usable state, so it can be run repeatedly. Cleanup operations should be handled in \c cleanup(), so they get run even if the test function fails and exits early. Alternatively, you can use RAII (resource acquisition is initialization), with cleanup operations called in destructors, to ensure they happen when the test function returns and the object moves out of scope. If \c{initTestCase()} fails, no test function will be executed. If \c{init()} fails, the following test function will not be executed, the test will proceed to the next test function. Example: \snippet code/doc_src_qtestlib.cpp 0 Finally, if the test class has a static public \c{void initMain()} method, it is called by the QTEST_MAIN macros before the QApplication object is instantiated. This was added in 5.14. For more examples, refer to the \l{Qt Test Tutorial}. \section1 Increasing Test Function Timeout QtTest limits the run-time of each test to catch infinite loops and similar bugs. By default, any test function call will be interrupted after five minutes. For data-driven tests, this applies to each call with a distinct data-tag. This timeout can be configured by setting the \c QTEST_FUNCTION_TIMEOUT environment variable to the maximum number of milliseconds that is acceptable for a single call to take. If a test takes longer than the configured timeout, it is interrupted, and \c qFatal() is called. As a result, the test aborts by default, as if it had crashed. To set \c QTEST_FUNCTION_TIMEOUT from the command line on Linux or macOS, enter: \badcode QTEST_FUNCTION_TIMEOUT=900000 export QTEST_FUNCTION_TIMEOUT \endcode On Windows: \badcode SET QTEST_FUNCTION_TIMEOUT=900000 \endcode Then run the test inside this environment. Alternatively, you can set the environment variable programmatically in the test code itself, for example by calling, from the \l{Creating a Test}{initMain()} special method of your test class: \badcode qputenv("QTEST_FUNCTION_TIMEOUT", "900000"); \endcode To calculate a suitable value for the timeout, see how long the test usually takes and decide how much longer it can take without that being a symptom of some problem. Convert that longer time to milliseconds to get the timeout value. For example, if you decide that a test that takes several minutes could reasonably take up to twenty minutes, for example on a slow machine, multiply \c{20 * 60 * 1000 = 1200000} and set the environment variable to \c 1200000 instead of the \c 900000 above. \if !defined(qtforpython) \section1 Building a Test You can build an executable that contains one test class that typically tests one class of production code. However, usually you would want to test several classes in a project by running one command. See \l {Chapter 1: Writing a Unit Test}{Writing a Unit Test} for a step by step explanation. \section2 Building with CMake and CTest You can use \l {Building with CMake and CTest} to create a test. \l{https://cmake.org/cmake/help/latest/manual/ctest.1.html}{CTest} enables you to include or exclude tests based on a regular expression that is matched against the test name. You can further apply the \c LABELS property to a test and CTest can then include or exclude tests based on those labels. All labeled targets will be run when \c {test} target is called on the command line. \note On Android, if you have one connected device or emulator, tests will run on that device. If you have more than one device connected, set the environment variable \c {ANDROID_DEVICE_SERIAL} to the \l {Android: Query for devices}{ADB serial number} of the device that you want to run tests on. There are several other advantages with CMake. For example, the result of a test run can be published on a web server using CDash with virtually no effort. CTest scales to very different unit test frameworks, and works out of the box with QTest. The following is an example of a CMakeLists.txt file that specifies the project name and the language used (here, \e mytest and C++), the Qt modules required for building the test (Qt5Test), and the files that are included in the test (\e tst_mytest.cpp). \quotefile code/doc_src_cmakelists.txt For more information about the options you have, see \l {Build with CMake}. \section2 Building with qmake If you are using \c qmake as your build tool, just add the following to your project file: \snippet code/doc_src_qtestlib.pro 1 If you would like to run the test via \c{make check}, add the additional line: \snippet code/doc_src_qtestlib.pro 2 To prevent the test from being installed to your target, add the additional line: \snippet code/doc_src_qtestlib.pro 3 See the \l{Building a Testcase}{qmake manual} for more information about \c{make check}. \section2 Building with Other Tools If you are using other build tools, make sure that you add the location of the Qt Test 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}. \endif \section1 Qt Test Command Line Arguments \section2 Syntax The syntax to execute an autotest takes the following simple form: \snippet 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 code/doc_src_qtestlib.qdoc 3 Runs the test function called \c toUpper with all available test data. \snippet 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 test data row called \c zero (if the specified test data doesn't exist, the associated test will fail and the available data tags are reported). \snippet code/doc_src_qtestlib.qdoc 5 Runs the \c testMyWidget function test, outputs every signal emission and waits 500 milliseconds after each simulated mouse/keyboard event. \section2 Options \section3 Logging Options The following command line options determine how test results are reported: \list \li \c -o \e{filename,format} \br Writes output to the specified file, in the specified format (one of \c txt, \c csv, \c junitxml, \c xml, \c lightxml, \c teamcity or \c tap). Use the special filename \c{-} (hyphen) to log to standard output. \li \c -o \e filename \br Writes output to the specified file. \li \c -txt \br Outputs results in plain text. \li \c -csv \br Outputs results as comma-separated values (CSV) suitable for import into spreadsheets. This mode is only suitable for benchmarks, since it suppresses normal pass/fail messages. \li \c -junitxml \br Outputs results as a \l{JUnit XML} document. \li \c -xml \br Outputs results as an XML document. \li \c -lightxml \br Outputs results as a stream of XML tags. \li \c -teamcity \br Outputs results in \l{TeamCity} format. \li \c -tap \br Outputs results in \l{Test Anything Protocol} (TAP) format. \endlist The first version of the \c -o option may be repeated in order to log test results in multiple formats, but no more than one instance of this option can log test results to standard output. If the first version of the \c -o option is used, neither the second version of the \c -o option nor the \c -txt, \c -xml, \c -lightxml, \c -teamcity, \c -junitxml or \c -tap options should be used. If neither version of the \c -o option is used, test results will be logged to standard output. If no format option is used, test results will be logged in plain text. \section3 Test Log Detail Options The following command line options control how much detail is reported in test logs: \list \li \c -silent \br Silent output; only shows fatal errors, test failures and minimal status messages. \li \c -v1 \br Verbose output; shows when each test function is entered. (This option only affects plain text output.) \li \c -v2 \br Extended verbose output; shows each \l QCOMPARE() and \l QVERIFY(). (This option affects all output formats and implies \c -v1 for plain text output.) \li \c -vs \br Shows all signals that get emitted and the slot invocations resulting from those signals. (This option affects all output formats.) \endlist \section3 Testing Options The following command-line options influence how tests are run: \list \li \c -functions \br Outputs all test functions available in the test, then quits. \li \c -datatags \br Outputs all data tags available in the test. A global data tag is preceded by ' __global__ '. \li \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. \li \c -keydelay \e ms \br Like -eventdelay, but only influences keyboard simulation and not mouse simulation. \li \c -mousedelay \e ms \br Like -eventdelay, but only influences mouse simulation and not keyboard simulation. \li \c -maxwarnings \e number \br Sets the maximum number of warnings to output. 0 for unlimited, defaults to 2000. \li \c -nocrashhandler \br Disables the crash handler on Unix platforms. On Windows, it re-enables the Windows Error Reporting dialog, which is turned off by default. This is useful for debugging crashes. \li \c -repeat \e n \br Run the testsuite n times or until the test fails. Useful for finding flaky tests. If negative, the tests are repeated forever. This is intended as a developer tool, and is only supported with the plain text logger. \li \c -skipblacklisted \br Skip the blacklisted tests. This option is intended to allow more accurate measurement of test coverage by preventing blacklisted tests from inflating coverage statistics. When not measuring test coverage, it is recommended to execute blacklisted tests to reveal any changes in their results, such as a new crash or the issue that caused blacklisting being resolved. \li \c -platform \e name \br This command line argument applies to all Qt applications, but might be especially useful in the context of auto-testing. By using the "offscreen" platform plugin (-platform offscreen) it's possible to have tests that use QWidget or QWindow run without showing anything on the screen. Currently the offscreen platform plugin is only fully supported on X11. \endlist \section3 Benchmarking Options The following command line options control benchmark testing: \list \li \c -callgrind \br Uses Callgrind to time benchmarks (Linux only). \li \c -tickcounter \br Uses CPU tick counters to time benchmarks. \li \c -eventcounter \br Counts events received during benchmarks. \li \c -minimumvalue \e n \br Sets the minimum acceptable measurement value. \li \c -minimumtotal \e n \br Sets the minimum acceptable total for repeated executions of a test function. \li \c -iterations \e n \br Sets the number of accumulation iterations. \li \c -median \e n \br Sets the number of median iterations. \li \c -vb \br Outputs verbose benchmarking information. \endlist \section3 Miscellaneous Options \list \li \c -help \br Outputs the possible command line arguments and gives some useful help. \endlist \section1 Qt Test Environment Variables You can set certain environment variables in order to affect the execution of an autotest: \list \li \c QTEST_DISABLE_CORE_DUMP \br Setting this variable to a non-zero value will disable the generation of a core dump file. \li \c QTEST_DISABLE_STACK_DUMP \br Setting this variable to a non-zero value will prevent Qt Test from printing a stacktrace in case an autotest times out or crashes. \li \c QTEST_FATAL_FAIL \br Setting this variable to a non-zero value will cause a failure in an autotest to immediately abort the entire autotest. This is useful to e.g. debug an unstable or intermittent failure in a test, by launching the test in a debugger. Support for this variable was added in Qt 6.1. \li \c QTEST_THROW_ON_FAIL (since 6.8) \br Setting this variable to a non-zero value will cause QCOMPARE()/QVERIFY() etc to throw on failure (as opposed to just returning from the immediately-surrounding function scope). \li \c QTEST_THROW_ON_SKIP (since 6.8) \br Same as \c QTEST_THROW_ON_FAIL, except affecting QSKIP(). \endlist \section1 Creating a Benchmark To create a benchmark, follow the instructions for creating a test and then add a \l QBENCHMARK macro or \l QTest::setBenchmarkResult() to the test function that you want to benchmark. In the following code snippet, the macro is used: \snippet code/doc_src_qtestlib.cpp 12 A test function that measures performance should contain either a single \c QBENCHMARK macro or a single call to \c setBenchmarkResult(). Multiple occurrences make no sense, because only one performance result can be reported per test function, or per data tag in a data-driven setup. Avoid changing the test code that forms (or influences) the body of a \c QBENCHMARK macro, or the test code that computes the value passed to \c setBenchmarkResult(). Differences in successive performance results should ideally be caused only by changes to the product you are testing. Changes to the test code can potentially result in misleading report of a change in performance. If you do need to change the test code, make that clear in the commit message. In a performance test function, the \c QBENCHMARK or \c setBenchmarkResult() should be followed by a verification step using \l QCOMPARE(), \l QVERIFY(), and so on. You can then flag a performance result as \e invalid if another code path than the intended one was measured. A performance analysis tool can use this information to filter out invalid results. For example, an unexpected error condition will typically cause the program to bail out prematurely from the normal program execution, and thus falsely show a dramatic performance increase. \section2 Selecting the Measurement Back-end 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 \li Name \li Command-line Argument \li Availability \row \li Walltime \li (default) \li All platforms \row \li CPU tick counter \li -tickcounter \li Windows, \macos, Linux, many UNIX-like systems. \row \li Event Counter \li -eventcounter \li All platforms \row \li Valgrind Callgrind \li -callgrind \li Linux (if installed) \row \li Linux Perf \li -perf \li Linux \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). The Linux Performance Monitoring solution is available only on Linux and provides many different counters, which can be selected by passing an additional option \c {-perfcounter countername}, such as \c {-perfcounter cache-misses}, \c {-perfcounter branch-misses}, or \c {-perfcounter l1d-load-misses}. The default counter is \c {cpu-cycles}. The full list of counters can be obtained by running any benchmark executable with the option \c -perfcounterlist. \note \list \li Using the performance counter may require enabling access to non-privileged applications. \li Devices that do not support high-resolution timers default to one-millisecond granularity. \endlist See \l {Chapter 5: Writing a Benchmark}{Writing a Benchmark} in the Qt Test Tutorial for more benchmarking examples. \section1 Using Global Test Data You can define \c{initTestCase_data()} to set up a global test data table. Each test is run once for each row in the global test data table. When the test function itself \l{Chapter 2: Data Driven Testing}{is data-driven}, it is run for each local data row, for each global data row. So, if there are \c g rows in the global data table and \c d rows in the test's own data-table, the number of runs of this test is \c g times \c d. Global data is fetched from the table using the \l QFETCH_GLOBAL() macro. The following are typical use cases for global test data: \list \li Selecting among the available database backends in QSql tests to run every test against every database. \li Doing all networking tests with and without SSL (HTTP versus HTTPS) and proxying. \li Testing a timer with a high precision clock and with a coarse one. \li Selecting whether a parser shall read from a QByteArray or from a QIODevice. \endlist For example, to test each number provided by \c {roundTripInt_data()} with each locale provided by \c {initTestCase_data()}: \snippet code/src_qtestlib_qtestcase_snippet.cpp 31 On the command-line of a test you can pass the name of a function (with no test-class-name prefix) to run only that one function's tests. If the test class has global data, or the function is data-driven, you can append a data tag, after a colon, to run only that tag's data-set for the function. To specify both a global tag and a tag specific to the test function, combine them with a colon between, putting the global data tag first. For example \snippet code/doc_src_qtestlib.qdoc 6 will run the \c zero test-case of the \c roundTripInt() test above (assuming its \c TestQLocale class has been compiled to an executable \c testqlocale) in each of the locales specified by \c initTestCase_data(), while \snippet code/doc_src_qtestlib.qdoc 7 will run all three test-cases of \c roundTripInt() only in the C locale and \snippet code/doc_src_qtestlib.qdoc 8 will only run the \c zero test-case in the C locale. Providing such fine-grained control over which tests are to be run can make it considerably easier to debug a problem, as you only need to step through the one test-case that has been seen to fail. */ /*! \page qtest-tutorial.html \brief A short introduction to testing with Qt Test. \nextpage {Chapter 1: Writing a Unit Test}{Chapter 1} \ingroup best-practices \title Qt Test Tutorial This tutorial introduces some of the features of the Qt Test framework. It is divided into six chapters: \list 1 \li \l {Chapter 1: Writing a Unit Test}{Writing a Unit Test} \li \l {Chapter 2: Data Driven Testing}{Data Driven Testing} \li \l {Chapter 3: Simulating GUI Events}{Simulating GUI Events} \li \l {Chapter 4: Replaying GUI Events}{Replaying GUI Events} \li \l {Chapter 5: Writing a Benchmark}{Writing a Benchmark} \li \l {Chapter 6: Skipping Tests with QSKIP}{Skipping Tests} \endlist */