Test Browser - Run MATLAB tests and view results - MATLAB (original) (raw)

Run MATLAB tests and view results

Since R2023a

Description

The Test Browser app enables you to run script-based, function-based, and class-based tests interactively.

You can use the test browser to:

• Create a test suite from files or folders.
• Run all or part of the specified tests.
• Access diagnostics and debug test failures.
• Customize a test run with options, such as running tests in parallel (requires Parallel Computing Toolbox™) or specifying a level of test output detail.
• Generate an HTML code coverage report for MATLAB® source code.

Overview of Test Browser Toolbar

This table provides a summary of the buttons on the toolbar of the test browser.

Open the Test Browser App

• MATLAB Toolstrip: On the Apps tab, underMATLAB, click the app icon.
• MATLAB command prompt: Enter testBrowser.

You can also open the test browser if you run tests using either of these options. These options do not support script-based tests:

• Right-click a test file in the Files panel and then select Run Tests.
• Open a test file in the MATLAB Editor or Live Editor and then run the tests using the Run Tests section of the Editor orLive Editor tab on the toolstrip.

Examples

expand all

Debug a test failure by using the test browser.

Create a function-based test file named sampleTest.m in your current folder. The file contains two tests that pass and one test that intentionally fails.

function tests = sampleTest tests = functiontests(localfunctions); end

function testA(testCase) verifyEqual(testCase,2+3,5) end

function testB(testCase) verifyLessThan(testCase,-1.1,-1.2) end

function testC(testCase) verifySubstring(testCase,"Hello World!","llo") end

Open the test browser.

Add the tests in the sampleTest.m file to the test browser. To add the tests, click the Add tests button on the toolbar and then select the test file. The test browser creates a test suite and displays the test names in a test tree.

Run the tests by clicking the Run current suite button on the toolbar. Two tests pass and one test fails due to a verification failure. If you click the node of testB in the test tree, the Test Diagnostics Viewer section at the bottom of theTest Browser panel displays the diagnostic information.

Click the line-number link provided for testB in the test tree or the hyperlinked path in the Test Diagnostics Viewer section to view the code that resulted in the failure. To debug the test failure, set a breakpoint on the highlighted line of code by clicking its line number. Then, runtestB by right-clicking its node in the test tree and selecting . MATLAB enters debug mode and enables debugging capabilities that you can use to investigate the cause of the test failure.

If you rerun the failed test in the Command Window with an appropriate actual value, the test passes.

Running sampleTest 10 verifyLessThan(testCase,-1.1,-1.2) K>> verifyLessThan(testCase,-1.3,-1.2) K>>

Exit debug mode and clear the breakpoint. Update the actual value in the test file, and save the file. Run the tests again by clicking the Run current suite button . All the tests pass.

Run parameterized tests by using the test browser.

In a file named cleanData.m in your current folder, create thecleanData function. The function accepts a numeric array and returns a cleaned and sorted version of the array. It vectorizes the array, removes theNaN, 0, and Inf entries, and finally sorts the vector.

function y = cleanData(X) y = X(:); % Vectorize the array y = rmmissing(y); % Remove NaN entries % Remove 0 and Inf entries idx = (y == 0 | y == Inf); y = y(~idx); % If the vector is empty, set it to eps if isempty(y) y = eps; end y = sort(y); % Sort the vector end

To test the cleanData function, create the parameterizedCleanDataTest test class in a file namedCleanDataTest.m in your current folder.

classdef CleanDataTest < matlab.unittest.TestCase properties (TestParameter) data = struct("empty",[],"scalar",0, ... "vector",[13 NaN 0],"matrix",[NaN 2 0; 1 Inf 3]); end

methods (Test)
    function sortTest(testCase,data)
        actual = cleanData(data);
        testCase.verifyTrue(issorted(actual))
    end
    function nonemptyTest(testCase,data)
        actual = cleanData(data);
        testCase.verifyNotEmpty(actual)
    end
end

end

Open the test browser.

Add the tests in CleanDataTest to the test browser. To add the tests, click the Add tests button on the toolbar and then select theCleanDataTest.m file. The test browser creates a test suite and displays the test names in a test tree. Each name includes information about parameterization.

Run the tests that use the "matrix" parameter name. To run the tests, first select the two tests by pressing the Ctrl key while making selections with the mouse. Then, right-click one of the selected nodes and select . In this example, both the tests pass.

To run the remaining tests, click the Not Run button on the toolbar. Then, click the Run current suite button . After running the tests, view the test results by clicking the Passed button on the toolbar. The statuses of tests that ran before the most recent test run appear dimmed in theTest Browser panel. (since R2025a)

You can customize a test run by selecting options in the Test Browser toolbar. Select an option to run your tests in parallel (requires Parallel Computing Toolbox).

In a file named TestRand.m in your current folder, create the parameterized TestRand class. The class results in 175 tests.

classdef TestRand < matlab.unittest.TestCase
properties (TestParameter) dim1 = createDimensionSizes; dim2 = createDimensionSizes; dim3 = createDimensionSizes; type = {'single','double'}; end

methods (Test)
    function testRepeatable(testCase,dim1,dim2,dim3)
        state = rng;
        firstRun = rand(dim1,dim2,dim3);
        rng(state)
        secondRun = rand(dim1,dim2,dim3);
        testCase.verifyEqual(firstRun,secondRun)
    end
    function testClass(testCase,dim1,dim2,type)
        testCase.verifyClass(rand(dim1,dim2,type),type)
    end
end

end

function sizes = createDimensionSizes % Create logarithmically spaced sizes up to 100 sizes = num2cell(round(logspace(0,2,5))); end

Open the test browser.

Add the tests in the TestRand class to the test browser. To add the tests, click the Add tests button on the toolbar and then select theTestRand.m file. The test browser creates a test suite and displays the test names in a test tree.

To run the tests in parallel, first click the Enable parallel test execution button on the toolbar. Then, click the Run current suite button . The test browser divides the test suite into groups and runs the groups on the available workers. You can see information about the groups in the Command Window. The test browser might vary the order and number of groups or which tests it includes in each group.

Split tests into 18 groups and running them on 6 workers.

Finished Group 1

Running TestRand .......... .. Done TestRand

Finished Group 3

Running TestRand .......... . Done TestRand

Finished Group 2

Running TestRand .......... .. Done TestRand

Finished Group 5

Running TestRand .......... . Done TestRand

Finished Group 4

Running TestRand .......... . Done TestRand

Finished Group 6

Running TestRand .......... . Done TestRand

Finished Group 7

Running TestRand .......... Done TestRand

Finished Group 9

Running TestRand .......... Done TestRand

Finished Group 8

Running TestRand .......... Done TestRand

Finished Group 11

Running TestRand ......... Done TestRand

Finished Group 10

Running TestRand .......... Done TestRand

Finished Group 12

Running TestRand ......... Done TestRand

Finished Group 13

Running TestRand ......... Done TestRand

Finished Group 15

Running TestRand ........ Done TestRand

Finished Group 14

Running TestRand ......... Done TestRand

Finished Group 16

Running TestRand ........ Done TestRand

Finished Group 17

Running TestRand ........ Done TestRand

Finished Group 18

Running TestRand ....... Done TestRand

Collect code coverage information and generate an HTML code coverage report for your source code when you run tests using the test browser.

In a file named cleanData.m in your current folder, create thecleanData function. The function accepts a numeric array and returns a cleaned and sorted version of the array. It vectorizes the array, removes theNaN, 0, and Inf entries, and finally sorts the vector.

function y = cleanData(X) y = X(:); % Vectorize the array y = rmmissing(y); % Remove NaN entries % Remove 0 and Inf entries idx = (y == 0 | y == Inf); y = y(~idx); % If the vector is empty, set it to eps if isempty(y) y = eps; end y = sort(y); % Sort the vector end

To test the cleanData function, create the parameterizedCleanDataTest test class in a file namedCleanDataTest.m in your current folder.

classdef CleanDataTest < matlab.unittest.TestCase properties (TestParameter) data = struct("empty",[],"scalar",0, ... "vector",[13 NaN 0],"matrix",[NaN 2 0; 1 Inf 3]); end

methods (Test)
    function sortTest(testCase,data)
        actual = cleanData(data);
        testCase.verifyTrue(issorted(actual))
    end
    function nonemptyTest(testCase,data)
        actual = cleanData(data);
        testCase.verifyNotEmpty(actual)
    end
end

end

Open the test browser.

Add the tests in CleanDataTest to the test browser. To add the tests, click the Add tests button on the toolbar, and then select theCleanDataTest.m file. The test browser creates a test suite and displays the test names in a test tree.

To generate a code coverage report for your source code incleanData.m:

1. Click the Open coverage settings button on the toolbar, and then select Enable coverage reporting.
2. In the Coverage Settings section, underSource, click the Add Files button and select the cleanData.m file.
3. In the Coverage Settings section, underReport, select Open coverage report after run if the check box is clear.

Run the tests by clicking the Run current suite button on the toolbar. The test browser runs the tests, generates an HTML code coverage report, and opens it after the test run. In this example, all the tests pass and the source code receives full function and statement coverage.

Related Examples

• Run Tests Using Test Browser

Version History

Introduced in R2023a

expand all

To rerun the tests from the most recent test run, click the drop-down arrow to the right of the Run current suite button on the toolbar and then select .

You can distinguish the test results produced in the most recent test run from those produced in earlier test runs. In the Test Browser panel, the statuses of tests that ran before the most recent test run appear dimmed.

If you run a specific test procedure using the Run Current Test button on the MATLAB Toolstrip, the test browser retains the tests already imported from the corresponding test file. In previous releases, the test browser removes the tests imported from the test file before running the specified test procedure.

You can use Test Browser menu options to:

• Add the tests from the currently open test file to the test browser. Click the drop-down arrow to the right of the Add tests button on the toolbar and then select .
• Clear the test results and return all the tests to a Not Run status . Click the three-dot button on the toolbar and then select .

You can run script-based, function-based, and class-based tests using the test browser in MATLAB Online™.

The behavior of the test browser has changed when a group of files includes both valid and invalid test files:

• If you add tests to the test browser by selecting multiple test files, then the test browser imports the tests in the valid test files. In previous releases, the test browser does not import any tests if the group of selected files includes invalid test files.
• If you run a group of both valid and invalid test files, then the test browser runs the tests in the valid test files. In previous releases, the test browser does not run any tests if the group includes invalid test files.

An invalid test file is a test file that the testing framework cannot run. Examples include a test file that contains syntax errors, a function-based test file that is missing local functions, and a file with a Test method that is passed an undefined parameterization property.