Testing: Questions and Answers

You are encouraged to add your own questions (and answers) here!

Q: How do I define a new unit test?

A: Use language-appropriate constructs, like GTest for C++. You can define a new file if need be, such as:

(in a BUILD.gn file)

executable("unittests") {
  output_name = "scenic_unittests"
  testonly = true
  sources = ["some_test.cc"],
  deps = [":some_dep"],
}

Q: What ensures it is run?

A: An unbroken chain of dependencies that roll up to your fx set command's universe of available packages (expandable using the --with flag), typically going through the all target of //<layer>/packages/tests/BUILD.gn, such as //garnet/packages/tests:all.

For example:

//src/ui/scenic:scenic_unittests

is an executable, listed under the "tests" stanza of

//src/ui/scenic:scenic_tests

which is a package, which is listed in a package group

//garnet/packages/tests:scenic

which is itself included in the catch-all group,

//garnet/packages/tests:all

Your product definition (typically one found in products/ may or may not transitively include this test group. If it doesn't, add it to your fx set command, like so:

fx set ... --with //garnet/packages/tests:scenic_tests

Typically, one just adds a new test to an existing binary, or a new test binary to an existing package.

Q: How do I run this unit test on a QEMU instance?

There's the easy way if your QEMU has networking, and the hard way if it doesn't.

A (with networking): In one terminal, start your QEMU instance with fx qemu -N. Next, on another terminal, type in fx run-test scenic_tests.

This invocation runs all the test executables in the scenic_tests package.

A (no networking): Start a QEMU instance (fx qemu), and then manually invoke the runtests command.

In the QEMU shell, type in run-test-component scenic_tests. The argument is a specific directory containing the test executables.

Note Well! Without networking, the files are loaded into the QEMU instance at startup. So after rebuilding a test, you'll need to shutdown and re-start the QEMU instance to see the rebuilt test.

To exit QEMU, dm shutdown.

Q: How do I run this unit test on my development device?

A: Either manual invocation, like in QEMU, or fx run-test to a running device.

Note that the booted device may not contain your binary at startup, but fx run-test will build the test binary, ship it over to the device, and run it, while piping the output back to your workstation terminal. Slick!

Make sure your device is running (hit Ctrl-D to boot an existing image) and connected to your workstation.

From your workstation, fx run-test scenic_tests will serially run through all test executables contained in the scenic_tests package.

To run just one test executable, fx run-test scenic_test -t scenic_unittests, where the argument to -t is the executable name.

You can automatically rebuild, install, and run your tests on every source file change with fx -i. For instance: fx -i run-test scenic_tests.

Q: Where are the test results captured?

A: The output is directed to your terminal.

There does exist a way to write test output into files (including a summary JSON file), which is how CQ bots collect the test output for automated runs.

Q: How to disable a test? How to find and run disabled tests?

A: There are several ways to do this. Whenever doing any of these, be sure to file a bug and reference that bug in a comment in the code that disables the test.

Tag the test as flaky

You can do this by adding "flaky" to the tags field in the test environment. This operates on the entire test target (which corresponds to an executable). It will prevent this target from running on the builders in the commit queue, and enable the target on special flaky builders that continue to run the test in CI. Be sure to note the bug in a comment in the BUILD.gn file. Example change.

If you want to disable only some tests that are part of a larger test target, you'll need to split the target into two GN targets, and tag one as flaky.

C++ googletest only: Prefix name with DISABLED

To disable a particular test inside of a larger test executable, you can mark it as disabled. Disabled tests are defined by having their name prefixed with DISABLED_. One way to find them is therefore simply git grep DISABLED_.

If running the test outputs YOU HAVE 1 DISABLED TEST, you can also pass the following flags to find out which test is disabled: fx run-test scenic_tests -- --gtest_list_tests --gtest_filter=*DISABLED_*.

To force-run disabled tests: fx run-test scenic_tests -- --gtest_also_run_disabled_tests.

Mark test disabled

Alternatively, you may also disable an entire test executable within a package containing several test executables. To do this, edit the BUILD.gn as follows: tests = [ { name = "scenic_unittests", disabled = true } ]. As a result, scenic_unittests will be put in a disabled sub-directory of /pkgfs/packages/<package_name>/0/test, and will not be run by the CQ system.

Comment out the test

To disable a particular test inside of a larger test executable, you can comment out the code that defines that test.

Q: How do I run a bunch of tests automatically? How do I ensure all dependencies are tested?

A: Upload your patch to Gerrit and do a CQ dry run.

Q: How do I run this unit test in a CQ dry run?

A: Clicking on CQ dry run (aka +1) will take your CL's properly defined unit test and run it on multiple bots, one for each build target (x86-64 versus arm64, release versus debug). Each job will have an output page showing all the tests that ran.

Q: How do I use some build time artifacts in my unit test?

A: The simplest artifact is just a file that is in your source directory. For this you just need to add it to resources attribute of the package definition of your unit test. For example, you may do something like this in your BUILD.gn:

rustc_binary("my-great-app") {
  with_unit_tests = true

  ...
}

test_package("my-great-app-tests") {
  deps = [
    ":my-great-app_test",
  ]

  resources = [
    {
      path = "source.zip"
      dest = "testing.zip"
    }
  ]

The file will be available as /pkg/data/testing.zip inside the environment where the test binary will be executed.

TODO: If you want an artifact that is generated as part of the build process, you should probably add the rule that generates the artifact to the data_deps array of the test_package rule. But I have not tried it yet. Update this section when you will try it :)

Send feedback