View Source mix test (Mix v1.18.0-dev)
Runs the tests for a project.
This task starts the current application, loads up
test/test_helper.exs
and then, requires all files matching the
test/**/*_test.exs
pattern in parallel.
A list of files and/or directories can be given after the task name in order to select the files to run:
$ mix test test/some/particular/file_test.exs
$ mix test test/some/particular/dir
Tests in umbrella projects can be run from the root by specifying
the full suite path, including apps/my_app/test
, in which case
recursive tests for other child apps will be skipped completely:
# To run all tests for my_app from the umbrella root
$ mix test apps/my_app/test
# To run a given test file on my_app from the umbrella root
$ mix test apps/my_app/test/some/particular/file_test.exs
Understanding test results
When you run your test suite, it prints results as they run with a summary at the end, as seen below:
$ mix test
Running ExUnit with seed: 646219, max_cases: 16
...
1) test greets the world (FooTest)
test/foo_test.exs:5
Assertion with == failed
code: assert Foo.hello() == :world!
left: :world
right: :world!
stacktrace:
test/foo_test.exs:6: (test)
........
Finished in 0.05 seconds (0.00s async, 0.05s sync)
1 doctest, 11 tests, 1 failure
For each test, the test suite will print a dot. Failed tests are printed immediately in the format described in the next section.
After all tests run, we print the suite summary. The first
line contains the total time spent on the suite, followed
by how much time was spent on async tests (defined with
use ExUnit.Case, async: true
) vs sync ones:
Finished in 0.05 seconds (0.00s async, 0.05s sync)
Developers want to minimize the time spent on sync tests whenever possible, as sync tests run serially and async tests run concurrently.
Finally, how many tests we have run, how many of them failed, how many were invalid, and so on.
Understanding test failures
First, it contains the failure counter, followed by the test name and the module the test was defined:
1) test greets the world (FooTest)
The next line contains the exact location of the test in the
FILE:LINE
format:
test/foo_test.exs:5
If you want to re-run only this test, all you need to do is to
copy the line above and paste it in front of mix test
:
$ mix test test/foo_test.exs:5
Then we show the error message, code snippet, and general information about the failed test:
Assertion with == failed
code: assert Foo.hello() == :world!
left: :world
right: :world!
If your terminal supports coloring (see the "Coloring" section below),
a diff is typically shown between left
and right
sides. Finally,
we print the stacktrace of the failure:
stacktrace:
test/foo_test.exs:6: (test)
Command line options
--all-warnings
(--no-all-warnings
) - prints all warnings, including previous compilations (default is true except on errors)-b
,--breakpoints
(since v1.17.0) - sets a breakpoint at the beginning of every test. The debugger goes line-by-line and can access all variables and imports (but not local functions). You can pressn
for the next line andc
for the next test. This automatically sets--trace
--color
- enables color in the output--cover
- runs coverage tool. See "Coverage" section below--exclude
- excludes tests that match the filter. This option may be given several times to apply different filters, such as--exclude ci --exclude slow
--exit-status
- use an alternate exit status to use when the test suite fails (default is 2).--export-coverage
- the name of the file to export coverage results to. Only has an effect when used with--cover
--failed
- runs only tests that failed the last time they ran--force
- forces compilation regardless of modification times--formatter
- sets the formatter module that will print the results. Defaults to ExUnit's built-in CLI formatter--include
- includes tests that match the filter. This option may be given several times to apply different filters, such as--include ci --include slow
--listen-on-stdin
- runs tests, and then listens on stdin. It will re-run tests once a newline is received. See the "File system watchers" section below--max-cases
- sets the maximum number of tests running asynchronously. Only tests from different modules run in parallel. Defaults to twice the number of cores--max-failures
- the suite stops evaluating tests when this number of test failures is reached. It runs all tests if omitted--max-requires
- sets the maximum number of test files to compile in parallel. Setting this to 1 will compile test files sequentially.--no-archives-check
- does not check archives--no-color
- disables color in the output--no-compile
- does not compile, even if files require compilation--no-deps-check
- does not check dependencies--no-elixir-version-check
- does not check the Elixir version frommix.exs
--no-start
- does not start applications after compilation--only
- runs only tests that match the filter--partitions
- sets the amount of partitions to split tests in. It must be a number greater than zero. If set to one, it acts a no-op. If more than one, then you must also set theMIX_TEST_PARTITION
environment variable with the partition to use in the current test run. See the "Operating system process partitioning" section for more information--preload-modules
- preloads all modules defined in applications--profile-require time
- profiles the time spent to require test files. Used only for debugging. The test suite does not run--raise
- immediately raises if the test suite fails, instead of continuing the execution of other Mix tasks--repeat-until-failure
(since v1.17.0) - sets the number of repetitions for running the suite until it fails. This is useful for debugging flaky tests within the same instance of the Erlang VM. For example,--repeat-until-failure 10000
repeats the test suite up to 10000 times until the first failure. This can be combined with--max-failures 1
to immediately stop if one test fails. However, if there is any leftover global state after running the tests, re-running the suite may trigger unrelated failures.--seed
- seeds the random number generator used to randomize the order of tests;--seed 0
disables randomization so the tests in a single file will always be ran in the same order they were defined in--slowest
- prints timing information for the N slowest tests. Includes time spent inExUnit.Callbacks.setup/1
. Automatically sets--trace
and--preload-modules
--slowest-modules
(since v1.17.0) - prints timing information for the N slowest modules. Includes time spent inExUnit.Callbacks.setup/1
. Automatically sets--trace
and--preload-modules
--stale
- runs only tests which reference modules that changed since the last time tests were ran with--stale
. You can read more about this option in the "The --stale option" section below--timeout
- sets the timeout for the tests--trace
- runs tests with detailed reporting. Automatically sets--max-cases
to1
. Note that in trace mode test timeouts will be ignored as timeout is set to:infinity
--warnings-as-errors
(since v1.12.0) - treats compilation warnings (from loading the test suite) as errors and returns a non-zero exit status if the test suite would otherwise pass. Note that failures reported by--warnings-as-errors
cannot be retried with the--failed
flag.This option only applies to test files. To treat warnings as errors during compilation and during tests, run:
MIX_ENV=test mix do compile --warnings-as-errors + test --warnings-as-errors
Configuration
These configurations can be set in the def project
section of your mix.exs
:
:test_coverage
- a set of options to be passed down to the coverage mechanism. See the "Coverage" section for more information:test_elixirc_options
- the compiler options to used when loading/compiling test files. By default it disables the debug chunk and docs chunk:test_paths
- list of paths containing test files. Defaults to["test"]
if thetest
directory exists, otherwise, it defaults to[]
. It is expected that all test paths contain atest_helper.exs
file:test_pattern
- a pattern to load test files. Defaults to*_test.exs
:warn_test_pattern
- a pattern to match potentially misnamed test files and display a warning. Defaults to*_test.ex
Coloring
Coloring is enabled by default on most Unix terminals. They are also available on Windows consoles from Windows 10, although it must be explicitly enabled for the current user in the registry by running the following command:
$ reg add HKCU\Console /v VirtualTerminalLevel /t REG_DWORD /d 1
After running the command above, you must restart your current console.
Filters
ExUnit provides tags and filtering functionality that allow developers to select which tests to run. The most common functionality is to exclude some particular tests from running by default in your test helper file:
# Exclude all external tests from running
ExUnit.configure(exclude: [external: true])
Then, whenever desired, those tests could be included in the run via the
--include
option:
$ mix test --include external:true
The example above will run all tests that have the external option set to
true
. It is also possible to include all examples that have a given tag,
regardless of its value:
$ mix test --include external
Note that all tests are included by default, so unless they are excluded
first (either in the test helper or via the --exclude
option) the
--include
option has no effect.
For this reason, Mix also provides an --only
option that excludes all
tests and includes only the given ones:
$ mix test --only external
Which is similar to:
$ mix test --include external --exclude test
It differs in that the test suite will fail if no tests are executed when the --only
option is used.
In case a single file is being tested, it is possible to pass one or more specific line numbers to run only those given tests:
$ mix test test/some/particular/file_test.exs:12
Which is equivalent to:
$ mix test --exclude test --include line:12 test/some/particular/file_test.exs
Or:
$ mix test test/some/particular/file_test.exs:12:24
Which is equivalent to:
$ mix test --exclude test --include line:12 --include line:24 test/some/particular/file_test.exs
If a given line starts a describe
block, that line filter runs all tests in it.
Otherwise, it runs the closest test on or before the given line number.
Coverage
Elixir provides built-in line-based test coverage via the --cover
flag.
The test coverages shows which lines of code and in which files were executed
during the test run.
Limitations
Coverage in Elixir has the following limitations:
Literals, such as atoms, strings, and numbers, are not traced by coverage. For example, if a function simply returns
:ok
, the atom:ok
itself is never taken into account by coverage;Macros, such as the ones defined by
defmacro/2
anddefguard/2
, and code invoked only by macros are never considered as covered, unless they are also invoked during the tests themselves. That's because macros are invoked at compilation time, before the test coverage instrumentation begins;
Configuration
The :test_coverage
configures the coverage tool and accepts the following options:
:output
- the output directory for cover results. Defaults to"cover"
.:tool
- a module specifying the coverage tool to use.:summary
- at the end of each coverage run, a summary of each module is printed, with results in red or green depending on whether the percentage is below or above a given threshold. The task will exit with status of 1 if the total coverage is below the threshold. The:summary
option allows you to customize the summary generation and defaults to[threshold: 90]
, but it may be set tofalse
to disable such reports.:export
- a filename to export results to instead of generating the coverage result on the fly. The.coverdata
extension is automatically added to the given file. This option is automatically set via the--export-coverage
option or when using process partitioning. Seemix test.coverage
to compile a report from multiple exports.:ignore_modules
- modules to ignore from generating reports and in summaries. It is a list of module names as atoms and regular expressions that are matched against the module names.:local_only
- by default coverage only tracks local calls, set this option to false if you plan to run coverage across nodes.
By default, a wrapper around OTP's cover
is used as the default coverage
tool. You can learn more about how it works in the docs for
mix test.coverage
. Your tool of choice can be given as follows:
def project() do
[
...
test_coverage: [tool: CoverModule]
...
]
end
CoverModule
can be any module that exports start/2
, receiving the
compilation path and the test_coverage
options as arguments.
It must return either nil
or an anonymous function of zero arity that
will run after the test suite is done.
Operating system process partitioning
While ExUnit supports the ability to run tests concurrently within the same Elixir instance, it is not always possible to run all tests concurrently. For example, some tests may rely on global resources.
For this reason, mix test
supports partitioning the test files across
different Elixir instances. This is done by setting the --partitions
option
to an integer, with the number of partitions, and setting the MIX_TEST_PARTITION
environment variable to control which test partition that particular instance
is running. This can also be useful if you want to distribute testing across
multiple machines.
For example, to split a test suite into 4 partitions and run them, you would use the following commands:
$ MIX_TEST_PARTITION=1 mix test --partitions 4
$ MIX_TEST_PARTITION=2 mix test --partitions 4
$ MIX_TEST_PARTITION=3 mix test --partitions 4
$ MIX_TEST_PARTITION=4 mix test --partitions 4
The test files are sorted upfront in a round-robin fashion. Note the partition
itself is given as an environment variable so it can be accessed in config files
and test scripts. For example, it can be used to setup a different database instance
per partition in config/test.exs
.
If partitioning is enabled and --cover
is used, no cover reports are generated,
as they only contain a subset of the coverage data. Instead, the coverage data
is exported to files such as cover/MIX_TEST_PARTITION.coverdata
. Once you have
the results of all partitions inside cover/
, you can run mix test.coverage
to
get the unified report.
The --stale option
The --stale
command line option attempts to run only the test files which
reference modules that have changed since the last time you ran this task with
--stale
.
The first time this task is run with --stale
, all tests are run and a manifest
is generated. On subsequent runs, a test file is marked "stale" if any modules it
references (and any modules those modules reference, recursively) were modified
since the last run with --stale
. A test file is also marked "stale" if it has
been changed since the last run with --stale
.
The --stale
option is extremely useful for software iteration, allowing you to
run only the relevant tests as you perform changes to the codebase.
File-system watchers
You can integrate mix test
with file system watchers through the command line
via the --listen-on-stdin
option. For example, you can use fswatch
or similar to emit newlines whenever there is a change, which will cause your test
suite to re-run:
$ fswatch lib test | mix test --listen-on-stdin
This can be combined with the --stale
option to re-run only the test files that
have changed as well as the tests that have gone stale due to changes in lib
.
Aborting the suite
It is possible to abort the test suite with Ctrl+\
, which sends a SIGQUIT
signal to the Erlang VM. ExUnit will intercept this signal to show all tests
that have been aborted and print the results collected so far.
This can be useful in case the suite gets stuck and you don't want to wait until the timeout times passes (which defaults to 30 seconds).