Testing on the Pyret track

Testing using the web editor

To run tests, click the "Run Tests" button and all the tests will be run.

Testing locally

Before working locally, see the installation docs to set up Pyret.

Each exercise has multiple tests. When you first download the exercise, only one test will be enabled. The recommended workflow is to run the test (which should initially fail) and then make changes to your code until that test passes. Then, enable the next test in the test file (see below for details) and repeat this process(modify the code, run tests, and move on) until all the tests pass. Some students preemptively enable all the tests before working on the code; this can be overwhelming with how many tests are failing and make it harder to solve the exercise.

Running tests

Each exercise will have a solution file and a test file. For instance, the Hello World exercise has both a hello-world.arr file and a hello-world-test.arr file. Your code will go into hello-world.arr. To run tests, execute the pyret command with the test file as an argument. For instance, running pyret hello-world-test.arr in the hello-world directory will run the currently enabled Hello World tests.

Enabling tests

Every test file has one or more tests, each test wrapped in its own function. At the bottom, there is a list of TestRun values representing the tests to be run. Each test in the list contains two values: the test function to run and an active value which controls whether the test will be executed. For instance, the etl-test.arr file contains four tests, with the first active and the rest inactive.

data TestRun: test(run, active) end

[list: 
  test(single-letter, true),
  test(single-score-multiple-letters, false),
  test(multiple-scores-multiple-letters, false),
  test(multiple-scores-different-numbers-of-letters, false)
].each(lam(t): when t.active: t.run() end end)

To enable additional tests, change the active value from false to true, and then rerun the test file with the pyret command.

provide

Tests on this track will import your file, allowing access to anything explicitly exported from your code.

To export variables, you need to add a provide statement at the beginning of your file.

The following snippets are three valid ways to export a, b, and c.

# using a list of bindings
provide a, b, c end

# using an object literal
provide {
  a: a,
  b: b,
  c: c
}
end

provide * is a shorthand for exporting all the top-level bindings except for custom data types However, it's generally not recommended because Pyret is strict about not allowing shadowing.

provide-types

Some exercises will require a custom data type to be exported for testing purposes. In those situations, you can use a provide-types statement. Since a data type has additional functions that might not be exported, it's advised to use provide-types * despite the shadowing concern.

provide-types *

data MyPoint:
  | two-dim(x, y)
  | three-dim(x, y, z)
end

All exercise stubs will have either provide or provide-types statements set up for your use.