Exercism Uiua Test Runner

The Docker image to automatically run tests on Uiua solutions submitted to Exercism.

Getting started

Build the test runner, conforming to the Test Runner interface specification. Update the files to match your track's needs. At the very least, you'll need to update bin/run.sh, Dockerfile and the test solutions in the tests directory

• Tip: look for TODO: comments to point you towards code that need updating
• Tip: look for OPTIONAL: comments to point you towards code that could be useful
• Tip: if it proves impossible for the Docker image to work on a read-only filesystem, remove the --read-only flag from the bin/run-in-docker.sh and bin/run-tests-in-docker.sh files. We don't yet enforce a read-only file system in production, but we might in the future!

Run the test runner

To run the tests of a single solution, do the following:

1. Open a terminal in the project's root
2. Run ./bin/run.sh <exercise-slug> <solution-dir> <output-dir>

Once the test runner has finished, its results will be written to <output-dir>/results.json.

Run the test runner on a solution using Docker

This script is provided for testing purposes, as it mimics how test runners run in Exercism's production environment.

To run the tests of a single solution using the Docker image, do the following:

1. Open a terminal in the project's root
2. Run ./bin/run-in-docker.sh <exercise-slug> <solution-dir> <output-dir>

Once the test runner has finished, its results will be written to <output-dir>/results.json.

Run the tests

To run the tests to verify the behavior of the test runner, do the following:

1. Open a terminal in the project's root
2. Run ./bin/run-tests.sh

These are golden tests that compare the results.json generated by running the current state of the code against the "known good" tests/<test-name>/expected_results.json. All files created during the test run itself are discarded.

When you've made modifications to the code that will result in a new "golden" state, you'll need to update the affected tests/<test-name>/expected_results.json file(s).

Run the tests using Docker

This script is provided for testing purposes, as it mimics how test runners run in Exercism's production environment.

To run the tests to verify the behavior of the test runner using the Docker image, do the following:

1. Open a terminal in the project's root
2. Run ./bin/run-tests-in-docker.sh

These are golden tests that compare the results.json generated by running the current state of the code against the "known good" tests/<test-name>/expected_results.json. All files created during the test run itself are discarded.

When you've made modifications to the code that will result in a new "golden" state, you'll need to update the affected tests/<test-name>/expected_results.json file(s).

Benchmarking

There are two scripts you can use to benchmark the test runner:

1. ./bin/benchmark.sh: benchmark the test runner code
2. ./bin/benchmark-in-docker.sh: benchmark the Docker image

These scripts can give a rough estimation of the test runner's performance. Bear in mind though that the performance on Exercism's production servers is often lower.