This repository contains the sandbox images used by the match runner. There is a compiler, as well as a runtime image for each supported programming language.
This guide assumes the user has python3 with access to the Riddles.io pypi server as well as the Google Cloud SDK with beta components and kubectl installed.
When running the scripts for the first time, ensure all dependencies are installed correctly by running the following command:
python3 -m venv .venv
source .venv/bin/activate
pip3 install -r requirements/requirements.txt
Next time you open your terminal, do not forget to activate your virtual environment by running source .venv/bin/activate after changing dir into docker-sandbox.
To build the containers, run following command:
bin/sandbox-scripts build
The script can also be used to publish the container to the Google Container Registry.
Run the following command to build and publish a specific container:
bin/sandbox-scripts publish <container_name>
In order to publish all containers, run the following command instead:
bin/sandbox-scripts publish all
The test suite can be invoked through the following command:
bin/sandbox-scripts test
This runs the test suite in its intirety. Markers can be added if you want to run only a subset of tests. For instance, running the tests for the Java compiler can be done as follows:
bin/sandbox-scripts test java and compiler
Similarly, running tests for both the OCaml and Reason compilers and runtimes can be done as follows:
bin/sandbox-scripts test ocaml or reason
Important: Running the test suite requires that the sandbox images have been built and are available on your local machine.
All that's needed to add tests for new implementations is
a simple mutation of the test/config.py file.
The config file contains a list of tuples consisting of the slug of the programming language and the command required to execute a binary/script.
Adding support for a language, say Java, is as simple as adding the following to the list:
('java', 'java -jar')
When testing for edge cases, specific language features or implementation
details, extra tests can be added to a sub folder in the test folder.
All files prefixed with test_ and ending in .py will be recognised
as valid tests.
Note that these tests should be marked according to the programming language being tested and whether the test targets the compiler or the runtime. Furthermore, all added files, folders and tests should have a unique name.
This repository uses the Github branching model with the following branch specification:
- master
- feature/FeatureName
The master branch is closed for writing, contributions should be submitted through a pull request from a feature branch.
Each pull request must describe the changes made and why these changes are necessary (for future reference) or reference an issue which does so.