Expedition is a lightweight roleplaying game that's fun for players - and storytellers. Anyone can learn to play in less than 5 minutes!
It's powered by a free companion app with hundreds of free adventures, with new quests every week.
For 1-6 players, 20+ minutes, ages 8+
All shared code goes in /shared
, all deployed code goes in /services.
Code in services can only reference shared code, and should not reference other services.
Contributions welcome! Earn loot points while practicing your coding skills on the bleeding edge of web technologies. If you're new to any of our tools or libraries, don't worry - we're happy to help and answer questions!
Not sure what to work on? Check out our open issues, especially those labeled with help wanted
.
Expedition requires a unix-based system like OSX or Linux. If you are on Windows, you can use the Windows Subsystem for Linux.
nvm install 8.11.3 --latest-npm
npm install -g yarn
yarn
yarn run app
, yarn run cards
or yarn run quests
-> open it in your browser at http://localhost:8080
yarn test
To build the Cordova app, you may need to install other global dependencies: npm install -g cordova webpack@4 webpack-cli@3 webpack-dev-server@3
Chrome is used for headless browser testing. How to install Chrome on WSL/Ubuntu cli
You can run anything in /services
by running the yarn command of the same name, for example the app by running yarn run app
.
For the services that depend on the API server, you can also run them against a local copy of the API server by running it as X-local
, for example yarn run app-local
.
See the readme at services/api/README.md
for instructions on configuring the API server.
yarn test
runs all tests in the repo; you can also run tests for a specific service as X-test
, for example yarn run app-test
.
Linting is run automatically to verify each commit; tests are run automatically to verify each push.
Linting errors? yarn run lint --fix
fixes most common linting issues automatically. For alphabetization issues, in Sublime Text you can select multiple lines of text then hit F5
to auto-sort them.
Reducer tests should test state changes via actions, specifically when there's logic involved / they aren't just glue
Use jasmine.objectContaining for more robust tests
Components: test branches for key content, i.e. if search results empty, make sure that resulting string contains "No results found"
Components: make sure that key interactivity works, i.e. that clicking the button calls the expected function (see app/AdvancedPlay tests / enzyme / using spies)
Containers: test when doing logic in mapping state or dispatch (i.e. not just glue) & test the logic (i.e. the combat container) See app/AdvancedPlayContainer, create a mock store, dispatch function, check that actions are dispatched with correct parameters.... this one goes too far and actually also tests the action
Actions: test actions with logic / not just glue (ie Quest load node) see app/actions/card - mock store, spy on outputs, expect the action with parameters to return an object containing. If plain function, call directly. If dispatch-wrapped action, call store.dispatch(action)
We use a Github bot called Renovate to keep our dependencies from getting stale. It automatically opens new PR's each time a dependency has a major version bump. In general, if the CI passes, it's probably safe to merge... but if you ever need to pull a branch down for local testing or changes, you can run git checkout origin/renovate/<BRANCH NAME>
, and then git push origin HEAD:renovate/<BRANCH NAME>
to push your changes back to the branch.
Services can be deployed via yarn run deploy
, which can also take a specific service and target as arguments (ie yarn run deploy app beta
). This requires AWS and Heroku permissions.
Reach us at contact@expeditiongame.com