TTM-frontend

This is an interactive map UI for displaying travel time matrices.

See the backend for serving the data.

Technology

Built on Vite, React and Typescript
Mapping is done with react-map-gl and maplibre
Axios for http requests

Development

The frontend will be available at http://localhost:5173.

Containerized

The development container is dev.Dockerfile.

To start the frontend run:

docker compose up

This starts the development container with the necessary ports and volumes setup to access the frontend with your browser and provide hot reload on any changes to the code. The default assumption is that you are serving travel time matrix geojson data on http://localhost:8080/. See dev.Dockerfile for changing the url to fetch data from.

Non-containerized

Install everything with:

npm install

Set the env var VITE_BACKEND_URL to reflect where you are serving the data from.

Start in development mode:

npm run dev -- --host

Building and deploying for production

A separate container image is used for building and serving the production build of the app (Dockerfile).

The production image is built and published automatically with github actions. It is hosted on github container registry and used by the deployment.

Deploying to rahti

The kubernetes configuration is in manifest.yml. It defines:

A Deployment that creates and runs a ReplicaSet of server pods (each serving the production build of the application)
A Service for exposing the deployment network
A Route for routing traffic from the internet to the Service

The command line tool oc is required for interacting with rahti.

After authenticating you can create a deployment with:

oc create -f manifest.yml

And delete it with:

oc delete all --selector app=ttm