STAC Collection Discovery Deployment for discover.maap-project.org

About

This is a deployment repo for MAAP's instance of the STAC Collection Discovery application.

The resources get deployed to the SMCE-MAAP AWS account by the deploy Github Actions workflow where you must specify the deployment environment (dev or test) and the API version to use.

The Github Action workflow will deploy:

• the stac-fastapi-collection-discovery API
  • Lambda function with an API gateway endpoint
  • alias for a url at maap-project.org
• the stac-collection-discovery client application
  • React app built by yarn and uploaded to S3, served by CloudFront
  • alias for a url at maap-project.org

Gotchas

• The corresponding alias records for maap-project.org are managed in the UAH-MAAP AWS account and must be manually entered/updated to redirect requests to the resources in this stack.

• We have to checkout the client app repo and build the React app then upload the contents to the S3 bucket after it is deployed. The deploy workflow pins Node to the exact version in client_app/.nvmrc and uses Corepack so the client build uses the toolchain the app expects.

Deployment environments

Each deployment requires several attributes to be specified, these are handled in the Github Environments and are passed as environment variables in the deploy step of the workflow.

The dev and test environments share some of these in common but the custom domain names differ between the two environments:

• *.dit.maap-project.org for the test environment
• *.maap-project.org for the dev environment

Local development

The AWS infrastructure is defined in a Python CDK application. You can install the Python dependencies for the project with this command from the Makefile (requires Node and uv):

make install

This will create a virtual environment if you don't already have one, will use an existing one if it is activated.

Non-Github Actions deployments

To deploy the resources to an AWS account from your local environment, you will need to define some variables in your environment (or in a .env-cdk file). Reference config.py for all of the possible options.

At a minimum you will want to define STAC_API_URLS which will define the list of STAC APIs that the Federated Collection Discovery API will search.

You can put this in .env-cdk which will get automatically used by the CDK app:

# comma separated list of STAC API urls:
STAC_API_URLS=https://stac.maap-project.org/,https://openveda.cloud/api/stac/,https://catalogue.dataspace.copernicus.eu/stac

To deploy the AWS resources:

make deploy

This will deploy the API and create the infrastructure for the client application, but you will still need to build the client application and upload it to the S3 bucket to complete the deployment! Check out the deploy workflow for inspiration if you want to do this locally.

TODO: Add docker-based client app build process

To destroy the stack:

make destroy