Causalist

Causalist is a tool designed for the organization and monitoring of legal cases in the German legal system. The application operates on AWS and comprises a Kotlin-based lambda backend along with a React-based frontend.

Please be aware that the deployed service is presently not accessible to the public.

Backend

The backend directory contains a Gradle project responsible for the backend service, which exposes a REST API for the frontend. This service, implemented in Kotlin, utilizes http4k for the API and http4k-connect for DynamoDB and S3 access.

To run the service locally, local instances of DynamoDB and S3 are required. There is a docker compose file that starts a local DynamoDB and a MinIO server (for S3). The scripts init-local-dynamodb.sh and init-local-s3.sh facilitate the initial creation of the required local DynamoDB tables and the local S3 bucket.

Execute the command gradle runShadow to launch the application at http://localhost:4000.

Visit http://localhost:4000/api/ to access a Swagger UI providing documentation for the service's API.

Frontend

Within the frontend directory, you'll find the source code for a single-page application that interfaces with the backend service. This application is developed in JavaScript, employing Vite, React, and Tailwind CSS.

To initiate a server delivering the frontend, execute the command pnpm run start. The frontend will be accessible at http://localhost:3000, while it will interact with the backend service at http://localhost:4000.

Infrastructure

In the infrastructure directory, you'll find the Terraform source code for deploying the service on AWS. Before the first deployment you need to:

• Create a Terraform workspace named prod (and optionally stage).
• Supply the necessary variables, as outlined in variables.tf, by creating a file named prod.tfvars (and optionally stage.tfvars).

To execute an AWS deployment, follow these steps:

1. Run gradle assemble in the backend directory.
2. Run pnpm run build in the frontend directory.
3. Execute scripts/deploy.sh prod or scripts/deploy.sh stage.

Note: Every AWS Lambda deployment creates a new function version, however terraform currently doesn't remove previous function versions. To get rid of unused functions versions, execute the script cleanup-lambda.sh.