Skip to content

README.md

Latest commit

 

History

History
98 lines (76 loc) · 3.45 KB

README.md

File metadata and controls

98 lines (76 loc) · 3.45 KB

KillrVideo v2 - Java Spring Boot Backend

Build KillrVideo Java backend

Date: June 2025

A reference backend for the KillrVideo sample application rebuilt for 2025 using Java, Spring Boot and DataStax Astra DB.

Overview

This repo demonstrates modern API best-practices with:

  • Spring Boot for typed request/response models
  • Role-based JWT auth
  • Restful Cassandra (Astra DB) Data API client via astra-db-java
  • Micro-service friendly layout – or run everything as a monolith

Prerequisites

  1. Java 21+ (use jenv or sdkman)
  2. Maven for dependency management:
<dependency>
    <groupId>com.datastax.astra</groupId>
    <artifactId>astra-db-java</artifactId>
    <version>2.0.0</version>
</dependency>
  1. A DataStax Astra DB serverless database – grab a free account.
  2. Docker or Podman (optional) if you want to run the individual services in containers.

Setup & Configuration

### clone
git clone git@github.com:KillrVideo/killrvideo-java-2025.git
cd killrvideo-java-2025

### build and install deps
mvn clean package

Database collections:

  1. Create a new keyspace named killrvideo_dataapi.
  2. Create the following non-vector-enabled collections:
  • comments
  • content_moderation
  • users
  • ratings
  • video_ratings
  1. Create the following vector-enabled collection:
  • videos (with a 384-dimensional vector)

Load data:

  1. AstraDB - KillrVideo collections loader
  2. OpenSearch - CSV-to-OpenSearch project

Environment variables (via export):

Variable Description
ASTRA_DB_API_ENDPOINT REST endpoint for your Astra DB instance
ASTRA_DB_APPLICATION_TOKEN Token created in Astra UI
ASTRA_DB_NAMESPACE killrvideo_dataapi
YOUTUBE_API_KEY A YouTube API key, used for grabbing new video metadata.
`HF_API_KEY A HuggingFace API key, used for generating embeddings without getting rate-limited...probably.
OPENSEARCH_HOST IP or hostname for your OpenSearch instance. Assumes https.
OPENSEARCH_PORT Port for your OpenSearch instance. Defaults to 9200 if null.
OPENSEARCH_USERNAME Username for your OpenSearch instance.
OPENSEARCH_PASSWORD Password for your OpenSearch instance.

Edit application.yml:

  • Generate and change the killrvideo.jwt.secret key (or use the default).
  • Generate your own keystore for your self-signed TLS certificate.

Running the Application

java -jar target/killrvideo-java-service-1.0.0-SNAPSHOT.jar --logging.level.com.killrvideo=DEBUG --logging.level.org.springframework.security=DEBUG

Building & Running with Docker or Podman

docker build -t killrvideo-2025-java .
docker run -p 8443:8443 -e ASTRA_DB_NAMESPACE=killrvideo_dataapi -e ASTRA_DB_APPLICATION_TOKEN=your_token -e ASTRA_DB_API_ENDPOINT=your_endpoint killrvideo-2025-java

Or

podman build -t killrvideo-2025-java .
podman run -p 8443:8443 -e ASTRA_DB_NAMESPACE=killrvideo_dataapi -e ASTRA_DB_APPLICATION_TOKEN=your_token -e ASTRA_DB_API_ENDPOINT=your_endpoint killrvideo-2025-java