Image Processing Microservice

Overview

A lightweight service for uploading an image from a client, validating and processing it, and uploading to a URL. Image processing is handled by a Redis task queue. Some supported operations include resizing, converting formats, generating thumbnails, and size optimization.

Features

• Accepts PNG, JPEG, GIF, WEBP, SVG, and HEIC image formats
• Authentication via bearer token
• Supports resizing, format conversion, size optimization, thumbnail generation
• Horizontally scalable
• Core processing library can be integrated directly in other apps
• Docker containers provided for easy deployment

Development

Install

PNPM workspace is used for project organization.

pnpm install

Run

A Redis instance must be available on IMAGE_API_REDIS_PORT, which defaults to 6379. An easy way to do that is:

docker run --name redis -d -p 6379:6379 redis:6.0

Run the server:

npm run dev:server

With the app running, API docs are available at http://localhost:4100/docs

Build

npm run build:server
npm run build:web

Test

Unit test

Run all unit tests

npm run test

Integration test

Redis must be running first

# Start server in another terminal
npm run dev:server

# Run E2E tests
npm run integration

Docker

# Build
docker build --progress=plain -t api:dev -f Dockerfile --target=dev .
# Run
docker run -p 4100:4100 api:dev

See build script for building specific images, which can be useful for testing.

Docker compose

# start app
docker compose up -d
# stop app
docker compose down
# re-build app
docker compose build

Rebuild one container without cache and restart

docker compose up -d --force-recreate --build image-api-server

Commit conventions

The commit message format is: <scope>: <short-summary> #<issue-number>

• scope: release | feat | fix | refactor | test | ci
• type (optional): server | web | example | docs
• short-summary: Short summary about this commit
• issue-number: The related issue number