David Alvarez e9c4d963ad
Build & Deploy / build-and-deploy (push) Failing after 49s
feat: add G1GC heap uncommit flags so container memory shrinks after uploads
-XX:MinHeapFreeRatio=10 -XX:MaxHeapFreeRatio=20 -XX:G1HeapUncommitPercent=5
After large game uploads, the JVM committed heap stays high. These flags tell
G1GC to release free heap pages back to the OS after GC cycles, so Docker
memory usage drops back toward baseline instead of staying at peak.
2026-06-04 11:25:02 +02:00
2026-05-28 23:14:35 +02:00

Dostalgia

A nostalgic DOS game hub. Upload your old DOS games, scrape artwork, and play them directly in the browser via js-dos (WebAssembly DOSBox).

Quick Start

# Build and run with Docker
docker build -t dostalgia .
docker run -p 8765:8765 -v $(pwd)/data:/data dostalgia

Open http://localhost:8765

Development (without Docker)

Terminal 1 — Frontend dev server:

cd frontend && npm install && npm run dev

Terminal 2 — Quarkus dev server:

mvn quarkus:dev

Open http://localhost:5173 (Vite proxies API calls to Quarkus on port 8765)

Production build (local)

cd frontend && npm install && npm run build
mvn package -DskipTests
java -jar target/quarkus-app/quarkus-run.jar

Architecture

  • Backend: Quarkus (Java 21, JAX-RS) — ~5 REST resource classes, zero database
  • Frontend: Svelte 5 SPA with hash-based routing
  • Emulation: js-dos v8 loaded from CDN, runs DOSBox in WebAssembly
  • Storage: JSON-per-game under /data/games/{id}/game.json

API

Method Path Description
GET /api/games List all games
GET /api/games/:id Get game details
PATCH /api/games/:id Update game metadata
DELETE /api/games/:id Delete a game
POST /api/upload Upload a game ZIP (multipart)
POST /api/games/:id/cover Upload cover art
GET /api/igdb/search?q= Search IGDB (placeholder)

Game JSON Schema

{
  "id": "doom",
  "title": "Doom",
  "year": 1993,
  "genre": "FPS",
  "developer": "id Software",
  "bundle_file": "doom.jsdos",
  "has_cover": true,
  "ready": true
}
S
Description
A nostalgic DOS game hub — upload, scrape artwork, and play your classic games in the browser
Readme MIT 5.6 MiB
Languages
Java 62.2%
Svelte 24.7%
JavaScript 8.8%
CSS 3.4%
Dockerfile 0.5%
Other 0.4%