Fusion

Lightweight, self-hosted friendly RSS aggregator and reader

19 followers

Lightweight, self-hosted friendly RSS aggregator and reader

19 followers

Visit website

A lightweight, self-hosted friendly RSS aggregator and reader.

Free

Launch tags:News•Newsletters•Open Source

Launch Team / Built With

CloudflareDeploy serverless functions globally in seconds

Promoted

Maker

📌

Hi, folks. As an RSS user, I tried Inoreader and Feedly, then ended up self-hosting a Miniflux instance on my homelab. A few months ago, I moved to another city and had to shut down my homelab for a long time, so I couldn't access my local miniflux. It was quite inconvenient. I decided to self-host my RSS aggregator on a tiny VPS or PaaS such as fly.io. However, Miniflux requires a PostgreSQL database, which may isn't suitable for a tiny VPS instance. So I built fusion with Golang and SQLite. It contains basic features such as Group, Bookmark, Search, Automatically feeds sniffing, Import/Export OPML file, etc. It uses about 80MB of Mem and negligible CPU usage (metrics here). Feel free to share your questions and suggestions. BTW, I also built an online tool to sniff RSS links from a URL. Here it is.

Report

2yr ago

This is cool, just pulled the code and gave a star. We are using shad-cn too!

Report

2yr ago

Maker

@gelks Glad to see you like it.

Report

2yr ago

SiteForge - AI Website Wireframer

congrats on the launch, rook1e. it’s interesting to see how your personal experience shaped fusion's development. considering its low resource requirements, does fusion support integrations with other popular services for enhanced functionality? how does it maintain performance with larger feed collections?

Report

2yr ago

Maker

Hi @mashy . The key to keeping resource usage low is using Go and SQLite. Additionally, we limit the number of concurrent "workers", so the usage will stay low even when refreshing a large number of feeds. You can check the metrics of my instance here, the small rise corresponds to 60+ feeds refreshing. We aim to keep fusion simple and clean, so there is no integration included currently. You can create an issue on our repo, we'll add it to plan if many users request it.

Report

2yr ago

Reviews