pyfeed

pyfeed is a pure-Python RSS/Atom feed parser and backend service. It fetches, merges, filters, and normalizes feeds into clean JSON for use in frontend applications—ideal for web UIs, dashboards, or minimalist aggregators.

---

Features

• RSS 2.0 and Atom 1.0 parsing
• JSON output with clean field normalization
• FastAPI backend with /feed, /merge, and /meta endpoints
• Multiple URL support with feed merging
• Duplicate item removal
• Filtering by keyword, category, and author
• Sorting (ascending or descending by publish date)
• Output limit and offset (for pagination)
• Preview mode: truncate long descriptions
• Per-item source_url tracking
• last_updated field in responses
• In-memory caching for performance
• CLI interface for quick local parsing

---

Installation

git clone https://github.com/jknndy/pyfeed
cd pyfeed
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

API Usage

GET /feed

Fetch, merge, and filter one or more feeds.

Query parameters:

url: one or more feed URLs

keyword: filter by keyword in title or description

category: filter by category

author: filter by author

limit: number of items to return

offset: skip N items

sort: asc or desc (default: desc)

format: json (default), text, or raw

preview: truncate description to N characters

Example:

curl "http://localhost:8000/feed?url=https://xkcd.com/rss.xml&limit=5&format=json"

POST /merge

Merge a list of feeds and return the result.

{
  "urls": ["https://xkcd.com/rss.xml", "https://hnrss.org/frontpage"]
}

GET /meta

Fetch basic metadata from a feed.

Example:

curl "http://localhost:8000/meta?url=https://xkcd.com/rss.xml"
Returns:
{
  "status": "ok",
  "title": "xkcd.com",
  "link": "https://xkcd.com/"
}

CLI Usage

python -m pyfeed.cli https://xkcd.com/rss.xml --json --limit 3

Testing

pytest tests