RSS2pdf

A minimal RSS/Atom to PDF script with zero third-party dependencies.

Features

• Reads feed sources from feeds.opml
• Fetches unseen entries from RSS/Atom feeds
• Creates a daily output folder: output/YYYY-MM-DD/
• Generates:
  • rss_digest.pdf
  • run.json (run summary + fetch errors)
• Prevents duplicate output across runs using .rss2pdf/state.json

Run Locally

python rss_to_pdf.py

Common example:

python rss_to_pdf.py --opml feeds.opml --max-per-feed 5 --timeout 10 --workers 8

CLI Options

• --opml: OPML file path (default: feeds.opml)
• --output-dir: Output root directory (default: output)
• --state-file: Dedup state file (default: .rss2pdf/state.json)
• --max-per-feed: Max unseen entries per feed
• --timeout: Per-feed request timeout (seconds)
• --workers: Number of concurrent feed fetch workers
• --summary-chars: Max summary text length per entry

Deduplication

Each exported entry is fingerprinted and stored in .rss2pdf/state.json. Future runs skip entries already recorded there.

GitHub Actions (Scheduled Runs)

Workflow file: .github/workflows/rss-fetch.yml

Current behavior:

• Manual trigger: workflow_dispatch
• Scheduled trigger: cron: 0 1 * * * (01:00 UTC daily)
• Auto-commits output/ and .rss2pdf/state.json back to the repository

To change schedule, edit the cron expression in .github/workflows/rss-fetch.yml.