Miniflux Block

A simple Python script to connect to a Miniflux instance via the API and set to "read" any feed entries that match words or phrases in a blocklist. It aims to be an effective way to cut 99% of the endless stream of utter rubbish some feeds contain, retaining the 1% I actually want to know about.

Miniflux's own solution to this problem, and other solutions I have found allow you to specify a blocklist on a per feed basis, which is annoying to maintain. This script allows you to set one blocklist to apply to all feeds instead.

Usage

Run the script at regular intervals via a cron job. I have it running every 5 minutes on my home server.

*/5 * * * * cd miniflux-block/ && /usr/bin/python3 miniflux-block/main.py

Configuration File

You must create a file named config.yml which contains the key/values below. A commented example file named config_template.yml is also provided in the repository.

miniflux_api_token: <api-token-value>
miniflux_base_url: https://<path-to-miniflux-instance>
miniflux_entries_limit: <max-number-of-entries> # Default is 100. Set to something high like 10000.
blocklist_regex:
  - \$\d+
  - \b(\% off)\b
  - \b(\d+ best)\b
  - \b(accessories)\b
  - \b(app store)\b
  - \b(best deals)\b

Blocklist

You can supply a list of regular expressions that you want the script to "mark as read" to the blocklist_regex list in config.yml.

Logging

When the script finds a match it will output to standard output (stdout). If you wish to keep a log, you should redirect stdout.

2022-01-29 16:01:02 - [INFO] - 7749 - 2022-01-28T16:16:00Z - Slashdot - Matched 'apple' - Samsung Led Smartphone Shipments For 2021, Beating Out Apple
2022-01-29 16:01:02 - [INFO] - 7781 - 2022-01-28T19:19:00Z - Slashdot - Matched 'apple' - Apple Finally Removing Python 2 In macOS 12.3