mrkoll-scraper

Tool to scrape and generate an RSS feed from the incredibly creepy mrkoll.se Swedish toplist. Each RSS entry represents the list as it looked at a given point in time, plus convenience links to search for the person's name on Duckduckgo, Flashback, and The Facebook.

Installation

pip install mrkoll-scraper

Usage

CLI

mrks --scrape

Scrapes the current list and saves it to the Shelve database mrks.db in the current working directory. This operation saves the raw scraped HTML as well as a generated feedgen FeedEntry. The data is saved in a dictionary with the current date as key, so multiple scrapes during the same day doesn't save a new entry, they only update the existing one.

mrks --regenerate

Iterates through the saved lists and re-generates FeedEntry's from the raw HTML, in case you've made some changes in the HTML template or so, and want them applied retroactively.

WSGI

mrks.wsgi contains a beautifully simple WSGI application, that simply outputs an RSS feed based on the data currently saved.

Cron

Crontab to run --scrape every Monday at midnight:

0 0 * * 1 cd /home/robert/mrkoll-scraper && /home/robert/mrkoll-scraper/venv/bin/mrks -s