This script runs a trawl of the courts.go.jp site to build Atom feeds of court judgments.

Cases are automatically divided among Civil, Criminal, IP, Labor, and Administrative feeds.

Running the script as node ./scraper.js 2>/dev/null without arguments collects cases from 30 days in the past.

Running the script with a single numeric argument, say as node ./scraper.js 2 2>/dev/null, will backtrack by the given number of 30-day intervals, and collect 30 days of cases from that point forward.

Feeds are save locally, in the directory from which the script is executed.

Saved feeds are treated as a cache, and only the most recent 50 cases in each feed are saved.

Executing the script with an argument is only needed to set up the initial feed content. Afterward, executing the script without arguments once or twice a week will keep the feeds fresh.