extract_urls.rb

#!/usr/bin/env ruby

require 'nokogiri'

unless ARGV.size > 0
  STDERR.puts "usage: #{$0} html-file"
  exit 1
end

doc = File.open(ARGV[0]) { |f| Nokogiri::HTML(f) }

puts doc.css('a').map { |a| a.attr('href') }

job_diff.rb

#!/usr/bin/env ruby

if ARGV.size < 2
  STDERR.puts "usage: #{$0} previous-jobs.txt current-jobs.txt"
  exit 1
end

def readfile(path)
  File.read(path).split("\n").reject { |line| line.strip == "" }
end

previous = ARGV[0]
current  = ARGV[1]

prev = readfile(previous)
curr = readfile(current)

added = curr - prev
removed = prev - curr

puts "# Added"
puts added
puts
puts "# Removed"
puts removed
puts
puts "# Run this line to remove"
puts "JobPosting.where(url: [#{removed.map(&:inspect).join(", ")}]).update_all(inactive: true)"