class Wco::ScrapeWsj

def initialize site: nil, max_count: 2
  site ||= Wco::Site.find_by slug: 'wsj'
  # out = HTTParty.get( site.origin ).response.body
  # out = `curl #{site.origin}`
  out = Nokogiri::HTML(out)
  out.css('h3').each_with_index do |headline, idx|
    if idx < max_count
      puts! headline.text, 'headline'
      h = Wco::Headline.create({
        date: Time.now.to_date,
        site: site,
        name: headline.text,
      })
      if h.persisted?
        print '.'
      else
        puts! h.errors.full_messages
      end
    end
  end
end