Download PDF files from a WebPage

get_pdf.rake

require 'nokogiri'
require 'open-uri'
require 'cocaine'

task :default do
    pattern = ':url'
    line = Cocaine::CommandLine.new('wget', pattern)

    base_url = 'http://www.cs.sjtu.edu.cn/~jdyu/teaching/SE/fall2012/'
    file = 'index.html'

    document = Nokogiri::HTML(open(base_url + file))
    document.css('a').each do |link|
        file = link['href']
        next unless file.end_with?('.pdf')
        line.run(:url => base_url + file)
    end
end

USAGE

gem install nokogiri
gem install cocaine
rake -f get_pdf`

多了个点- -