require 'open-uri'
ARRAY_LIMIT = 10000
urls = ['http://www.baidu.com']
while true
  url = urls.first
  urls.shift 1 
  begin
    open(url) do |page|
      page_content = page.read();
      next if urls.length > ARRAY_LIMIT
      links = page_content.scan(/<a href=\"(.*?)\"/).flatten
      links.each do |l|
        if l[0,4] == 'http'
          urls << l 
        elsif l[0,1] =='.'
          urls << url+l[1,l.length]
        elsif l[0,1] == '/'
          urls << url[/http:\/\/[^\/]+/]+l ;
        end
      end
      urls.uniq!
    end
  rescue
    print "超时\n";
    next
  end
  print(urls.length);
  print "\n";
end

目的也很简单，就是想给一个网址，按网址上的链接依次往下爬，尽可能的多爬取。 所以就写了上面的简单的代码。 想在里面加入多进程或多线程，要怎么做？