Get aozora bunko

aozoraget.rb

#aozoraget.rb

require 'rubygems'
require 'hpricot'
require 'open-uri'

for i in 1..13
	toc = "http://www.aozora.gr.jp/index_pages/sakuhin_a#{i}.html"
	puts "Opening: " + toc
	html = IO.read(toc)
	html.scan(/<a href="\.\.\/(cards\/.+?)"/){|str|
		card =  "http://www.aozora.gr.jp/" + str[0]
		IO.read(card).scan(/<a href="\.(\/files\/.+?\.html)"/){|s|
			card =~ /(.+)\/.+?\.html/
			content_url = $1 + s[0]
			content_url =~ /.+\/(.+?\.html)/
			outfile = $1
			open("./database/"+outfile,'w'){|out|
				puts outfile
				out.write IO.read(content_url)
			}
			break
		}
	}
end