getting MM9's text from webmysteries.jp.

OBSOLETE

#
# This command was obsolete. see also https://github.com/tdtds/aozoragen
#

#!/usr/bin/env ruby
#
# getting text from webmysteries.jp's novel.
#
require 'open-uri'
require 'nokogiri'
require 'cgi'

class WebMysteries
	def get( *indexes )
		indexes.flatten.each do |index|
			get_pages( index ) do |text|
				yield text
			end
		end
	end

:private
	def get_pages( index )
		begin
			pages = []
			html = Nokogiri( open( index, 'r', &:read ) )
			(html / 'ul.pageNavi a').each do |a|
				pages << a.attr( 'href' )
			end
			pages.shift	# delete current page
			begin
				(html / 'noscript param[name="FlashVars"]')[0].attr( 'value' ).scan( /entry=(\d+)/ ) do |i|
					yield get_text( i[0] )
				end
			end while html = Nokogiri( open( pages.shift, 'r', &:read ) )
		rescue TypeError
			# ignore open nil
		end
	end

	def get_text( xml_id )
		result = ''
		open( "http://www.webmysteries.jp/entry_xml_data/#{xml_id}.xml" ) do |fx|
			CGI::unescape( fx.read ).scan( %r|<entryBody>(.*?)</entryBody>|m ) do |entry|
				result << entry[0].gsub( %r|<.*?>|m, "" )
			end
		end
		result
	end
end

if __FILE__ == $0
	WebMysteries::new.get( ARGV ) do |text|
		puts text
	end
end