multithread scrape

gistfile1.rb

#!/usr/bin/ruby
# coding: utf-8
require 'uri'
#require 'iconv'
require 'open-uri'
require 'thread'


def scrape(from,to,n)
    begin
        file = File.open("data#{n}.txt","w+")
        num = 1

        for num in from..to
            page = open(@url + num.to_s)
            text = page.read; nil
            text = text.to_s.scan(@regexp)

            if( @regexp_title =~ text.to_s )
                print "#{n}:#{num.to_s}:#{$1},#{$2},"
                file << "#{$1},#{$2},"
            end

            if( @regexp_jingdian =~ text.to_s )
                print "#{$1}\n"
                file << "#{$1}\n"
            end
        end

        file.close
    rescue Timeout::Error
        logger.debug "Time out"
    end
end

if $0 == __FILE__
    @url = 'http://jingdian.tuniu.com/fengjing/'
    @regexp = /\<h1\>.*span\>/
    @regexp_jingdian = /\<h1\>(.*)\<\/h1\>/
    @regexp_jingdian1 = /title=\".*\"\>(.*)\<\/a\>\//
    @regexp_jingdian2 = /href=.*"\>(.*)\<\/a\>"/
    @regexp_title = /href=.*"\>(.*)\<\/a\>\/<a.*"\>(.*)\<\/a\>/


    # from to 10376..38144
    threads = []
    1.upto(40) do |n|
        threads << Thread.new {
            # do task
            if n != 40
                tem_from = 10376 + (n-1) * 694
                tem_to = tem_from + 693
                scrape(tem_from, tem_to, n)
            else
                tem_from = 10376 + (n-1) * 694
                tem_to = 38144
                scrape(tem_from, tem_to, n)
            end
        }
    end

    threads.each { |t| t.join }

    puts "shit"

end