ruby访问网页（处理中文url）

  1 require 'open-uri'
  2 require 'net/http'
  3 require 'iconv'
  4 require 'uri'
  5
  6 src1_file = "place_file.txt"
  7 src2_file = "things_file.txt"
  8 save_file = open("key_word_rank_baidu.txt","w")
  9 max_num = '100'
10
11 File.open(src1_file).readlines.each do |place|
12   File.open(src2_file).readlines.each do |thing|
13     enc_pl = URI.escape(Iconv.conv('gbk','utf-8',place))
14     enc_th = URI.escape(Iconv.conv('gbk','utf-8',thing))
15     query = enc_pl.to_s.chomp+"+"+enc_th.to_s.chomp
16     url='http://www.baidu.com/s?wd='+query+'&rn='+max_num
17     target=open(url).read.scan(/<font\scolor=#008000>.*?<\/font>/)
18     puts target
19     rank = 0
20     if target != nil
21       for val in target
22         if (val =~ /<font\scolor=#008000>.*?\s[0-9]+K\s.*?<\/font>/) != nil
23           rank += 1
24           if (val =~ /xxxxx/) != nil
25             save_file.puts val
26             save_file.puts query
27             save_file.puts rank.to_s
28             break
29           end
30         end
31       end
32     end
33   end
34 end
35
36 email_to = "[e-mail]"
37 file = "key_word_rank.txt"
38
39 sendmail_cmd = "cat #{file} | mailx -s daodaorank #{email_to}"
40 pipe = IO.popen(sendmail_cmd)