Created
August 23, 2013 06:47
-
-
Save RaymondChou/6316235 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
require 'net/http' | |
require 'nokogiri' | |
http = Net::HTTP.new('trade.jd.com', 80) | |
# 登录后的cookie | |
cookie = 'aview=843.783520|843.791660|843.771412|877.674903|688.813736|877.674899|688.823463|688.768334; atw=688.813736.14|843.783520.11|877.674903.4; pin=4077039-235606; npin=4077039-235606; logintype=qq; unick=freezestart; ceshi3.com=A64E8CE1156ECFCB00EC559A187D88322AB429A3EFAD42082ED22DD481E4F1AA3BAED95977D278EF2735A6F543D424395E486E7E719225E15665BF29C4C51D760E33948B1C2D3C6B3A8B98081AD52E451E58C4E95F7C9ECDCEB8F819FB683131C337CEE1F37760EDF0D9A574157844D653335B21A7EA51E2C7281CADDBD302C2E9AC84770278B1940620C238809C8FAB92D2E45AD0C562F20C35F58E55742274; eb=0; reWids4=1015970968; vpflag=\"\"; cn=1; _jzqco=%7C%7C%7C%7C1371459554633%7C1.417866817.1367743079141.1371459876094.1371460190862.1371459876094.1371460190862.0.0.0.25.25; ipLoc-djd=12-904-3379; ipLocation=%u6c5f%u82cf; track=30275297-8610-4032-8bef-c1fc7b1c6d71; __jda=122270672.133917276.1364047666.1368619380.1371459558.7; __jdb=122270672.5.133917276|7.1371459558; __jdc=122270672; __jdv=122270672|www.jd.com|-|referral|-' | |
headers = { | |
'Cookie' => cookie, | |
'Referer' => 'http://trade.jd.com', | |
'Content-Type' => 'application/x-www-form-urlencoded' | |
} | |
# 省 | |
path = '/dynamic/consignee/getProvinces.action' | |
data = '' | |
resp, _ = http.post(path, data, headers) | |
resp = resp.body.force_encoding('GBK').encode('UTF-8') | |
province_array ||= [] | |
doc = Nokogiri::HTML(resp) | |
fh = File.new("province.txt", "w") | |
fh.puts 'province_id|province_name' | |
doc.css('option').each_with_index do |option,i| | |
next if i == 0 | |
province = option.content.gsub('*','') | |
province_id = option['value'] | |
province_array << province_id | |
fh.puts "#{province_id}|#{province}" | |
end | |
fh.close | |
# 市 | |
path = '/dynamic/consignee/getCitys.action' | |
city_array ||= [] | |
fh = File.new("city.txt", "w") | |
fh.puts 'city_id|province_id|city_name' | |
province_array.each do |province_id| | |
data = "consigneeParam.provinceId=#{province_id}" | |
resp, _ = http.post(path, data, headers) | |
resp = resp.body.force_encoding('GBK').encode('UTF-8') | |
doc = Nokogiri::HTML(resp) | |
doc.css('option').each_with_index do |option,i| | |
next if i == 0 | |
city = option.content.gsub('*','') | |
city_id = option['value'] | |
city_array << city_id | |
fh.puts "#{city_id}|#{province_id}|#{city}" | |
end | |
end | |
fh.close | |
# 区 | |
path = '/dynamic/consignee/getCountys.action' | |
county_array ||= [] | |
fh = File.new("county.txt", "w") | |
fh.puts 'county_id|city_id|county_name' | |
city_array.each do |city_id| | |
data = "consigneeParam.cityId=#{city_id}" | |
resp, _ = http.post(path, data, headers) | |
resp = resp.body.force_encoding('GBK').encode('UTF-8') | |
doc = Nokogiri::HTML(resp) | |
doc.css('option').each_with_index do |option,i| | |
next if i == 0 | |
county = option.content.gsub('*','') | |
county_id = option['value'] | |
county_array << county_id | |
fh.puts "#{county_id}|#{city_id}|#{county}" | |
end | |
end | |
fh.close | |
# 街道 | |
path = '/dynamic/consignee/getTowns.action' | |
fh = File.new("town.txt", "w") | |
fh.puts 'town_id|county_id|town_name' | |
county_array.each do |county_id| | |
data = "consigneeParam.countyId=#{county_id}" | |
resp, _ = http.post(path, data, headers) | |
resp = resp.body.force_encoding('GBK').encode('UTF-8') | |
doc = Nokogiri::HTML(resp) | |
doc.css('option').each_with_index do |option,i| | |
next if i == 0 | |
town = option.content.gsub('*','') | |
town_id = option['value'] | |
fh.puts "#{town_id}|#{county_id}|#{town}" | |
end | |
end | |
fh.close |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment