Created
December 21, 2011 00:48
-
-
Save shawn0lds442/1504001 to your computer and use it in GitHub Desktop.
Web Scraper for Data Mining
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 'mechanize' | |
require 'active_record' | |
require 'nokogiri' | |
require 'sqlite3' | |
require 'csv' | |
ActiveRecord::Base.establish_connection(:adapter => "sqlite3", :database => "db") | |
ActiveRecord::Schema.define do | |
create_table "locations", :force => true do |t| | |
t.string "name" | |
t.string "address" | |
t.string "city" | |
t.string "zip" | |
t.string "phone" | |
t.string "website" | |
t.datetime "created_at" | |
t.datetime "updated_at" | |
end | |
create_table "zip_codes", :force => true do |t| | |
t.string "zip" | |
t.decimal "lat", :precision => 15, :scale => 10 | |
t.decimal "lng", :precision => 15, :scale => 10 | |
t.string "city" | |
t.string "state" | |
t.string "the_type" | |
t.datetime "created_at" | |
t.datetime "updated_at" | |
end | |
end | |
class Location < ActiveRecord::Base | |
end | |
class ZipCode < ActiveRecord::Base | |
end | |
CSV.foreach("zip_codes.csv") do |row| | |
ZipCode.create( | |
:zip => row[0], | |
:lat => row[1], | |
:lng => row[2], | |
:city => row[3], | |
:state => row[4], | |
:the_type => row[5] | |
) | |
puts "stored #{row[0]}" | |
end | |
zip_codes = ZipCode.find(:all, :select => "distinct(city)") | |
zip_codes.each do |zip_code| | |
agent = Mechanize.new | |
page = agent.post('http://www.geosprawl.com/Default.aspx?alias=www.geosprawl.com/norvell', | |
[ | |
["Searching", 1], | |
["city", "#{zip_code.city}"], | |
["cname", "BizSearchResult"], | |
["country", ""], | |
["fullcrit", ""] , | |
["m", "40"], | |
["state", ""], | |
["tag", "All"], | |
["zip", ""] | |
]) | |
doc = Nokogiri::HTML.parse(page.body) | |
data = [] | |
result_div = doc.at_css(".repeaterscroll") | |
result_div.css("tr").each do |row| | |
id = row.attribute("id").to_s | |
if (id) | |
if id =~ /BusinessName/ | |
@hash = Hash.new | |
@hash[:name] = (row.css("td").first.text).lstrip.rstrip | |
end | |
if id =~ /FullAddress/ | |
@hash[:address] = (row.css("td").first.text).lstrip.rstrip | |
end | |
if id =~ /Phone/ | |
@hash[:phone] = (row.css("td").first.text).lstrip.rstrip | |
end | |
if id =~ /Website/ | |
@hash[:website] = (row.css("td").first.text).lstrip.rstrip | |
data << @hash | |
puts "scraped #{zip_code.zip}" | |
end | |
end | |
end | |
num = 0 | |
data.each do |data_hash| | |
data_hash[:city] = data_hash[:address].split(' ')[-3] | |
data_hash[:zip] = (/[0-9]{5}/.match(data_hash[:address])).to_s | |
Location.create(data_hash) | |
puts "stored #{num}" | |
num = num + 1 | |
end | |
end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment