Created
January 23, 2011 14:00
-
-
Save pierot/792096 to your computer and use it in GitHub Desktop.
Makro Resto + Fuel scraping
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 'sinatra' | |
require 'builder' | |
require 'nokogiri' | |
require 'uri' | |
require 'open-uri' | |
$: << File.join(File.dirname(__FILE__), '..', 'app') | |
configure do | |
set :root, File.join(File.dirname(__FILE__), '..') | |
set :locations, { "deurne" => 0, | |
"machelen" => 1, | |
"alleur" => 2, | |
"eke" => 3, | |
"sint-pieters-leeuw" => 4} | |
set :target_fuel, 'http://www.makro.be/Content/assortiment/benzinestation/benzineprijzen/1/index.jsp?stat=' | |
set :targets_resto, { 'nl-BE' => 'http://www.makro.be/Content/assortiment/restaurant/1/index.html', | |
'fr-BE' => 'http://www.makro.be/Content/assortiment/restaurant/2/index.html'} | |
end | |
error do | |
'Sorry there was a nasty error - ' + request.env['sinatra.error'].name | |
end | |
not_found do | |
'Sorry - Not Found' | |
end | |
get '/' do | |
'<script src="https://gist.github.com/792096.js?file=makro_nokogiri.rb"></script>' | |
end | |
get '/resto/:lang' do | |
redirect '/' unless ['fr-BE', 'nl-BE'].include? params[:lang] | |
uri = URI.parse(settings.targets_resto[params[:lang]]) | |
html = open(uri) | |
doc = Nokogiri::HTML(html.read.gsub('<!-- *** div header end *** //-->', '')) | |
img = doc.css('.boxContent p img').first | |
builder do |xml| | |
xml.instruct! :xml, :version => '1.0' | |
xml.makro do | |
xml.error(img.nil? ? '1' : '0') | |
xml.img "#{uri.scheme}://#{uri.host}/#{img['src']}" unless img.nil? | |
end | |
end | |
end | |
get '/location/:loc' do | |
redirect '/' if settings.locations[params[:loc]].nil? | |
doc = Nokogiri::HTML(open("#{settings.target_fuel}#{settings.locations[params[:loc]]}")) | |
fuels_prices = [] | |
doc.css('div.boxContent table[width="337"] tr:first-child').each do |tr| | |
fuels_prices << { :officieel => tr.next.css('td')[0].content, | |
:korting => tr.next.css('td')[1].content, | |
:makro => tr.next.css('td')[2].content} | |
end | |
p fuels_prices | |
builder do |xml| | |
xml.instruct! :xml, :version => '1.0' | |
xml.makro do | |
xml.error(fuels_prices.empty? ? '1' : '0') | |
unless fuels_prices.empty? | |
xml.fuels do | |
['diesel', 'eurosuper', 'superplus'].each_with_index do |fuel, i| | |
# xml[fuel] do | |
xml.tag! fuel do | |
xml.officieel fuels_prices[i][:officieel] | |
xml.korting fuels_prices[i][:korting] | |
xml.makro fuels_prices[i][:makro] | |
end | |
end | |
end | |
end | |
end | |
end | |
end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment