Last active
August 29, 2015 14:05
-
-
Save knagode/b0be5225e028d1d3c152 to your computer and use it in GitHub Desktop.
Rails Copybara scrap example
This file contains hidden or 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 'rails_helper' | |
require 'rspec/rails' | |
feature "scrapper" do | |
it "gets all companies which export to austria" do | |
# first lets visit setting where we will include financial data inside our search results | |
visit "http://www.sloexport.si/default.asp?MenuID=279&Settings=true" | |
sleep(0.3) | |
find(:xpath, ".//input[@value='8']").click | |
find(:xpath, ".//input[@name='btnResultSettings']").click | |
country = "Austria" | |
visit "http://www.sloexport.si/default.asp?MenuID=279&advancedSearch=true" | |
find_by_id('dialog-drzave-link').click | |
sleep(0.3) | |
find(:xpath, ".//input[@desc='" + country + "']").click | |
find_by_id('btnChoose').click | |
find(:css, "input.button_adv_search").click | |
# go trough every page | |
url = current_url | |
while true do | |
page.all(:css, '.company').each do |el| | |
company = Company.new | |
company.name = el.all(:css, ".CompName").first.text | |
company.address = el.find(:css, ".address").text | |
objects = el.all(:css, "b") | |
if objects.count > 0 | |
income = el.all(:css, "b").first.text | |
if income.length > 2 | |
company.income = income.gsub(',', '').split(".").first.to_i | |
else | |
company.income = -1; | |
end | |
else | |
company.income = -1; | |
end | |
objs = el.all(:css, ".contacts") | |
if objs.count > 0 | |
company.full_contacts = objs.first.text | |
emails = objs.first.all(:css, "a") | |
if emails.count > 0 | |
company.email = emails.first.text | |
end | |
end | |
company.country = country | |
company.save! | |
puts company.id.to_s + ": " + company.name | |
company.reload | |
end | |
if have_css('a.searchnext') | |
page.all(:css, "a.searchnext").first.click | |
if url == current_url # it seems like we reached the last page | |
break | |
end | |
url = current_url | |
else | |
break | |
end | |
end | |
end | |
end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment