Created
May 13, 2014 16:45
-
-
Save malev/40cc453e1c6c15db102f to your computer and use it in GitHub Desktop.
Scrapper for list of names in BA
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
# encoding: UTF-8 | |
require 'open-uri' | |
require 'nokogiri' | |
require 'csv' | |
def gen_url(offset=0) | |
if offset == 0 | |
"http://www.buenosaires.gob.ar/areas/registrocivil/nombres/busqueda/buscador_nombres.php?&menu_id=16082" | |
else | |
"http://www.buenosaires.gob.ar/areas/registrocivil/nombres/busqueda/buscador_nombres.php?offset=#{offset}&menu_id=16082" | |
end | |
end | |
def scrap | |
range = (0..9600) | |
range.step(150).each_slice(10) do |group| | |
group.each do |offset| | |
threads = [] | |
threads << Thread.new{ | |
puts "Working on #{offset}" | |
File.open("#{offset}.html", "w") do |file| | |
file.write(open(gen_url(offset)).read) | |
end | |
} | |
threads.each(&:join) | |
puts "finish with group" | |
end | |
end | |
end | |
output = [] | |
headers = ["Name","years.appearing","count.male","count.female","prob.gender","obs.male","est.male","upper,lower"] | |
def get_gender(a) | |
if a == "M" | |
'male' | |
elsif a == "F" | |
'female' | |
else | |
'unknown' | |
end | |
end | |
CSV.open("output.csv", "w:iso-8859-1") do |csv| | |
csv << headers | |
Dir['html/*.html'].sort.each do |file| | |
range = (0..9600) | |
range.step(150).each do |num| | |
filename = "html/#{num}.html" | |
if File.exists?(filename) | |
html = open(filename, "r:iso-8859-1").read | |
doc = Nokogiri::HTML(html) | |
doc.css('table.contenido tbody tr').each do |trow| | |
begin | |
csv << [trow.children[0].text, nil,nil,nil,get_gender(trow.children[1].text)] | |
rescue => e | |
puts e, trow.children[0].text | |
end | |
end | |
end | |
end | |
end | |
end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment