bengaluru-food-census/parse_restaurant.rb

27 lines
571 B
Ruby

require 'nokogiri'
require "csv"
restaurants = []
CSV.open("database.csv", "wb") do |csv|
csv << ["url", "title", "location", "address", "cuisine"]
Dir.glob("html/restaurants-*.html") do |file|
page = Nokogiri::HTML(open(file))
page.css('.plr10').each do |div|
links = div.css('a')
spans = div.css('span')
title = links[0].text
location = links[1].text
address = spans[1].text
cuisine = spans[0].text
url = links[0]['href']
csv << [url, title, location, address, cuisine]
puts url
end
end
end