Question

I’m trying to scrape the following website, since the XML is malformed and does not contain all of the data I need:

http://www.cafebonappetit.com/menu/your-cafe/pitzer

When I fetch the document with Mechanize, however, I only get:

{meta_refresh}
{title "Collins  | Claremont McKenna Cafés | Café Bon Appétit"}
{iframes}
{frames}
{links
 #<Mechanize::Page::Link "Welcome" "http://www.cafebonappetit.com/">
 #<Mechanize::Page::Link "Our Approach" "javascript://">
 #<Mechanize::Page::Link
 "Kitchen Principles"
 "http://www.cafebonappetit.com/our-approach/kitchen-principles">
 .....
 }

Unfortunately, I obviously need to get at what is in the tables (I guess they are iFrames). Any thoughts?

Thanks!

Answer 1

Here’s a simple mech + Nokogiri script that scrapes the menu items.

require 'rubygems'
require 'mechanize'
require 'pp'

agent = Mechanize.new
url   = "http://www.cafebonappetit.com/menu/your-cafe/pitzer"
page  = agent.get(url)

#Grab each daily menu
page.search('div#menu-items > table.my-day-menu-table').each do |menu|
  day = menu.xpath('preceding-sibling::div[1]/a').text.strip
  puts day
  fare = []
  #Collect the menu items
  menu.xpath('tr').each do |item|
    fare << item.xpath('td/strong').map(&:text).join(": ")
  end
  pp fare
end

Result (excerpt):

Sunday, May 6th, 2012
["Brunch",
 "chef's table: custom omelet bar",
 "main plate: chicken sanchez",
 "meatless chicken and sauce",
 "options: banana pancakes",
 "stocks: beed barley",
 "vegetable minestrone",
 "Lunch",
 "main plate: steamed broccoli",
 "Dinner",
 "chef's table: pasta bar",
 "farm to fork: sauteed rainbow chard",
 "options: mozzarella sticks",
 "ovens: pizza bar",
 "main plate: roasted herb chicken",
 "baked ziti pasta",
 "steamed carrots and parsnips"]