interface.rb

require "yaml"
require_relative "scraper"

# Get the top 5 urls
urls = top_5_links

# Scrape the movie info for each of them
movie_infos = urls.map do |url|
  puts "Scraping #{url}...."
  movie_info(url)
end

File.open("movies.yml", "wb") do |file|
  file << YAML.dump(movie_infos)
end

scraper.rb

require "nokogiri"
require "httparty" # gem install httparty

def top_5_links
  top_url = "https://www.imdb.com/chart/top"
  # Open the link as an html string
  html_string = HTTParty.get(top_url)
  # Parse the html string
  doc = Nokogiri::HTML.parse(html_string)
  # Identify the first 5 <a> elements from the page

  elements = doc.search(".lister-list .titleColumn a").first(5)
  elements.map do |element|
    # Get the href from each element and put them into an array
    "https://www.imdb.com#{element.attr("href")}"
  end
end

def movie_info(url)
  # Open the url
  html_string = HTTParty.get(url)
  # Parse the html string
  doc = Nokogiri::HTML.parse(html_string)
  meta_elements = doc.at(".ipc-metadata-list").children

  # find the element with title
  title = doc.at("h1").text
  # find the element with director
  director = meta_elements.first.at("a").text
  # find the element with cast
  cast = meta_elements.last.search("li a").map { |el| el.text }
  # find the element with storyline
  storyline = doc.at('[data-testid="plot"]').text
  # find the element with year
  year = doc.at('.ipc-link').text.to_i

  # return the movie info hash
  { title: title, cast: cast, storyline: storyline, year: year, director: director }
end