Extract Van Gogh paintings

Gemfile

@@ -0,0 +1,7 @@
+source "https://rubygems.org"
+
+ruby ">= 3.0"
+
+gem "byebug", ">= 11.1"
+gem "ferrum", "~> 0.17.1"
+gem "rspec", ">= 3.12"

Gemfile.lock

@@ -0,0 +1,49 @@
+GEM
+  remote: https://rubygems.org/
+  specs:
+    addressable (2.8.7)
+      public_suffix (>= 2.0.2, < 7.0)
+    base64 (0.3.0)
+    byebug (12.0.0)
+    concurrent-ruby (1.3.5)
+    diff-lcs (1.6.2)
+    ferrum (0.17.1)
+      addressable (~> 2.5)
+      base64 (~> 0.2)
+      concurrent-ruby (~> 1.1)
+      webrick (~> 1.7)
+      websocket-driver (~> 0.7)
+    public_suffix (6.0.2)
+    rspec (3.13.1)
+      rspec-core (~> 3.13.0)
+      rspec-expectations (~> 3.13.0)
+      rspec-mocks (~> 3.13.0)
+    rspec-core (3.13.5)
+      rspec-support (~> 3.13.0)
+    rspec-expectations (3.13.5)
+      diff-lcs (>= 1.2.0, < 2.0)
+      rspec-support (~> 3.13.0)
+    rspec-mocks (3.13.5)
+      diff-lcs (>= 1.2.0, < 2.0)
+      rspec-support (~> 3.13.0)
+    rspec-support (3.13.5)
+    webrick (1.9.1)
+    websocket-driver (0.8.0)
+      base64
+      websocket-extensions (>= 0.1.0)
+    websocket-extensions (0.1.5)
+
+PLATFORMS
+  arm64-darwin-23
+  ruby
+
+DEPENDENCIES
+  byebug (>= 11.1)
+  ferrum (~> 0.17.1)
+  rspec (>= 3.12)
+
+RUBY VERSION
+   ruby 3.3.0p0
+
+BUNDLED WITH
+   2.5.3

extract_van_gogh_paintings.rb

@@ -0,0 +1,38 @@
+require 'json'
+require 'ferrum'
+require 'byebug'
+
+def extract_paintings(file_path)
+  browser = Ferrum::Browser.new(headless: true)
+  browser.goto("file://#{File.expand_path(file_path)}")
+
+  paintings = []
+
+  browser.css('div.iELo6').each do |painting_div|
+    link_tag = painting_div.at_css('a')
+    google_link = link_tag&.attribute('href')&.strip
+    google_link = "https://www.google.com#{google_link}" if google_link && !google_link.start_with?("http")
+
+    thumbnail_tag = link_tag.at_css('img.taFZJe')
+    image = thumbnail_tag&.attribute('data-src')&.strip || thumbnail_tag&.attribute('src')&.strip
+
+    painting_parent_div = painting_div.at_css('div.KHK6lb')
+    painting_name = painting_parent_div.at_css('div.pgNMRc')&.text&.strip
+    extensions_text = painting_parent_div.at_css('div.cxzHyb')&.text&.strip
+
+    painting_hash = { name: painting_name }
+    painting_hash[:extensions] = [extensions_text] unless extensions_text.nil? || extensions_text.empty?
+    painting_hash[:link] = google_link
+    painting_hash[:image] = image
+
+    paintings << painting_hash
+  end
+
+  browser.quit
+  { artworks: paintings }
+end
+
+File.write(
+  'van-gogh-paintings-extracted.json',
+  JSON.pretty_generate(extract_paintings('files/van-gogh-paintings.html'))
+)

spec/extract_van_gogh_paintings_spec.rb

@@ -0,0 +1,68 @@
+require 'json'
+
+RSpec.describe 'Van Gogh Paintings Extraction' do
+  let(:extracted_data) { JSON.parse(File.read('./van-gogh-paintings-extracted.json'))['artworks'] }
+
+  it 'matches the expected array' do
+    expected_data = JSON.parse(File.read('./files/expected-array.json'))['artworks']
+    expect(extracted_data).to eq(expected_data)
+  end
+
+  it 'ensures each artwork has required keys in correct order' do
+    extracted_data.each do |artwork|
+      keys = artwork.keys
+      expect(keys.first).to eq('name')
+      expect(keys.last).to eq('image')
+      expect(keys).to include('link')
+    end
+  end
+
+  it 'ensures extensions key is omitted when empty' do
+    extracted_data.each do |artwork|
+      if artwork['extensions'].nil? || artwork['extensions'].empty?
+        expect(artwork.key?('extensions')).to be false
+      end
+    end
+  end
+
+  it 'ensures extensions is an array if present' do
+    extracted_data.each do |artwork|
+      if artwork.key?('extensions')
+        expect(artwork['extensions']).to be_an(Array)
+      end
+    end
+  end
+
+  it 'ensures all Google links start with https://www.google.com' do
+    extracted_data.each do |artwork|
+      expect(artwork['link']).to start_with('https://www.google.com')
+    end
+  end
+
+  it 'ensures images are present for each artwork' do
+    extracted_data.each do |artwork|
+      expect(artwork['image']).not_to be_nil
+      expect(artwork['image']).not_to be_empty
+    end
+  end
+
+  it 'ensures name is always present' do
+    extracted_data.each do |artwork|
+      expect(artwork['name']).not_to be_nil
+      expect(artwork['name']).not_to be_empty
+    end
+  end
+
+  it 'ensures image is either a URL or a base64 data string' do
+    extracted_data.each do |artwork|
+      expect(artwork['image']).to match(/^https?:\/\/|^data:/)
+    end
+  end
+
+  it 'does not include unexpected keys' do
+    allowed_keys = %w[name extensions link image]
+    extracted_data.each do |artwork|
+      expect(artwork.keys - allowed_keys).to be_empty
+    end
+  end
+end