Add link and page PORO helpers (#3)

betagouv · Feb 14, 2025 · 8a3623c · 8a3623c
2 parents 4a775f5 + aaf2c55
commit 8a3623c
Show file tree

Hide file tree

Showing 4 changed files with 245 additions and 0 deletions.
diff --git a/app/models/link.rb b/app/models/link.rb
@@ -0,0 +1,8 @@
+Link = Data.define(:href, :text) do
+  def initialize(href:, text:)
+    super(href: href.to_s, text: text&.squish || "")
+  end
+
+  def to_str = href
+  def ==(other) = href == other.href
+end
diff --git a/app/models/page.rb b/app/models/page.rb
@@ -0,0 +1,34 @@
+require "net/http"
+
+class Page < Data.define(:url, :root)
+  CACHE_TTL = 10.minutes
+
+  def initialize(url:, root: nil)
+    super(url: URI.parse(url), root: URI.parse(root || url))
+  end
+
+  def path = url.to_s.delete_prefix(root.to_s)
+  def root? = url == root
+  def html = Rails.cache.fetch(url, expires_in: CACHE_TTL) { Net::HTTP.get(URI.parse(url)) }
+  def dom = Nokogiri::HTML(html)
+  def css(selector) = dom.css(selector)
+  def title = dom.title&.squish
+  def text = dom.text&.squish
+  def headings = dom.css("h1,h2,h3,h4,h5,h6").collect(&:text).collect(&:squish)
+  def internal_links = links.select { |link| link.href.start_with?(root) }
+  def external_links = links - internal_links
+
+  def links
+    dom.css("a[href]:not([href^='#']):not([href^=mailto]):not([href^=tel])").collect do |link|
+      href = link["href"]
+      uri = URI.parse(href)
+      if uri.relative?
+        relative_path = href.start_with?("/") ? href[1..-1] : href
+        uri = URI.parse(root.to_s.chomp("/") + "/" + relative_path)
+      end
+      uri.fragment = nil
+      uri.query = nil
+      Link.new(uri, link.text)
+    end.compact
+  end
+end
diff --git a/spec/models/link_spec.rb b/spec/models/link_spec.rb
@@ -0,0 +1,47 @@
+require "rails_helper"
+
+RSpec.describe Link do
+  describe "#initialize" do
+    it "creates a Link with href and text" do
+      link = Link.new(href: "https://example.com", text: "Example")
+      expect(link.href).to eq("https://example.com")
+      expect(link.text).to eq("Example")
+    end
+
+    it "converts href to string" do
+      link = Link.new(href: URI("https://example.com"), text: "Example")
+      expect(link.href).to eq("https://example.com")
+    end
+
+    it "squishes text" do
+      link = Link.new(href: "https://example.com", text: "  Example  Text  ")
+      expect(link.text).to eq("Example Text")
+    end
+
+    it "handles nil text" do
+      link = Link.new(href: "https://example.com", text: nil)
+      expect(link.text).to eq("")
+    end
+  end
+
+  describe "#to_str" do
+    it "returns the href" do
+      link = Link.new(href: "https://example.com", text: "Example")
+      expect(link.to_str).to eq("https://example.com")
+    end
+  end
+
+  describe "#==" do
+    it "considers links equal if they have the same href" do
+      link1 = Link.new(href: "https://example.com", text: "Example 1")
+      link2 = Link.new(href: "https://example.com", text: "Example 2")
+      expect(link1).to eq(link2)
+    end
+
+    it "considers links different if they have different hrefs" do
+      link1 = Link.new(href: "https://example1.com", text: "Example")
+      link2 = Link.new(href: "https://example2.com", text: "Example")
+      expect(link1).not_to eq(link2)
+    end
+  end
+end
diff --git a/spec/models/page_spec.rb b/spec/models/page_spec.rb
@@ -0,0 +1,156 @@
+require "rails_helper"
+
+RSpec.describe Page do
+  let(:root) { "https://example.com" }
+  let(:url) { "https://example.com/about" }
+  let(:parsed_url) { URI.parse(url) }
+  let(:page) { described_class.new(url:, root:) }
+  let(:html_content) do
+    <<~HTML
+      <!DOCTYPE html>
+      <html>
+        <head>
+          <title>Example Page</title>
+        </head>
+        <body>
+          <h1>Main Heading</h1>
+          <h2>Sub Heading</h2>
+          <p>Some content</p>
+          <a href="/contact">Contact</a>
+          <a href="https://external.com">External</a>
+          <a href="tel:123456">Phone</a>
+          <a href="mailto:[email protected]">Email</a>
+          <a href="#section">Section</a>
+          <a href="relative/path">Relative</a>
+        </body>
+      </html>
+    HTML
+  end
+
+  before do
+    allow(Net::HTTP).to receive(:get).and_return(html_content)
+  end
+
+  describe "#path" do
+    it "returns the path portion of the URL" do
+      expect(page.path).to eq("/about")
+    end
+
+    context "when URL is the root URL" do
+      let(:url) { root }
+
+      it "returns an empty string" do
+        expect(page.path).to eq("")
+      end
+    end
+  end
+
+  describe "#root?" do
+    it "returns false when URL is not the root URL" do
+      expect(page.root?).to be false
+    end
+
+    context "when URL is the root URL" do
+      let(:url) { root }
+
+      it "returns true" do
+        expect(page.root?).to be true
+      end
+    end
+  end
+
+  describe "#html" do
+    it "fetches and caches the page content" do
+      expect(Rails.cache).to receive(:fetch)
+        .with(parsed_url, expires_in: described_class::CACHE_TTL)
+        .and_yield
+
+      expect(page.html).to eq(html_content)
+      expect(Net::HTTP).to have_received(:get)
+        .with(Addressable::URI.parse(url))
+        .once
+    end
+  end
+
+  describe "#dom" do
+    it "returns a Nokogiri::HTML document" do
+      expect(page.dom).to be_a(Nokogiri::HTML::Document)
+    end
+  end
+
+  describe "#css" do
+    it "forwards CSS selector queries to the DOM" do
+      expect(page.css("h1").first.text).to eq("Main Heading")
+    end
+  end
+
+  describe "#title" do
+    it "returns the page title" do
+      expect(page.title).to eq("Example Page")
+    end
+  end
+
+  describe "#text" do
+    it "returns the full text content" do
+      expect(page.text).to include("Main Heading", "Sub Heading", "Some content")
+    end
+  end
+
+  describe "#headings" do
+    it "returns an array of text, one line for each heading" do
+      expect(page.headings).to eq(["Main Heading", "Sub Heading"])
+    end
+  end
+
+  describe "#links" do
+    it "returns a hash of URLs and their link texts" do
+      expected_links = [
+        Link.new("https://example.com/contact", "Contact"),
+        Link.new("https://external.com", "External"),
+        Link.new("https://example.com/relative/path", "Relative"),
+      ]
+      expect(page.links).to eq(expected_links)
+    end
+
+    it "excludes mailto and tel links" do
+      expect(page.links.collect(&:text)).not_to include("Phone", "Email")
+    end
+
+    it "excludes fragment-only links" do
+      expect(page.links.collect(&:text)).not_to include("Section")
+    end
+
+    it "resolves relative URLs" do
+      expect(page.links.collect(&:href)).to include("https://example.com/relative/path")
+    end
+
+    it "strips fragments and query parameters from URLs" do
+      html_with_params = html_content.gsub(
+        '<a href="https://external.com">',
+        '<a href="https://external.com?param=1#section">'
+      )
+      allow(Net::HTTP).to receive(:get).and_return(html_with_params)
+
+      expect(page.links.collect(&:href)).to include("https://external.com")
+    end
+  end
+
+  describe "#internal_links" do
+    it "returns only links that start with the root URL" do
+      expected_internal_links = [
+        Link.new("https://example.com/contact", "Contact"),
+        Link.new("https://example.com/relative/path", "Relative"),
+      ]
+      expect(page.internal_links).to eq(expected_internal_links)
+    end
+  end
+
+  describe "#external_links" do
+    it "returns only links that don't start with the root URL" do
+      expected_external_links = [
+        Link.new("https://external.com", "External")
+      ]
+      expect(page.external_links).to eq(expected_external_links)
+    end
+  end
+end