Merge pull request #59 from alphagov/parts-body

Handle body content in `parts`

Gemfile

@@ -11,6 +11,7 @@ gem "railties", RAILS_GEMS_VERSION
 
 gem "bootsnap", require: false
 gem "govuk_app_config"
+gem "loofah"
 gem "oj"
 
 # Gems specific to the document sync worker that aren't required for the main Rails API app

Gemfile.lock

@@ -451,6 +451,7 @@ DEPENDENCIES
   govuk_test
   json_schemer
   jsonpath
+  loofah
   oj
   plek
   railties (= 7.0.7.2)

lib/document_sync_worker/document/content_with_multiple_types.rb

@@ -0,0 +1,22 @@
+module DocumentSyncWorker::Document
+  class ContentWithMultipleTypes
+    def initialize(content_with_multiple_types)
+      @content_with_multiple_types = content_with_multiple_types
+    end
+
+    def html_content
+      @content_with_multiple_types.find { _1["content_type"] == "text/html" }&.dig("content")
+    end
+
+    def text_content
+      Loofah
+        .document(html_content)
+        .to_text(encode_special_chars: false)
+        .squish
+    end
+
+    def summarized_text_content(length: 75, omission: "…", separator: " ")
+      text_content.truncate(length, omission:, separator:)
+    end
+  end
+end

lib/document_sync_worker/document/publish.rb

@@ -96,7 +96,13 @@ def historic?
       def parts
         document_hash
           .dig("details", "parts")
-          &.map { { slug: _1["slug"], title: _1["title"] } } || []
+          &.map do
+            {
+              slug: _1["slug"],
+              title: _1["title"],
+              body: ContentWithMultipleTypes.new(_1["body"]).summarized_text_content,
+            }
+          end
       end
     end
   end

spec/lib/document_sync_worker/document/publish_spec.rb

@@ -327,7 +327,7 @@
       context "when the document has no parts" do
         let(:parts) { nil }
 
-        it { is_expected.to be_empty }
+        it { is_expected.to be_nil }
       end
 
       context "when the document has parts" do
@@ -336,17 +336,44 @@
             {
               "title" => "Part 1",
               "slug" => "/part-1",
-              "body" => "Part 1 body",
+              "body" => [
+                {
+                  "content" => "Part 1 body",
+                  "content_type" => "text/simples",
+                },
+                {
+                  "content" => "<div class=\"lipsum\">Lorem ipsum dolor sit amet, consectetur adipiscing elit. Curabitur <blink>tincidunt sem erat</blink>, eget blandit urna porta ac. Mauris lobortis tincidunt dui at pharetra.</div>",
+                  "content_type" => "text/html",
+                },
+              ],
             },
             {
               "title" => "Part 2",
               "slug" => "/part-2",
-              "body" => "Part 2 body",
+              "body" => [
+                {
+                  "content" => "I have no HTML content :(",
+                  "content_type" => "text/simples",
+                },
+              ],
             },
           ]
         end
 
-        it { is_expected.to eq([{ title: "Part 1", slug: "/part-1" }, { title: "Part 2", slug: "/part-2" }]) }
+        it "contains the expected titles" do
+          expect(extracted_parts.map { _1[:title] }).to eq(["Part 1", "Part 2"])
+        end
+
+        it "contains the expected slugs" do
+          expect(extracted_parts.map { _1[:slug] }).to eq(%w[/part-1 /part-2])
+        end
+
+        it "contains the expected body with HTML stripped and truncated" do
+          expect(extracted_parts.map { _1[:body] }).to eq([
+            "Lorem ipsum dolor sit amet, consectetur adipiscing elit. Curabitur…",
+            "",
+          ])
+        end
       end
     end
   end

spec/lib/document_sync_worker_integration_spec.rb

@@ -31,7 +31,7 @@
           668cd623-c7a8-4159-9575-90caac36d4b4 c31256e8-f328-462b-993f-dce50b7892e9
         ],
         locale: "en",
-        parts: [],
+        parts: nil,
       )
 
       expect(result[:content]).to start_with("<div class=\"govspeak\"><p>The government has")
@@ -64,11 +64,31 @@
         ],
         locale: "en",
         parts: [
-          { slug: "warnings-and-insurance", title: "Warnings and insurance" },
-          { slug: "entry-requirements", title: "Entry requirements" },
-          { slug: "safety-and-security", title: "Safety and security" },
-          { slug: "health", title: "Health" },
-          { slug: "getting-help", title: "Getting help" },
+          {
+            slug: "warnings-and-insurance",
+            title: "Warnings and insurance",
+            body: "The Foreign, Commonwealth & Development Office (FCDO) provides advice…",
+          },
+          {
+            slug: "entry-requirements",
+            title: "Entry requirements",
+            body: "This advice reflects the UK government’s understanding of current rules…",
+          },
+          {
+            slug: "safety-and-security",
+            title: "Safety and security",
+            body: "Terrorism There is a high threat of terrorist attack globally affecting UK…",
+          },
+          {
+            slug: "health",
+            title: "Health",
+            body: "Before you travel check that: your destination can provide the healthcare…",
+          },
+          {
+            slug: "getting-help",
+            title: "Getting help",
+            body: "The Foreign, Commonwealth & Development Office (FCDO) cannot provide…",
+          },
         ],
       )
 
@@ -101,7 +121,7 @@
           3dbeb4a3-33c0-4bda-bd21-b721b0f8736f
         ],
         locale: "en",
-        parts: [],
+        parts: nil,
       )
 
       expect(result[:content]).to start_with("<div class=\"govspeak\"><p>In the UEFA Champions")
@@ -131,7 +151,7 @@
         content_purpose_supergroup: "other",
         part_of_taxonomy_tree: [],
         locale: "en",
-        parts: [],
+        parts: nil,
       )
 
       expect(result[:content]).to be_blank