Merge pull request #138 from alphagov/title-desc-content

Add title and description to indexable content
alphagov · Dec 7, 2023 · bd3960a · bd3960a
2 parents 990d5c2 + 7483d38
commit bd3960a
Show file tree

Hide file tree

Showing 2 changed files with 20 additions and 8 deletions.
diff --git a/app/models/concerns/publishing_api/content.rb b/app/models/concerns/publishing_api/content.rb
@@ -2,7 +2,14 @@ module PublishingApi
   module Content
     # All the possible keys in the message hash that can contain the primary unstructured document
     # content that we want to index, represented as JsonPath path strings.
+    #
+    # Note that this also indexes some things that are in metadata attributes (e.g. title,
+    # description) as it seems that the engine is better at picking up on them if they are in the
+    # unstructured content.
     INDEXABLE_CONTENT_VALUES_JSON_PATHS = %w[
+      $.title
+      $.description
+
       $.details.acronym
       $.details.attachments[*]['title','isbn','unique_reference','command_paper_number','hoc_paper_number']
       $.details.body

diff --git a/spec/integration/document_synchronization_spec.rb b/spec/integration/document_synchronization_spec.rb
@@ -33,7 +33,7 @@
           part_of_taxonomy_tree: %w[37d0fa26-abed-4c74-8835-b3b51ae1c8b2],
           locale: "en",
         },
-        content: a_string_starting_with("<div class=\"govspeak\"><p>The UK was represented remotely"),
+        content: a_string_including("<div class=\"govspeak\"><p>The UK was represented remotely"),
         payload_version: 12_345,
       )
     end
@@ -87,7 +87,7 @@
             },
           ],
         },
-        content: a_string_starting_with("<h1>Warnings and insurance</h1>\n<p>The Foreign"),
+        content: a_string_including("<h1>Warnings and insurance</h1>\n<p>The Foreign"),
         payload_version: 12_345,
       )
     end
@@ -117,7 +117,7 @@
           ],
           locale: "en",
         },
-        content: a_string_starting_with("<div class=\"govspeak\"><p>In the UEFA Champions"),
+        content: a_string_including("<div class=\"govspeak\"><p>In the UEFA Champions"),
         payload_version: 12_345,
       )
     end
@@ -141,7 +141,7 @@
           content_purpose_supergroup: "guidance_and_regulation",
           locale: "en",
         },
-        content: a_string_matching(/\A<h2 id="section-6-1">6\.1\. Structure.+<\/table>\n\n\z/m),
+        content: a_string_matching(/<h2 id="section-6-1">6\.1\. Structure.+<\/table>\n\n/m),
         payload_version: 12_345,
       )
     end
@@ -166,7 +166,7 @@
           content_purpose_supergroup: "other",
           locale: "en",
         },
-        content: a_string_starting_with("LAA\n<div class=\"govspeak\"><p>We provide civil"),
+        content: a_string_including("LAA\n<div class=\"govspeak\"><p>We provide civil"),
         payload_version: 12_345,
       )
     end
@@ -192,7 +192,9 @@
           part_of_taxonomy_tree: %w[f3caf326-fe33-410f-b7f4-553f4011c81e],
           locale: "en",
         },
-        content: a_string_starting_with(<<~TEXT.chomp),
+        content: a_string_including(<<~TEXT.chomp),
+          Directgov 2010 and beyond: revolution not evolution, a report by Martha Lane Fox
+          A report from the Digital Champion Martha Lane Fox with recommendations for the future of Directgov.
           Directgov 2010 and Beyond: Revolution Not Evolution - Letter from Martha Lane Fox to Francis Maude
           Francis Maude's reply to Martha Lane Fox's letter
           Directgov Strategic Review - Executive Summary
@@ -221,7 +223,10 @@
           content_purpose_supergroup: "other",
           locale: "en",
         },
-        content: "",
+        content: <<~TEXT.chomp,
+          UK help and services in Switzerland
+          Services if you're visiting, studying, working or living in Switzerland. Includes information about trading with and doing business in the UK and Switzerland, and your rights after the UK’s exit from the EU.
+        TEXT
         payload_version: 12_345,
       )
     end
@@ -245,7 +250,7 @@
           content_purpose_supergroup: "other",
           locale: "en",
         },
-        content: "Brighton & Hove City Council",
+        content: a_string_including("Brighton & Hove City Council"),
         payload_version: 17,
       )
     end