Add primary key to preview_cards_statuses join table (includes deduplication migration) (#25243)

db/post_migrate/20230803082451_add_unique_index_on_preview_cards_statuses.rb

@@ -0,0 +1,39 @@
+# frozen_string_literal: true
+
+class AddUniqueIndexOnPreviewCardsStatuses < ActiveRecord::Migration[6.1]
+  disable_ddl_transaction!
+
+  def up
+    add_index :preview_cards_statuses, [:status_id, :preview_card_id], name: :preview_cards_statuses_pkey, algorithm: :concurrently, unique: true
+  rescue ActiveRecord::RecordNotUnique
+    deduplicate_and_reindex!
+  end
+
+  def down
+    remove_index :preview_cards_statuses, name: :preview_cards_statuses_pkey
+  end
+
+  private
+
+  def deduplicate_and_reindex!
+    deduplicate_preview_cards!
+
+    safety_assured { execute 'REINDEX INDEX preview_cards_statuses_pkey' }
+  rescue ActiveRecord::RecordNotUnique
+    retry
+  end
+
+  def deduplicate_preview_cards!
+    # Statuses should have only one preview card at most, even if that's not the database
+    # constraint we will end up with
+    duplicate_ids = select_all('SELECT status_id FROM preview_cards_statuses GROUP BY status_id HAVING count(*) > 1;').rows
+
+    duplicate_ids.each_slice(1000) do |ids|
+      # This one is tricky: since we don't have primary keys to keep only one record,
+      # use the physical `ctid`
+      safety_assured do
+        execute "DELETE FROM preview_cards_statuses p WHERE p.status_id IN (#{ids.join(', ')}) AND p.ctid NOT IN (SELECT q.ctid FROM preview_cards_statuses q WHERE q.status_id = p.status_id LIMIT 1)"
+      end
+    end
+  end
+end

db/post_migrate/20230803112520_add_primary_key_to_preview_cards_statuses_join_table.rb

@@ -0,0 +1,20 @@
+# frozen_string_literal: true
+
+class AddPrimaryKeyToPreviewCardsStatusesJoinTable < ActiveRecord::Migration[6.1]
+  disable_ddl_transaction!
+
+  def up
+    safety_assured do
+      execute 'ALTER TABLE preview_cards_statuses ADD PRIMARY KEY USING INDEX preview_cards_statuses_pkey'
+    end
+  end
+
+  def down
+    safety_assured do
+      # I have found no way to demote the primary key to an index, instead, re-create the index
+      execute 'CREATE UNIQUE INDEX CONCURRENTLY preview_cards_statuses_pkey_tmp ON preview_cards_statuses (status_id, preview_card_id)'
+      execute 'ALTER TABLE preview_cards_statuses DROP CONSTRAINT preview_cards_statuses_pkey'
+      execute 'ALTER INDEX preview_cards_statuses_pkey_tmp RENAME TO preview_cards_statuses_pkey'
+    end
+  end
+end

db/schema.rb

@@ -10,7 +10,7 @@
 #
 # It's strongly recommended that you check this file into your version control system.
 
-ActiveRecord::Schema[7.0].define(version: 2023_07_24_160715) do
+ActiveRecord::Schema[7.0].define(version: 2023_08_03_112520) do
   # These are extensions that must be enabled in order to support this database
   enable_extension "plpgsql"
 
@@ -805,7 +805,7 @@ ActiveRecord::Schema[7.0].define(version: 2023_07_24_160715) do
     t.index ["url"], name: "index_preview_cards_on_url", unique: true
   end
 
-  create_table "preview_cards_statuses", id: false, force: :cascade do |t|
+  create_table "preview_cards_statuses", primary_key: ["status_id", "preview_card_id"], force: :cascade do |t|
     t.bigint "preview_card_id", null: false
     t.bigint "status_id", null: false
     t.index ["status_id", "preview_card_id"], name: "index_preview_cards_statuses_on_status_id_and_preview_card_id"