Build statistics block from new data format

Previously we're expecting statistics data to be represented in three
columns, one of which would contain the variable name.

That is no longer the case. The variable name is coming from the Y-axis
header. Now the first column represents the x-axis and all of the
others represent the values on the y-axes.

The statistics block is still able to display multiple lines on a line
graph with the same x-axis.

The first header value is being ignored because that is the value for
the x-axis. The rows method is concerned with collating the values to
plot the data against the y-axis.

Also updates the tests to handle files where some data points are
missing, or provided as decimal values. And updates the landing page
factory to accept four attachments to test these different scenarios.

We no longer need to symbolize the keys in the CSV as we need to be able
to match against the header value which is a string, and pass the header
values for the y-axis to the chart component as a string.

app/models/landing_page/block/statistics.rb

@@ -11,25 +11,16 @@ def x_axis_keys
     end
 
     def rows
-      rows = []
-
-      csv_rows.each do |row|
-        variable_name = row.values.second
-        value = row.values.last
-
-        existing_row = rows.find { |item| item[:label].include?(variable_name) }
-
-        if existing_row.present?
-          existing_row[:values] << value.to_i
-        else
-          rows << {
-            label: variable_name,
-            values: [value.to_i],
-          }
+      csv_headers[1..].map do |header|
+        values = csv_rows.map do |row|
+          row[header].to_f
         end
-      end
 
-      rows
+        {
+          label: header,
+          values:,
+        }
+      end
     end
 
     def attachment
@@ -39,16 +30,24 @@ def attachment
   private
 
     def csv_rows
-      @csv_rows ||= begin
-        rows = if attachment
-                 CSV.new(URI.parse(attachment.url).open, headers: true).map(&:to_h)
-               else
-                 # SCAFFOLDING
-                 csv_file_path = Rails.root.join("#{STATISTICS_DATA_PATH}/#{data['csv_file']}")
-                 CSV.read(csv_file_path, headers: true).map(&:to_h)
-               end
-        rows.each(&:deep_symbolize_keys!)
-      end
+      @csv_rows ||= opened_csv.map(&:to_h)
+    end
+
+    def csv_headers
+      opened_csv.headers
+    end
+
+    def opened_csv
+      @opened_csv ||= attachment ? csv_from_url : csv_from_file
+    end
+
+    def csv_from_url
+      CSV.parse(URI.parse(attachment.url).open, headers: true)
+    end
+
+    def csv_from_file
+      csv_file_path = Rails.root.join("#{STATISTICS_DATA_PATH}/#{data['csv_file']}")
+      CSV.read(csv_file_path, headers: true)
     end
   end
 end

spec/factories/content_items.rb

@@ -28,6 +28,28 @@
               title: "Data Two",
               url: "https://www.asset.test.gov.uk/data_two.csv",
             },
+            {
+              accessible: false,
+              attachment_type: "document",
+              content_type: "text/csv",
+              file_size: 123,
+              filename: "data_three.csv",
+              id: 12_347,
+              preview_url: "https://www.asset.test.gov.uk/data_three.csv/preview",
+              title: "Data Three",
+              url: "https://www.asset.test.gov.uk/data_three.csv",
+            },
+            {
+              accessible: false,
+              attachment_type: "document",
+              content_type: "text/csv",
+              file_size: 123,
+              filename: "data_four.csv",
+              id: 12_348,
+              preview_url: "https://www.asset.test.gov.uk/data_four.csv/preview",
+              title: "Data Four",
+              url: "https://www.asset.test.gov.uk/data_four.csv",
+            },
           ],
         }
       end

spec/models/content_item_spec.rb

@@ -77,7 +77,7 @@
 
   describe "#attachments" do
     it "loads the attachment data from the content item" do
-      expect(subject.attachments.count).to eq(2)
+      expect(subject.attachments.count).to eq(4)
       expect(subject.attachments[0].title).to eq("Data One")
     end
   end

spec/models/landing_page/block/statistics_spec.rb

@@ -14,60 +14,106 @@
   before do
     stub_request(:get, "https://www.asset.test.gov.uk/data_one.csv").to_return(status: 200, body: File.read("spec/fixtures/landing_page_statistics_data/data_one.csv"), headers: {})
     stub_request(:get, "https://www.asset.test.gov.uk/data_two.csv").to_return(status: 200, body: File.read("spec/fixtures/landing_page_statistics_data/data_two.csv"), headers: {})
+    stub_request(:get, "https://www.asset.test.gov.uk/data_three.csv").to_return(status: 200, body: File.read("spec/fixtures/landing_page_statistics_data/data_three.csv"), headers: {})
+    stub_request(:get, "https://www.asset.test.gov.uk/data_four.csv").to_return(status: 200, body: File.read("spec/fixtures/landing_page_statistics_data/data_four.csv"), headers: {})
   end
 
   describe "#x_axis_keys" do
     it "gets all of the x-axis data points" do
       expected_keys = %w[
-        2024-01-01
-        2024-02-01
-        2024-03-01
-        2024-04-01
-        2024-05-01
-        2024-06-01
+        1/6/2013
+        1/6/2014
+        1/6/2015
+        1/6/2016
+        1/6/2017
+        1/6/2018
+        1/6/2019
       ]
 
       expect(subject.x_axis_keys).to eq(expected_keys)
     end
+  end
 
-    it "gets all of the unique x-axis data points" do
-      expected_keys = %w[
-        2024-01-01
-        2024-02-01
-        2024-03-01
+  describe "#rows" do
+    it "gets the rows for one line of data with decimals" do
+      expected_rows = [
+        {
+          label: "Percentage of people being kinder to one and other",
+          values: [
+            0.5,
+            0.6,
+            0.7,
+            0.5,
+            0.6,
+            0.7,
+            0.55,
+          ],
+        },
       ]
-      blocks_hash["csv_file"] = "data_two.csv"
 
-      expect(subject.x_axis_keys).to eq(expected_keys)
+      expect(subject.rows).to eq(expected_rows)
     end
-  end
 
-  describe "#rows" do
-    it "gets the rows for one line of data" do
+    it "gets the rows for one line of data with whole numbers" do
+      blocks_hash["csv_file"] = "data_three.csv"
+
       expected_rows = [
         {
-          label: "variable_name",
-          values: [10, 11, 12, 13, 14, 15],
+          label: "The number of people working smarter not harder",
+          values: [
+            45_775.0,
+            47_518.0,
+            50_546.0,
+            56_042.0,
+            45_768.0,
+            34_530.0,
+            29_585.0,
+          ],
         },
       ]
 
       expect(subject.rows).to eq(expected_rows)
     end
 
-    it "gets the rows for multiple lines of data" do
+    it "gets the rows for one line of data when some rows are missing" do
+      blocks_hash["csv_file"] = "data_two.csv"
+
+      expected_rows = [
+        {
+          label: "Percentage of people being kinder to one and other",
+          values: [
+            0.5,
+            0.6,
+            0.7,
+            0.5,
+            0.6,
+            0.7,
+            0.5,
+            0.0,
+            0.0,
+            0.5,
+            0.6,
+          ],
+        },
+      ]
+
+      expect(subject.rows).to eq(expected_rows)
+    end
+
+    it "gets the rows for two lines of data" do
+      blocks_hash["csv_file"] = "data_four.csv"
+
       expected_rows = [
         {
-          label: "variable_name",
-          values: [10, 11, 12],
+          label: "Generation X",
+          values: [10.0, 20.0, 25.0, 30.0],
         },
         {
-          label: "variable_name_two",
-          values: [13, 14, 15],
+          label: "Millenials",
+          values: [15.0, 25.0, 30.0, 35.0],
         },
       ]
 
-      blocks_hash["csv_file"] = "data_two.csv"
-
       expect(subject.rows).to eq(expected_rows)
     end
   end