Added option to return spreadsheet headers when parsing

Convert EOC script now finds its columns using the headers

app/routines/process_spreadsheet.rb

@@ -3,7 +3,7 @@ class ProcessSpreadsheet
 
   protected
 
-  def exec(filename:, offset:, pad_cells: true, &block)
+  def exec(filename:, offset: 1, pad_cells: true, return_headers: false, &block)
     raise ArgumentError, 'A block must be provided' if block.nil?
 
     if File.extname(filename) == 'csv'
@@ -14,14 +14,17 @@ def exec(filename:, offset:, pad_cells: true, &block)
       method = :each_row_streaming
     end
 
+    args = []
     pad_to_size = 0 if pad_cells
     klass.new(filename).public_send(method).each_with_index do |row, row_index|
-      if pad_cells
+      if return_headers && row_index == 0
+        args << row.map(&:to_s)
+      elsif pad_cells
         row += [nil] * (pad_to_size - row.length) if pad_to_size > row.length
         pad_to_size = row.length
       end
 
-      block.call(row, row_index) if row_index >= offset
+      block.call(*args.concat([row, row_index])) if row_index >= offset
     end
   end
 end

lib/tasks/exercises/tag.rake

@@ -1,6 +1,6 @@
 namespace :exercises do
   namespace :tag do
-    desc "add a module to exercises that are tagged with another module"
+    desc 'add a module to exercises that are tagged with another module'
     task :import_from_module_map, [:file] => :environment do |t, args|
       Tag.transaction do
         CSV.foreach(args[:file]) do |src_uuid, new_uuid|
@@ -24,7 +24,7 @@ namespace :exercises do
     # filename, [skip_first_row]
     # Example: rake exercises:tag:spreadsheet[tags.xlsx]
     #          will tag exercises based on tags.xlsx
-    desc "tags exercises using a spreadsheet"
+    desc 'tags exercises using a spreadsheet'
     task :spreadsheet, [:filename, :skip_first_row] => :environment do |t, args|
       # Output import logging info to the console (except in the test environment)
       original_logger = Rails.logger
@@ -44,7 +44,7 @@ namespace :exercises do
     # filename, book_uuid, [skip_first_row]
     # Example: rake exercises:tag:assessments[tags.xlsx,12345]
     #          will tag exercises based on tags.xlsx with the book_uuid 12345
-    desc "tags exercises for OpenStax Assessments using a spreadsheet"
+    desc 'tags exercises for OpenStax Assessments using a spreadsheet'
     task :assessments, [:filename, :book_uuid, :skip_first_row] => :environment do |t, args|
       # Output import logging info to the console (except in the test environment)
       original_logger = Rails.logger
@@ -78,36 +78,34 @@ namespace :exercises do
         has_units = ActiveModel::Type::Boolean.new.cast args[:has_units]
         chapters = has_units ? parts.flat_map { |part| get_parts.call(part) } : parts
 
-        chapter_uuid_by_page_uuid = {}
-        chapters.each do |chapter|
-          chapter.all_pages.each do |page|
-            chapter_uuid_by_page_uuid[page.uuid] = chapter.uuid
-          end
-        end
+        chapter_uuids = chapters.map(&:uuid)
 
         Rails.logger.info { "Processing \"#{args[:filename]}\"" }
 
         output_filename = "#{book.slug}.csv"
+        downcased_headers = nil
+        chapter_index = nil
+        exercise_id_index = nil
         CSV.open(output_filename, 'w') do |csv|
           csv << [ 'Exercise UID', 'Tags...' ]
 
-          ProcessSpreadsheet.call(filename: args[:filename], offset: 2) do |row, index|
-            next Rails.logger.warn do
-              "Skipped row #{index + 1} because column 5 is blank"
-            end if row[4].blank?
+          ProcessSpreadsheet.call(filename: args[:filename], return_headers: true) do |headers, row, index|
+            downcased_headers ||= headers.map(&:downcase)
 
-            next Rails.logger.warn do
-              "Skipped row #{index + 1} because column 4 says it is not in Exercises"
-            end if row[3].to_s == 'no'
+            chapter_index ||= downcased_headers.find_index { |header| header.include? 'chapter' }
+            raise ArgumentError, 'Could not find Chapter column' if chapter_index.nil?
+
+            exercise_id_index ||= downcased_headers.find_index do |header|
+              header.include?('assessment') || header.include?('exercise')
+            end
+            raise ArgumentError, 'Could not find Assessment ID column' if exercise_id_index.nil?
 
-            page_uuid = row[0].to_s.strip
-            chapter_uuid = chapter_uuid_by_page_uuid[page_uuid]
-            next Rails.logger.warn do
-              "Skipped row #{index + 1} because page with UUID \"#{page_uuid}\" was not found"
-            end if chapter_uuid.nil?
+            chapter = row[chapter_index].to_s.strip
+            # The value in the Chapter column may be a UUID or a chapter number
+            chapter_uuid = chapter_uuids.include?(chapter) ? chapter : chapter_uuids[Integer(chapter) - 1]
 
             csv << [
-              row[4].to_s.strip,
+              row[exercise_id_index].to_s.strip,
               "assessment:practice:https://openstax.org/orn/book:subbook/#{
                 args[:book_uuid]}:#{chapter_uuid}"
             ]