add one-a

app/services/dese/four_d_loader.rb

@@ -6,8 +6,10 @@ module Dese
       CSV.parse(File.read(filepath), headers: true) do |row|
         score = likert_score(row:)
         unless valid_likert_score(likert_score: score)
+          school = School.find_by_dese_id(row['DESE ID']) || School.new(name: 'Non-MCIEA school',
+                                                                        dese_id: row['DESE ID'])
           puts "Invalid score: #{score}
-        for school: #{School.find_by_dese_id(row['DESE ID']).name}
+        for school: #{school.name}
         admin data item #{admin_data_item(row:)} "
           next
         end

app/services/dese/four_d_scraper.rb

@@ -24,7 +24,7 @@ module Dese
       browser.select(id: 'ctl00_ContentPlaceHolder1_ddReportType').select(/School/)
       browser.select(id: 'ctl00_ContentPlaceHolder1_ddYear').select(text: range)
       browser.button(id: 'btnViewReport').click
-      sleep 5  # Sleep to prevent hitting mass.edu with too many requests
+      sleep 2  # Sleep to prevent hitting mass.edu with too many requests
       document = Nokogiri::HTML(browser.html)
       document.css('tr')
     end
@@ -45,15 +45,17 @@ module Dese
           next if dese_id.nil? || dese_id.zero?
 
           raw_likert_score = calculate(cells: items)
-          items << raw_likert_score
           likert_score = raw_likert_score
           likert_score = 5 if raw_likert_score > 5
           likert_score = 1 if raw_likert_score < 1
           likert_score = likert_score.round(2)
-          items << likert_score
+          output = []
-          items << id
+          output << raw_likert_score
-          items << range
+          output << likert_score
-          csv << items
+          output << id
+          output << range
+          output << items
+          csv << output.flatten
         end
       end
     end

app/services/dese/one_a_scraper.rb

@@ -0,0 +1,109 @@
+require 'watir'
+require 'csv'
+
+module Dese
+  class OneAScraper
+    def initialize(filepath: Rails.root.join('data', 'admin_data', 'dese', 'one_a.csv'))
+      url = 'https://profiles.doe.mass.edu/statereport/teacherdata.aspx'
+      browser = Watir::Browser.new
+      write_headers(filepath:)
+      academic_years = AcademicYear.all
+      academic_years.each do |academic_year|
+        document = scrape(browser:, url:, range: academic_year.range)
+        id = 'a-exp-i1'
+        write_csv(document:, filepath:, range: academic_year.range, id:) unless document.nil?
+      end
+      browser.close
+    end
+
+    def scrape(browser:, url:, range:)
+      browser.goto(url)
+
+      return unless browser.option(text: 'School').present?
+      return unless browser.option(text: range).present?
+
+      browser.select(id: 'ctl00_ContentPlaceHolder1_ddReportType').select(text: 'School')
+      browser.select(id: 'ctl00_ContentPlaceHolder1_ddYear').select(text: range)
+      browser.button(id: 'ctl00_ContentPlaceHolder1_btnViewReport').click
+      sleep 2  # Sleep to prevent hitting mass.edu with too many requests
+      Nokogiri::HTML(browser.html)
+    end
+
+    def write_headers(filepath:)
+      CSV.open(filepath, 'w') do |csv|
+        headers = ['Raw likert calculation', 'Likert Score', 'Admin Data Item', 'Academic Year', 'School Name', 'DESE ID', 'Total # of Teachers(FTE)', 'Percent of Teachers Licensed',
+                   'Student/Teacher Ratio', 'Percent of Experienced Teachers', 'Percent of Teachers without Waiver or Provisional License', 'Percent Teaching in-field']
+        csv << headers
+      end
+    end
+
+    def write_csv(document:, filepath:, range:, id:)
+      table = document.css('tr')
+      headers = document.css('.sorting')
+      header_hash = headers.each_with_index.map { |header, index| [header.text, index] }.to_h
+      experienced_teacher_index = header_hash['Percent of Experienced Teachers']
+      dese_id_index = header_hash['School Code']
+
+      CSV.open(filepath, 'a') do |csv|
+        table.each do |row|
+          items = row.css('td').map(&:text)
+          dese_id = items[1].to_i
+          next if dese_id.nil? || dese_id.zero?
+
+          raw_likert_score = items[experienced_teacher_index].to_f * 4 / 80 if experienced_teacher_index.present?
+          raw_likert_score ||= 'NA'
+          likert_score = raw_likert_score
+          if likert_score != 'NA'
+            likert_score = 5 if likert_score > 5
+            likert_score = 1 if likert_score < 1
+            likert_score = likert_score.round(2)
+          end
+
+          output = []
+          output << raw_likert_score
+          output << likert_score
+          output << 'a-exp-i1'
+          output << range
+          output << items
+          output = output.flatten
+          csv << output
+        end
+      end
+
+      in_field_index = header_hash['Percent Teaching In-Field']
+
+      CSV.open(filepath, 'a') do |csv|
+        table.each do |row|
+          items = row.css('td').map(&:text)
+          dese_id = items[dese_id_index].to_i
+          next if dese_id.nil? || dese_id.zero?
+
+          percent_in_field = items[in_field_index].to_f if in_field_index.present?
+          if in_field_index.present? && percent_in_field.present? && !percent_in_field.zero?
+            raw_likert_score = percent_in_field * 4 / 95
+          end
+          raw_likert_score ||= 'NA'
+          likert_score = raw_likert_score
+          if likert_score != 'NA'
+            likert_score = 5 if likert_score > 5
+            likert_score = 1 if likert_score < 1
+            likert_score = likert_score.round(2)
+          end
+
+          output = []
+          output << raw_likert_score
+          output << likert_score
+          output << 'a-exp-i3'
+          output << range
+          output << items
+          output = output.flatten
+          csv << output
+        end
+      end
+    end
+
+    def calculate(cells:)
+      cells[5].to_f * 4 / 95
+    end
+  end
+end

spec/services/dese/four_d_loader_spec.rb

@@ -1,5 +1,5 @@
 require 'rails_helper'
-RSpec.describe Dese::FourDLoader, type: :model do
+RSpec.describe Dese::FourDLoader do
   let(:path_to_admin_data) { Rails.root.join('spec', 'fixtures', 'sample_four_d_data.csv') }
   let(:ay_2020_21) { AcademicYear.find_by_range '2020-21' }
   let(:ay_2018_19) { AcademicYear.find_by_range '2018-19' }

spec/services/dese/four_d_scraper_spec.rb

@@ -1,6 +1,6 @@
 require 'rails_helper'
 require 'fileutils'
-RSpec.describe type: :model do
+RSpec.describe Dese::FourDScraper do
   let(:academic_years) do
     [
       create(:academic_year, range: '2020-21'),
@@ -13,6 +13,7 @@ RSpec.describe type: :model do
   before :each do
     academic_years
   end
+
   xcontext 'Creating a new FourDScraper' do
     it 'creates a csv file with the scraped data' do
       FileUtils.mkdir_p 'tmp/spec/dese'

spec/services/dese/one_a_scraper_spec.rb

@@ -0,0 +1,76 @@
+require 'rails_helper'
+require 'fileutils'
+require 'csv'
+
+RSpec.describe Dese::OneAScraper do
+  let(:academic_years) do
+    [
+      create(:academic_year, range: '2021-22'),
+      create(:academic_year, range: '2020-21')
+      # create(:academic_year, range: '2019-20'),
+      # create(:academic_year, range: '2018-19'),
+      # create(:academic_year, range: '2017-18'),
+      # create(:academic_year, range: '2016-17')
+    ]
+  end
+  let(:filepath) { Rails.root.join('tmp', 'spec', 'dese', 'one_a.csv') }
+  before do
+    FileUtils.mkdir_p 'tmp/spec/dese'
+  end
+
+  before :each do
+    academic_years
+  end
+
+  xcontext 'Creating a new Scraper' do
+    it 'creates a csv file with the scraped data' do
+      Dese::OneAScraper.new(filepath:)
+      expect(filepath).to exist
+    end
+
+    it 'has the correct headers' do
+      headers = File.open(filepath) do |file|
+        headers = file.first
+      end.split(',')
+      expect(headers).to eq ['Raw likert calculation', 'Likert Score', 'Admin Data Item', 'Academic Year',
+                             'School Name', 'DESE ID', 'Total # of Teachers(FTE)', 'Percent of Teachers Licensed',
+                             'Student/Teacher Ratio', 'Percent of Experienced Teachers',
+                             'Percent of Teachers without Waiver or Provisional License', "Percent Teaching in-field\n"]
+    end
+
+    it 'has the right likert score results for a-exp-i1' do
+      results = CSV.parse(File.read(filepath), headers: true).map do |row|
+        row['Likert Score'].to_f
+      end
+
+      expect(results.take(20)).to eq [3.7, 4, 4.31, 3.95, 3.99, 3.69, 1.69,
+                                      4.5, 4.21, 4.1, 5, 4.2, 4.51, 3.97, 4.35,
+                                      4.38, 4.08, 4, 4.12, 5]
+    end
+    it 'has the right likert score results for a-exp-i3' do
+      results = CSV.parse(File.read(filepath), headers: true).map do |row|
+        next unless row['Admin Data Item'] == 'a-exp-i3' && row['Academic Year'] == '2020-21'
+
+        row['Likert Score'].to_f
+      end.flatten.compact
+
+      expect(results.take(20)).to eq [3.68, 4.21, 4.01, 3.73, 4.21, 4.21, 2.48, 4.06, 4.21, 4.21, 4.21, 4.21, 4.21,
+                                      4.21, 4.21, 4.1, 4, 4.07, 3.32, 4.21]
+    end
+
+    it 'has the latest academic_year' do
+      academic_year = CSV.parse(File.read(filepath), headers: true).map do |row|
+        break row['Academic Year']
+      end
+
+      expect(academic_year).to eq '2021-22'
+    end
+    it 'has both admin data items in the file' do
+      results = CSV.parse(File.read(filepath), headers: true).map do |row|
+        row['Admin Data Item']
+      end
+
+      expect(results.uniq).to eq %w[a-exp-i1 a-exp-i3]
+    end
+  end
+end