Merge pull request #796 from pulibrary/airtable-job

Add an airtable staff list job and URL

app/controllers/staff_directory_controller.rb

@@ -16,4 +16,16 @@ def removed
       format.text { send_data differ.ids.join(","), filename: "removed-staff.txt" }
     end
   end
+
+  def pul_staff_report
+    if Flipflop.air_table_staff_list?
+      job = AirTableStaff::StaffListJob.new
+      job.run
+      respond_to do |format|
+        format.csv { send_data job.read_most_recent_report, filename: "pul-staff-report.csv" }
+      end
+    else
+      render plain: 'Airtable-based staff list is turned off.  Go to /features to turn it back on.'
+    end
+  end
 end

app/models/air_table_staff/staff_list_job.rb

@@ -0,0 +1,41 @@
+# frozen_string_literal: true
+module AirTableStaff
+  # This class is responsible for running and
+  # recording the results of Airtable-based
+  # CSV file generation
+  class StaffListJob < LibJob
+    def initialize(filename: nil)
+      super(category: 'AirTableStaffDirectory')
+      @report_filename = filename if filename
+    end
+
+    private
+
+    def handle(data_set:)
+      if Flipflop.air_table_staff_list?
+        return most_recent_dataset if most_recent_dataset && recent_enough?(most_recent_dataset&.data_file)
+        write_csv_to_disk
+        data_set.data_file = report_filename
+      else
+        data_set.data = 'Airtable-based staff list is typically scheduled for this time, but it is turned off.  Go to /features to turn it back on.'
+      end
+      data_set.report_time = Time.zone.now
+      data_set
+    end
+
+    def write_csv_to_disk
+      File.open(report_filename, 'w') { |file| file.write(CSVBuilder.new.to_csv) }
+    end
+
+    def report_filename
+      @report_filename ||= begin
+        date_str = Time.zone.now.strftime('%Y%m%d%H%M')
+        File.join(Rails.configuration.staff_directory['report_directory'], "library_staff_#{date_str}.csv")
+      end
+    end
+
+    def recent_enough?(filename)
+      File.mtime(filename) > 2.hours.ago
+    end
+  end
+end

config/features.rb

@@ -35,4 +35,8 @@
   feature :meter_files_sent_to_recap,
     default: false,
     description: "Send only the configured number of files to recap"
+
+  feature :air_table_staff_list,
+    default: false,
+    description: "Generate a staff list CSV based on the data in Airtable"
 end

config/routes.rb

@@ -8,6 +8,7 @@
   get '/data_sets/latest/:category', to: 'data_sets#latest', defaults: { format: 'text' }
 
   get '/staff-directory', to: 'staff_directory#index', defaults: { format: 'csv' }
+  get '/pul-staff-report', to: 'staff_directory#pul_staff_report', defaults: { format: 'csv' }
   get '/removed-staff', to: 'staff_directory#removed', defaults: { format: 'text' }
 
   get '/library-events', to: 'library_events#index', defaults: { format: 'csv' }

docs/air_table_staff/README.md

@@ -0,0 +1,28 @@
+# Web Staff
+  This generates the staff report from data in an Airtable.  It
+  creates a CSV that can then be used by the OIT WDS-hosted
+  library website.
+
+## Sequence of events
+
+The steps to maintain this list, as illustrated in the sequence diagram below.
+1. Library Administration staff make updates in the Airtable UI.
+1. Lib jobs requests a staff list from the Airtable API
+1. The Airtable API returns a paginated JSON response
+1. Lib-jobs publishes a CSV of the Airtable data
+1. Drupal downloads the CSV
+1. Drupal displays the staff list
+
+```mermaid
+---
+title: Airtable-based Staff Report Generation by lib-jobs
+---
+sequenceDiagram
+    actor Library Administration
+    Library Administration->>Airtable UI: make updates
+    Lib jobs->>Airtable API: request staff list
+    Airtable API-->+Lib jobs: return paginated json
+    Lib jobs->>Lib jobs: publish a CSV of the Airtable data
+    Drupal->>Lib jobs: download the CSV
+    Drupal->>Drupal: display the staff list
+```

spec/models/air_table_staff/staff_list_job_spec.rb

@@ -0,0 +1,101 @@
+# frozen_string_literal: true
+
+require 'rails_helper'
+
+RSpec.describe AirTableStaff::StaffListJob, type: :model do
+  before do
+    stub_request(:get, 'https://api.airtable.com/v0/appv7XA5FWS7DG9oe/Synchronized%20Staff%20Directory%20View?view=Grid%20view')
+      .with(
+     headers: {
+       'Authorization' => 'Bearer FAKE_AIRTABLE_TOKEN'
+     }
+   )
+      .to_return(status: 200, body: File.read(file_fixture('air_table/records_no_offset.json')), headers: {})
+  end
+
+  context 'job is turned off' do
+    before do
+      allow(Flipflop).to receive(:air_table_staff_list?).and_return(true)
+    end
+    describe('CSV file generation') do
+      let(:file_path) { Pathname.new(Rails.root.join('tmp', "airtable_staff.csv")) }
+      let(:first_row) do
+        [
+          '123', 'ab123', '(123) 123-1234', 'Phillip Librarian', 'Librarian', 'Phillip', '[email protected]', '123 Stokes', 'Stokes', 'Stokes', nil, nil, nil, 'Virtual Reality'
+        ]
+      end
+
+      around do |example|
+        File.delete(file_path) if File.exist?(file_path)
+        example.run
+        File.delete(file_path) if File.exist?(file_path)
+      end
+
+      it 'creates a CSV file' do
+        job = described_class.new(filename: file_path)
+        job.run
+        expect(File.exist?(file_path)).to be true
+      end
+
+      it 'the CSV file has a header row and a data row' do
+        job = described_class.new(filename: file_path)
+        job.run
+        expect(CSV.read(file_path).length).to eq(2)
+      end
+
+      it 'the first row of data is correct' do
+        job = described_class.new(filename: file_path)
+        job.run
+        expect(CSV.read(file_path).second).to eq(first_row)
+      end
+
+      context 'when run at a particular time' do
+        let(:run_time) { Time.zone.local(2022, 3, 14, 15, 9, 26) }
+        before do
+          allow(Time).to receive(:now).and_return(run_time)
+        end
+        it 'records that time in the database' do
+          job = described_class.new(filename: file_path)
+          job.run
+          expect(DataSet.order('created_at').last.report_time).to eq(run_time)
+        end
+      end
+
+      context 'when the process has already been run in the past hour' do
+        let(:original_data_set) do
+          FactoryBot.create(:data_set,
+                                                    category: 'AirTableStaffDirectory',
+                                                    data_file: 'library_staff_20221109.csv')
+        end
+        before do
+          original_data_set.save
+          allow(File).to receive(:exist?).and_return true
+          allow(File).to receive(:mtime).and_return 1.minute.ago
+        end
+
+        it 'returns the existing dataset' do
+          job = described_class.new(filename: file_path)
+          expect { job.run }.not_to change { DataSet.count }
+        end
+      end
+    end
+  end
+
+  context 'job is turned off' do
+    before do
+      allow(Flipflop).to receive(:air_table_staff_list?).and_return(false)
+    end
+    it 'logs that it is turned off' do
+      stafflist_job = described_class.new(filename: 'should_not_exist.csv')
+      stafflist_job.run
+      data_set = DataSet.last
+      expect(data_set.data).to eq('Airtable-based staff list is typically scheduled for this time, but it is turned off.  Go to /features to turn it back on.')
+    end
+    it 'does not create a CSV file' do
+      stafflist_job = described_class.new(filename: 'should_not_exist.csv')
+      stafflist_job.run
+
+      expect(File.exist?('should_not_exist.csv')).to be false
+    end
+  end
+end