Merge pull request #793 from pulibrary/airtable

Basic functionality for airtable staff list

app/models/air_table_staff/csv_builder.rb

@@ -0,0 +1,19 @@
+# frozen_string_literal: true
+
+module AirTableStaff
+  # This class is responsible for creating a CSV out of the
+  # data from Airtable
+  class CSVBuilder
+    def to_csv
+      @csv ||= CSV.generate do |csv|
+        # Add the headers...
+        csv << StaffDirectoryMapping.new.to_a
+
+        # Then add the data
+        AirTableStaff::RecordList.new.to_a.each do |record|
+          csv << record.to_a
+        end
+      end
+    end
+  end
+end

app/models/air_table_staff/json_value_extractor.rb

@@ -0,0 +1,22 @@
+# frozen_string_literal: true
+module AirTableStaff
+  # This class is responsible for extracting a single
+  # value from a json hash, based on the criteria in
+  # the field hash
+  class JsonValueExtractor
+    def initialize(json:, field:)
+      @json = json
+      @field = field
+    end
+
+    def extract
+      raw_value = json[field[:airtable_field]]
+      transformer = field[:transformer]
+      transformer ? transformer.call(raw_value) : raw_value
+    end
+
+    private
+
+    attr_reader :json, :field
+  end
+end

app/models/air_table_staff/record_list.rb

@@ -0,0 +1,51 @@
+# frozen_string_literal: true
+module AirTableStaff
+  # This class is responsible for maintaining a list
+  # of staff records taken from the Airtable API
+  class RecordList
+    def initialize
+      @base_url = "https://api.airtable.com/v0/appv7XA5FWS7DG9oe/Synchronized%20Staff%20Directory%20View?view=Grid%20view"
+      @token = LibJobs.config[:airtable_token]
+    end
+
+    # The library staff list is split into several pages.
+    # For each page (except the last), Airtable gives us an
+    # offset, which is how we request the next page.
+    def to_a(offset: nil)
+      @as_array ||= begin
+        json = get_json(offset:)
+        records = json[:records].map do |row|
+          AirTableStaff::StaffDirectoryPerson.new(row[:fields])
+        end
+        offset = json[:offset]
+
+        # If we have an offset, call this method recursively
+        # (to fetch additional pages of data), until airtable
+        # no longer gives us an offset
+        records += to_a(offset:) if offset
+        records
+      end
+    end
+
+    private
+
+    attr_reader :base_url, :token
+
+    def get_json(offset: nil)
+      JSON.parse(response(offset:).body, symbolize_names: true)
+    end
+
+    def response(offset: nil)
+      url = url_with_optional_offset(offset:)
+      request = Net::HTTP::Get.new(url)
+      request["Authorization"] = "Bearer #{token}"
+      Net::HTTP.start(url.hostname, url.port, { use_ssl: true }) do |http|
+        http.request(request)
+      end
+    end
+
+    def url_with_optional_offset(offset: nil)
+      URI.parse(offset ? "#{base_url}&offset=#{offset}" : base_url)
+    end
+  end
+end

app/models/air_table_staff/staff_directory_mapping.rb

@@ -0,0 +1,36 @@
+# frozen_string_literal: true
+module AirTableStaff
+  # This class is responsible for describing the fields in the
+  # airtable and the resultant CSV, and how they are related
+  class StaffDirectoryMapping
+    # A mapping of:
+    #  - an airtable field (airtable_field)
+    #  - a column name we want for our csv (our_field)
+    #  - an optional "transformer" lambda for converting the airtable values
+    #    to the desired format for the csv
+    # rubocop:disable Metrics/MethodLength
+    def fields
+      [
+        { airtable_field: :'University ID', our_field: :puid },
+        { airtable_field: :netid, our_field: :netid },
+        { airtable_field: :'University Phone', our_field: :phone },
+        { airtable_field: :'pul:Preferred Name', our_field: :name },
+        { airtable_field: :'Last Name', our_field: :lastName },
+        { airtable_field: :'First Name', our_field: :firstName },
+        { airtable_field: :Email, our_field: :email },
+        { airtable_field: :Address, our_field: :address },
+        { airtable_field: :'pul:Building', our_field: :building },
+        { airtable_field: :Division, our_field: :department },
+        { airtable_field: :'pul:Department', our_field: :division },
+        { airtable_field: :'pul:Unit', our_field: :unit },
+        { airtable_field: :'pul:Team', our_field: :team },
+        { airtable_field: :'Area of Study', our_field: :areasOfStudy, transformer: ->(areas) { areas&.join('//') } }
+      ]
+    end
+    # rubocop:enable Metrics/MethodLength
+
+    def to_a
+      @as_array ||= fields.pluck(:our_field)
+    end
+  end
+end

app/models/air_table_staff/staff_directory_person.rb

@@ -0,0 +1,22 @@
+# frozen_string_literal: true
+module AirTableStaff
+  # This class is responsible for extracting information about
+  # a person from the airtable staff directory JSON, according
+  # to the mapping from the StaffDirectoryMapping class.
+  class StaffDirectoryPerson
+    def initialize(json)
+      @json = json
+      @mapping = StaffDirectoryMapping.new
+    end
+
+    def to_a
+      @array_version ||= mapping.fields.map do |field|
+        JsonValueExtractor.new(field:, json:).extract
+      end
+    end
+
+    private
+
+    attr_reader :json, :mapping
+  end
+end

config/config.yml

@@ -28,6 +28,7 @@ defaults: &defaults
   alma_region: <%= ENV["ALMA_REGION"] %>
   alma_api_key: <%= ENV["ALMA_CONFIG_API_KEY"] || "1234" %>
   alma_bib_api_key: <%= ENV["ALMA_BIB_API_KEY"] || "1234" %>
+  airtable_token: <%= ENV["AIRTABLE_TOKEN"] || 'FAKE_AIRTABLE_TOKEN' %>
 development:
   <<: *defaults
 

spec/fixtures/files/air_table/records_no_offset.json

@@ -0,0 +1,21 @@
+{
+    "records": [
+        {
+            "id": "recrhIUhcJw3lUlRE",
+            "createdTime": "2024-01-30T22:16:16.000Z",
+            "fields": {
+                "Address": "123 Stokes",
+                "Area of Study": ["Virtual Reality"],
+                "Division": "Stokes",
+                "pul:Building": "Stokes",
+                "Last Name": "Librarian",
+                "First Name": "Phillip",
+                "University ID": "123",
+                "netid": "ab123",
+                "University Phone": "(123) 123-1234",
+                "pul:Preferred Name": "Phillip Librarian",
+                "Email": "[email protected]"
+            }
+         }
+    ]
+}

spec/fixtures/files/air_table/records_with_offset.json

@@ -0,0 +1,21 @@
+{
+    "records": [
+        {
+            "id": "recrhIUhcJw3lUlRE",
+            "createdTime": "2025-01-30T22:16:16.000Z",
+            "fields": {
+                "Address": "Firestone A floor",
+                "Area of Study": ["Cinema history", "Robots"],
+                "Division": "Special Collections",
+                "pul:Building": "Firestone",
+                "Last Name": "Carmant",
+                "First Name": "Drema",
+                "University ID": "456",
+                "netid": "zz99",
+                "University Phone": "(123) 555-5555",
+                "pul:Preferred Name": "Drema Carmant",
+                "Email": "[email protected]"
+            }
+         }
+    ], "offset": "naeQu2ul/Ash6eiQu"
+}

spec/models/air_table_staff/csv_builder_spec.rb

@@ -0,0 +1,22 @@
+# frozen_string_literal: true
+require 'rails_helper'
+
+RSpec.describe AirTableStaff::CSVBuilder 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
+  it 'creates a CSV object with data from the HTTP API' do
+    expected = <<~END_CSV
+            puid,netid,phone,name,lastName,firstName,email,address,building,department,division,unit,team,areasOfStudy
+            123,ab123,(123) 123-1234,Phillip Librarian,Librarian,Phillip,[email protected],123 Stokes,Stokes,Stokes,,,,Virtual Reality
+        END_CSV
+    directory = described_class.new
+    expect(directory.to_csv).to eq(expected)
+  end
+end

spec/models/air_table_staff/json_value_extractor_spec.rb

@@ -0,0 +1,19 @@
+# frozen_string_literal: true
+require 'rails_helper'
+
+RSpec.describe AirTableStaff::JsonValueExtractor do
+  context 'when there is no transformer lambda provided in field config' do
+    it 'extracts the text verbatim' do
+      json = { cat: 'tabby cat' }
+      field = { airtable_field: :cat }
+      expect(described_class.new(json:, field:).extract).to eq('tabby cat')
+    end
+  end
+  context 'when there is a transformer lambda provided in field config' do
+    it 'extracts the text verbatim' do
+      json = { cat: 'tabby cat' }
+      field = { airtable_field: :cat, transformer: ->(value) { "My favorite cat is #{value}" } }
+      expect(described_class.new(json:, field:).extract).to eq('My favorite cat is tabby cat')
+    end
+  end
+end

spec/models/air_table_staff/record_list_spec.rb

@@ -0,0 +1,74 @@
+# frozen_string_literal: true
+require 'rails_helper'
+
+BASE_AIRTABLE_URL = 'https://api.airtable.com/v0/appv7XA5FWS7DG9oe/Synchronized%20Staff%20Directory%20View?view=Grid%20view'
+
+RSpec.describe AirTableStaff::RecordList do
+  context 'when the airtable response is not paginated' do
+    before do
+      stub_request(:get, BASE_AIRTABLE_URL)
+        .with(
+         headers: {
+           'Authorization' => 'Bearer FAKE_AIRTABLE_TOKEN'
+         }
+       )
+        .to_return(status: 200, body: File.read(file_fixture('air_table/records_no_offset.json')), headers: {})
+    end
+    it 'creates an array with data from a single call to the HTTP API' do
+      list = described_class.new.to_a
+
+      expect(list.length).to eq(1)
+
+      first_person = list[0].to_a
+      expect(first_person[0]).to eq('123') # puid
+      expect(first_person[3]).to eq('Phillip Librarian') # name
+
+      expect(WebMock).to have_requested(:get, BASE_AIRTABLE_URL).once
+    end
+  end
+  context 'when the airtable response is paginated' do
+    before do
+      stub_request(:get, BASE_AIRTABLE_URL)
+        .with(
+         headers: {
+           'Authorization' => 'Bearer FAKE_AIRTABLE_TOKEN'
+         }
+       )
+        .to_return(status: 200, body: File.read(file_fixture('air_table/records_with_offset.json')), headers: {})
+      stub_request(:get, "#{BASE_AIRTABLE_URL}&offset=naeQu2ul/Ash6eiQu")
+        .with(
+         headers: {
+           'Authorization' => 'Bearer FAKE_AIRTABLE_TOKEN'
+         }
+       )
+        .to_return(status: 200, body: File.read(file_fixture('air_table/records_no_offset.json')), headers: {})
+    end
+    it 'creates an array with data from multiple calls to the HTTP API' do
+      list = described_class.new.to_a
+
+      expect(list.length).to eq(2)
+
+      first_person = list[0].to_a
+      expect(first_person[0]).to eq('456') # puid
+      expect(first_person[3]).to eq('Drema Carmant') # name
+
+      second_person = list[1].to_a
+      expect(second_person[0]).to eq('123') # puid
+      expect(second_person[3]).to eq('Phillip Librarian') # name
+
+      expect(WebMock).to have_requested(:get, BASE_AIRTABLE_URL).once
+      expect(WebMock).to have_requested(:get, "#{BASE_AIRTABLE_URL}&offset=naeQu2ul/Ash6eiQu").once
+    end
+    context 'when we run it multiple times' do
+      it 'gives us the same data without additional network calls' do
+        list = described_class.new
+        first_time = list.to_a
+        second_time = list.to_a
+
+        expect(first_time).to eq(second_time)
+        expect(WebMock).to have_requested(:get, BASE_AIRTABLE_URL).once
+        expect(WebMock).to have_requested(:get, "#{BASE_AIRTABLE_URL}&offset=naeQu2ul/Ash6eiQu").once
+      end
+    end
+  end
+end

spec/models/air_table_staff/staff_directory_person_spec.rb

@@ -0,0 +1,43 @@
+# frozen_string_literal: true
+require 'rails_helper'
+
+RSpec.describe AirTableStaff::StaffDirectoryPerson do
+  describe '#to_a' do
+    it 'uses the order from the mapping' do
+      json = {
+        'pul:Preferred Name': 'Sage Archivist',
+        'University ID': '987654321',
+        'netid': 'ab1234',
+        'University Phone': '(609) 555-1234',
+        'Last Name': 'Archivist',
+        'First Name': 'Phoenix',
+        'Email': '[email protected]',
+        'Address': '123 Lewis Library',
+        'pul:Building': 'Stokes Library',
+        'Division': 'ReCAP',
+        'pul:Department': 'Cataloging and Metadata Services',
+        'pul:Unit': 'Rare Books Cataloging Team',
+        'pul:Team': 'IT, Discovery and Access Services',
+        'Area of Study': ['Chemistry', 'African American Studies']
+      }
+      expected = [
+        '987654321', # puid
+        'ab1234', # netid
+        '(609) 555-1234', # phone
+        'Sage Archivist', # name
+        'Archivist', # lastName
+        'Phoenix', # firstName
+        '[email protected]', # email
+        '123 Lewis Library', # address
+        'Stokes Library', # building
+        'ReCAP', # department
+        'Cataloging and Metadata Services', # division
+        'Rare Books Cataloging Team', # unit
+        'IT, Discovery and Access Services', # team
+        'Chemistry//African American Studies' # areasOfStudy
+      ]
+
+      expect(described_class.new(json).to_a).to eq(expected)
+    end
+  end
+end