From f7d1bbf16ae7d591ac4b7d14725c1798f087cae7 Mon Sep 17 00:00:00 2001
From: Kenneth Lee <kenfodder@gmail.com>
Date: Tue, 10 Dec 2024 17:03:08 +0000
Subject: [PATCH] CAPT-2040 normalize names for payroll csv export

---
 app/models/payroll/name_normalizer.rb       | 39 ++++++++++++++
 app/models/payroll/payment_csv_row.rb       | 12 +++++
 spec/models/payroll/name_normalizer_spec.rb | 57 +++++++++++++++++++++
 3 files changed, 108 insertions(+)
 create mode 100644 app/models/payroll/name_normalizer.rb
 create mode 100644 spec/models/payroll/name_normalizer_spec.rb

diff --git a/app/models/payroll/name_normalizer.rb b/app/models/payroll/name_normalizer.rb
new file mode 100644
index 0000000000..1498179dad
--- /dev/null
+++ b/app/models/payroll/name_normalizer.rb
@@ -0,0 +1,39 @@
+module Payroll
+  class NameNormalizer
+    def self.normalize(name)
+      new(name).normalize
+    end
+
+    def initialize(name)
+      @name = name
+    end
+
+    def normalize
+      return nil if @name.nil?
+
+      @name
+        .then { |n| remove_typical_disallowed_chars(n) }
+        .then { |n| transliterate(n) }
+        .then { |n| final_cleanup(n) }
+    end
+
+    private
+
+    # Things we allow from `NameFormatValidator` and typically found in names
+    # but Payroll provider doesn't accept
+    # Also handles curly apostrophe commonly found
+    def remove_typical_disallowed_chars(name)
+      name.gsub(/[,.;\-'‘’\s]/, "")
+    end
+
+    # Attempt to replace things like `è` with `e`
+    def transliterate(name)
+      I18n.transliterate(name)
+    end
+
+    # Just remove anything missed that aren't allowed by Payroll
+    def final_cleanup(name)
+      name.gsub(/[^A-Za-z]/, "")
+    end
+  end
+end
diff --git a/app/models/payroll/payment_csv_row.rb b/app/models/payroll/payment_csv_row.rb
index cf34456281..52b17c29f6 100644
--- a/app/models/payroll/payment_csv_row.rb
+++ b/app/models/payroll/payment_csv_row.rb
@@ -27,6 +27,18 @@ def to_a
 
     private
 
+    def first_name
+      NameNormalizer.normalize(model.first_name)
+    end
+
+    def middle_name
+      NameNormalizer.normalize(model.middle_name)
+    end
+
+    def surname
+      NameNormalizer.normalize(model.surname)
+    end
+
     def title
       TITLE
     end
diff --git a/spec/models/payroll/name_normalizer_spec.rb b/spec/models/payroll/name_normalizer_spec.rb
new file mode 100644
index 0000000000..5d87f8c87d
--- /dev/null
+++ b/spec/models/payroll/name_normalizer_spec.rb
@@ -0,0 +1,57 @@
+require "rails_helper"
+
+RSpec.describe Payroll::NameNormalizer do
+  describe ".normalize" do
+    subject { described_class.normalize(name) }
+
+    context "nil" do
+      let(:name) { nil }
+      it { is_expected.to be_nil }
+    end
+
+    context "empty string" do
+      let(:name) { "" }
+      it { is_expected.to eq "" }
+    end
+
+    context "blank string" do
+      let(:name) { "   " }
+      it { is_expected.to eq "" }
+    end
+
+    context "name with nothing to change" do
+      let(:name) { "John" }
+      it { is_expected.to eq "John" }
+    end
+
+    context "name curly quotes" do
+      let(:name) { "O’Something" }
+      it { is_expected.to eq "OSomething" }
+    end
+
+    context "name accents and spaces" do
+      let(:name) { "Óscar Hernández" }
+      it { is_expected.to eq "OscarHernandez" }
+    end
+
+    context "name with multiple spaces" do
+      let(:name) { "Chan Chiu Bruce" }
+      it { is_expected.to eq "ChanChiuBruce" }
+    end
+
+    context "name with emojis spaces" do
+      let(:name) { "Thumbs 👍 Up " }
+      it { is_expected.to eq "ThumbsUp" }
+    end
+
+    context "name with semi-colon" do
+      let(:name) { "Samuel;" }
+      it { is_expected.to eq "Samuel" }
+    end
+
+    context "name that has it all" do
+      let(:name) { "Jámes', Ryan, O’Hughes 👍" }
+      it { is_expected.to eq "JamesRyanOHughes" }
+    end
+  end
+end