Saif - gatekeeper exercise initial commit

jaffle_shop/models/final/final/finance/_models.yml

@@ -0,0 +1,23 @@
+version: 2
+
+exposures:
+  - name: fnl_finance
+    description: Used in Inksacio app
+    type: application
+    url: https://inksacio.eks.octopus.engineering/my_certification_dashboard/
+    owner:
+      email: saif.widyatmoko@octoenergy.com
+    depends_on:
+      - ref('wh_customers')
+
+models:
+  - name: fnl_finance
+    description: This table contains the total value of orders per customer
+    columns:
+      - name: customer_id
+        description: This is a unique identifier for a customer
+        tests:
+          - unique
+          - not_null
+      - name: customer_lifetime_value
+        description: This is the total value of orders for a customer

jaffle_shop/models/final/final/finance/fnl_finance.sql

@@ -0,0 +1,4 @@
+SELECT 
+  customers.customer_id
+  , customers.customer_lifetime_value
+FROM {{ ref('wh_customers') }} AS customers

jaffle_shop/models/final/final/sales/_models.yml

@@ -0,0 +1,23 @@
+version: 2
+
+exposures:
+  - name: fnl_sales
+    description: Used in Inksacio app
+    type: application
+    url: https://inksacio.eks.octopus.engineering/my_certification_dashboard/
+    owner:
+      email: saif.widyatmoko@octoenergy.com
+    depends_on:
+      - ref('wh_customers')
+
+models:
+  - name: fnl_sales
+    description: This table contains the number of new customers per month
+    columns:
+      - name: month_start_date
+        description: This is a date representing the start of a month
+        tests:
+          - unique
+          - not_null
+      - name: new_customer_count
+        description: This is the number of new customers in a given month

jaffle_shop/models/final/final/sales/fnl_sales.sql

@@ -0,0 +1,5 @@
+SELECT 
+  DATE_TRUNC('MONTH', customers.first_order) AS month_start_date
+  , COUNT(customers.customer_id) AS new_customer_count
+FROM {{ ref('wh_customers') }} AS customers
+GROUP BY DATE_TRUNC('MONTH', customers.first_order)

jaffle_shop/models/staging/_models.yml

@@ -0,0 +1,67 @@
+version: 2
+
+models:
+  - name: stg_customers 
+    meta:
+      owner: 'saif.widyatmoko@octoenergy.com'
+    description: |
+      This model contains customer data with hashed PII fields for privacy protection.
+      It is used to ensure that sensitive information is not exposed in the data warehouse.
+    columns:
+      - name: customer_id
+        description: Primary key of the table
+        tests:
+          - unique
+          - not_null
+      - name: first_name_hashed
+        description: Hash of the first name for PII protection
+      - name: last_name_hashed
+        description: Hash of the last name for PII protection
+  - name: stg_customers_pii
+    meta:
+      owner: 'saif.widyatmoko@octoenergy.com
+    description: |
+      This model contains customer data with sensitive fields that are not hashed.
+      It is used for internal processes where PII is required, but access is restricted.
+    columns:
+      - name: customer_id
+        description: Primary key of the table
+        tests:
+          - unique
+          - not_null
+      - name: first_name
+        description: First name of the customer
+        meta:
+          sensitive: true
+      - name: last_name
+        description: Last name of the customer
+        meta:
+          sensitive: true
+  - name: stg_payments
+    meta: 
+      owner: 'saif.widyatmoko@octoenergy.com'
+    description: |
+      This model contains payment data with methods and amounts.
+    columns:
+      - name: payment_id
+        tests:
+          - unique
+          - not_null
+      - name: payment_method
+        tests:
+          - accepted_values:
+              values: ['credit_card', 'coupon', 'bank_transfer', 'gift_card']
+  - name: stg_orders
+    meta:
+      owner: 'saif.widyatmoko@octoenergy.com'
+    description: |
+      This model contains order data with customer and payment information.
+    columns:
+      - name: order_id
+        tests:
+          - unique
+          - not_null
+      - name: status
+        tests:
+          - accepted_values:
+              values: ['placed', 'shipped', 'completed', 'return_pending', 'returned']

jaffle_shop/models/staging/stg_customers.sql

@@ -1,22 +1,3 @@
-with source as (
-
-    {#-
-    Normally we would select from the table here, but we are using seeds to load
-    our data in this project
-    #}
-    select * from {{ ref('raw_customers') }}
-
-),
-
-renamed as (
-
-    select
-        id as customer_id,
-        first_name,
-        last_name
-
-    from source
-
-)
-
-select * from renamed
+SELECT 
+    {{ hash_sensitive_columns('stg_customers_pii')}}
+FROM {{ ref('stg_customers_pii')}}

jaffle_shop/models/staging/stg_customers_pii.sql

@@ -0,0 +1,12 @@
+WITH source AS (
+
+    SELECT * FROM {{ ref('raw_customers') }}
+
+)
+
+SELECT
+    id AS customer_id
+    , first_name
+    , last_name
+
+FROM source

jaffle_shop/models/staging/stg_orders.sql

@@ -1,23 +1,13 @@
-with source as (
+WITH source AS (
 
-    {#-
-    Normally we would select from the table here, but we are using seeds to load
-    our data in this project
-    #}
-    select * from {{ ref('raw_orders') }}
-
-),
-
-renamed as (
-
-    select
-        id as order_id,
-        user_id as customer_id,
-        order_date,
-        status
-
-    from source
+    SELECT * FROM {{ ref('raw_orders') }}
 
 )
 
-select * from renamed
+SELECT
+    id AS order_id
+    , user_id AS customer_id
+    , order_date
+    , status
+
+FROM source