docs for athena staging

models/staging/athena/schema.yml

@@ -1,2 +1,269 @@
-#TODO Update to the fact that this only has athena stuff now
+version: 2
-#TODO update tests to check that there's only athena stuff
+
+models:
+  - name: stg_athena__verifications
+    description: |
+      Records of each transaction that happens in the Athena API. Records are
+      mutable and can get updated.
+
+      The table originally receives both records from Athena and edeposit APIs,
+      but we force only keeping Athena (V1) records here.
+    columns:
+      - name: id_verification
+        data_type: character varying
+        description: "Unique id for the specific transaction."
+        tests:
+          - unique
+          - not_null
+
+      - name: id_booking
+        data_type: text
+        description: ""
+        tests:
+          - not_null:
+              where: created_at_utc > '2024-04-01T00:00:00Z' and verification_status != 'InsufficientInformation'
+
+      - name: id_user_partner
+        data_type: text
+        description: The unique ID of the partner calling the API.
+        tests:
+          - not_null
+
+      - name: id_accommodation
+        data_type: text
+        description: An ID for the listing related to the booking.
+
+      - name: version
+        data_type: text
+        description: |
+          Indicates whether the verification is for V1 (Athena) or V2 
+          (e-deposit).
+        tests:
+          - not_null
+          - accepted_values:
+              values:
+                - "V1"
+
+      - name: nightly_fee_local
+        data_type: numeric
+        description: |
+          The fee per night to be charged for this verification, in the currency
+          of the user. Note that the Athena/e-deposit user also has a configured
+          nightly fee. It's unclear at this point which one has priority for
+          billing.
+        tests:
+          - dbt_expectations.expect_column_values_to_be_between:
+              min_value: 0
+              max_value: 100
+              strictly: true
+
+      - name: verification_status
+        data_type: text
+        description: |
+          This field shows the outcome of the Verification itself.
+        tests:
+          - not_null
+          - accepted_values:
+              values:
+                - "Approved"
+                - "Flagged"
+                - "Rejected"
+                - "InsufficientInformation"
+
+      - name: verification_status_reason
+        data_type: text
+        description: Not used, ignore.
+
+      - name: email_flag
+        data_type: text
+        description: |
+          Null if the email shows no issues, otherwise it details the problems
+          attached to the given email.
+
+      - name: phone_flag
+        data_type: text
+        description: |
+          Null if the phone number shows no issues, otherwise it details the 
+          problems attached to the given phone number.
+        tests:
+          - accepted_values:
+              values:
+                - "Phone number not reachable"
+                - "Not a real phone number"
+                - "Phone number is disposable"
+
+      - name: watch_list
+        data_type: text
+        description: ""
+        tests:
+          - accepted_values:
+              values:
+                - "Match"
+                - "No Match"
+
+      - name: channel
+        data_type: text
+        description: ""
+
+      - name: checkin_at_utc
+        data_type: timestamp without time zone
+        description: ""
+
+      - name: checkout_at_utc
+        data_type: timestamp without time zone
+        description: ""
+
+      - name: is_cancelled
+        data_type: boolean
+        description: |
+          Indicates if the booking has been cancelled or not. At the source, 
+          null and false values have the same meaning, so here we turn nulls
+          into false to keep thing simple.
+        tests:
+          - not_null
+          - accepted_values:
+              values:
+                - true
+                - false
+
+      - name: cancelled_at_utc
+        data_type: timestamp without time zone
+        description: If the booking was cancelled, when did we learn about it.
+        tests:
+          - not_null:
+              where: is_cancelled = true
+
+      - name: user_email
+        data_type: text
+        description: |
+          The email of the Athena/e-deposit partner user.
+
+      - name: guest_email
+        data_type: text
+        description: ""
+
+      - name: guest_last_name
+        data_type: text
+        description: ""
+
+      - name: guest_first_name
+        data_type: text
+        description: ""
+
+      - name: guest_telephone
+        data_type: text
+        description: ""
+
+      - name: company_name
+        data_type: text
+        description: ""
+
+      - name: property_manager_name
+        data_type: text
+        description: ""
+
+      - name: property_manager_email
+        data_type: text
+        description: ""
+
+      - name: listing_name
+        data_type: text
+        description: ""
+
+      - name: listing_town
+        data_type: text
+        description: ""
+
+      - name: listing_address
+        data_type: text
+        description: ""
+
+      - name: listing_country
+        data_type: text
+        description: ""
+
+      - name: listing_postcode
+        data_type: text
+        description: ""
+
+      - name: pets_allowed
+        data_type: boolean
+        description: ""
+
+      - name: level_of_protection_amount
+        data_type: integer
+        description: Ignore.
+
+      - name: level_of_protection_currency
+        data_type: text
+        description: Ignore.
+
+      - name: attachments
+        data_type: text
+        description: ""
+
+      - name: status_updated_at_utc
+        data_type: timestamp without time zone
+        description: |
+          Timestamp of the last time the record was modified before screening 
+          happened. Only relevant for V1 records.
+        tests:
+          - not_null:
+              where: version = 'V1'
+
+      - name: status_updated_date_utc
+        data_type: timestamp without time zone
+        description: |
+          Date of the last time the record was modified before screening 
+          happened. Only relevant for V1 records.
+        tests:
+          - not_null:
+              where: version = 'V1'
+
+      - name: updated_at_utc
+        data_type: timestamp without time zone
+        description: |
+          Timestamp of the last edit of the record, as set by the 
+          Athena/e-deposit application.
+        tests:
+          - not_null
+
+      - name: updated_date_utc
+        data_type: timestamp without time zone
+        description: |
+          Date of the last edit of the record, as set by the 
+          Athena/e-deposit application.
+        tests:
+          - not_null
+
+      - name: athena_creation_at_utc
+        data_type: timestamp without time zone
+        description: |
+          A client-provided timestamp of when the booking was created. It's an
+          optional field, only relevant for 'V1' records.
+
+      - name: athena_creation_date_utc
+        data_type: timestamp without time zone
+        description: |
+          A client-provided date of when the booking was created. It's an
+          optional field, only relevant for 'V1' records.
+
+      - name: created_at_utc
+        data_type: timestamp without time zone
+        description: |
+          The internal application timestamp of when this record was created.
+        tests:
+          - not_null
+
+      - name: created_date_utc
+        data_type: timestamp without time zone
+        description: |
+          The internal application date of when this record was created.
+        tests:
+          - not_null
+
+      - name: cosmos_db_timestamp_utc
+        data_type: timestamp with time zone
+        description: The internal Cosmos DB timestamp of the last record update.
+        tests:
+          - not_null