Add missing primary key to media_ad_views

CHANGELOG

@@ -1,10 +1,11 @@
 snowplow-media-player 0.7.0 (2023-xx-xx)
 ---------------------------------------
 ## Summary
-This release adds a more robust unique media identifier. This fixes an issue where duplicate `media_id` values could occur in the media stats table as a result of incorrect tracking implementation (e.g. sharing the same media label across different media types).
+This release adds a more robust unique media identifier. This fixes an issue where duplicate `media_id` values could occur in the media stats table as a result of incorrect tracking implementation (e.g. sharing the same media label across different media types). This release also fixes the incremental materialization of the media_ad_views table by adding a unique primary key.
 
-## Features
-Add unique media identifier (close #59)
+## Fixes
+- Add unique media identifier (close #59)
+- Add missing primary key to media_ad_views
 
 ## Under the hood
 

docs/markdown/snowplow_media_player_common_cols.md

@@ -3,15 +3,15 @@ A UUID for each event e.g. `c6ef3124-b53a-4b13-a233-0088f79dcbcb`.
 {% enddocs %}
 
 {% docs col_media_identifier %}
-The surrogate key generated from `media_id`, `media_label`, `media_type` and `media_player_type` to create a unique media element identifier.
+The surrogate key generated from `player_id`, `media_label`, `media_type` and `media_player_type` to create a unique media element identifier.
 {% enddocs %}
 
 {% docs col_player_id %}
 The HTML id attribute of the media content. It is the `player_id` in case of YouTube and `html_id` in case of HTML5.
 {% enddocs %}
 
 {% docs col_play_id %}
-The surrogate key generated from `page_view_id`, `media_id`, `media_label`, `media_type` and `media_player_type` to create a unique play event identifier.
+The surrogate key generated from `page_view_id`, `player_id`, `media_label`, `media_type` and `media_player_type` to create a unique play event identifier.
 {% enddocs %}
 
 {% docs col_page_view_id %}
@@ -301,7 +301,7 @@ The number of pageviews with audio plays of any duration.
 {% enddocs %}
 
 {% docs col_last_base_tstamp %}
-The start_tstamp of the last processed page_view across all media_ids to be used as a lower limit for subsequent incremental runs.
+The start_tstamp of the last processed page_view across all media_identifiers to be used as a lower limit for subsequent incremental runs.
 {% enddocs %}
 
 {% docs col_player_current_time %}

docs/markdown/snowplow_media_player_macro_docs.md

@@ -137,10 +137,10 @@ The query for the player_id column.
 ```sql
 select
     ...,
-    {{ media_id_field(
+    {{ player_id_field(
         youtube_player_id='a.contexts_com_youtube_youtube_1[0]:playerId',
         media_player_id='a.contexts_org_whatwg_media_element_1[0]:htmlId::varchar'
-    ) }} as media_id
+    ) }} as player_id
     from {{ var('snowplow__events') }} as a
 ```
 

models/base/scratch/base_scratch.yml

@@ -44,7 +44,7 @@ models:
         description: '{{ doc("col_domain_userid") }}'
       - name: media_identifier
         description: '{{ doc("col_media_identifier") }}'
-      - name: media_id
+      - name: player_id
         description: '{{ doc("col_player_id") }}'
       - name: media_label
         description: '{{ doc("col_media_label") }}'

models/media_ad_views/media_ad_views.yml

@@ -5,6 +5,13 @@ models:
     +tags: "snowplow_media_player_incremental"
     description: '{{ doc("table_base") }}'
     columns:
+      - name: media_ad_view_id
+        description: The primary key of this table
+        tags:
+          - primary-key
+        tests:
+          - unique
+          - not_null
       - name: media_ad_id
         description: '{{ doc("col_media_ad_id") }}'
         tests:

models/media_ad_views/scratch/base_scratch.yml

@@ -4,6 +4,13 @@ models:
   - name: snowplow_media_player_media_ad_views_this_run
     description: '{{ doc("table_media_ad_views_this_run") }}'
     columns:
+      - name: media_ad_view_id
+        description: The primary key of this table
+        tags:
+          - primary-key
+        tests:
+          - unique
+          - not_null
       - name: media_ad_id
         description: '{{ doc("col_media_ad_id") }}'
         tests:

models/media_ad_views/scratch/snowplow_media_player_media_ad_views_this_run.sql

@@ -70,8 +70,11 @@ events_this_run as (
 
 )
 
-select *
-    {% if target.type in ['databricks', 'spark'] -%}
-      , date(prep.viewed_at) as viewed_at_date
-    {%- endif %}
-  from prep
+select
+  {{ dbt_utils.generate_surrogate_key(['p.play_id', 'p.ad_break_id', 'p.media_ad_id']) }} as media_ad_view_id
+  , p.*
+  {% if target.type in ['databricks', 'spark'] -%}
+    , date(p.viewed_at) as viewed_at_date
+  {%- endif %}
+
+from prep as p

models/media_ad_views/snowplow_media_player_media_ad_views.sql

@@ -8,6 +8,7 @@ You may obtain a copy of the Snowplow Personal and Academic License Version 1.0
 {{
   config(
     materialized= "incremental",
+    unique_key= 'media_ad_view_id',
     upsert_date_key='last_event',
     sort = 'last_event',
     dist = 'media_ad_id',