Add missing primary key to media_ad_views

snowplow · Nov 13, 2023 · a25ebd6 · a25ebd6
1 parent e706777
commit a25ebd6
Show file tree

Hide file tree

Showing 5 changed files with 27 additions and 8 deletions.
diff --git a/CHANGELOG b/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
 

diff --git a/models/media_ad_views/media_ad_views.yml b/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:

diff --git a/models/media_ad_views/scratch/base_scratch.yml b/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:

diff --git a/models/media_ad_views/scratch/snowplow_media_player_media_ad_views_this_run.sql b/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
diff --git a/models/media_ad_views/snowplow_media_player_media_ad_views.sql b/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',