Refactor album and genre models with gold fact review addition (#16)

andrewcrosher · Jan 26, 2025 · 1c6a21e · 1c6a21e
1 parent 49c6b4b
commit 1c6a21e
Show file tree

Hide file tree

Showing 7 changed files with 104 additions and 101 deletions.
diff --git a/models/albums_properties.yml b/models/albums_properties.yml
@@ -52,54 +52,6 @@ models:
         tests:
           - not_null
 
-  - name: silver_dim_album
-    description: a dimension table of all albums reviewed
-    columns:
-      - name: youtube_id
-        description: The PK of the table, the YouTube album ID
-        data_type: string
-        tests:
-          - not_null
-          - unique
-      - name: spotify_id
-        description: The Spotify album ID
-        data_type: string
-      - name: apple_music_id
-        description: The Apple Music album ID
-        data_type: string
-      - name: tidal_id
-        description: The Tidal album ID
-        data_type: string
-      - name: amazon_music_id
-        description: The Amazon Music album ID
-        data_type: string
-      - name: qobuz_id
-        description: The Qobuz album ID
-        data_type: string
-      - name: deezer_id
-        description: The Deezer album ID
-        data_type: string
-      - name: album
-        description: The name of the album
-        data_type: string
-        tests:
-          - not_null
-      - name: artist
-        description: The name of the album artist
-        data_type: string
-        tests:
-          - not_null
-      - name: release_date
-        description: Year that the album was released
-        data_type: string
-        tests:
-          - not_null
-      - name: wikipedia_url
-        description: URL of the Wikipedia article for this album
-        data_type: string
-        tests:
-          - not_null
-
   - name: silver_dim_date
     description: a date dimension, materialised as a table using the dbt_date.get_date_dimension macro
     columns:
@@ -212,47 +164,6 @@ models:
         data_type: date
         quote: true
 
-  - name: silver_dim_genre
-    description: a dimension table of all genres
-    columns:
-      - name: genre
-        description: The genre of the album
-        data_type: string
-      - name: subgenre
-        description: The subgenre of the album
-        data_type: string
-
-  - name: gold_fact_review
-    columns:
-      - name: Album
-        description: The album being rated
-        data_type: string
-        quote: true
-      - name: Artist
-        description: The album artist
-        data_type: string
-        quote: true
-      - name: Release Year
-        description: The year of release
-        data_type: int
-        quote: true
-      - name: My Rating
-        description: The rating I gave
-        data_type: bigint
-        quote: true
-      - name: Global Rating
-        description: The global rating
-        data_type: double
-        quote: true
-      - name: My Relative Rating
-        description: My rating indexed against the global rating
-        data_type: decimal(5,2)
-        quote: true
-      - name: Review Date
-        description: The date of the review
-        data_type: date
-        quote: true
-
 exposures:
   - name: 1001_albums_review_dashboard
     type: dashboard

diff --git a/models/gold_fact_review.sql b/models/gold_fact_review.sql
@@ -2,7 +2,7 @@ select
     da.album as `Album`,
     da.artist as `Artist`,
     cast(da.release_date as int) as `Release Year`,
-    da.genres as `Genres`,
+    dg.genres as `Genres`,
     fr.my_rating as `My Rating`,
     fr.global_rating as `Global Rating`,
     cast(
@@ -15,5 +15,8 @@ from
 inner join
     {{ ref("silver_dim_album") }} as da
     on fr.youtube_id = da.youtube_id
+inner join
+    {{ ref("silver_dim_genre") }} as dg
+    on fr.youtube_id = dg.youtube_id
 order by
     fr.review_date desc
diff --git a/models/gold_fact_review.yml b/models/gold_fact_review.yml
@@ -0,0 +1,36 @@
+version: 2
+models:
+  - name: gold_fact_review
+    columns:
+      - name: Album
+        description: The album being rated
+        data_type: string
+        quote: true
+      - name: Artist
+        description: The album artist
+        data_type: string
+        quote: true
+      - name: Release Year
+        description: The year of release
+        data_type: int
+        quote: true
+      - name: Genres
+        description: The genres of the album
+        data_type: string
+        quote: true
+      - name: My Rating
+        description: The rating I gave
+        data_type: bigint
+        quote: true
+      - name: Global Rating
+        description: The global rating
+        data_type: double
+        quote: true
+      - name: My Relative Rating
+        description: My rating indexed against the global rating
+        data_type: decimal(5,2)
+        quote: true
+      - name: Review Date
+        description: The date of the review
+        data_type: date
+        quote: true
diff --git a/models/silver_dim_abum.yml b/models/silver_dim_abum.yml
@@ -0,0 +1,49 @@
+version: 2
+models:
+  - name: silver_dim_album
+    description: a dimension table of all albums reviewed
+    columns:
+      - name: youtube_id
+        description: The PK of the table, the YouTube album ID
+        data_type: string
+        tests:
+          - not_null
+          - unique
+      - name: spotify_id
+        description: The Spotify album ID
+        data_type: string
+      - name: apple_music_id
+        description: The Apple Music album ID
+        data_type: string
+      - name: tidal_id
+        description: The Tidal album ID
+        data_type: string
+      - name: amazon_music_id
+        description: The Amazon Music album ID
+        data_type: string
+      - name: qobuz_id
+        description: The Qobuz album ID
+        data_type: string
+      - name: deezer_id
+        description: The Deezer album ID
+        data_type: string
+      - name: album
+        description: The name of the album
+        data_type: string
+        tests:
+          - not_null
+      - name: artist
+        description: The name of the album artist
+        data_type: string
+        tests:
+          - not_null
+      - name: release_date
+        description: Year that the album was released
+        data_type: string
+        tests:
+          - not_null
+      - name: wikipedia_url
+        description: URL of the Wikipedia article for this album
+        data_type: string
+        tests:
+          - not_null
diff --git a/models/silver_dim_album.sql b/models/silver_dim_album.sql
@@ -10,7 +10,6 @@ with cte_all as (
         h.album.name as album,
         h.album.artist,
         h.album.releasedate as release_date,
-        h.album.genres,
         h.album.wikipediaurl as wikipedia_url,
         h.generatedat as reviewed_on
     from
@@ -38,7 +37,6 @@ select
     album,
     artist,
     release_date,
-    genres,
     wikipedia_url
 from
     cte_dedup
diff --git a/models/silver_dim_genre.sql b/models/silver_dim_genre.sql
@@ -16,15 +16,7 @@ cte_dedup as (
         cte_all
     order by
         reviewed_on
-),
-
-cte_genres as (
-    select distinct
-        genre,
-        subgenres
-    from
-        cte_dedup lateral view explode(genres) as genre
 )
 
 select *
-from cte_genres
+from cte_dedup
diff --git a/models/silver_dim_genre.yml b/models/silver_dim_genre.yml
@@ -0,0 +1,14 @@
+version: 2
+models:
+  - name: silver_dim_genre
+    description: a dimension table of all genres
+    columns:
+      - name: youtube_id
+        description: The YouTube album ID
+        data_type: string
+      - name: genres
+        description: The genre of the album
+        data_type: string
+      - name: subgenre
+        description: The subgenre of the album
+        data_type: string