Merge pull request #95 from openedx/cag/video-performance-v2

feat: move video engagement to an mv

models/video/fact_video_engagement.sql

@@ -1,112 +1,49 @@
 with
-    subsection_counts as (
+    subsection_engagement as (
         select
             org,
             course_key,
-            course_run,
-            section_with_name,
-            subsection_with_name,
-            actor_id,
-            item_count,
-            count(distinct video_id) as videos_viewed,
-            case
-                when videos_viewed = 0
-                then 'No videos viewed yet'
-                when videos_viewed = item_count
-                then 'All videos viewed'
-                else 'At least one video viewed'
-            end as engagement_level,
-            username,
-            name,
-            email
-        from {{ ref("fact_video_engagement_per_subsection") }}
-        group by
-            org,
-            course_key,
-            course_run,
-            section_with_name,
-            subsection_with_name,
+            'subsection' as content_level,
             actor_id,
-            item_count,
-            username,
-            name,
-            email
+            subsection_block_id as block_id,
+            engagement_level as section_subsection_video_engagement
+        from {{ ref("subsection_video_engagement") }}
     ),
-    section_counts as (
+    section_engagement as (
         select
             org,
             course_key,
-            course_run,
-            section_with_name,
-            '' as subsection_with_name,
-            actor_id,
-            sum(item_count) as item_count,
-            sum(videos_viewed) as videos_viewed,
-            case
-                when videos_viewed = 0
-                then 'No videos viewed yet'
-                when videos_viewed = item_count
-                then 'All videos viewed'
-                else 'At least one video viewed'
-            end as engagement_level,
-            username,
-            name,
-            email
-        from subsection_counts
-        group by
-            org,
-            course_key,
-            course_run,
-            section_with_name,
-            subsection_with_name,
+            'section' as content_level,
             actor_id,
-            username,
-            name,
-            email
+            section_block_id as block_id,
+            engagement_level as section_subsection_video_engagement
+        from {{ ref("section_video_engagement") }}
     ),
-    all_counts as (
-
-        select
-            org,
-            course_key,
-            course_run,
-            section_with_name as section_with_name,
-            subsection_with_name as subsection_with_name,
-            subsection_with_name as section_subsection_name,
-            'subsection' as content_level,
-            actor_id as actor_id,
-            engagement_level as section_subsection_video_engagement,
-            username,
-            name,
-            email
-        from subsection_counts
+    video_engagement as (
+        select *
+        from subsection_engagement
         union all
-        select
-            org,
-            course_key,
-            course_run,
-            section_with_name as section_with_name,
-            subsection_with_name as subsection_with_name,
-            section_with_name as section_subsection_name,
-            'section' as content_level,
-            actor_id as actor_id,
-            engagement_level as section_subsection_video_engagement,
-            username,
-            name,
-            email
-        from section_counts
+        select *
+        from section_engagement
     )
 select
-    ac.org as org,
-    ac.course_key as course_key,
-    ac.course_run as course_run,
-    ac.section_with_name as section_with_name,
-    ac.subsection_with_name as subsection_with_name,
-    ac.section_subsection_name as section_subsection_name,
-    ac.content_level as content_level,
-    ac.actor_id as actor_id,
-    ac.section_subsection_video_engagement as section_subsection_video_engagement,
-    ac.username as username,
-    ac.name as name,
-    ac.email as email
-from all_counts ac
+    ve.org as org,
+    ve.course_key as course_key,
+    course_blocks.course_run as course_run,
+    course_blocks.display_name_with_location as section_subsection_name,
+    ve.content_level as content_level,
+    ve.actor_id as actor_id,
+    ve.section_subsection_video_engagement as section_subsection_video_engagement,
+    users.username as username,
+    users.name as name,
+    users.email as email
+from video_engagement ve
+join
+    {{ ref("dim_course_blocks") }} course_blocks
+    on (
+        ve.org = course_blocks.org
+        and ve.course_key = course_blocks.course_key
+        and ve.block_id = course_blocks.block_id
+    )
+left outer join
+    {{ ref("dim_user_pii") }} users on toUUID(ve.actor_id) = users.external_user_id

models/video/schema.yml

@@ -201,52 +201,6 @@ models:
         data_type: string
         description: "The unique identifier for the section block"
 
-  - name: fact_video_engagement_per_subsection
-    description: "A dataset with one record representing a video viewed by a learner and the section and subsection that video belongs to"
-    columns:
-      - name: viewed_on
-        data_type: date
-        description: "The date on which the video was viewed"
-      - name: org
-        data_type: string
-        description: "The organization that the video belongs to"
-      - name: course_key
-        data_type: string
-        description: "The course key for the course"
-      - name: course_run
-        data_type: String
-        description: "The course run for the course"
-      - name: section_with_name
-        data_type: string
-        description: "The name of the section this subsection belongs to, with section_number prepended"
-      - name: subsection_with_name
-        data_type: string
-        description: "The name of the subsection, with section_number prepended"
-      - name: item_count
-        data_type: uint64
-        description: "The number of videos in this subsection"
-      - name: actor_id
-        data_type: string
-        description: "The xAPI actor identifier"
-      - name: video_id
-        data_type: string
-        description: "The xAPI object identifier"
-      - name: graded
-        data_type: Boolean
-        description: "Whether the block is graded"
-      - name: username
-        data_type: String
-        description: "The username of the learner"
-      - name: name
-        data_type: String
-        description: "The full name of the learner"
-      - name: email
-        data_type: String
-        description: "The email address of the learner"
-      - name: course_order
-        data_type: Int32
-        description: "The sort order of this block in the course across all course blocks"
-
   - name: fact_video_engagement
     description: "A dataset with one record representing the engagement of a learner with a video"
     columns:
@@ -353,3 +307,90 @@ models:
       - name: email
         data_type: string
         description: "The email address of the learner"
+
+  - name: section_video_engagement
+    description: "A record per course per section per learner with their engagement level"
+    columns:
+      - name: org
+        data_type: string
+        description: "The organization that the course belongs to"
+      - name: course_key
+        data_type: string
+        description: "The course key for the course"
+      - name: actor_id
+        data_type: string
+        description: "The xAPI actor identifier"
+      - name: section_block_id
+        data_type: string
+        description: "The unique identifier for the section block"
+      - name: engagement_level
+        data_type: string
+        description: "The engagement level of the learner with the section"
+        tests:
+          - accepted_values:
+              values:
+                - 'No videos viewed yet'
+                - 'All videos viewed'
+                - 'At least one video viewed'
+
+  - name: subsection_video_engagement
+    description: "A record per course per subsection per learner with their engagement level"
+    columns:
+      - name: org
+        data_type: string
+        description: "The organization that the course belongs to"
+      - name: course_key
+        data_type: string
+        description: "The course key for the course"
+      - name: actor_id
+        data_type: string
+        description: "The xAPI actor identifier"
+      - name: subsection_block_id
+        data_type: string
+        description: "The unique identifier for the subsection block"
+      - name: engagement_level
+        data_type: string
+        description: "The engagement level of the learner with the subsection"
+        tests:
+          - accepted_values:
+              values:
+                - 'No videos viewed yet'
+                - 'All videos viewed'
+                - 'At least one video viewed'
+
+  - name: watched_video_segments
+    description: "A dataset with one record per learner per video segment watched in a course"
+    columns:
+      - name: org
+        data_type: string
+        description: "The organization that the course belongs to"
+      - name: course_key
+        data_type: string
+        description: "The course key for the course"
+      - name: actor_id
+        data_type: string
+        description: "The xAPI actor identifier"
+      - name: video_id
+        data_type: string
+        description: "The xAPI object identifier"
+      - name: start_position
+        data_type: int64
+        description: "The start position of the segment"
+      - name: end_position
+        data_type: int64
+        description: "The end position of the segment"
+      - name: started_at
+        data_type: datetime
+        description: "The time the video segment was started"
+      - name: ended_at
+        data_type: datetime
+        description: "The time the video segment was ended"
+      - name: emission_time
+        data_type: datetime
+        description: "The time the event was emitted"
+      - name: end_type
+        data_type: string
+        description: "The type of event that ended the video segment"
+      - name: video_duration
+        data_type: int64
+        description: "The duration of the video in seconds"