Prevent duplicates in web vital measurements (Close #215)

snowplow · Jul 27, 2023 · f7b00c6 · f7b00c6
1 parent 7b98224
commit f7b00c6
Show file tree

Hide file tree

Showing 2 changed files with 7 additions and 7 deletions.
diff --git a/models/optional_modules/core_web_vitals/default/snowplow_web_vital_measurements.sql b/models/optional_modules/core_web_vitals/default/snowplow_web_vital_measurements.sql
@@ -14,9 +14,9 @@
 with prep as (
 
   select
-    page_url,
-    device_class,
-    geo_country,
+    coalesce(page_url, 'unknown_page_url') as page_url,
+    coalesce(device_class, 'unknown_device_class') as device_class,
+    coalesce(geo_country, 'unknown_geo_country') as geo_country,
     concat(cast({{ date_trunc('day', 'derived_tstamp') }} as {{ type_string() }}),'.000') as time_period,
     lcp,
     fid,

diff --git a/models/optional_modules/core_web_vitals/snowflake/snowplow_web_vital_measurements.sql b/models/optional_modules/core_web_vitals/snowflake/snowplow_web_vital_measurements.sql
@@ -9,9 +9,9 @@
 with measurements as (
 
   select
-    page_url,
-    device_class,
-    geo_country,
+    coalesce(page_url, 'unknown_page_url') as page_url,
+    coalesce(device_class, 'unknown_device_class') as device_class,
+    coalesce(geo_country, 'unknown_geo_country') as geo_country,
     cast( {{ dbt.date_trunc('day', 'derived_tstamp') }} as {{ dbt.type_string() }}) as time_period,
     count(*) as page_view_count,
     grouping_id(page_url, device_class) as id_url_and_device,
@@ -30,7 +30,7 @@ with measurements as (
 
   where cast(derived_tstamp as date) >= {{ dateadd('day', '-'+var('snowplow__cwv_days_to_measure')|string, date_trunc('day', snowplow_utils.current_timestamp_in_utc())) }}
 
-  group by grouping sets ((), (page_url, device_class), (device_class), (cast( {{ dbt.date_trunc('day', 'derived_tstamp') }} as {{ dbt.type_string() }})), (cast( {{ dbt.date_trunc('day', 'derived_tstamp') }} as {{ dbt.type_string() }}), device_class), (geo_country), (geo_country, device_class))
+  group by grouping sets ((), (ßpage_url, device_class), (device_class), (cast( {{ dbt.date_trunc('day', 'derived_tstamp') }} as {{ dbt.type_string() }})), (cast( {{ dbt.date_trunc('day', 'derived_tstamp') }} as {{ dbt.type_string() }}), device_class), (geo_country), (geo_country, device_class))
 
 )