From 325082814520ed3bc66f5be4abeff42c9d313485 Mon Sep 17 00:00:00 2001
From: Claire Carroll <carroll.claire.e@gmail.com>
Date: Thu, 23 Apr 2020 11:59:01 -0400
Subject: [PATCH] Handle hard deletes in snapshots

---
 .../macros/materializations/snapshot/snapshot.sql     |  6 +++---
 .../macros/materializations/snapshot/strategies.sql   | 11 +++++++++--
 2 files changed, 12 insertions(+), 5 deletions(-)

diff --git a/core/dbt/include/global_project/macros/materializations/snapshot/snapshot.sql b/core/dbt/include/global_project/macros/materializations/snapshot/snapshot.sql
index a0166931f34..d64b36bae9b 100644
--- a/core/dbt/include/global_project/macros/materializations/snapshot/snapshot.sql
+++ b/core/dbt/include/global_project/macros/materializations/snapshot/snapshot.sql
@@ -110,10 +110,10 @@
         select
             'update' as dbt_change_type,
             snapshotted_data.dbt_scd_id,
-            source_data.dbt_valid_from as dbt_valid_to
+            coalesce(source_data.dbt_valid_from, {{ snapshot_get_time() }}) as dbt_valid_to
 
-        from source_data
-        join snapshotted_data on snapshotted_data.dbt_unique_key = source_data.dbt_unique_key
+        from snapshotted_data
+        left join source_data  on snapshotted_data.dbt_unique_key = source_data.dbt_unique_key
         where snapshotted_data.dbt_valid_to is null
         and (
             {{ strategy.row_changed }}
diff --git a/core/dbt/include/global_project/macros/materializations/snapshot/strategies.sql b/core/dbt/include/global_project/macros/materializations/snapshot/strategies.sql
index 108d3f01709..2ddfbde216d 100644
--- a/core/dbt/include/global_project/macros/materializations/snapshot/strategies.sql
+++ b/core/dbt/include/global_project/macros/materializations/snapshot/strategies.sql
@@ -68,7 +68,10 @@
     {% set updated_at = config['updated_at'] %}
 
     {% set row_changed_expr -%}
-        ({{ snapshotted_rel }}.{{ updated_at }} < {{ current_rel }}.{{ updated_at }})
+        ({{ snapshotted_rel }}.{{ updated_at }} < {{ current_rel }}.{{ updated_at }}
+            -- this indicates that the source record was hard-deleted
+            or {{ current_rel }}.{{ primary_key }} is null
+        )
     {%- endset %}
 
     {% set scd_id_expr = snapshot_hash_arguments([primary_key, updated_at]) %}
@@ -148,12 +151,16 @@
     {%- if column_added -%}
         TRUE
     {%- else -%}
-    {%- for col in check_cols -%}
+    -- this indicates that the source record was hard-deleted
+    {{ current_rel }}.{{ primary_key }} is null) or
+    {% for col in check_cols -%}
         {{ snapshotted_rel }}.{{ col }} != {{ current_rel }}.{{ col }}
         or
         ({{ snapshotted_rel }}.{{ col }} is null) != ({{ current_rel }}.{{ col }} is null)
         {%- if not loop.last %} or {% endif -%}
     {%- endfor -%}
+
+
     {%- endif -%}
     )
     {%- endset %}