fix: Alter destination table along with mv on schema changes

CHANGELOG.md

@@ -1,3 +1,8 @@
+### Release [1.9.6]
+
+#### Bugs
+* Alter destination table along with mv using `on_schema_changes` ([#534](https://github.com/ClickHouse/dbt-clickhouse/pull/534))
+
 ### Release [1.9.5], 2025-10-20
 
 #### Bugs
@@ -18,7 +23,6 @@
 * All documentation is now hosted on [clickhouse.com/docs](https://clickhouse.com/docs/integrations/dbt). The README in this repository has been updated to reflect this change and now includes a quick start guide and links to the full documentation. ([#526](https://github.com/ClickHouse/dbt-clickhouse/pull/526)).
 * List of supported/tested ClickHouse versions has been updated to include only [actively supported versions](https://github.com/ClickHouse/ClickHouse/blob/master/SECURITY.md) ([#517](https://github.com/ClickHouse/dbt-clickhouse/pull/517)).
 
-
 ### Release [1.9.3], 2025-09-08
 
 #### Bugs

dbt/include/clickhouse/macros/materializations/materialized_view.sql

@@ -86,6 +86,15 @@
         select 1
       {%- endcall %}
 
+       {%- set on_schema_change = incremental_validate_on_schema_change(config.get('on_schema_change'), default='ignore') -%}
+       {{ log('on_schema_change strategy for destination table of  MV: ' + on_schema_change, info=True) }}
+       {%- if on_schema_change != 'ignore' -%}
+        {%- set column_changes = adapter.check_incremental_schema_changes(on_schema_change, existing_relation, sql) -%}
+        {% if column_changes %}
+          {% do clickhouse__apply_column_changes(column_changes, existing_relation) %}
+          {% set existing_relation = load_cached_relation(this) %}
+        {% endif %}
+      {%- endif %}
       -- try to alter view first to replace sql, else drop and create
       {{ clickhouse__update_mvs(target_relation, cluster_clause, refreshable_clause, views) }}
 

tests/integration/adapter/materialized_view/test_materialized_view.py

@@ -27,6 +27,7 @@
        materialized='materialized_view' if var('run_type', '') != 'view_conversion' else 'view',
        engine='MergeTree()',
        order_by='(id)',
+       on_schema_change='sync_all_columns',
        schema='catchup' if var('run_type', '') == 'catchup' else 'custom_schema',
         **({'catchup': False} if var('run_type', '') == 'catchup' else {})
 ) }}
@@ -53,7 +54,8 @@
         when name like 'Dade' and age != 11 then 'crash override'
         when name like 'Kate' then 'acid burn'
         else 'N/A'
-    end as hacker_alias
+    end as hacker_alias,
+    id as id2
 from {{ source('raw', 'people') }}
 where department = 'engineering'
 
@@ -245,6 +247,19 @@ def test_update_incremental(self, project):
         )
         assert len(result) == 2
 
+        # assert that the destination table is updated with the new column
+        table_description_after_update = project.run_sql(
+            f"DESCRIBE TABLE {schema}.hackers", fetch="all"
+        )
+        assert any(col[0] == "id2" and col[1] == "Int32" for col in table_description_after_update)
+
+        # run again without extended schema, to make sure table is updated back without the id2 column
+        run_dbt()
+        table_description_after_revert_update = project.run_sql(
+            f"DESCRIBE TABLE {schema}.hackers", fetch="all"
+        )
+        assert not any(col[0] == "id2" for col in table_description_after_revert_update)
+
     def test_update_full_refresh(self, project):
         schema = quote_identifier(project.test_schema + "_custom_schema")
         # create our initial materialized view

tests/integration/adapter/materialized_view/test_multiple_materialized_views.py

@@ -28,6 +28,7 @@
        materialized='materialized_view',
        engine='MergeTree()',
        order_by='(id)',
+       on_schema_change='sync_all_columns',
        schema='custom_schema_for_multiple_mv',
 ) }}
 
@@ -70,7 +71,8 @@
         when name like 'Dade' and age != 11 then 'crash override'
         when name like 'Kate' then 'acid burn'
         else 'N/A'
-    end as hacker_alias
+    end as hacker_alias,
+    id as id2 
 from {{ source('raw', 'people') }}
 where department = 'engineering'
 --mv1:end
@@ -82,7 +84,8 @@
     id,
     name,
     -- sales people are not cool enough to have a hacker alias
-    'N/A' as hacker_alias
+    'N/A' as hacker_alias,
+    id as id2
 from {{ source('raw', 'people') }}
 where department = 'sales'
 --mv2:end
@@ -219,6 +222,17 @@ def test_update_incremental(self, project):
         assert result[0][0] == "crash_override"
         assert result[1][0] == "zero cool"
 
+        # assert that the destination table is updated with the new column
+        table_description_after_update = project.run_sql(f"DESCRIBE {schema}.hackers", fetch="all")
+        assert any(col[0] == "id2" and col[1] == "Int32" for col in table_description_after_update)
+
+        # run again without extended schema, to make sure table is updated back without the id2 column
+        run_dbt()
+        table_description_after_revert_update = project.run_sql(
+            f"DESCRIBE TABLE {schema}.hackers", fetch="all"
+        )
+        assert not any(col[0] == "id2" for col in table_description_after_revert_update)
+
     def test_update_full_refresh(self, project):
         schema = quote_identifier(project.test_schema + "_custom_schema_for_multiple_mv")
         # create our initial materialized view