Skip to content

Commit

Permalink
ADAP-972: Fix issue where materialized views were being mapped as tab…
Browse files Browse the repository at this point in the history
…les in catalog queries (#996)

* changelog
* add test demonstrating issue
* update catalog query to correctly identify materialized views

(cherry picked from commit f6e26d0)
  • Loading branch information
mikealfare authored and github-actions[bot] committed Oct 31, 2023
1 parent 381b859 commit b0935e6
Show file tree
Hide file tree
Showing 4 changed files with 109 additions and 14 deletions.
6 changes: 6 additions & 0 deletions .changes/unreleased/Fixes-20231030-222134.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
kind: Fixes
body: Assign the correct relation type to materialized views in catalog queries
time: 2023-10-30T22:21:34.401675-04:00
custom:
Author: mikealfare
Issue: "995"
40 changes: 26 additions & 14 deletions dbt/include/bigquery/macros/catalog.sql
Original file line number Diff line number Diff line change
Expand Up @@ -7,30 +7,42 @@
{%- else -%}

{%- set query -%}
with tables as (
with materialized_views as (
select
project_id as table_database,
dataset_id as table_schema,
table_id as original_table_name,
table_catalog as project_id,
table_schema as dataset_id,
table_name as table_id
from {{ information_schema.replace(information_schema_view='MATERIALIZED_VIEWS') }}
),
tables as (
select
tables.project_id as table_database,
tables.dataset_id as table_schema,
tables.table_id as original_table_name,

concat(project_id, '.', dataset_id, '.', table_id) as relation_id,
concat(tables.project_id, '.', tables.dataset_id, '.', tables.table_id) as relation_id,

row_count,
size_bytes as size_bytes,
tables.row_count,
tables.size_bytes as size_bytes,
case
when type = 1 then 'table'
when type = 2 then 'view'
when materialized_views.table_id is not null then 'materialized view'
when tables.type = 1 then 'table'
when tables.type = 2 then 'view'
else 'external'
end as table_type,

REGEXP_CONTAINS(table_id, '^.+[0-9]{8}$') and coalesce(type, 0) = 1 as is_date_shard,
REGEXP_EXTRACT(table_id, '^(.+)[0-9]{8}$') as shard_base_name,
REGEXP_EXTRACT(table_id, '^.+([0-9]{8})$') as shard_name
REGEXP_CONTAINS(tables.table_id, '^.+[0-9]{8}$') and coalesce(type, 0) = 1 as is_date_shard,
REGEXP_EXTRACT(tables.table_id, '^(.+)[0-9]{8}$') as shard_base_name,
REGEXP_EXTRACT(tables.table_id, '^.+([0-9]{8})$') as shard_name

from {{ information_schema.replace(information_schema_view='__TABLES__') }}
from {{ information_schema.replace(information_schema_view='__TABLES__') }} tables
left join materialized_views
on materialized_views.project_id = tables.project_id
and materialized_views.dataset_id = tables.dataset_id
and materialized_views.table_id = tables.table_id
where (
{%- for schema in schemas -%}
upper(dataset_id) = upper('{{ schema }}'){%- if not loop.last %} or {% endif -%}
upper(tables.dataset_id) = upper('{{ schema }}'){%- if not loop.last %} or {% endif -%}
{%- endfor -%}
)
),
Expand Down
33 changes: 33 additions & 0 deletions tests/functional/adapter/catalog_tests/files.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,33 @@
MY_SEED = """
id,value,record_valid_date
1,100,2023-01-01 00:00:00
2,200,2023-01-02 00:00:00
3,300,2023-01-02 00:00:00
""".strip()


MY_TABLE = """
{{ config(
materialized='table',
) }}
select *
from {{ ref('my_seed') }}
"""


MY_VIEW = """
{{ config(
materialized='view',
) }}
select *
from {{ ref('my_seed') }}
"""


MY_MATERIALIZED_VIEW = """
{{ config(
materialized='materialized_view',
) }}
select *
from {{ ref('my_table') }}
"""
44 changes: 44 additions & 0 deletions tests/functional/adapter/catalog_tests/test_relation_types.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,44 @@
from dbt.contracts.results import CatalogArtifact
from dbt.tests.util import run_dbt
import pytest

from tests.functional.adapter.catalog_tests import files


class TestCatalogRelationTypes:
@pytest.fixture(scope="class", autouse=True)
def seeds(self):
return {"my_seed.csv": files.MY_SEED}

@pytest.fixture(scope="class", autouse=True)
def models(self):
yield {
"my_table.sql": files.MY_TABLE,
"my_view.sql": files.MY_VIEW,
"my_materialized_view.sql": files.MY_MATERIALIZED_VIEW,
}

@pytest.fixture(scope="class", autouse=True)
def docs(self, project):
run_dbt(["seed"])
run_dbt(["run"])
yield run_dbt(["docs", "generate"])

@pytest.mark.parametrize(
"node_name,relation_type",
[
("seed.test.my_seed", "table"),
("model.test.my_table", "table"),
("model.test.my_view", "view"),
("model.test.my_materialized_view", "materialized view"),
],
)
def test_relation_types_populate_correctly(
self, docs: CatalogArtifact, node_name: str, relation_type: str
):
"""
This test addresses: https://github.com/dbt-labs/dbt-bigquery/issues/995
"""
assert node_name in docs.nodes
node = docs.nodes[node_name]
assert node.metadata.type == relation_type

0 comments on commit b0935e6

Please sign in to comment.