You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I understand this is more of a stretch, but if sqlmesh format could format the sql included in tests it would be a killer feature.
For example you can define test data like this:
test_attributed_queries:
model: staging.attributed_queries
inputs:
source.query_history:
# Use sql entry, so we can parse_json the query_tags correctly. Have to put all columns in the right order.
query: |
-- no query tag
SELECT
'123' as query_id,
null as query_type,
'svc_bob' as user_name,
null as role_name,
null as warehouse_id,
'common' as warehouse_name,
'X-Small' as warehouse_size,
'STANDARD' as warehouse_type,
null as cluster_number,
null as query_tag,
cast('2024-06-01 01:05:00 +0000' as timestamptz) as start_time,
null as total_elapsed_time,
null as compilation_time,
5 as execution_time,
null as credits_used_cloud_services
It should be possible for the formatter to extract the sql from the test, format it, and put it back.
The text was updated successfully, but these errors were encountered:
I understand this is more of a stretch, but if sqlmesh format could format the sql included in tests it would be a killer feature.
For example you can define test data like this:
It should be possible for the formatter to extract the sql from the test, format it, and put it back.
The text was updated successfully, but these errors were encountered: