Hi, team. I am trying to do an integration using appsotre. Related threads and issues are below. • https://airbytehq.slack.com/archives/C021JANJ6TY/p1661245951575079?thread_ts=1661161170.202709&cid=C021JANJ6TY • https://discuss.airbyte.io/t/normalization-failed-during-appstore-dbt-run-for-the-sales-report/2340 The date format in appsotre sales reports is USA (mm/dd/yyyy), not ISO 8601. Therefore, CAST does not work when Destination is set to BigQuery. I think it would be better to output the date type as a string type. Or is it possible to convert using Custom schema type transformation? Are there any other good solutions?

Hi. If you can use cast- use instead format_date or parse_date to make it custom or iso 8601 compliant you can aslo use pendulum to format it

@Eli Sigal Thanks for the reply. Which process are you saying should handle those conversions? Sure, there are FORMAT_DATE and PARSE_DATE functions on BigQuery. Can I control them in the Basic Normalization stage?

you can control them if you write a custom normalization model in dbt or dont use the normalization and just map the json your self using dbt and bigquery instead. or you can just do it in the connector (though if this is an airbyte connector I would create a copy since an update will delete your custom changes)

@Eli Sigal Thanks.

you can control them if you write a custom normalization model in dbt or dont use the normalization and just map the json your self using dbt and bigquery instead.

Sorry, I do not understand the difference between these. Could you give a brief explanation? Also, is there any documentation I should look at when implementing dbt + bq?

You have the option for each connection in airbyte to use or not use a normalization step- if you dont use it - airbyte always by default creates a RAW table with the output as json. you can create a custom dbt model that will parse that json

Hello Soshi Nakachi仲地早司, it's been a while without an update from us. Are you still having problems or did you find a solution?