Hello, when using DBT software-defined-asset, is there a way to trigger a full refresh of my incremental models ?

cc @owen

hi @Timothee Vandeput! when configuring your dbt_cli_resource, you can pass in

{... (other config options) ..., "full-refresh": True}

@Dagster Bot discussion How to trigger a full refresh of dbt incremental models with software-defined assets

Question in the thread has been surfaced to GitHub Discussions for future discoverability: https://github.com/dagster-io/dagster/discussions/8873

Hello Owen, thanks for the help ! It's working when I pass the full-refresh option from the ressource definition in my repository. Example :

"dbt": dbt_cli_resource.configured(

{"project_dir": DBT_PROJECT_DIR, "profiles_dir": DBT_PROFILES_DIR, "full-refresh": True}

),

I would like to do it from the launchpad as well. I tried with :

resources:

dbt:

config:

full-refresh: True

But it's not picking the flag. Any idea on how I should do it from the launchpad ?

ah it's not picking up the flag because by default, once you call .configured(), that's basically the “finalized” config for that resource. the nicest to handle this would be something like https://docs.dagster.io/concepts/configuration/configured#partially-filling-the-configuration

Thanks a lot for the support @owen. For reference, I managed to do it with the following code :

@configured(dbt_cli_resource, config_schema={"full-refresh": bool})

def dbt_configuration(config):

return {"project_dir": DBT_PROJECT_DIR, "profiles_dir": DBT_PROFILES_DIR, "full-refresh": config["full-refresh"]}

And then specifying my new configuration for DBT in my ressource definition :

"dbt": dbt_configuration

I can now use the following configuration in the launchpad :

resources:

dbt:

config:

full-refresh: true

nice! that looks great