[CT-2904] [Bug] Should clone with --full-refresh respect the model config?

[rudeb0y]

Is this a new bug in dbt-core?

• I believe this is a new bug in dbt-core
• I have searched the existing issues, and I could not find an existing issue for this bug

Current Behavior

given the config:

  config(
    materialized = 'incremental',
    unique_key = 'event_key',
    on_schema_change = 'fail',
    incremental_strategy = 'delete+insert',
    transient=false,
    full_refresh=false,
    )
}}

when using the dbt clone -s <the_model> --full-refresh - the model is cloned.

Expected Behavior

Not sure what I'd want here. Personally I was pleased to see it not respected.. but it probably should be? Or it's a mix of concepts and it should be some --force flag or something with dbt clone to say "I don't care that it exists, clone it"

Steps To Reproduce

create an incremental model..

set full_refresh=false in it's config..

run dbt clone --full-refresh.

Which database adapter are you using with dbt?

dbt-snowflake==1.6.0

[jtcohen6]

@rudeb0y Thanks for trying out v1.6 & clone! :)

I don't think this was an intentional omission, but I'm with you that being able to pass --full-refresh and override the full_refresh: false config is actually quite nice. I think this speaks to a larger behavior change that we might want to make around the interop of the full refresh config & flag:

• [CT-1301] Behavior change: full_refresh: true config should respect --full-refresh, -f flag #6013

That aside: It's a little bit odd to think about setting full_refresh: true|false on a specific model for the purposes of configuring its clone behavior. I'm not sure what the use case would be for doing it. But for the sake of consistency, clone probably ought to respect that config.

[rudeb0y]

Ok cool - basically I have full_refresh: false to protect some models, in-case someone does a dbt run --full-refresh and broke an incremental model unintentionally. It has a rollup element so the full_refresh output is not the same as the incremental and hence want that preserved (It's not great, should get same output regardless but we cannot at this time).

That said once build is done I still want to clone that incremental model into a new env hence why it was in this case.. fine!👍