Shouldn't we sync the documentation even if tables schemas have mismatches between dbt manifest and metabase? about dbt-metabase HOT 3 CLOSED

Alright so I've looked more in details at the code and I understood what was going on, I had some warnings on non-existing columns like this message :
WARNING Column colname not found in metabase.py:282 SCHEMA.TABLE model

Which cause are_model_compatible = False here, and then it stops the computation in MetabaseClient because sync=True per default.

What I do not understand about this behaviour is that even for only one column mismatch the sync will not be run at all even if the majority of the tables schema matches. Maybe I just do not see the obvious reason ahha but why should we have the default sync=True instead of sync=False (as it was before) ?
Thanks and have a great day :)

PS: btw I (Pycharm actually 😅) saw a micro typo I'll open a really small MR to clean it.

from dbt-metabase.

Intended behaviour is maintaining all columns in the models, hence sync is enabled by default (it was mistakenly turned off a few builds ago). You can run it in "best efforts" mode to export what it can by disabling it, but that's not encouraged.

from dbt-metabase.

Ok thanks that's more clear, have a great day :)

from dbt-metabase.