Daniel Reeves

Also, I should say, if you'd like to contribute to this, you certainly can, but we need to think through implementation. My first thought is we could do something like...

Coming back to this a whole year later. `commands_before_options` will be released as part of 1.9.0 and is in this PR: #252 ```python import rich_click as click @click.group @click.rich_config({"commands_before_options": True})...

Closing this because panel ordering is 100% feature-complete via #254. The `1.9.0dev0` prerelease is not fully done re: command/panel ordering, but `1.9.0dev1` fixes all the issues (when that's available).

@pankajastro ~~I am wondering if there is a simpler fix. I think the best solution is just that `super().__init__(**kwargs)` needs to be at the top of the `AbstractDbtBaseOperator` init call...

Reopened in light of: https://apache-airflow.slack.com/archives/C059CC42E9W/p1718538236076319

Yeah I believe this is a memory issue. "Zombie tasks" is the error message you get on Astronomer if your celery worker sigkills. This happens because Cosmos spins up subprocesses...

@w0ut0 My most opinionated suggestion is to use the `DBT_RUNNER` invocation mode. I'm sure the Astronomer people would like it if I suggested using bigger workers since that's more money...

@benc-db This change doesn't do anything in 1.7, only in 1.8. So yes, it should be rebased to that branch (didn't notice it!). I can rebase to that when I...

Swapped to `2.0.latest`.

That's interesting, I did not notice that. In the `dbt-adapters` library for 1.8 beta, the `list_relations_without_caching` doesn't appear to use `agate`. I'm not aware of if this was different in...