Open kokorin opened 8 months ago
Hi @kokorin ! Thanks for opening this issue, and sorry about the delayed response.
This is a cool use case! Makes sense to me overall to save this information. I'd use separate fields rather than incorporate it into existing ones. I think also that since this information is passed as external variables to dbt, then the feature is generic and doesn't necessarily require explicit logic for dbt Cloud / github actions, so I think you should be fine there.
Feel free to open a PR that adds this capability, lmk if further guidance is needed!
Is your feature request related to a problem? Please describe.
We run DBT with Elementary package in Airflow. Airflow DAG contains several DBT build tasks: core models first, then integration models (based on core). We want to build separate Elementary reports for core module and for each integration module:
For that we need to get
dbt_invocation
UUID from DBT Invocations table and use it withedr report --select invocation_id:XXXXXXXXXXXXX
.It would be useful to have Airflow
task_id
andtry_number
in DBT Invocations table to simplifydbt_invocation
lookupDescribe the solution you'd like A clear and concise description of what you want to happen.
Describe alternatives you've considered It's possible to include
task_id
andtry_number
intojob_id
orjob_run_id
, but Elementary uses the latter vars to buildjob_url
andjob_run_url
.Additional context N/A
Would you be willing to contribute this feature? Yes, but I don't have DBT Cloud and have limited experience with Github Actions.