eezec
05/06/2021, 3:25 PMdouwe_maan
05/06/2021, 3:30 PMtransform/models
2. create a separate dbt package in a separate repository, and manually add it as a transform plugin in meltano.yml, or directly to transform/packages.yml
Option 1 is easiest, unless you expect wanting to reuse these dbt models in multiple Meltano/dbt projectsdouwe_maan
05/06/2021, 3:31 PMeezec
05/06/2021, 4:55 PMeezec
05/06/2021, 4:55 PMeezec
05/06/2021, 4:56 PMmeltano add transform --custom dbt-my-customtapdouwe_maan
05/06/2021, 5:04 PMmeltano add --custom transform is not currently supported. If you've created your own dbt package, the best approach right now is to add it directly in transform/packages.ymleezec
05/06/2021, 5:12 PMeezec
05/06/2021, 5:13 PMdouwe_maan
05/06/2021, 5:13 PMdouwe_maan
05/06/2021, 5:15 PMalso, it found that dbt does receive the target postgres var i've set, i had to add them in a .env file inserted to the docker container via the directive env_file in docker-composeHmm, that shouldn't be necessary. How were you running
dbt ? Just as part of meltano elt --transform=run ? The variables in https://gitlab.com/meltano/files-dbt/-/blob/master/bundle/transform/profile/profiles.yml should automatically be populated based on the configuration env vars for the loader: https://meltano.com/docs/integration.html#pipeline-environment-variableseezec
05/06/2021, 6:18 PMeezec
05/06/2021, 6:21 PMdouwe_maan
05/06/2021, 6:50 PMPG_ADDRESS should be populated as well: https://gitlab.com/meltano/meltano/blob/master/src/meltano/core/bundle/discovery.yml#L1404