Redshift: attribution_overview can't identify type for spend

[jethron]

Describe the bug

When building attribution_overview with no snowplow__spend_source defined, the spend column in the CTEs is always the null value; when this model tries to aggregate the values via min(spend), it is unable to find an appropriate type and produces a ERROR: failed to find conversion function from "unknown" to text error.

Steps to reproduce

Run the attribution models on Redshift without specifying a snowplow__spend_source configuration.

Expected results

The model should run successfully, producing 0/null for spend.

Actual results

The model fails to build, producing an error.

Screenshots and log output

Which database are you using dbt with?

• postgres
• redshift
• bigquery
• snowflake
• other (specify: ____________)

Additional context

Casting the null to something more specific in the CTEs seems to resolve the issue:
cast(null as double precision) as spend

Are you interested in contributing the fix?

If you like.

[agnessnowplow]

Thank you for raising this issue and apologies for not getting back to you earlier on this. The day this was raised we released v0.2.0 which I just checked should already fix this issue. An upgrade should solve this therefore I'm closing this for now.