[Feature] Create Error Message if "fields_to_include" is not recognized

[klmopenspace]

Is there an existing feature request for this?

• I have searched the existing issues

Describe the Feature

I was trying to ingest "Quota_ARR__c" from the opportunity into dbt given that it is a formula field in Salesforce. However, no matter what I did, I couldn't get it to work. Despite writing the write syntax (see below), the opportunity table would load, but without "Quota_ARR__c".

{{ salesforce_formula_utils.sfdc_formula_view( source_table='OPPORTUNITY', fields_to_include=['quota_arr__c'], full_statement_version=false) }}

Because of this, I looked at the Fivetran UI schema to see how it was spelled (which was the same as the way as I wrote it above) & looked at the Salesforce API. However, when I looked at the fivetran_formula table ingested into Snowflake, I saw it was written with 1 underscore rather than 2.

Given this, I highly recommend you create a safeguard that if it is incorrectly written, that the code snippet doesn't run and/or creates an error. I spent hours trying to figure out why this wasn't running until finally realizing it should be 1 underscore, not 2

Describe alternatives you've considered

No response

Are you interested in contributing this feature?

• Yes.
• Yes, but I will need assistance and will schedule time during your office hours for guidance.
• No.

Anything else?

No response

[fivetran-joemarkiewicz]

Hi @klmopenspace thanks for raising the troubleshooting issue you encountered and providing a possible resolution in this feature. I think this could be something we fold into a future update to the package.

My one question would be around what your preferred method of raising a warning would be? We could provide a few options:

• Make the model flat out fail if the fields_to_include do not find a match. This would ensure the model only succeeds if there are in fact fields to map to in the source.
• Provide a CLI warning informing about the fields that do not find a match. This would allow the model to succeed, but would only provide a CLI warning that X field did not find a match.

Let me know your thoughts!