Keep unused columns in success .csvs when importing with Bulk API 2.0

[bchfundraising]

Is your feature request related to a problem? Please describe.
With SOAP or Bulk API, when importing data, the success.csv file contains all columns regardless of whether they were used or not, but Bulk API 2.0 only returns columns that were imported (and the new ID and sf_created columns)

Describe the solution you'd like
Keep all columns that were in the original import .csv file

Describe alternatives you've considered
Using other SOAP or Bulk API instead

Additional context
Dataloader version: 61.0.0
OS: MacOS Sonoma 14.4.1
Configs: configs.zip

To describe why this is useful, one of my workflow involves me having a flat file of Opportunities, GAU Allocation and Payments details at the ready. I'd import the Opportunities first to create them, then using the success file, I can proceed to import GAU allocations and Payments directly without extra steps.

Our Opportunities don't always have Unique References, so I can't use an xlookup to ensure I'm mapping the correct opportunities to the Payments and GAU on subsequent imports. Whilst I know the success file is meant to return the records in the same order of import - as a data person - I just cannot make that assumption just in case a bug creeps in one day and causes the order to change.

Maintaining the behaviour of SOAP and Bulk API behaviour would allow us to use this without concern in the future.

Edit: added more context info, not that I think it's relevant

[ashitsalesforce]

Hi @bchfundraising , Bulk API 2.0 returns success and error results separately to facilitate upload retry on failed rows more readily (client code does not have to separate failed rows from success rows). However, it means that the information regarding which row in the uploaded CSV is lost because there is no unique id in the import CSV to match a successful or failed insert row with. We plan to enhance Bulk API 2.0 in the future to be able to identify which row in the uploaded CSV failed. Data Loader will be enhanced after that to support your request.

[ashitsalesforce]

Hi @bchfundraising regarding your point:

Our Opportunities don't always have Unique References, so I can't use an xlookup to ensure I'm mapping the correct opportunities to the Payments and GAU on subsequent imports

How do you map successfully imported Opportunities to Payments and GAU sobjects if Opportunities do not have unique references and you do not want to rely on import order?