Epic: improve testing around core/critical E2E functions

[jadudm]

At a glance

In order to have confidence in the workbook migration code
as a product owner
I want more testing around what was previously exploratory/nice-to-have app code.

workbook_creation.py

https://github.com/GSA-TTS/FAC/blob/main/backend/dissemination/workbooklib/workbook_creation.py

Tasks

Preview Give feedback

1. E2E improve: Mappings #2673
   eng +1
   
2. E2E improve: setup_sac() #2674
   eng +1
   
3. E2E improve: workbook_loader() #2675
   eng +1
   
4. E2E improve: Create Transformation Utils for Common Transformations #2833
   eng +1
   
5. Make Use of Relative Imports #2890
   eng +1
   

excel_creation.py

Tasks

Preview Give feedback

1. E2E improve: Paths, names #2676
   +0
   
2. E2E improve: Remove test_pfix()
3. E2E improve: set_single_cell_range() #2677
   +0
   
4. E2E improve: col_to_ndx() #2678
   +0
   
5. E2E improve: set_range() #2679
   +0
   
6. E2E improve: dbkey_to_test_report() #2680
   +0

The code around test tables is problematic, and I speak about it in the E2E documentation. There are ways to improve this dramatically, but it may not be worth it. In which case, despite how this code looks... not touching it might be the right thing to do.

sac_creation.py

Probably all of it. At the least, we need a way to map

historic field -> named range -> api_field

as well as provide a migration function. In essence, it would be good to have a dictionary of FieldMap structures (just like the workbooks) that do the work of creating the general info object. However, we would then take the FieldMap structures and, instead of filling a workbook, use them to create a JSON object that represents the general_info and audit_info objects in the SAC.

That would be the best way to clean this whole thing up. And, it would then be easy to integrate the general info form into the API end-to-end testing.