Odd nesting/flattening for the BUM Survey

[KELSEYDOWLING7]

Hi Jake, I noticed a weird data issues when working on the Biospecimen Survey QCs. They're both only in nih-nci-dceg-connect-prod-6d04.FlatConnect.bioSurvey_v1_JP and only seem to be happening with a few hundred participants from 2022, so I don't know if these can even be resolved but I wanted to ask if this type of occurrence could be fixed in flattening. Low priority and can be looked at whenever you get the chance.

1. For one variable 980800222...

we would expect to have this for the first loop:

• D_980800222_1_1_D_173502329_1_1,
• D_980800222_1_1_D_366972678_1_1

Instead, it has six different versions:

• D_980800222_1_1_D_173502329_1,
• D_980800222_1_1_D_173502329_1_1,
• D_980800222_1_1_D_173502329_1_1u25_20required
• D_980800222_1_1_D_366972678_1,
• D_980800222_1_1_D_366972678_1_1,
• D_980800222_1_1_D_366972678_1_1u25_20required

2. The other variable 715581797...

we would expect to have this for the third loop:

• D_715581797_3_3_D_652923023_3_3,
• D_715581797_3_3_D_701387353_3_3

Instead, we have all of these:

• D_715581797_3_3,
• D_715581797_3_3_D_652923023_3,
• D_715581797_3_3_D_652923023_3_3,
• D_715581797_3_3_D_701387353_3,
• D_715581797_3_3_D_701387353_3_3

The biggest problem here is that the unnested D_715581797_3_3 variable is an array

[jacobmpeters]

Thanks for creating this issue. I'll add this to my ToDos and will reach out if I have questions.