Masking a column based on user login

I know this answer is wayyyy overdue, but the solution to the problem has been provided by combining two answers into one...

Step 1) Create a mapping Table that maps the UserPrincipalName to your permisisons. let's say

UPN = alex@contoso
Permission = Masked

UPN = manager@contoso

Permission = All

then create a Role in RLS to filter the Permission Table based onn UPN=UserPrincipalName(). so their Login will automatically filter that table. 

then in your Table you can use a calculated field with a Lookup for Min(Permission)="All" in the permission table to display the data or not. you won't be using the UserPrincipalName in the field at all. that already did its magic when the user opened the report. 

Even better would be to separate out the sensitive data into a separate table with a 1-1 match on the orginal one and use relationships to filter the data already at Row Level. In that way the user can't even circumvent the masking if they start using Self-Service features like QnA or online report editing. 

i.e. 

Original:

ID = 1

Name=Alex

Creditcard = 1234-4567-9012-4444

New:

Employee:

ID = 1

Name=Alex

Sensitive:

ID = 1

Creditcard = 1234-4567-9012-4444

Permission = "HR"

Permissions:

UPN=alex@contoso

Permission = "HR"

Add a m-n relationship between Permissions and Senstive based on the Permission Column with security filter applied.

Add a 1-1 relationship between Employee and Sensitive based on ID column

Then either just drag the fields together on the fly, or add a new calculated column that shows a mask "****-****-****-****" when no Related(Creditcard) is found.

Don't forget to apply the RLS with the UPN on the Permission table though...

Hope this all makes sense!