Let's imagine I am working with a file that has the following data structure:
Region / Supplier# / PO# / Account code # / Category Level 1 / Category Level 2 / Amount

In an ideal world, the file should come with all the data, however, some of the data are missing: category level 1 and category level 2 need to be added in some rows.
The way they can be added is by defining rules that depends on the region, the supplier# and the account code#.
For example:
If (region = 1 and supplier# = 123456 and accountcode# = 1234) then category1 = 12 and category2 = 45

The way I am planning to solve this issue is to first filter the data that are already classified by using the "filter" box. So I will end up with a file missing all category level 1 and 2 data.
Then, I will introduce a new file that will contain the "rules" with the following structure: Region / Supplier# /Account code # / Category Level 1 / Category Level 2
I will then merge the two files, take all data but the cat level 1 and 2 from the original file and cat level 1 and 2 from the new file. The link between the 2 files will be on region, supplier# and account code#.
I'll then merge the new file with the correct rows from the old ones. So far, so good (please correct me if I'm wrong).

Regarding the Region / Supplier# /Account code # / Category Level 1 / Category Level 2 new file, everything is fine if all data are entered. For example, region=region1, supplier=123456, accountcode=1234, etc...
Now, my question is, let's imagine the rule is on region and account code and works for any supplier. That would mean leaving the supplier field empty and the link to the previous table would be broken. So, how would you handle that so that the administrator only has to create 1 new file?

At that point, I am wondering if I am clear or not, so please don't hesitate to ask me questions if needed.

Thank you.