I have the following recordset:
| ProductType | Currency | Book |
|---|---|---|
| Product_1 | USD | EPQ001 |
| Product_1 | GBP | EPQ001 |
| Product_2 | USD | EPQ001 |
| Product_1 | EUR | EPQ001 |
| Product_3 | ZAR | EPQ001 |
| Product_3 | USD | EPQ001 |
| Product_1 | CHF | EPQ001 |
| Product_1 | USD | EPQ001 |
| Product_1 | HKD | EPQ002 |
| Product_1 | HKD | EPQ002 |
| Product_1 | INR | EPQ002 |
I would like to group the recordset by all three columns, like so:
| ProductType | Currency | Book |
|---|---|---|
| Product_1 | USD,GBP,EUR,CHF | EPQ001 |
| Product_2 | USD | EPQ001 |
| Product_3 | ZAR,USD | EPQ001 |
| Product_1 | HKD,INR | EPQ002 |
As you can see, groupings occur at all levels and the list of currencies per ProductType / Book categorisation is cast into a string. Duplicates are removed
I've been trying to use the np.where function to group and can successfully group ProductType and Book columns and successfully group the Currency and Book columns, but am lost on how to merge both of the resulting lists so that the Currency appears as a string. Any ideas?