JOIN: 'Same' code, different results?

[#1]

Hello all,

I was recently struggling with a piece of code joining customer data to sales data.

I finally managed to get it work, but I still wonder about it.

Below you see two versions of what I consider to be the same code.

V1 is working as expected, V2 not.

(Simplified you can imagine Sales having three columns: TransactionID, Revenue, CustomerID)

V2 'ignored' the 'distinct', thus the number of transactions got multiplied.

My question: What is the (internal) difference?

// - V1 -----------------------------------------------
C:
LOAD Distinct
       CustomerID,
       CustomerName
Resident Customers;

left join(Sales)
load * Resident C;
drop table C;

// - V2 -----------------------------------------------
left join (Sales)
LOAD Distinct
       CustomerID,
       CustomerName
Resident Customers;

Thanks for any ideas,

Thilo