5.3 Iteration in groups

 

5.2 Early terminated iterative operation


1. Calculate the cumulative sales of each salesperson.

2. Number the orders of each salesperson from 1 and replace ORDERID with corresponding number.

SPL

A B
1 =file(“SALES.csv”).import@tc()
2 =A1.sort(SELLERID,ORDERDATE) /Sorting
3 =A2.(iterate(~~+AMOUNT,0;SELLERID)) /Cumulative sales
4 =A2.derive(iterate(~~+AMOUNT,0;SELLERID):CUM_AMOUNT) /Cumulative sales
5 =A2.derive(cum(AMOUNT;SELLERID):CUM_AMOUNT) /Cumulative sales
6 =A2.(iterate(~~+#,0;SELLERID)) /Number the orders of salesperson
7 =A2.run(ORDERID=iterate(~~+#,0;SELLERID)) /Number the orders of salesperson
8 =A2.run(ORDERID=seq(SELLERID)) /Number the orders of salesperson

A3: The iterate(x,a;G) is to calculate the cumulative sales. In this function, G is the name of field. When G changes, it will record the current ~~ and then start a new iteration again, which is equivalent to A2.group@o(SELLERID).conj(~.(iterate(~~+AMOUNT,0))).

A4: Take the result of iterate() as a new column in A2.

A5: The cum(x;G) is to calculate the cumulative sales. When G changes, it will start to calculate again and produce the same result as A4.

A6: The iterate() is to number by SELLERID from 1. When SELLERID changes, it will start to number from 1 again.

A7: Change the ORDERID of A2 to the value of A6.

A8: The seq()has the same role as the iterate() in A6, and will produce the same result as A7.

SQL

SQL doesn’t provide a similar method.

Python

Python doesn’t provide a similar method.


6.1 Positioning operations
Example codes for comparing SPL, SQL, and Python