Sometimes you got business requirments to create new view or dashboard and to do this you need to convert rows to columns. You can transpose them on two ways: using CASE or PIVOT function. In this post I will show you how to convert rows to columns at Teradata on these two ways.
To explain you how to use pivot function, first we create PRODUCT table and insert some samples records in it.
CREATE TABLE TUTORIAL.PRODUCT ( PRODUCER_NME VARCHAR(100), PRODUCT_NME VARCHAR(100), R_YR INTEGER, R_MTH BYTEINT, SOLD_NO NUMBER(7,2) );
And this is the data preview:
PIVOT function appeared in the 16 version of Teradata database and allow us quickly tranpose rows into columns. Let’s look at the example below!
SEL * FROM TUTORIAL.PRODUCT PIVOT ( SUM(SOLD_NO) FOR R_MTH IN ( 1 AS MTH_1, 2 AS MTH_2, 3 AS MTH_3) ) piv;
R_MTH in my table is the atribute which I want to see as columns, but as value for this attributes I want to see sum of the SOLD_NO values. I can write SEL *, because my table not contain any additional attributes – in case when your table have a lot of columns, probably you will must limit them.
The screenshot below presents like my data look after using the PIVOT function.
You can get the same results as using the pivot function using the CASE statment. CASE statment is used more often than PIVOT so you can feel more comfortable using it. But remember what is the better, quicker and shorter depends how many attributes, calculation and conditions you need to use. And in most cases the PIVOT function wins 🙂
SEL PRODUCER_NME, PRODUCT_NME, R_YR, SUM(CASE WHEN R_MTH = 1 THEN SOLD_NO ELSE NULL END) AS MTH_1, SUM(CASE WHEN R_MTH = 2 THEN SOLD_NO ELSE NULL END) AS MTH_2, SUM(CASE WHEN R_MTH = 3 THEN SOLD_NO ELSE NULL END) AS MTH_3 FROM TUTORIAL.PRODUCT GROUP BY PRODUCER_NME, PRODUCT_NME, R_YR;
The SELECT statment above can give you the same result like the PIVOT function.
If you want to download all the scripts using in this post, you can download the Product.txt file below:
If you enjoyed this post please add the comment below or share this post on your Facebook, Twitter, LinkedIn or another social media webpage.
Thanks in advanced!