select statement

[wujee]

Dear all,

I like to find out how to write a query that will place a null on a particular column on duplicate records. For example, I have a complicated join statement that returns a set of data like the followings:

UID ENTRY_TM DOC_NUM TYPE DSSN AMT IP_AMT
&714104&74 28-Sep-16 DJTS16H475 E4 3801 828.25 451.5
&714104&74 28-Sep-16 DJTS16H475 E4 3801 828.25 376.75

Most of the columns are duplicated except the last one. So I am required to return a null value on one duplicated column such as the following:
UID ENTRY_TM DOC_NUM TYPE DSSN AMT IP_AMT
&714104&74 28-Sep-16 DJTS16H475 E4 3801 828.25 451.5
&714104&74 28-Sep-16 DJTS16H475 E4 3801 null 376.75

I know I can write a cursor to get to what I want; however, is there a way to get the same result in plain select query without function or cursor?


Thanks,
Tina

[gandolf989]

But in 11g there is a lead lag function that will look at other rows as a part of analytics.
Here is a link that might help.

https://www.techonthenet.com/oracle/functions/lag.php

[wujee]

Thanks gandolf989!

I am able to use lag function with decode to get what I need. Thanks again!

[gandolf989]

Your welcome.