Skip to main content
Announcements
NEW: Seamless Public Data Sharing with Qlik's New Anonymous Access Capability: TELL ME MORE!
cancel
Showing results for 
Search instead for 
Did you mean: 
danielnevitt
Creator
Creator

How to show duplicates?

Hi,

I have the following script and I would like to write an expression in a straight table to show where there are duplicate records based on CO_CD, PRICE1 and CONTRACT_MONTHYEAR being the same (all of these fields the same, rather than duplicate records within each field).

 

// Start of [QlikviewPhysical.xml] LOAD statements

RECORD:
LOAD TRADE_TYPE,
TRADE_STATUS_IND,
STRATEGY_NUM,
TRADE_NUM,
DESK_CD,
OFFICE_CD,
PRICE1,
EXTENDED_AMT,
CO_CD,
BS_IND,
Date(Date#(CONTRACT_DT, 'MM/DD/YYYY')) AS CONTRACT_DT,
Week(Date#(CONTRACT_DT, 'MM/DD/YYYY')) AS CONTRACT_WEEK,
Month(Date#(CONTRACT_DT, 'MM/DD/YYYY')) AS CONTRACT_MONTH,
Year(Date#(CONTRACT_DT, 'MM/DD/YYYY')) AS CONTRACT_YEAR,
MonthName(Date#(CONTRACT_DT, 'MM/DD/YYYY')) AS CONTRACT_MONTHYEAR
FROM V:\ACL\ACL_Import\Qlikview\QlikviewPhysical.xml (XmlSimple, Table is [RECORDS/RECORD]);

// End of [QlikviewPhysical.xml] LOAD statements

ls this possible?

Any help would be much appreciated.

Regards,

Daniel

7 Replies
Not applicable

Daniel

I would create this key in your script, i.e.

CO_CD & '-' & PRICE1 & '-' & CONTRACT_MONTHYEAR as %KEY;

Then in your straight table, use the formula Count(%KEY).  This will work as long as you have each of these fields as dimensions.


Let me know how you get on.


Steve

danielnevitt
Creator
Creator
Author

Hi Steve,

Thank you very much for your reply.  That idea works perfectly and I have now created a straight table showing the count of %KEY.

Are you able to able the best way to restrict the table to only show where the %KEY count of distinct TRADE_NUM is greater than 2?

Thanks,

Daniel

Not applicable

There's a couple of ways of doing this.  You could simply create an expression that only shows items above 2, such as;

=if(count(%KEY)<=2,0,count(%KEY))

Which will only ever return the duplicates.  Or you could create a Calculated Dimension, using the formula;

=if(count(%KEY)<=2, 'Not Duplicate', 'Duplicate')

This second option will give you more flexibility to filter on the Calculated Dimension, so you can see everything, just the Duplicates, or just the Non-Duplicates.

Can you mark my answer correct if I get it right?  And let me know how it goes.


Regards


Steve

danielnevitt
Creator
Creator
Author

Hi Steve,

That first idea works great, thanks for your suggestion.

Are you able to advise how I can use that idea and only show DISTINCT TRADE_NUM?

Regards,

Daniel

Not applicable

Should you only be getting one record for each TRADE_NUM?

If so just use the same formula for a calculated dimension;

=if(count(TRADE_NUM)<=2, 'Not Duplicate', 'Duplicate')

Again, TRADE_NUM will have to be a dimension for this to work.  Does this work for you?

danielnevitt
Creator
Creator
Author

Hi Steve,

Each trade number can have multiple settlements against it.

Therefore I want to only show duplicates where the TRADE_NUM is distinct.

When I input the above as a calculated dimension, I receive //Error in calculation in my table.

Regards,

Daniel


Not applicable

Daniel

Without doing any testing, I believe you can change the formula to also be dependent upon whether there is only one distinct TRADE_NUM, like so;

=if(count(%KEY)<=2 AND count(DISTINCT TRADE_NUM) = 1 , 0 , count(%KEY))

Alternatively, you could use the below;

=if( (count(%KEY)/Count(TRADE_NUM)) <= 2,0, (count(%KEY)/Count(TRADE_NUM)) )

Regards


Steve