Skip to main content
Announcements
Have questions about Qlik Connect? Join us live on April 10th, at 11 AM ET: SIGN UP NOW
cancel
Showing results for 
Search instead for 
Did you mean: 
paulyeo11
Master
Master

QS how to reduce the data ?

Hi All

I have a QS doc , i like to know how can i reduce the data to 2% ?

Binary 'lib://vFOLDER/QV4QS_v18.qvw'; 

I am using Binary load the data , but i like to keep the data confidential , so i want to only read 2% of the data , so that when i post the QVF file , other will not be able to know my original data source

Paul

17 Replies
paulyeo11
Master
Master
Author

Hi Andy

I know my field name , BEAND_='BECKHOFF', but i don't know where to insert them into below script :-

NewFilteredTable:

NOCONCATENATE

LOAD *

RESIDENT sales WHERE Your filter Criteria;

DROP sales;

ogster1974
Partner - Master II
Partner - Master II

NewFilteredTable:

NOCONCATENATE

LOAD *

RESIDENT sales WHERE BEAND_='BECKHOFF';

DROP sales;

Note you can change the name NewFilteredTable if you want to something more appropriate to.

paulyeo11
Master
Master
Author

Hi Andy

I have try it manage to filter the brand = BECKHOFF , but when it run till last line of script , it return error msg :-

drop.png

ogster1974
Partner - Master II
Partner - Master II

‌DROP TABLES sales; should do the trick.

IF you get syntax type errors in future searching the community for similar examples and customising them will yeld quicker results for you and also aid your understanding of what you can do in script.

REgards

Andy

paulyeo11
Master
Master
Author

Hi Andy

Finally it work now , Thank you very much. Thank you for your advise.

Paul

paulyeo11
Master
Master
Author

‌HI Andy

i am wonder why the reduce script need to add filter by Brand = BRCKHOFF ? As my  QVW file have many field , I imagine that reduce script should just random take 10 record from the total record and don't need add filter.

Hope you can share with me.

PAul

ogster1974
Partner - Master II
Partner - Master II

This is where you Paul need to be driving these decisions not guys on the community.  The reason that was chosen as a filter was your call as you know your data.

You can filter your data on any field or fields in the table to get the right subset of data both volume and content,

Reason I would use a filter approach rather than a random approach is because at some point you are going to need to test the quality of your results i.e. does the value on the chart, table filter etc.. match what i'm expecting.

For these kinds of tests you need to be sure of your base content.  Random records in my experience equals random results.  fine for demoing functionality but not for testing.

Regards

Andy

paulyeo11
Master
Master
Author

Hi Andy

i have try to change BRAND=HAKKO , and i noice that it only filter all record belong to HAKKO , so now i see the advantage of this approach compare to random filter .

Thank you very much for your advise.

Paul