Do not input private or sensitive data. View Qlik Privacy & Cookie Policy.
Skip to main content

Announcements
Qlik Open Lakehouse is Now Generally Available! Discover the key highlights and partner resources here.
cancel
Showing results for 
Search instead for 
Did you mean: 
Not applicable

Cannot get distinct count right by dimension

Hi,

I am trying to get a count by "status" in the table below - in pie char

I am expecting - 2 pass and 2 fail - for start time at 2pm (first table) and 1 pass and 3 fail at 3 pm.

Please help!

test_suite_id test_case_id status start_time

suite1               case1     pass          9/12/2012 2pm

suite1               case2     pass          9/12/2012 2pm

suite1               case3     fail             9/12/2012 2pm

suite1               case4     fail             9/12/2012 2pm

suite1               case1     pass          9/12/2012 3pm

suite1               case2     fail              9/12/2012 3pm

suite1               case3     fail             9/12/2012 3pm

suite1               case4     fail             9/12/2012 3pm

6 Replies
giakoum
Partner - Master II
Partner - Master II

Hope this helps!

Not applicable
Author

Thanks so much for the quick response. It gives me the correct graph, however when  I check show values on data points, on the pie chart it shows me 1, 1 for the 3 pm case where it should show 3/1. How can I see the right counts?

qlikview-pie-chart-image.png

Not applicable
Author

Also, by default I have a date range selected, so for each test suite, within a date range, I want the count for the latest date (max on date). If I use the answer in your response, I get it right for one date selection, however, it doesn't work for a date range

giakoum
Partner - Master II
Partner - Master II

max date seems to be working and values I added in the legend. Value on data points acts strange, and it is Sunday today..

whiteline
Master II
Master II

Hi, sapte123.

Why not just create a pie chart with 'status' as dimension and Count(status) as expression ?

Not applicable
Author

Thanks! max works for start_time. But in my actual data I have more test suites and each having different test cases. So the moment I add this data to your qvw, the pass fail counts are off. I am expecting - 16 fail and 6 pass. I get the accurate results only when I select the time.

So what I am looking for is without any filter, I want a max(Start_time) - group by test_suite, test_case

suite2,               case1,     pass,          9/12/2012 2pm
suite2,               case2,     pass,          9/12/2012 2pm

suite2,               case3,     fail,             9/12/2012 2pm

suite2,               case4,     fail,             9/12/2012 2pm

suite2,               case1,     pass,          9/12/2012 3pm
suite2,               case2,     fail,              9/12/2012 3pm

suite2,               case3,     fail,             9/12/2012 3pm

suite2,               case4,     fail,             9/12/2012 3pm

suite2,               case4,     fail,          9/12/2012 4pm

suite2,               case2,     fail,              9/12/2012 4pm

suite2,               case3,     fail,             9/12/2012 4pm

]