Forum Discussion
oren
7 years agoHelper III
Ranking by groups
Hi
i what to create a Mesure that rankink rows for groups of values order by number -
same as in sql row_number over (partition by X,Y,Z order by L)
is it possible?
thank you very mutch
oren
Please ignore the previous formula.
This one works:
Sorting_Rank = RANKX ( FILTER( ALL( 'Table'[A], 'Table'[B], 'Table'[C] ), 'Table'[A] = MAX('Table'[A]) && 'Table'[B] = MAX('Table'[B]) ), CALCULATE(SUM('Table'[C])), , ASC, Dense )
6 Replies
- AnkitBISolution Sage
Please try like below.
Measure 11 = RANKX(FILTER(all(Table1),Table1[Column1] = max(Table1[Column1]) && Table1[Column2] = max(Table1[Column2])),CALCULATE(sum(Table1[Column3])),,ASC,Skip)
Regards
AJ
Do Like Post if response seems good and Worth liking.
Do Mark as Solution if response resolved your Issue.- orenHelper III
it dosent work- i am getting only "1" ( but maybe i am doing somthing not right)
thanks
- v-chuncz-msftCommunity Support
- AnkitBISolution Sage
Never used ISONORAFTER before but will definately check. Thanks for Suggestion.
- themistoklisCommunity Champion
Try the following formula:
Sorting Rank = IF ( HASONEVALUE ( 'Table'[A]), RANKX ( ALL ( 'Table'[A], 'Table'[B] ), SUM('Table'[C]), , ASC, Dense) ) - themistoklisCommunity Champion
Please ignore the previous formula.
This one works:
Sorting_Rank = RANKX ( FILTER( ALL( 'Table'[A], 'Table'[B], 'Table'[C] ), 'Table'[A] = MAX('Table'[A]) && 'Table'[B] = MAX('Table'[B]) ), CALCULATE(SUM('Table'[C])), , ASC, Dense )