I have a couple of questions on the Subset Editor in Arc. Not sure if there is an FAQ or documentation to answer these.
What does the “All Elements” right click option do? I thought it would be the same as the button on the toolbar, but it doesn’t seem to be.
If an element is displayed multiple times (same or different MUN) and I select the duplicates to remove, it removes all occurrences of the element. Is there a way to just remove the duplicates?
Until this gets implemented you can manually edit the MDX and just wrap the Distinct() function around the entire set. If there are any duplicates then only the first instance will be kept.
However, be aware that the Distinct function works on members NOT elements. Duplicate members with the same MUN will be removed but if you have duplicate elements which are identified with different MUNs due to a different rollup path these won’t be seen as duplicates (becasue from a MDX perspective they aren’t.)
e.g.
If this is wrapped in Distinct there will still be 2 members in the set as there are 2 unique rollup paths to the element 2023.
Regarding removing duplicates in Arc Subset Editor, rather than selecting and removing duplicates, sometimes I just select the members I want to keep (including one of the duplicate members) and use the Keep button. This, of course, only makes sense with smaller sets.
Working on version 5.3.0 of Arc, when clicking on Remove Duplicate, it does not seem to remove any duplicate elements. Quoting @cw-ch-scott here in THIS THREAD, I understand that this can be due to the DISTINCT function.
This is causing confusion for the user who has multiple elements visible in the set, and when clicking on remove duplicates, nothing is happening.
Would it be possible to change this to a UNION or INTERSECT function so that whatever the user multi-select in the set, it will force a removal of any duplicate elements?
There’s a distinction between ELEMENTS and MEMBERS
Consol1^1000b and Consol2^1000b may both identify the same element 1000b but these are still 2 unique members from the point of view of MDX.
The DISTINCT function is removing duplicate members, but as the selected members are not duplicates nothing happens. Everything is in fact functioning as intended.
I agree from a technical MDX evaluation perspective these members are unique in each consolidator, and the DISTINCT statement is functioning as intended.
From a non-technical user perspective - who does not know about unique members vs elements - the function name “Remove Duplicate” does not correlate to the output.
Would it be possible to expand the above to 2 buttons?
Remove Duplicate Members → DISTINCT function
Remove Duplicate Elements → UNION or INTERSECT function