I have a group of items that's being created in a rule and its format is lik

RE: I have a group of items that's being created in a rule and its format is lik

Michael Tan — Sat, 12 Apr 2014 14:55:14 GMT

For whatever data set you have returned, you can get the unique combination of month and year by merging the month and year arrays and taking the union of that array with itself. See rule below:

union(merge(a,b),merge(a,b))

For your example data set this would yield three combinations to check for (3-2013, 3-2014, and 4-2014).
You can use this with the expression Jesse mentioned to dynamically iterate over the original data set.

RE: I have a group of items that's being created in a rule and its format is lik

aswinb — Thu, 10 Apr 2014 22:45:09 GMT

Yes Jesse, I am trying to do this dynamically for each month in different years and so a monthOfInterest and yearOfInterest can't be provided. There will be two given sets of dates and between those I have to count the number of items that were created each month and in each year that happens between the time intervals that the user selects.

RE: I have a group of items that's being created in a rule and its format is lik

jesse.triplett — Thu, 10 Apr 2014 22:23:12 GMT

If you're looking for a specific month and year you can use wherecontains.
EG You could return the indices that match the year of interest, the indices that match the month of interest and intersect them, and count the result. Alternatively you could do it one line like below.

count(wherecontains(monthOfInterest,touniformstring(index(items,wherecontains(yearOfInterest,touniformstring(items.year))).month)))

If you're looking to do this dynamically for all combinations of unique month/year, you'll have to make some expression rules and apply over them, but the same principles apply.

RE: I have a group of items that's being created in a rule and its format is lik

larson.thune — Thu, 10 Apr 2014 21:46:06 GMT

Wrap your CDT in todatasubset().totalCount