Aah, I see what you are getting at, well, that's a probably a bit more complex to figure out, but I think it should be doable. I think, I would get all topics together with their associated replies in your date range. I'd probably transform the data into an array of (nested) objects, where each object is a topic/discussion and the replies are "within" the object, so everything that belongs together is already where it should be before analysis. Then I'd loop through the discussions, filtering out all replies that were posted outside your date-range of interest, actually I would make a copy of the original nested objects-array and literally delete all the reply objects from it if they are not of "interest". After you have done that, simply loop again and filter out topics/discussions that do not have any replies left.
🤯 while I am still trying to wrap my mind around this, I think it sounds feasible.
I'm sorry for the delayed response, I missed your reply until now, but still would like to work this out. Thanks!
