I'd like to know if there's a way to issue a query or create a filter with a list of strings, (ie. "roses", "petunias", "orchids") on a variety of fields (ie. "title", "summary", "genus", "tags") and create an aggregation which tells me how many docs had a hit for each of the strings in the query or filter. So, if I found "roses" in any one of the specified fields, the doc hits that bucket. If I find "petunias" in the same doc for the specified fields, it also hits that bucket and so on.
It really doesn't have to be a query if there's a way to specify a filter on multiple fields for multiple values. That's what I was hoping to do with the Filters Aggregation but with that it looks like I would have to do an individual "terms" filter for every term I'm interested in for every field I'm interested in and then sum up all the times I saw "roses" in the "title" plus in the "summary" plus "tags" and so on for each term I was interested in. Besides that, I'm assuming that approach would become less and less performant as the number of terms and fields to check grew. Is there a better way? I'm currently using ES 1.7.x.