Thank you for your feedback. While it is unlikely we’ll address this suggestion in the near future, we’ll reassess based on the number of votes it receives.
There are major performance implications for supporting this which we will need to work through.
Azure Search Product TeamAnonymous commented
I vote for this as well. It's a dumb limitation. I tried to use it to solve a problem I'm having where I have more than 3 million unique facet values and then querying on them will kill any Azure Search server with a single request. I know 3 million unique facets seems like a lot but it's really not when you have 12 billion records. Problem happens when I add query $facet=joefacet:count 30000000
The query is super fast so I created a work around for for the facet limitation but it relied on being able to page through a sorted result set and then I ran into the 10000 skip limitation. I'm about ready to just switch to Lucene, which really seems to be the foundation for Azure Search.
The "ID in the doc" solution won't work for me because off the need I have to sort on the "facet" field so I mimic the counting off facets that doesn't currently work in Azure with over 3 million unique facet values.
1 voteAnonymous shared this idea ·
This is currently not on our roadmap but will revisit this in future planning cycles.
Thank you for your suggestion.Anonymous supported this idea ·