If filtering to a particular string and the block doesn't contain that string, we can skip aggregating that block. This might help certain use cases for redbull.
Basically, we would prioritize unpacking that string column first and then check filter against the string table.
This may or may not work well.