The project i’m working on is creating a knowledge graph of sorts, we’re pulling in data from variety of disparate sources, typically deeply nested JSON data, and it’s not immediately clear to me what the schema needs to be ahead-of-time, until we see more use cases, we only have a vague sense of which parts of the JSON will end up being useful, is this a good fit for datomic? I tend to think of schemaless databases as the ‘solution’ here, where retrospectively you can decide on what your indexes will be, speed up particular queries , as you acquire more requirements/use cases on the sorts of queries that matter. If not datomic, what would you suggest, also needs to support graph query operations. If datomic, the strategy is to break apart these deeply nested JSON structures at ingest time try and mapping to datomic primitives, e.g. json arrays into db.cardinality/many or how would you go about that?

how do i pass multiple rules in a single query?

(d/q '[:find ?t (count ?t2) (count ?t3)
       :in $ % ?p1 ?p2 
       :where
       (played2 ?p1 ?p2 ?t)
       (wins ?p1 ?p2 ?t2)
       (wins ?p2 ?p1 ?t3)
       [(= ?t ?t2 ?t3)]]
     atp wins played2 "Roger Federer" "Novak Djokovic" )