You have some schema there that models an alternative. Either alt or or?

yes, each schema always accumulates to the :path (path in schema) even if they do not contribute to the :in (path in value). Each Schema defines it’s own meaning for :path, but by convention, numbers represent child indexes, starting from 0. e.g. • [:tuple :int :int] has indexes 0 & 1 • [:maybe :int] has just index 0

(def schema [:map [:x [:tuple :int [:maybe [:map [:y [:and :int [:> 6]]]]]]]])

(mu/subschemas schemas)
;[{:path [], :in [], :schema [:map [:x [:tuple :int [:maybe [:map [:y [:and :int [:> 6]]]]]]]]}
; {:path [:x], :in [:x], :schema [:tuple :int [:maybe [:map [:y [:and :int [:> 6]]]]]]}
; {:path [:x 0], :in [:x 0], :schema :int}
; {:path [:x 1], :in [:x 1], :schema [:maybe [:map [:y [:and :int [:> 6]]]]]}
; {:path [:x 1 0], :in [:x 1], :schema [:map [:y [:and :int [:> 6]]]]}
; {:path [:x 1 0 :y], :in [:x 1 :y], :schema [:and :int [:> 6]]}
; {:path [:x 1 0 :y 0], :in [:x 1 :y], :schema :int}
; {:path [:x 1 0 :y 1], :in [:x 1 :y], :schema [:> 6]}]

(mu/get-in schema [:x 1 0 :y])
; [:and :int [:> 6]]

(mu/get-in *1 [0])
; :int

there are also utilities for going from between the two, mu/path->in and mu/in->paths (could be many)

Thanks! I'm trying to piece together something to derive different subscription and event paths for a re-frame db given my database schema. Given my pretty basic skillset, this is a bit of a stretch goal for me, but this is really helpful!

:in is the path in the value (e.g. the db), did you try that?