This Document contains information on how to implement reverse indexes and range scans in lambdaroyal-memory.

reverse index lookup and reverse range scans traverse the data from the key with the highest rank to the lowests. rank here refers to the pos of a document key within a sorting index|index set. So for monotonic (strictly goes here naturally since a key is unique) increasing numeric keys the newest document has the highest rank whereas the oldest document has the lowest rank.

Functions that should be able to perform the reverse lookup

range scan

(require '[lambdaroyal.memory.core.tx :refer :all])
(select tx :stock-order :reverse)

Additional benefits from this implementation

index scans that checks for equality

(require '[lambdaroyal.memory.core.tx :refer :all])
(select tx :stock-order [:ident] = ["gaga"])

index scans that presume just one attribute in an index

(require '[lambdaroyal.memory.core.tx :refer :all])
(select tx :stock-order :ident = "gaga")
(select tx :stock-order :ident >= "gaga")