Module drive_document_average_query 

A query to compute the average of an integer property across documents using CountSumTree / ProvableCountProvableSumTree (PCPS) elements. Averages are NOT computed server-side; the response carries a (count, sum) pair (atomic per group) and the client divides. See book/src/drive/average-index-examples.md for the worked example contract. DriveDocumentAverageQuery — Drive’s average-query surface.

Parallels crate::query::drive_document_sum_query for the averaging surface. Averages are NOT computed server-side: every response carries a (count, sum) pair (atomic per group), and the client divides to obtain the average. This preserves full precision and lets callers pick their own representation (integer-truncated, floating-point, decimal); pre-dividing on the server would force a choice that loses information.

The grovedb primitive backing this is AggregateCountAndSumOnRange (added in grovedb PR 670 alongside the ProvableCountProvableSumTree / PCPS element variant) — one root-hash-committed traversal returns both metrics together. See book/src/drive/average-index-examples.md for the design and the grades-contract worked example.

Wired end-to-end: the dispatcher routes prove-true requests to the PCPS / primary-key proof executors, and prove-false requests to the joint single-walk count-and-sum dispatcher at crate::query::drive_document_count_and_sum_query. Both paths read (count, sum) from each visited count-sum-bearing element in a single grovedb walk — see the drive_dispatcher module docstring for the routing details and crate::query::drive_document_count_and_sum_query for the no-prove perf / atomicity contract.

Modules

drive_dispatcher

Average-query dispatcher entry point.

Structs

AverageEntry

A single per-key average entry. Carries BOTH count and sum so the client can divide; the server intentionally doesn’t pre-divide (see the module docstring).

DocumentAverageRequest

Server-side request input for the average dispatcher. Mirrors crate::query::drive_document_sum_query::DocumentSumRequest — same fields, same semantics; the response carries (count, sum) pairs instead of a single sum.

Enums

AverageMode

What kind of average-query the dispatcher should run. Parallels crate::query::drive_document_sum_query::SumMode.

DocumentAverageResponse

Server-side response from the average dispatcher. Parallels crate::query::drive_document_sum_query::DocumentSumResponse — same outer shape; the Aggregate and Entries payloads carry (count, sum) instead of just sum.