Internal Current Implementation Details

Source and installed layout

The R package installs one runtime extension artifact for each declared exact DuckDB engine release (currently v1.5.0 through v1.5.4, generated from the build manifest) and selects the matching version at runtime. The user-facing policy and diagnostics are covered in DuckDB Version Compatibility:

rducks_extension/build/v1.5.4/rducks.duckdb_extension

The source and vendored dependency inputs live in the source checkout under tools/ext, not under inst/rducks_extension:

tools/ext/rducks_extension.c
tools/ext/src/*.c
tools/ext/duckdb_capi/v*/*.h
tools/ext/third_party/*

configure and configure.win build the extension variants from tools/ext and copy only the runtime artifacts into the installed package footprint. Because Rducks uses C_STRUCT_UNSTABLE, rducks_enable() queries the target connection and loads only an artifact for that exact DuckDB engine version.

Package and extension split

Rducks is split across R code and a loaded DuckDB extension:

• R code validates descriptors, records connection-local defaults, preserves R closures, starts managed IPC workers, and exposes diagnostics.
• The DuckDB extension registers SQL functions, stores native catalog metadata, owns extension-side connections, handles DuckDB callbacks, and moves data through DuckDB vectors directly or through Quack wire-codec byte payloads.

DuckDB function kind, Rducks scalar-UDF evaluation mode, and scalar-UDF execution plan are separate axes. Aggregates and table functions do not run through the scalar-UDF execution-plan matrix.

Runtime scopes

There are three important scopes:

• R process/package scope: recorded R-thread token, process-local provider store, release queues, weak-reference finalizers, and diagnostic helpers.
• DuckDB database runtime/catalog scope: SQL functions, evaluator handles, preserved closures, scalar-UDF evaluator/transport metadata, native runtime backend, and counters.
• DBI connection attachment scope: default execution plan for future registrations, finalizer bookkeeping, and the R-side registry view.

rducks_release(con) releases the connection attachment. It is not an SQL unregister operation.

R-thread discipline

DuckDB worker threads must not call the R API. Rducks records the calling R thread when the extension is enabled. Paths that need R evaluation either run on that recorded thread or use an explicit cross-process worker plan.

For in-process concurrent scalar-UDF execution, DuckDB worker callbacks submit synchronous requests to an extension-owned queue. The recorded R thread drains that queue, performs all R API work, and hands owned result data back to the waiting worker-side callback. This is a safety and liveness mechanism, not parallel R execution.

Native destructors that discover preserved R objects on the wrong thread queue release work. Safe drain points include enable/release calls, registration, scalar-UDF execution, and metadata/stat queries.

Scalar UDF registration and execution

rducks_register_scalar_udf() creates DuckDB scalar UDF catalog entries and stores native metadata for the selected transport. Declared args pin the SQL input signature. Omitted args registers a DuckDB varargs ANY function; during DuckDB bind, Rducks records the concrete logical argument types for that call and uses those effective types during execution.

The current scalar-UDF data paths are:

inproc (direct):
  DuckDB chunk -> native C materialization to SEXP -> R closure
  -> native C result conversion -> DuckDB output

ipc (wire):
  DuckDB chunk -> Quack wire bytes -> NNG worker process
  -> R closure in worker -> Quack wire result bytes -> DuckDB output

The in-process direct path is the reference implementation. The ipc path must match its SQL type, NULL, error, and result semantics; unsupported combinations fail instead of falling back. Before any worker result is written back, the wire decoder validates the decoded column count and per-column types against the declared signature.

VARIANT is runtime-gated rather than version-gated. The extension asks the SQL binder for the canonical type (SELECT NULL::VARIANT), validates its physical STRUCT(keys, children, values, data) metadata, inspects a populated query-produced VARIANT chunk, and creates an owned chunk to verify that generic struct/list vector accessors expose every nested child on both execution shapes. The canonical type is cloned for UDF signatures; both direct and wire marshallers reuse a synthetic descriptor for the validated physical storage. A missing or changed layout leaves the capability disabled, including for nested and dynamic signatures.

Aggregates

rducks_register_aggregate() registers DuckDB aggregate functions. Aggregate state is an R object preserved by the extension, not an untyped pointer to R memory inside DuckDB. Row-wise callbacks use update(state, ...), combine(left, right), and finalize(state). Chunk callbacks are available for batch-oriented state updates.

Aggregate callbacks require the recorded R thread. They are intentionally not controlled by scalar-UDF execution plans.

Table functions

rducks_register_table() infers the SQL argument count from the R function’s formals, registers those inputs as DuckDB ANY, converts actual SQL bind values to R values, and calls the R function during DuckDB bind on the recorded R thread.

A finite result is materialized once during bind by filling DuckDB output vectors directly from the returned R columns. A rducks_table_stream() result uses a bind-time prototype for schema and a scan-time next_batch(batch_size) callback for successive batches. Projection-aware copying writes only requested columns into DuckDB output chunks.

Query streams

rducks_query_stream() is an R-side consumer API, not a table function. It uses DuckDB’s native streaming-result/data-chunk APIs through a dedicated extension-owned query-stream connection. That dedicated connection keeps query streaming separate from the runtime connection used for dynamic scalar, table, and aggregate registration.

A query stream fetches DuckDB chunks and materializes each one into a data-frame batch directly from DuckDB vectors. The current implementation supports one active native query stream per caller/runtime connection.

IPC provider lifecycle

The native IPC provider uses NNG for request/reply and local mirai processes for the default managed worker lifecycle. During ipc registration, Rducks starts workers, launches the worker loop, pings endpoints, and registers the UDF payload with each worker.

Client pools are native extension objects; worker lifecycle is R-side provider state. rducks_release(con) closes local client pools when the last attachment to a runtime is released, then stops Rducks-managed local workers and removes the provider. Caller-supplied external endpoints are not stopped by Rducks.

Rducks does not call NNG’s global nng_fini() during ordinary runtime cleanup. Socket and pool shutdown are the quiesce points so an R session can release one connection and later register more IPC UDFs.

Ownership and copying model

The implementation favors explicit ownership boundaries:

• Borrowed DuckDB vectors and chunks are used only inside the callback frame that supplied them.
• Quack wire payloads are owned byte buffers and are used for process/transport boundaries.
• R closures are preserved while native metadata can call them.
• Same-host mori sharing is available only for selected long-lived globals, not for SQL chunk data.
• Output vectors and chunks are copied into owned R or DuckDB containers before the borrowed source lifetime ends.

These rules are part of the package semantics. Changes that add allocations, thread hops, or zero-copy paths need a protection and lifetime audit.