execute

Execute reactive resolution

Description

Resolve one or more output resource specifications through the reactive execution engine.

Return the current status of an asynchronous execution started through workspace$.internal$resolver$resolve.async().

Return a compact execution summary for an asynchronous run.

Return the tracked execution dependency graph and node states.

Return the most recently started execution-engine run tracked by workspace.

Usage

execute(
  outputs,
  inputs = NULL,
  .execute = TRUE,
  .transform = NULL,
  .action = NULL,
  .status = NULL
)

execute.status(execution)

execute.summary(execution)

execute.graph(execution)

execute.last()

Arguments

`outputs`	Output specification(s): single `ResourceSpec`, list/vector of `ResourceSpec`, or `ResourceSpecSet`.
`inputs`	Optional input specification(s) that are available to the execution engine.
`.execute`	Logical; when `TRUE` run execution-engine resolution, otherwise return requested specs without scheduling.
`.transform`	Optional function applied to the loaded output value list before returning (only when `.execute = TRUE`).
`.action`	Optional execution-engine action label used in live execution status.
`.status`	Optional execution-status mode override (`“auto”`, `“yes”`, `“no”`). Defaults to `options(“PolyGenius.execution.status”)`.
`execution`	Execution object returned by `workspace$.internal$resolver$resolve.async()`.

Details

When .execute = FALSE, this helper does not schedule work and returns a ResourceSpecSet containing requested outputs plus any provided inputs.

Value

When .execute = TRUE, resolved output resource value(s) loaded from cache (single value for one output, list for multiple outputs). When .execute = FALSE, a ResourceSpecSet containing ⁠.arg outputs⁠ and ⁠.arg inputs⁠ (if supplied).

A single status string (“pending”, “running”, “completed”, or “failed”).

A named list with execution-engine progress fields such as status, queued/running counts, and failure details.

A named list with:

nodes: node table with id, type, label, status, and reason.
edges: dependency edges (from -> to with input name).
blocked.by: subset of edges explaining blocked nodes via failed/blocked upstream nodes.

Execution object or NULL when no execution has run yet.