Struct RunKickRequest 

pub struct RunKickRequest {
    pub init_ctx_override: Option<Value>,
    pub task_input_override: Option<TaskInputSpec>,
    pub timeout_secs: Option<u64>,
    pub detach: bool,
    pub operator_sid: Option<String>,
}

Request body for POST /v1/tasks/:id/runs (issue #19 ST4) — every field is optional, and the body itself is optional (see task_rekick’s Option<Json<Self>> parameter); a caller that sends no body, or {}, or omits a field gets exactly today’s rekick behavior for that layer.

Fields

init_ctx_override: Option<Value>

Per-Run override for the flow-ir initial ctx. Merged on top of TaskRecord.input_ctx (itself already merged on top of Blueprint.default_init_ctx at original launch time) via merge_init_ctx_3layer — Run wins on key collision, same shallow-merge / non-Object-fully-replaces rule as every other layer in the cascade. None (absent field, or no body at all) is a no-op: the BP+Task merge alone seeds this kick, identical to pre-#19 rekick.

task_input_override: Option<TaskInputSpec>

Per-Run override for the Task-level canonical fields (project_root / work_dir / task_metadata). None falls back to TaskRecord.task_input_spec (the spec resolved and snapshotted at original POST /v1/tasks time); Some replaces it wholesale for this kick only — the stored TaskRecord.task_input_spec is never mutated by a rekick.

timeout_secs: Option<u64>

Per-Run ceiling (seconds) for this kick’s synchronous dispatch await (issue #35 ST3 — GH #33 Guard 2 parity). Some(0) is rejected (400). None falls back to AppState.sync_timeout_secs (the server-wide default), same cascade as TaskLaunchRequest.timeout_secs (lib.rs:818-826).

detach: bool

GH #37: opt into the detached (asynchronous) rekick — same semantics as TaskLaunchRequest.detach. false (default) keeps the synchronous dispatch; true spawns the flow eval as a detached background task bounded by the run TTL alone and returns 202 Accepted with status: "running" immediately. Mutually exclusive with timeout_secs (400 when combined).

operator_sid: Option<String>

Per-Run pin to a live Operator session (rekick parity with POST /v1/tasks’ operator_sid — see crate::TaskLaunchRequest::operator_sid for the full disconnected-vs-unknown / last-write-wins contract). Resolved before any Task/Run store write: an unknown sid fails fast with a 400, never silently falling back to the BP-level alias lookup. Some(sid) becomes this kick’s operator_backend_id (this handler carries no other Operator-override field) and is persisted verbatim into RunRecord.operator_sid; None (absent field, or no body at all) preserves the pre-existing Operator-default rekick path byte-for-byte.

Trait Implementations

impl Debug for RunKickRequest

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for RunKickRequest

fn default() -> RunKickRequest

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for RunKickRequest

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl JsonSchema for RunKickRequest

fn schema_name() -> Cow<'static, str>

The name of the generated JSON Schema.

fn schema_id() -> Cow<'static, str>

Returns a string that uniquely identifies the schema produced by this type.

fn json_schema(generator: &mut SchemaGenerator) -> Schema

Generates a JSON Schema for this type.

fn inline_schema() -> bool

Whether JSON Schemas generated for this type should be included directly in parent schemas, rather than being re-used where possible using the $ref keyword.