Struct Stmts 

pub struct Stmts {

    pub upsert: String,
    pub select_body: String,
    pub select_status: String,
    pub select_submitted: String,
    pub delete: String,
    pub list: String,
    pub purge_runs: String,
    pub purge_idem: String,
    pub select_orphans: String,
    pub renew_leases: String,
    pub insert_idem: String,
    pub select_idem: String,
    pub takeover_idem: String,
    pub delete_idem_by_run: String,
    pub select_pending: String,
    pub claim_one: String,
    pub reclaim_select: String,
    pub reclaim_requeue: String,
    pub reclaim_fail: String,
    pub finalize_owned: String,
    pub cancel_pending: String,
    pub request_cancel: String,
    pub pending_cancellations: String,
    pub heartbeat_instance: String,
    pub live_instances: String,
    pub prune_instances: String,
    pub insert_shard: String,
    pub claim_shards_select: String,
    pub claim_shard_one: String,
    pub renew_shard_leases: String,
    pub reclaim_shards_select: String,
    pub reclaim_shard_requeue: String,
    pub reclaim_shard_fail: String,
    pub finalize_shard: String,
    pub shard_progress: String,
    pub pending_shard_cancellations: String,
    pub select_sharded_parents: String,
    pub finalize_sharded_parent: String,
    pub delete_shards_by_run: String,
    pub purge_orphan_shards: String,
}

Available on crate feature serve and (crate features serve-history-postgres or serve-history-sqlite) only.

Prepared-statement text for a backend, built once per dialect at connect time.

Fields

upsert: String

(cancel_requested is intentionally NOT written by upsert — it is set only via request_cancel and cleared by reclaim_requeue; it defaults to NULL on insert.)

select_body: String

select_status: String

select_submitted: String

delete: String

list: String

purge_runs: String

purge_idem: String

select_orphans: String

Select non-terminal runs whose owning instance’s lease has expired (or is unset) — the orphans this instance may safely fail. Param: now.

renew_leases: String

Extend the lease of this instance’s own non-terminal runs (heartbeat). Params: new_lease_expiry, instance_id.

insert_idem: String

select_idem: String

takeover_idem: String

delete_idem_by_run: String

Delete the idempotency claim(s) that point at a given run — used when a run is deleted so a replay of the key starts fresh rather than 404-ing on the missing record (#146 M8). Scoped by run_id, so a newer run that re-claimed the same key keeps its claim.

select_pending: String

Cluster dispatcher: fetch oldest pending runs up to a given limit.

claim_one: String

Cluster dispatcher: atomically claim a pending run (set owner + running).

reclaim_select: String

Cluster reclaimer: select expired running runs for requeue/fail evaluation. NOTE: 'queued' is the single-instance status; cluster runs flow pending → running, so the failover reclaimer covers 'running' only.

reclaim_requeue: String

Cluster reclaimer: requeue an expired running run back to pending.

reclaim_fail: String

Cluster reclaimer: fail an expired running run that cannot be requeued.

finalize_owned: String

Finalize a run owned by this instance (terminal status update).

cancel_pending: String

Cancel a pending run directly (transition pending → cancelled).

request_cancel: String

Request cancellation of an in-flight run owned by another instance.

pending_cancellations: String

List run IDs owned by this instance that have a pending cancellation request.

heartbeat_instance: String

Upsert this instance’s membership heartbeat into faucet_serve_instances.

live_instances: String

List instances whose last heartbeat is at or after a given threshold.

prune_instances: String

Prune instances whose last heartbeat is before a given threshold.

insert_shard: String

Idempotent shard insert (ON CONFLICT (run_id, shard_id) DO NOTHING).

claim_shards_select: String

Select claimable pending shards joined to their run body, largest first.

claim_shard_one: String

Atomically claim one pending shard for this instance.

renew_shard_leases: String

Heartbeat this instance’s running shards.

reclaim_shards_select: String

Select expired-lease running shards for requeue/fail evaluation.

reclaim_shard_requeue: String

Requeue an expired running shard back to pending (attempt++).

reclaim_shard_fail: String

Fail an expired running shard that exhausted its attempts (poison).

finalize_shard: String

Owner-fenced terminal write for one shard.

shard_progress: String

Status counts for a run’s shards.

pending_shard_cancellations: String

Distinct run_ids for which THIS instance owns a running shard whose parent run has a pending cancellation request (cross-instance shard cancel, F10). Param: instance_id.

select_sharded_parents: String

Select run_ids of sharded parents (candidates to finalize once all their shards are terminal, F11).

finalize_sharded_parent: String

Status-fenced terminal write for a sharded parent (F11). A benign double-finalize across instances is a no-op: the guard requires the parent to still be sharded. Does NOT re-arm owner/lease.

delete_shards_by_run: String

Delete a run’s shard rows (paired with delete so a deleted run leaves no orphaned shard rows behind, F25). Param: run_id.

purge_orphan_shards: String

Purge shard rows whose parent run no longer exists (run-record purged by retention, F25). No params — a set-difference against faucet_serve_runs.

Implementations

impl Stmts

pub fn new(dialect: Dialect) -> Self