Struct SeaOrmPipelineRunRepository 

pub struct SeaOrmPipelineRunRepository { /* private fields */ }

SeaORM-backed implementation of PipelineRunRepository.

Wraps a shared DatabaseConnection. All methods write or query the pipeline_runs table using the “new row per status transition” pattern, matching both Python’s writing pattern and the cross-SDK audit trail requirement.

Implementations

impl SeaOrmPipelineRunRepository

pub fn new(db: Arc<DatabaseConnection>) -> Self

Create a new repository backed by the given database connection.

Trait Implementations

impl PipelineRunRepository for SeaOrmPipelineRunRepository

fn log_pipeline_run<'life0, 'life1, 'async_trait>( &'life0 self, pipeline_run_id: Uuid, pipeline_id: Uuid, pipeline_name: &'life1 str, dataset_id: Option<Uuid>, status: PipelineRunStatus, run_info: Option<Value>, ) -> Pin<Box<dyn Future<Output = Result<Uuid, DatabaseError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Insert one row representing a status transition. Returns the new row’s primary key (pipeline_runs.id), which is a freshly generated UUIDv4.

fn latest_status<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, dataset_ids: &'life1 [Uuid], pipeline_name: &'life2 str, ) -> Pin<Box<dyn Future<Output = Result<HashMap<Uuid, PipelineRunStatus>, DatabaseError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Latest status per dataset for a given pipeline name.

fn list_recent<'life0, 'async_trait>( &'life0 self, dataset_id: Option<Uuid>, limit: u32, ) -> Pin<Box<dyn Future<Output = Result<Vec<PipelineRunRow>, DatabaseError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Recent runs for the activity router, with optional dataset filter.

fn list_recent_with_attribution<'life0, 'async_trait>( &'life0 self, dataset_id: Option<Uuid>, limit: u32, ) -> Pin<Box<dyn Future<Output = Result<Vec<PipelineRunWithAttributionRow>, DatabaseError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Recent runs with attribution (dataset + owner). Powers GET /api/v1/activity/pipeline-runs. Single SELECT joining pipeline_runs ⨝ datasets ⨝ users (LEFT JOIN both ways so orphaned runs whose dataset has been deleted still surface).

fn reset_orphans<'life0, 'life1, 'async_trait>( &'life0 self, reason: &'life1 str, ) -> Pin<Box<dyn Future<Output = Result<u64, DatabaseError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Restart-orphan reset: rewrite any row stuck in INITIATED / STARTED without a more recent successor to ERRORED with the given reason.

fn set_payload_field<'life0, 'life1, 'async_trait>( &'life0 self, run_id: Uuid, key: &'life1 str, value: Value, ) -> Pin<Box<dyn Future<Output = Result<(), DatabaseError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Upsert a single payload field for a run. Concurrent calls with the same (run_id, key) are last-write-wins per row; calls with different keys do not contend.

fn get_payload<'life0, 'async_trait>( &'life0 self, run_id: Uuid, ) -> Pin<Box<dyn Future<Output = Result<Map<String, Value>, DatabaseError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Read all payload fields for a run as a serde_json::Map. Returns an empty map (not None) when the run has no payload events; returns Err only on actual DB failures.

fn get_pipeline_run<'life0, 'async_trait>( &'life0 self, pipeline_run_id: Uuid, ) -> Pin<Box<dyn Future<Output = Result<Option<PipelineRun>, DatabaseError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Return the latest row for pipeline_run_id (ordered by created_at DESC).

fn get_pipeline_run_by_dataset<'life0, 'life1, 'async_trait>( &'life0 self, dataset_id: Uuid, pipeline_name: &'life1 str, ) -> Pin<Box<dyn Future<Output = Result<Option<PipelineRun>, DatabaseError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Return the latest run for (dataset_id, pipeline_name) by created_at.

fn get_pipeline_runs_by_dataset<'life0, 'async_trait>( &'life0 self, dataset_id: Uuid, ) -> Pin<Box<dyn Future<Output = Result<Vec<PipelineRun>, DatabaseError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Return one latest row per distinct pipeline_name that has runs for dataset_id. Result order is unspecified.