Skip to Content
This documentation is provided with the HEAT environment and is relevant for this HEAT instance only.
RunnersSystem Utilsnode-output-query

node-output-query

Platform reserved (system-utils only). Runs a read-only query against the Core API PostgreSQL database and emits a single JSON artifact listing all matching NodeOutput IDs. There is no pagination in v1: large result sets (for example 170k+ IDs) are returned in one payload for downstream bulk ingest.

When to use it

Use as the discovery step before system-bulk-tabular-writer:

node-output-query → system-bulk-tabular-writer

Typical criteria: all outputs from an upload node named csv-input or json-input, optionally scoped by project, session template, simulation name, or session creation window.

Configuration

KeyRequiredPurpose
sourceNodeNameyesNode instance name to match (for example csv-input).
sessionTemplateNamenoCase-insensitive match on parent session template name.
projectId / projectNamenoScope to one project.
sessionCreatedFrom / sessionCreatedTonoISO datetimes; filter on session createdAt.
simulationNamenoCase-insensitive match on Session.SimulationName.
includeStaticBackedSessionsnoDefault false; excludes sessions backing active platform static instances.

Output shape

{ "sourceNodeName": "csv-input", "criteria": { "sourceNodeName": "csv-input" }, "totalCount": 172000, "outputIds": [1, 2, 3] }

The artifact is stored in HEAT Managed Object Store like other system-utils JSON outputs.

Limitations

  • Queries core Postgres only; does not connect to optional bulk analytics Postgres.
  • Not for trainee-facing dashboards; platform/static bulk ingest pipelines only.
  • Very large outputIds arrays produce a large JSON blob and long downstream runtime.