Results API

scinr.newton.results.DocumentResult dataclass

Result of processing a single document through a pipeline stage.

Attributes:

Name	Type	Description
document_name	str	The Neo4j document_name (or filename stem) of the processed document.
nodes_processed	int	Number of nodes (or files) successfully processed within this document. For stages 0-2, this is 1 for success, 0 for failure. For stages 3-4, this is the number of StructureNodes processed. For the tabular stage, this is the number of tables (sheets) written.
nodes_failed	int	Number of nodes (or files) that failed processing within this document. For the tabular stage, the number of sheets with errors.
errors	list[str]	List of error messages for this document. Empty on full success.

scinr.newton.results.StageResult dataclass

Aggregated result of running a single pipeline stage.

Attributes:

Name	Type	Description
stage	str	Stage identifier. One of: 'preprocess', 'extraction', 'ingestion', 'annotation', 'entity_extraction', 'tabular'.
success	bool	True if total_failed == 0 and no global errors occurred.
documents	list[DocumentResult]	Per-document results. One DocumentResult per file or document processed.
total_processed	int	Sum of nodes_processed across all DocumentResult entries.
total_failed	int	Sum of nodes_failed across all DocumentResult entries.
duration_seconds	float	Wall-clock time in seconds for the entire stage.
errors	list[str]	Global stage-level errors not attributable to a specific document.

scinr.newton.results.PipelineResult dataclass

Aggregated result of a full run_pipeline() invocation.

Attributes:

Name	Type	Description
success	bool	True only if every executed stage succeeded (no failures or errors).
total_duration_seconds	float	Total wall-clock time for the entire pipeline run.
stages_executed	list[str]	Ordered list of stage names that were actually run (skipped stages are not included).
preprocess	StageResult | None	StageResult for Stage 0, or None if the stage was not executed.
extraction	StageResult | None	StageResult for Stage 1, or None if the stage was not executed.
ingestion	StageResult | None	StageResult for Stage 2, or None if the stage was not executed.
annotation	StageResult | None	StageResult for Stage 3, or None if the stage was not executed.
entity_extraction	StageResult | None	StageResult for Stage 4, or None if the stage was not executed.
tabular	StageResult | None	StageResult for the tabular pipeline, or None if not executed.

scinr.newton.results.DeletionResult dataclass

Result of a delete_document() call — full Document + cascade + GC deletion.

Attributes:

Name	Type	Description
path	str | None	The Document path that was targeted for deletion, or None when the deletion was selected by job_id instead.
version	int | None	The specific version requested, or None if all versions were targeted.
job_id	str | list[str] | tuple[str, ...] | None	The job_id selector (one value or several) that was targeted, or None when the deletion was selected by path instead.
tenant_id	str | None	The tenant the deletion was scoped to (always applied), or None when it targeted public documents.
created_by_user_id	str | list[str] | tuple[str, ...] | None	The created_by_user_id filter applied to the match, or None if no such filter was requested.
found	bool	True if at least one matching Document existed before deletion. When False, all counters below are 0 and no delete or GC queries were executed.
versions_deleted	list[int]	Sorted list of integer versions that matched and were deleted. Empty when found is False.
documents_deleted	int	Number of :Document nodes deleted (the matched Document(s) plus any reached via IS_COMPOSED_OF*).
structure_nodes_deleted	int	Number of :StructureNode nodes deleted.
info_units_deleted	int	Number of :InfoUnit nodes deleted.
model_decisions_deleted	int	Number of :ModelDecision nodes deleted.
proposed_models_deleted	int	Number of :ProposedModel nodes deleted.
proposed_fields_deleted	int	Number of :ProposedField nodes deleted.
extraction_results_deleted	int	Number of :ExtractionResult nodes deleted.
gc_entity_model_instance_deleted	int	:Entity/:ModelInstance nodes this deletion left orphaned, and deleted. Only the nodes that hung from the deleted :ExtractionResult nodes are checked, not the whole tenant (see collect_orphans() for that).
gc_entity_model_instance_passes	int	Number of GC rounds run over :Entity/:ModelInstance candidates: 0 when there was none; one more each time a deleted node turned what it pointed at into candidates.
gc_labeled_entity_deleted	int	:LabeledEntity nodes this deletion left orphaned, and deleted.
gc_labeled_entity_passes	int	Number of GC rounds run over :LabeledEntity candidates.
raw_files_deleted	int	Number of raw_file_ids for which deletion was attempted against the storage backend (GridFS + metadata), for the raw_file_ids referenced by the deleted Document(s) and their descendants. Idempotent — includes ids that were already absent, since delete() returns None either way.
converted_pages_deleted	int	Number of ConvertedPageRecord (converted Markdown pages) deleted from the storage layer for the same raw_file_ids.

scinr.newton.results.OrphanCollectionResult dataclass

Result of a collect_orphans() call — sweep of every orphan of one tenant.

Attributes:

Name	Type	Description
tenant_id	str | None	The tenant that was swept (API value: None = public).
gc_entity_model_instance_deleted	int	:Entity/:ModelInstance nodes of the tenant deleted because no :ExtractionResult reached them.
gc_entity_model_instance_passes	int	Iterations of the Entity/ModelInstance pass (each one checks every node of the tenant; it stops at the first iteration that deletes nothing, capped at GC_MAX_PASSES).
gc_labeled_entity_deleted	int	:LabeledEntity nodes of the tenant deleted because nothing pointed at them.
gc_labeled_entity_passes	int	Iterations of the LabeledEntity pass (same rule).

scinr.newton.results.FreezeResult dataclass

Result of a freeze_document() call — snapshot export + graph reduction.

Attributes:

Name	Type	Description
path	str | None	The Document path targeted, or None when selected by job_id.
version	int | None	The specific version requested, or None if all versions were targeted.
job_id	str | list[str] | tuple[str, ...] | None	The job_id selector (one value or several) as passed by the caller, or None when selected by path.
tenant_id	str | None	The tenant the freeze was scoped to (API value: None = public).
created_by_user_id	str | list[str] | tuple[str, ...] | None	The created_by_user_id filter as passed, or None.
found	bool	True if at least one Document matched. When False, every counter is 0 and nothing was exported, stored or mutated.
mode	Literal['freeze', 'backup']	"freeze" (delete_after_export=True: the subtree was reduced and the Document(s) marked frozen=true) or "backup" (delete_after_export=False: only the snapshot was stored and last_backup_blob_id/last_backup_at set; the graph is intact).
frozen_blob_id	str | None	Id of the stored snapshot in the freeze backend (in both modes), or None when nothing was found.
versions_frozen	list[int]	Sorted versions of the Documents matched by the selector (IS_COMPOSED_OF* descendants not included).
documents_frozen	int	Number of :Document nodes in the snapshot — frozen (mode="freeze") or backed up (mode="backup") — including the IS_COMPOSED_OF* descendants of the matched ones.
structure_nodes_deleted	/ structure_nodes_kept	:StructureNode nodes removed / left in place (keep_structure_nodes=True).
info_units_deleted	int	:InfoUnit nodes removed (always, in mode="freeze").
model_decisions_deleted	/ model_decisions_kept	:ModelDecision nodes removed / kept (keep_annotations=True).
proposed_models_deleted	/ proposed_fields_deleted	:ProposedModel / :ProposedField nodes removed (keep_annotations=False).
extraction_results_deleted	/ extraction_results_kept	:ExtractionResult nodes removed / kept (keep_extraction_results=True).
gc_*	/ extraction_results_kept	Garbage-collection counters, same meaning as in DeletionResult (the orphans this freeze caused; always 0 with keep_extraction_results=True, which deletes no :ExtractionResult).

Every *_deleted / *_kept / gc_* counter is 0 in mode="backup".

scinr.newton.results.RestoreResult dataclass

Result of a restore_document() call — subtree rebuilt from its snapshot.

Attributes:

Name	Type	Description
path	str | None	The Document path targeted, or None when selected by job_id.
version	int | None	The specific version requested, or None if all versions were targeted.
job_id	str | list[str] | tuple[str, ...] | None	The job_id selector as passed by the caller, or None.
tenant_id	str | None	The tenant the restore was scoped to (API value: None = public).
created_by_user_id	str | list[str] | tuple[str, ...] | None	The created_by_user_id filter as passed, or None.
found	bool	True if at least one frozen Document matched. When False, every counter is 0 and nothing was read or written.
versions_restored	list[int]	Sorted versions of the frozen Documents matched by the selector (folder descendants restored with them not included).
documents_restored	int	Number of :Document nodes restored (matched ones plus the IS_COMPOSED_OF* descendants frozen in the same snapshot), stubs and recreated ones alike.
documents_recreated	int	How many of them no longer existed in the graph (deleted, or restored from a backup) and were recreated.
frozen_blob_ids	list[str]	Distinct snapshots the documents were restored from.
snapshots_deleted	int	Snapshots deleted from the freeze backend because no other Document of the tenant still references them.
nodes_created	int	Nodes created from the snapshot (recreated :Document nodes included).
nodes_reused	int	Snapshot nodes that already existed and were reused via MERGE (kept :StructureNode trees, shared :ModelInstance/:Entity/:LabeledEntity, catalog nodes are not counted).
relationships_created	int	Relationships recreated from the snapshot (those that already existed with the same properties are not duplicated).
gc_*	int	Garbage-collection counters of a sweep of the whole tenant (same meaning as in OrphanCollectionResult); 0 when the restore deleted no stale :ExtractionResult (the GC is then skipped).