API Reference

Public functions

`openbasement.extract(graph, template, entity=None, transforms=None, merge_same_as=None)`

Extract structured data from an RDF graph using a template.

Parameters:

Name	Type	Description	Default
`graph`	`Graph`	An rdflib Graph loaded with RDF data.	required
`template`	`str \| dict`	Template source -- built-in name (e.g. "eu_procedure"), path to YAML file, or a dict.	required
`entity`	`str \| None`	Optional entity name to extract. If None, extracts the first (root) entity defined in the template.	`None`
`transforms`	`dict \| None`	Optional dict of custom transform name -> callable. These are merged with built-in transforms (custom takes precedence). Templates reference transforms by name via the "transform" field option.	`None`
`merge_same_as`	`bool \| None`	If True, group owl:sameAs-equivalent instances and merge their triples into one entity. If False, extract each URI as a separate entity. If None (default), uses the template's same_as_merge setting (which defaults to True).	`None`

Returns:

Type	Description
`list[dict]`	List of extracted entity dicts.

`openbasement.load_template(source)`

Load and normalize a template from various sources.

Parameters:

Name	Type	Description	Default
`source`	`str \| Path \| dict`	One of: - str name of a built-in template (e.g. "eu_procedure") - Path to a YAML file - dict with template content	required

Returns:

Type	Description
`dict`	Normalized template dict.

`openbasement.list_builtin_templates()`

List available built-in template names.

`openbasement.audit`

Drift detection: compare template predicates against actual graph content.

`audit(graph, template)`

Compare a template against actual predicates in a graph.

For each entity type in the template, finds all instances and checks: - Which template predicates are missing from the graph - Which graph predicates are not covered by any template field or relation - Overall coverage percentage

Parameters:

Name	Type	Description	Default
`graph`	`Graph`	An rdflib Graph to audit.	required
`template`	`dict`	A normalized template dict (output of load_template).	required

Returns:

Type	Description
`dict[str, Any]`	Dict with keys: "entities": per-entity audit results "summary": overall coverage stats

Built-in transforms

Name	Description
`year_from_date`	Extracts the first 4 characters (year) from a date string, e.g. `"2023-07-05"` -> `"2023"`
`uri_local_name`	Extracts the local name from a URI after `#` or `/`, e.g. `"http://example.org/foo#Bar"` -> `"Bar"`

`openbasement.transforms.apply_transform(value, transform_name, custom_transforms=None)`

Apply a named transform to a value.

Looks up the transform name in custom_transforms first, then built-ins. Raises ValueError if the name is not found in either.