Definition Extraction¶

The unformat parameter extracts structured values from a rule's cprmv:definition text using the Python parse library's pattern syntax.

When to use it¶

Rule definitions in Dutch law frequently embed domain-specific values in natural language:

een alleenstaande van 18, 19 of 20 jaar: € 337,98;

The unformat parameter allows you to pull these values out as named fields in the response, ready for direct use in downstream processing or decision model inputs — without needing a separate text-parsing step.

Pattern syntax¶

Use {fieldname:param_value} placeholders in the pattern string. The :param_value type annotation is required; it matches any non-empty string.

Pattern: {situatie:param_value}: € {norm:param_value}

Applied to BWBR0015703_2025-07-01_0, Artikel 20, lid 1, onderdeel a., whose definition is een alleenstaande van 18, 19 of 20 jaar: € 337,98;, the selected rule in the cprmv-json response becomes:

{
    "http://www.w3.org/1999/02/22-rdf-syntax-ns#type": "https://standaarden.open-regels.nl/standards/cprmv/0.4.2#Rule",
    "https://standaarden.open-regels.nl/standards/cprmv/0.4.2#id": "onderdeel a.",
    "https://standaarden.open-regels.nl/standards/cprmv/0.4.2#definition": "een alleenstaande van 18, 19 of 20 jaar: € 337,98;",
    "http://cprmv.open-regels.nl/situatie": "een alleenstaande van 18, 19 of 20 jaar",
    "http://cprmv.open-regels.nl/norm": "337,98;",
    "http://cprmv.open-regels.nl/rulesetid": "BWBR0015703_2025-07-01_0",
    "http://cprmv.open-regels.nl/rule_id_path": "BWBR0015703_2025-07-01_0, Artikel 20, lid 1, onderdeel a."
}

The named fields, plus the rulesetid and rule_id_path context fields, are added to the selected rule as triples. Their predicate URIs are formed from the namespace of the rule set's URI (here http://cprmv.open-regels.nl/) and the field name. In the full response this rule object sits under the rule set's cprmv:hasPart.

Behaviour when the pattern does not match¶

If the parse search does not find the pattern in the definition text, or the selected rule has no cprmv:definition, unformat has no effect and the standard response is returned without the extracted fields.

Non-breaking spaces¶

The API normalises cprmv:definition text before matching: every run of whitespace, including non-breaking spaces (\u00A0), is collapsed to a single regular space. This ensures patterns work consistently regardless of how the source XML encoded whitespace.

URL encoding the pattern¶

The unformat value must be URL-encoded in requests constructed manually:

Character	Encoded
`{`	`%7B`
`}`	`%7D`
`:`	`%3A`
	`%20`
`€`	`%E2%82%AC`

Example full URL:

GET /rules/BWBR0015703_2025-07-01_0%2C%20Artikel%2020%2C%20lid%201%2C%20onderdeel%20a.
    ?unformat=%7Bsituatie%3Aparam_value%7D%3A%20%E2%82%AC%20%7Bnorm%3Aparam_value%7D

FastAPI's Swagger UI handles encoding automatically.

Works with any output format¶

unformat works with any format: the values extracted from the definition are added as additional triples on the selected rule, so they appear in the RDF serialisations too (e.g. turtle, n3, json-ld). Combine unformat with format=turtle to get the extracted situatie/norm values as triples in the Turtle output.

Note

The interactive Swagger description of the unformat parameter still states that it implies cprmv-json; that text is stale — the cross-format behaviour above is what the API does.