Search Extracts

Extract paragraphs matching a term from cases matching filters.

Applies strict/exclude filters on authorities, then searches their paragraphs for the term. Single JOIN query — no intermediate ID set.

With an empty term ("citation mode"), returns the paragraphs that contain the cited authority instead (authority_citations.paragraph_ref → authority_chunks). Requires a 'cites' filter. Citing cases with no recorded pinpoint paragraph are omitted but counted in cases_without_pinpoint.

post/search/extracts

Request

Base URL: https://api.casenode.ai
URL: https://api.casenode.ai/search/extracts
Auth: none declared

Request body

termstring

alsostring[]

Additional terms — paragraph must match ALL

prec'exact' | 'fuzzy'

Precision for term: 'exact' = literal phrase, 'fuzzy' = typo-tolerant word match.

also_precstring[]

Per-term precision for also, aligned by index. Entries missing (shorter than also) default to 'exact'.

mode'keyword' | 'semantic'

group_by'case' | 'flat'

limitinteger

offsetinteger

target_courtstring nullable

sort'relevance' | 'citations' | 'date'

Order of the matching paragraphs, by their case, with the options /search takes: 'relevance' (how well the paragraph matches), 'citations' (most-cited case first), 'date' (newest judgment first). A non-relevance sort re-orders the best-matching paragraphs, as /search does.

cited_by_mininteger nullable

Only paragraphs from cases cited at least this many times. The same floor as a cited_by_min filter.

Example request

{
  "filters": [
    {
      "field": "court",
      "value": "EWCA-Civ"
    }
  ],
  "limit": 5,
  "term": "burden of proof"
}

Response

Successful Response

unknown required

Changes

No recorded changes to this operation across all 1 revision of this API.