Search library

​

Searches reference metadata, PDF full text, and annotations in your Paperpile library. Use plain terms, quoted phrases, field prefixes (title:, author:, abstract:, journal:, year:, publisher:, and pdf:), or not: exclusions. Each result carries referenceId, an available citekey, and a resultType: reference_metadata includes matched metadata and matchedFields; fulltext and pdf_annotation include a highlighted snippet plus file and page IDs. fulltext results include charOffsets, half-open UTF-8 byte bounds in normalized PDF page text for get_text_content; annotation results include annotationId and no offsets. A full-text passage whose file cannot be identified is returned as reference_metadata with its snippet and charOffsets. A non-empty q is required. folder_id and label_id can scope the search but cannot be combined, and search_full_text=false restricts matching to metadata.

Query Parameters

  • q
    Type: string
  • folder_id
    Type: string
  • label_id
    Type: string
  • limit
    Type: integer
    min:  
    1
    max:  
    100

    Integer numbers.

  • offset
    Type: integer
    min:  
    0

    Integer numbers.

  • count
    Type: "include" or "skip" or "only"enum
  • include_trashed
    Type: boolean
  • search_full_text
    Type: boolean

Responses

  • 200

    Search results include data and resultCount, plus total when count=include; count=only returns only the total. total counts matching references, while resultCount counts returned entries. Entries contain matched metadata or highlighted snippets with available location fields.

  • application/json
  • application/json
  • application/json
  • application/json
Request Example for get/search
curl 'https://stage-api.paperpile.com/v1/search?limit=20&offset=0&count=include&include_trashed=false&search_full_text=true' \
  --header 'Authorization: Bearer YOUR_SECRET_TOKEN'
No Body

Search results include data and resultCount, plus total when count=include; count=only returns only the total. total counts matching references, while resultCount counts returned entries. Entries contain matched metadata or highlighted snippets with available location fields.