`get_document_text`

Document text · salt-okunur

Extract text from a PDF/HTML/text attachment (in memory; nothing is stored). Binary
    formats (xlsx, zip) and scanned PDFs return `text: null` with a reason. Prefer
    search_documents to pull only the relevant part of long reports.

Parametreler

ad	tip	zorunlu	varsayılan	açıklama
`attachment_id`	`string` minLength=1	evet	—	From get_disclosure_documents / get_disclosure attachments
`pages`	`string`	—	—	PDF page selection like '1-3,7' (1-based). Default: all pages, subject to max_chars.
`max_chars`	`integer` maximum=200000, minimum=1000	—	`30000`

Girdi şeması

{
  "properties": {
    "attachment_id": {
      "description": "From get_disclosure_documents / get_disclosure attachments",
      "minLength": 1,
      "title": "Attachment Id",
      "type": "string"
    },
    "pages": {
      "anyOf": [
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "description": "PDF page selection like '1-3,7' (1-based). Default: all pages, subject to max_chars.",
      "title": "Pages"
    },
    "max_chars": {
      "default": 30000,
      "maximum": 200000,
      "minimum": 1000,
      "title": "Max Chars",
      "type": "integer"
    }
  },
  "required": [
    "attachment_id"
  ],
  "title": "get_document_textArguments",
  "type": "object"
}

Çıktı şeması

{
  "additionalProperties": true,
  "title": "get_document_textDictOutput",
  "type": "object"
}