Skip to main content
function extractStructuredDataFromFile(file, options): Promise<any>

​
Examples

import { extractStructuredDataFromFile } from "@intuned/sdk/ai-extractors";

const movie = await extractStructuredDataFromFile({
  source: {
    type: "url",
    data: "<file url>"
  },
  type: "pdf",
  // pages array is optional, do not pass it if you want to include all pages in the process
  pages: [1, 2]
}, {
  label: "extract_movie",
  dataSchema: {
    type: "object",
    properties: {
      "name": {
        type: "string",
        description: "movie name"
      },
      revenue: {
        type: "string",
        description: "movie revenue"
      }
    }
  }
})

​
Parameters

• file: ImageFile | PdfFile | SpreadsheetFile the file you want to extract the data from, • options • options.dataSchema: JsonSchema the json schema of the data you’re trying to extract. • options.label: string a label for this extraction process, used for billing and monitoring • options.prompt?: string optional, a prompt to guide the extraction process and provide more context. • options.strategy?: MarkdownFileStrategy | ImageFileStrategy optional, the strategy to use for extraction. use IMAGE if the info you’re trying to extract is visual and cannot be converted to markdown. Defaults to MARKDOWN strategy with gpt4-turbo model.

​
Returns

Promise<any>