Unstructured File Loader

​

Node Details

• Name: UnstructuredFile_DocumentLoaders
• Type: Document
• Category: Document Loaders
• Version: 3.0

​

Input Parameters

​

Main Parameters

1. File Path (optional)

   • Type: string
   • Description: Path to the file to be processed. This will be deprecated in future releases.

2. Files Upload

   • Type: file
   • Description: Files to be processed. Multiple files can be uploaded.
   • Supported file types: .txt, .text, .pdf, .docx, .doc, .jpg, .jpeg, .eml, .html, .htm, .md, .pptx, .ppt, .msg, .rtf, .xlsx, .xls, .odt, .epub

3. Unstructured API URL

   • Type: string
   • Default: “http://localhost:8000/general/v0/general”
   • Description: The URL for the Unstructured API.

4. Strategy

   • Type: options
   • Options: Hi-Res, Fast, OCR Only, Auto
   • Default: “auto”
   • Description: The strategy to use for partitioning PDF/image.

​

Additional Parameters

5. Encoding (optional)

   • Type: string
   • Default: “utf-8”
   • Description: The encoding method used to decode the text input.

6. Skip Infer Table Types (optional)

   • Type: multiOptions
   • Default: [“pdf”, “jpg”, “png”]
   • Description: Document types to skip table extraction with.

7. Hi-Res Model Name (optional)

   • Type: options
   • Options: chipper, detectron2_onnx, yolox, yolox_quantized
   • Default: “detectron2_onnx”
   • Description: The name of the inference model used when strategy is hi_res.

8. Chunking Strategy (optional)

   • Type: options
   • Options: None, By Title
   • Default: “by_title”
   • Description: Strategy to chunk the returned elements.

9. OCR Languages (optional)

   • Type: multiOptions
   • Description: The languages to use for OCR.

10. Source ID Key (optional)

    • Type: string
    • Default: “source”
    • Description: Key used to get the true source of document.

11. Coordinates (optional)

    • Type: boolean
    • Default: false
    • Description: If true, return coordinates for each element.

12. XML Keep Tags (optional)

    • Type: boolean
    • Description: Whether to keep XML tags in the output.

13. Include Page Breaks (optional)

    • Type: boolean
    • Description: When true, include page break elements when the filetype supports it.

14. Multi-Page Sections (optional)

    • Type: boolean
    • Description: Whether to treat multi-page documents as separate sections.

15. Combine Under N Chars (optional)

    • Type: number
    • Description: Combine elements until a section reaches a length of n chars.

16. New After N Chars (optional)

    • Type: number
    • Description: Cut off new sections after reaching a length of n chars (soft max).

17. Max Characters (optional)

    • Type: number
    • Default: 500
    • Description: Cut off new sections after reaching a length of n chars (hard max).

18. Additional Metadata (optional)

    • Type: json
    • Description: Additional metadata to be added to the extracted documents.

19. Omit Metadata Keys (optional)

    • Type: string
    • Description: List of metadata keys to omit from the output.

​

Output

The node outputs an array of IDocument objects, each representing a processed document with its content and metadata.

​

Credentials

• Unstructured API Key (optional): API key for accessing the Unstructured.io API.

​

Use Cases

1. Extracting text and structure from complex document formats like PDFs or scanned images.
2. Processing multiple files of different types in a single workflow.
3. Preparing unstructured data for further NLP tasks such as summarization, question answering, or information extraction.

​

Notes

• This node is particularly useful for handling documents with complex layouts or mixed content types.
• The Unstructured.io API provides advanced document processing capabilities, including OCR for images and scanned documents.
• Users can fine-tune the processing by