The file to text node allows you to parse text from a file.

​
Node Inputs

  1. File to parse: The file you want to parse text from.
    • Type: File

For providing the file, you have two options:

  1. If toggle is on Upload: Upload a file by clicking the upload button
  2. If toggle is on Variable: Reference files from other nodes.

​
Node Parameters

On the face of the node:

  1. File Parser: The file parser you want to use to parse the inputted file. The available options are: Default (Basic OCR), Llama Parse, and Textract.

In the gear:

  1. Chunk Text: Chunk the text with the below dimensions
  2. Chunk Size (if Chunk Text is set to True): The size of each chunks in tokens (1 token = ~4 characters; default is 1024 tokens)
  3. Chunk Overlap (if Chunk Text is set to True): The number of tokens of overlap between chunks (default is 400 tokens)

​
Node Outputs

  1. Processed Text: The text from the file
    • Type: Text
    • Example usage: {{file_to_text_0.processed_text}}

​
Example

The below example shows a pipeline that takes a pdf file called _10_K-Q4-2023-As-Filled.pdf containing financial statements and parses the text from it.

  1. Input Node: Contains the file you want to parse text from.
  2. File to Text Node: Parses the text from the file.
    • File to parse: {{input_0.file}}
  3. Output: The parsed text
    • Output: {{file_to_text_0.processed_text}}