Lazily initialize the tokenizer for token counting.

Get user information from Auth0 userinfo endpoint.

Advanced paragraph grouping using multiple heuristics.

Initialize docling's pre-trained models.

Initialize paths to downloaded models.

Initialize Nanonets OCR model from local cache.

Initialize all required models.

Check if text is a list item.

Check if text might be a table row.

Check if the text looks like a URL. Args: text: Text to check Returns: True if text look

Organize table data into proper structure using row/col indices from tf_responses.

Parse markdown content into sections based on headers.

Post-process text to improve readability.

Process blockquotes.

Process fenced code blocks.

Process a CSV file and return a conversion result. Args: file_path: Path to the CSV file to process

Process .doc files using pypandoc.

Process .docx files using python-docx with improved table extraction.

Process an Excel file and return a conversion result. Args: file_path: Path to the Excel file to process

Download and process a file from URL. Args: url: URL to download from file_info: Information about the file

Process markdown headers.

Process horizontal rules.

Process image file with OCR capabilities. Args: file_path: Path to the image file Returns:

Process ordered and unordered lists.

Process paragraphs by wrapping non-empty lines in <p> tags.

Process PDF file by converting to images and using OCR. Args: file_path: Path to the PDF file Return

Process .ppt files using pypandoc.

Process .pptx files using python-pptx.

Process markdown tables.

Process tables using the table structure model.

Process a web page URL. Args: url: URL to process Returns: ConversionResult containing t

Process PDF using OCR after converting pages to images.

Search for a query in the document.

Setup MCP server handlers.

Setup local processors based on GPU preferences.

Start local server to handle OAuth callback with limited ports for Auth0 whitelist.

Analyze the content structure of the markdown.

Calculate area of the element.

Perform browser-based authentication. Args: force_reauth: Force re-authentication even if cached credentials exi

Check if this processor can handle the given file. Args: file_path: Path to the file to check Return

Check if the processor can handle the file.

Clear cached authentication credentials.

Create DocumentExtractor with proper error handling for processing mode.

Create additional sample files for testing.

Create a sample text file for testing.

Download models from HF and package for S3.

Download models synchronously before starting the app.

Download sample files for each supported format using reliable sources.

Download sample files for each supported format.

Exchange authorization code for access token directly with Auth0.

Convert markdown text to HTML. Args: markdown_text: The markdown content to extract Returns:

Export tables as CSV format. Args: table_index: Which table to export (0-based index). Default is 0 (first table).

Export as structured JSON using Nanonets model with specific prompt.

Export as structured JSON with optional field extraction. Args: specified_fields: Optional list of specific fields to ext

Extract important fields and their values from document content using Ollama. Args: markdown_content: Raw markdown conten

Extract specified fields from markdown content. Args: markdown_content: The markdown content to extract from

Export as HTML.

Extract text from image using Nanonets OCR.

Extract text using Nanonets OCR.

Extract text with layout awareness using Nanonets OCR.

Convert a URL to internal format. Args: url: URL to extract Returns: ConversionResult co

Extract data according to a JSON schema from markdown content. Args: markdown_content: The markdown content to extract fr

Generate a detailed test report.

Generate a comprehensive test report.

Get list of available OCR providers. Returns: List of available provider names

Get detailed GPU information. Returns: Dictionary with GPU information

Handle logout command.

Join text words into proper sentences and paragraphs.

Main CLI function.

CLI entry point for authentication.

Main setup function.

Main test function.

Main test function.

Main test function.

Main test function.

Run the test.

Main test function.

Main test function.

Run all tests.

Pre-download OCR models by running a dummy prediction.

Print supported formats in a nice format.

Print version information.

Process image file with OCR capabilities. Args: file_path: Path to the image file Returns:

Create a lazy CloudConversionResult that will make API calls on demand. Args: file_path: Path to the file to process

Process a single input item and return result with metadata.

Get the Nanonets processor.