API Reference¶

Detailed documentation for Kreuzberg's public API.

Core Components¶

• Extraction Functions - Functions for text extraction (Guide)
• Types - Data structures for results and configuration (Guide)
• OCR Configuration - OCR engine settings (Guide)
• Extractor Registry - Document extractor management (Guide)
• Exceptions - Error handling (Examples)

Public API¶

All components documented in this section are exported directly from the kreuzberg package and can be imported as follows:

1

from kreuzberg import extract_file, ExtractionConfig, TesseractConfig  # etc.

API Overview¶

Kreuzberg's API has four main components:

1. Extraction Functions: Extract text from documents
2. Configuration Objects: Control extraction behavior
3. Result Objects: Contain extracted text and metadata
4. OCR Backends: Pluggable OCR engines

Examples¶

Async API (Recommended)¶

1
2
3
4
5
6
7

from kreuzberg import extract_file, ExtractionConfig

# Basic usage
result = await extract_file("document.pdf")

# With configuration
result = await extract_file("document.pdf", config=ExtractionConfig(force_ocr=True))

Sync API¶

1
2
3
4

from kreuzberg import extract_file_sync

# Basic usage
result = extract_file_sync("document.pdf")