Kreuzberg: open source tool to extract text from any document

General Introduction

Kreuzberg is a library to simplify the text extraction of PDF files , designed to provide simple , hassle-free text extraction solution . The library is particularly well suited for RAG (Retrieval-Augmented Generation) services that require text extraction.Kreuzberg supports local operation, is easy to control and inexpensive. It combines a variety of open source and commercial options to provide flexible text extraction capabilities.

Function List

• PDF Text Extraction: Extract text content from PDF files.
• Image/PDF OCR: Optical character recognition of images and PDFs using Tesseract-OCR.
• Non-PDF Text Extraction: Extraction of text in other formats via Pandoc.
• local operation: Support local installation and operation, easy to control and manage.
• Open source and free: Based on the MIT license open source, free to use.

Using Help

Installation process

1. Installing Python Packages::

   pip install kreuzberg

2. Installation of system dependencies::
   • Pandoc: for non-PDF text extraction (GPL v2.0 license, used as CLI only).
   • Tesseract-OCR: OCR for images and PDFs (Apache license).

Guidelines for use

1. Basic use::
   • Import the library and initialize it: python
from kreuzberg import Kreuzberg
extractor = Kreuzberg()

   • Extract PDF text: python
text = extractor.extract_text('path/to/pdf/file.pdf')
print(text)

2. OCR function::
   • Perform OCR on images or PDFs: python
ocr_text = extractor.ocr('path/to/image_or_pdf')
print(ocr_text)

3. Non-PDF Text Extraction::
   • Use Pandoc to extract text in other formats: python
other_text = extractor.extract_text('path/to/other/file')
print(other_text)


Detailed function operation flow

1. PDF Text Extraction::
   • Make sure the PDF file path is correct.
   • utilizationextract_textmethod to extract the text.
   • Process the extracted text data for subsequent operations.
2. OCR function::
   • Install and configure Tesseract-OCR.
   • utilizationocrmethod for OCR processing of images or PDFs.
   • Get and process OCR results.
3. Non-PDF Text Extraction::
   • Install and configure Pandoc.
   • utilizationextract_textmethod to extract text in other formats.
   • Process the extracted text data for subsequent operations.

Through the above steps, users can easily get started with Kreuzberg text extraction operations to meet a variety of text processing needs.