vhtml 0.2.2

vhtml (ang. Visual HyperText Markup Language) - Optical character recognition and HTML layout analysis library

vHTML - Visual HTML Generator

A modular system for converting PDF documents to HTML with OCR and layout analysis.

Features

• PDF to image conversion with preprocessing (denoise, deskew)
• Document layout analysis and segmentation
• OCR with multi-language support (Polish, English, German)
• Language detection and confidence scoring
• HTML generation with embedded images and metadata
• Batch processing capabilities
• Command-line interface

Installation

Prerequisites

• Python 3.8+
• Tesseract OCR
• Poppler utilities

Using Poetry (Recommended)

# Clone the repository
git clone https://github.com/fin-officer/vhtml.git
cd vhtml

# Install with Poetry
make install

Manual Installation

# Install system dependencies
sudo apt-get update
sudo apt-get install -y tesseract-ocr tesseract-ocr-pol tesseract-ocr-eng tesseract-ocr-deu poppler-utils

# Install Python dependencies
pip install poetry
poetry install

Validate Installation

To verify that all dependencies are correctly installed:

make validate

or

python scripts/validate_installation.py

Usage

Command Line Interface

# Process a single PDF file
poetry run python -m vhtml.main /path/to/document.pdf -o output_directory

# Process a directory of PDF files
poetry run python -m vhtml.main /path/to/pdf_directory -b -o output_directory

# Process and open in browser
poetry run python -m vhtml.main /path/to/document.pdf -v

Integration Test

# Run the integration test with your PDF file
poetry run python scripts/test_integration.py /path/to/document.pdf -v

Python API

from vhtml.main import DocumentAnalyzer

# Initialize the analyzer
analyzer = DocumentAnalyzer()

# Process a document
html_path = analyzer.analyze_document("document.pdf", "output_dir")

# Print the path to the generated HTML
print(f"Generated HTML: {html_path}")

Examples

Generate Standalone HTML

Generate a standalone HTML file with all images, JS, and JSON embedded:

poetry run python examples/pdf2html.py

• Input: Folder with HTML, images, JS, and JSON (e.g., output/mhtml_example/Invoice-30392B3C-0001)
• Output: Standalone HTML (e.g., output/html_example/Invoice-30392B3C-0001_standalone.html)

Generate MHTML (Web Archive)

Generate a fully self-contained MHTML file for browser archiving:

poetry run python examples/pdf2mhtml.py

• Input: PDF(s) in invoices/ (or other test files)
• Output: MHTML file (e.g., output/mhtml_example/Invoice-30392B3C-0001.mhtml)

---

• See examples/html.py and examples/mhtml.py for usage patterns and batch processing.
• Both scripts demonstrate how to use the vHTML API for document conversion and archiving.

Core Components

• PDFProcessor: Handles PDF to image conversion and preprocessing
• LayoutAnalyzer: Analyzes document layout and segments content blocks
• OCREngine: Performs OCR with language detection and confidence scoring
• HTMLGenerator: Generates HTML with embedded images and styling
• DocumentAnalyzer: Integrates all components into a complete workflow

Project Structure

vhtml/
├── vhtml/
│   ├── core/
│   │   ├── pdf_processor.py
│   │   ├── layout_analyzer.py
│   │   ├── ocr_engine.py
│   │   └── html_generator.py
│   └── main.py
├── scripts/
│   ├── validate_installation.py
│   └── test_integration.py
├── docs/
│   ├── ARCHITECTURE.md
│   ├── IMPLEMENTATION.md
│   └── PROJECT_STRUCTURE.md
├── Makefile
├── pyproject.toml
└── README.md

Development

# Setup development environment
make setup

# Run tests
make test

# Format code
make format

# Lint code
make lint

# Build package
make build

Documentation

For more detailed information, see the documentation files:

• Architecture
• Implementation
• Project Structure

License

This project is licensed under the Apache 2.0 License - see the LICENSE file for details.