Document Processor

A modern desktop application for parsing and processing documents (PDF, DOCX, TXT) with intelligent image context extraction.

Built with Rust + Tauri + Svelte for maximum performance and minimal footprint.

Features

• Multi-format parsing: PDF, DOCX, TXT (.doc files are accepted but routed through the DOCX parser — pure legacy .doc binary support is not implemented; RTF is not supported)
• Image extraction with context: Images are extracted with surrounding text preserved
• Document classification: Automatic detection of document types (umowa, pozew, ustawa, etc.)
• Watch folder: Automatic processing of new documents
• SQLite database: Fast search and organization
• Modern UI: Dark theme, responsive design
• Linux desktop: built and distributed for Linux (deb / AppImage); the Tauri toolchain compiles on Windows + macOS but no installers are produced by the default bundle config — adding nsis / dmg targets to tauri.conf.json is welcomed via a PR

Installation

Prerequisites

Linux (Ubuntu/Debian)

sudo apt install -y libwebkit2gtk-4.1-dev libappindicator3-dev librsvg2-dev patchelf libssl-dev

Windows (build-from-source only — no shipped installer)

The Tauri toolchain compiles on Windows but tauri.conf.json only defines Linux bundle targets, so npm run tauri build produces a debug executable rather than a packaged installer. To get an .msi or .exe, add nsis / msi to bundle.targets in a fork. Prerequisites if you want to try:

• Install WebView2
• Install Visual Studio Build Tools

Build from source

# Install Rust
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh

# Clone and build
cd document-processor
npm install
npm run tauri build

Development

npm run tauri dev

Usage

GUI Application

1. Launch the application
2. Drag & drop documents or click to browse
3. Set a "Watch Folder" for automatic processing
4. View processed documents with extracted text and images

Output Structure

Processed documents land under the app's data directory:

<project-root>/dane/
├── documents.db                       # SQLite index of every processed file
└── przetworzone/<document-id>/        # one folder per document
    ├── document.md                    # Human-readable markdown
    ├── document.json                  # Structured data for AI
    ├── images/
    │   ├── img_001.png                # Extracted images
    │   ├── img_001.json               # Image metadata + context
    │   └── thumb_001.png              # Thumbnails
    └── original.<ext>                 # Original file copy

<project-root> is the directory the launcher selects on startup (the working directory under development; /opt/document-processor on a packaged install). dane/ and przetworzone/ are Polish legacy identifiers (Polish for "data" and "processed") — see CONTRIBUTING for the rename policy.

Image Context

Each image includes:

• context_before: 200 characters of text before the image
• context_after: 200 characters after
• position_marker: Page and position reference
• ocr_text: Text extracted from image (if applicable)
• ai_description: AI-generated description (when available)

Integrations

Claude Code skills (external)

Two companion Claude Code skills live outside this repo and call the processor over the command line. They are not bundled here — install them separately in your Claude Code configuration:

• /parse — parses a document and extracts text + images.

  /parse ~/Documents/contract.pdf
  

• /document-upload-analyzer — analyses document upload methods in a web application.

  /document-upload-analyzer
  

Architecture

document-processor/
├── src/                  # Svelte frontend
│   ├── App.svelte       # Main component
│   ├── main.js          # Entry point
│   └── styles.css       # Global styles
├── src-tauri/
│   └── src/
│       ├── main.rs      # Tauri entry point
│       ├── parser.rs    # Document parsing logic
│       ├── db.rs        # SQLite database
│       └── watcher.rs   # Folder watching
└── package.json

Technologies

• Backend: Rust (lopdf, pdf-extract, image, rusqlite)
• Frontend: Svelte 5
• Framework: Tauri 2
• Database: SQLite (rusqlite with bundled)
• Build: Vite

License

Dual-licensed:

• AGPLv3 for open source, personal, and internal use — see LICENSE.
• Commercial license for SaaS, embedded use, or proprietary modifications — see LICENSE-COMMERCIAL.md.