Spaces:
Sleeping
Sleeping
| title: Smart PDF Chapter Splitter | |
| emoji: 📚 | |
| colorFrom: gray | |
| colorTo: yellow | |
| sdk: gradio | |
| sdk_version: 6.5.1 | |
| app_file: app.py | |
| pinned: false | |
| license: mit | |
| short_description: 'Split large PDFs (books) into clean, per-chapter files ' | |
| # 📚 Smart PDF Chapter Splitter | |
| Split large PDFs (books, manuals, technical documents) into clean, per-chapter files — **fast, local, and deterministic**. | |
| This tool uses **PDF bookmarks (Table of Contents)** to extract chapters with **near-perfect accuracy** for professionally published documents. | |
| --- | |
| ## ✨ Features | |
| - 📖 Splits PDFs into individual chapter files | |
| - ⚙️ Uses **embedded bookmarks** (no AI, no guesswork) | |
| - 🚀 Extremely fast (local processing) | |
| - 🧼 Safe filenames (cross-platform) | |
| - 📂 Batch-ready and automation-friendly | |
| --- | |
| ## 🧠 How It Works | |
| Most modern PDFs contain an internal **Table of Contents (bookmarks)**. | |
| This Space: | |
| 1. Reads the PDF outline | |
| 2. Identifies top-level chapters | |
| 3. Calculates page ranges | |
| 4. Exports each chapter as its own PDF | |
| > ✅ Deterministic | |
| > ❌ No OCR | |
| > ❌ No AI hallucinations | |
| --- | |
| ## 📊 Accuracy Expectations | |
| | PDF Type | Accuracy | | |
| |-------|---------| | |
| | Digital-first published books | ⭐⭐⭐⭐⭐ (~100%) | | |
| | Technical manuals | ⭐⭐⭐⭐⭐ | | |
| | Semi-digital PDFs | ⭐⭐⭐⭐ | | |
| | Scanned PDFs (no bookmarks) | ❌ Not supported | | |
| --- | |
| ## 🏗️ Ideal Use Cases | |
| - 📚 Published books (Springer, O’Reilly, Wiley, Packt…) | |
| - ⚙️ Engineering manuals | |
| - 🧾 Technical specifications | |
| - 🏭 PLM & documentation pipelines | |
| - 📂 Large PDF libraries | |
| --- | |
| ## 🚫 Limitations | |
| This tool **requires bookmarks**. | |
| If your PDF: | |
| - Is scanned | |
| - Has no outline | |
| - Has broken TOC metadata | |
| ➡️ You will need **OCR or AI-based structure detection** (not included here). | |
| --- | |
| ## 🛠️ Tech Stack | |
| - **Python** | |
| - **PyMuPDF (fitz)** | |
| - Local execution (no cloud dependency) | |
| Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference | |