PDF Splitter MCP Server

# PDF Splitter MCP Tests This directory contains comprehensive tests for the PDF Splitter MCP server. ## Test Files - **`pdf-processor.test.ts`** - Unit tests for the PDFProcessor class - **`mcp-server.test.ts`** - Integration tests for the MCP server - **`image-extraction.test.ts`** - Tests for image extraction functionality - **`image-validation.test.ts`** - Comprehensive image validation tests using Sharp - **`page-rendering.test.ts`** - Tests for page rendering at different DPIs and formats - **`japanese-pdf.test.ts`** - Tests for handling PDFs with CJK characters - **`example-usage.ts`** - Executable example demonstrating all features - **`example-image-extraction.ts`** - Example demonstrating image extraction ## Running Tests ### Run all tests ```bash npm test # or bun test ``` ### Run tests in watch mode ```bash npm run test:watch # or bun test --watch ``` ### Run specific test file ```bash bun test pdf-processor.test.ts ``` ### Run the example ```bash bun run test/example-usage.ts ``` ## Test Coverage The tests cover all major functionality: 1. **PDF Loading** - Local file loading - URL loading (using Adobe PDF spec as example) - Error handling for invalid paths/URLs 2. **Page Extraction** - Single page extraction - Range extraction - Boundary validation 3. **Text Search** - Case-insensitive search (default) - Case-sensitive search - Multi-page results 4. **Document Outline** - TOC extraction with page numbers - Formatted outline output - Handling PDFs without outlines 5. **PDF Information** - Metadata extraction - PDF listing - Error handling 6. **Image Extraction** - List embedded images with metadata - Extract images as base64-encoded data - Extract specific images by page and index - Handle PDFs without images 7. **Page Rendering** - Render pages at different DPI settings (72-300) - Support for PNG and JPEG formats - Batch rendering of multiple pages - Maintain aspect ratios - Performance optimization 8. **Regex Search** - Search with regular expressions - Case-sensitive regex matching - Complex pattern matching - Invalid regex handling ## Test PDF The tests use the Adobe PDF 32000 specification document as a real-world example: - URL: https://opensource.adobe.com/dc-acrobat-sdk-docs/pdfstandards/PDF32000_2008.pdf - This is a large, well-structured PDF with: - Multiple pages (700+) - Complete document outline/TOC - Rich metadata - Searchable text content ## Notes - URL tests require internet connection - URL tests have extended timeout (30 seconds) for downloading large PDFs - The example script provides a practical demonstration of all features