KnowledgeBase/server/pdf_to_images.py

import fitz  # PyMuPDF
import sys
import os
import json

def convert_pdf_to_images(pdf_path, output_dir, zoom=2.0, quality=85):
    """
    Converts PDF pages to images.
    zoom: 2.0 means 200% scaling (approx 144 DPI if original is 72 DPI)
    """
    try:
        if not os.path.exists(output_dir):
            os.makedirs(output_dir)

        doc = fitz.open(pdf_path)
        images = []
        
        # Matrix for scaling (DPI control)
        mat = fitz.Matrix(zoom, zoom)
        
        for i in range(len(doc)):
            page = doc.load_page(i)
            pix = page.get_pixmap(matrix=mat, colorspace=fitz.csRGB)
            
            output_path = os.path.join(output_dir, f"page-{i+1}.jpg")
            # In newer PyMuPDF, save() doesn't take quality. Use tobytes instead.
            img_bytes = pix.tobytes("jpg", jpg_quality=quality)
            with open(output_path, "wb") as f:
                f.write(img_bytes)
            
            images.append({
                "path": output_path,
                "pageIndex": i + 1,
                "size": os.path.getsize(output_path)
            })
            
        doc.close()
        return {
            "success": True,
            "images": images,
            "totalPages": len(images)
        }
    except Exception as e:
        return {
            "success": False,
            "error": str(e)
        }

if __name__ == "__main__":
    if len(sys.argv) < 3:
        print(json.dumps({"success": False, "error": "Usage: python pdf_to_images.py <pdf_path> <output_dir> [zoom] [quality]"}))
        sys.exit(1)
        
    pdf_path = sys.argv[1]
    output_dir = sys.argv[2]
    zoom = float(sys.argv[3]) if len(sys.argv) > 3 else 2.0
    quality = int(sys.argv[4]) if len(sys.argv) > 4 else 85
    
    result = convert_pdf_to_images(pdf_path, output_dir, zoom, quality)
    print(json.dumps(result))