tools/python/pdf_to_image.py

#!/usr/bin/env python3
"""PDF to Image (JPG/PNG) converter using pdf2image (poppler)."""
import argparse, json, sys, os, zipfile
from pathlib import Path

def convert(input_path, output_path, fmt='jpg', dpi=150):
    from pdf2image import convert_from_path
    
    images = convert_from_path(input_path, dpi=dpi, fmt=fmt)
    
    if len(images) == 1:
        ext = 'jpg' if fmt.lower() in ('jpg', 'jpeg') else fmt.lower()
        if not output_path:
            output_path = str(Path(input_path).with_suffix(f'.{ext}'))
        save_fmt = 'JPEG' if ext in ('jpg', 'jpeg') else ext.upper()
        images[0].save(output_path, save_fmt, quality=95)
        return output_path
    else:
        out_dir = Path(output_path).parent
        base = Path(input_path).stem
        ext = 'jpg' if fmt.lower() in ('jpg', 'jpeg') else fmt.lower()
        zip_path = str(out_dir / f'{base}_images.zip')
        
        with zipfile.ZipFile(zip_path, 'w', zipfile.ZIP_DEFLATED) as zf:
            for i, img in enumerate(images):
                img_name = f'{base}_page_{i+1}.{ext}'
                img_path = str(out_dir / img_name)
                save_fmt = 'JPEG' if ext in ('jpg', 'jpeg') else ext.upper()
                img.save(img_path, save_fmt, quality=95)
                zf.write(img_path, img_name)
                os.remove(img_path)
        
        return zip_path

def main():
    parser = argparse.ArgumentParser(description='Convert PDF to images')
    parser.add_argument('--input', required=True)
    parser.add_argument('--output', required=True)
    parser.add_argument('--format', default='jpg', choices=['jpg', 'png'])
    parser.add_argument('--dpi', type=int, default=150)
    args = parser.parse_args()
    try:
        result = convert(args.input, args.output, args.format, args.dpi)
        print(json.dumps({"success": True, "output": result, "message": f"PDF converted to {args.format.upper()} images"}))
    except Exception as e:
        print(json.dumps({"success": False, "output": "", "message": str(e)}))
        sys.exit(1)

if __name__ == '__main__':
    main()