paddleocr-skills 1.0.0

package/templates/paddleocr-vl/scripts/paddleocr-vl/configure.py

@@ -0,0 +1,270 @@
+#!/usr/bin/env python3
+"""
+PaddleOCR-VL Configuration Wizard
+
+Supports two modes:
+1. Interactive mode (default): python configure.py
+2. CLI mode: python configure.py --api-url URL --token TOKEN
+
+Interactive configuration for PaddleOCR-VL API credentials.
+Saves configuration to .env file in project root.
+"""
+
+import argparse
+import os
+import sys
+from pathlib import Path
+
+
+def save_config(api_url: str, token: str, project_root: Path, quiet: bool = False) -> bool:
+    """
+    Save configuration to .env file
+
+    Args:
+        api_url: VL API URL
+        token: VL access token
+        project_root: Project root directory
+        quiet: If True, suppress output messages
+
+    Returns:
+        True if successful, False otherwise
+    """
+    env_file = project_root / ".env"
+
+    # Read existing .env if it exists
+    existing_config = {}
+    if env_file.exists():
+        if not quiet:
+            print(f"Found existing .env file: {env_file}")
+            overwrite = input("Overwrite? [Y/n]: ").strip().lower()
+            if overwrite == 'n':
+                print("Configuration cancelled")
+                return False
+
+        with open(env_file, 'r', encoding='utf-8') as f:
+            for line in f:
+                line = line.strip()
+                if line and not line.startswith('#') and '=' in line:
+                    key, value = line.split('=', 1)
+                    existing_config[key.strip()] = value.strip()
+
+    # Update configuration
+    existing_config['VL_API_URL'] = api_url
+    existing_config['VL_TOKEN'] = token
+
+    # Write to .env file
+    try:
+        with open(env_file, 'w', encoding='utf-8') as f:
+            # Write header
+            f.write("# PaddleOCR Skills Configuration\n")
+            f.write("# Generated by configuration wizard\n")
+            f.write("\n")
+
+            # Group PP-OCRv5 configs
+            if any(k.startswith('API_URL') or k.startswith('PADDLE_OCR_TOKEN') or k.startswith('PPOCRV5') for k in existing_config.keys()):
+                f.write("# ========================================\n")
+                f.write("# PP-OCRv5 Configuration\n")
+                f.write("# ========================================\n")
+                for key in ['API_URL', 'PADDLE_OCR_TOKEN', 'PPOCRV5_API_URL', 'PPOCRV5_TOKEN']:
+                    if key in existing_config:
+                        f.write(f"{key}={existing_config[key]}\n")
+                f.write("\n")
+
+            # Group PaddleOCR-VL configs
+            f.write("# ========================================\n")
+            f.write("# PaddleOCR-VL Configuration\n")
+            f.write("# ========================================\n")
+            f.write(f"VL_API_URL={existing_config['VL_API_URL']}\n")
+            f.write(f"VL_TOKEN={existing_config['VL_TOKEN']}\n")
+            f.write("\n")
+
+            # Write other configs
+            other_keys = [k for k in existing_config.keys()
+                         if k not in ['API_URL', 'PADDLE_OCR_TOKEN', 'PPOCRV5_API_URL',
+                                      'PPOCRV5_TOKEN', 'VL_API_URL', 'VL_TOKEN']]
+            if other_keys:
+                f.write("# ========================================\n")
+                f.write("# Other Configuration\n")
+                f.write("# ========================================\n")
+                for key in other_keys:
+                    f.write(f"{key}={existing_config[key]}\n")
+
+        if not quiet:
+            print(f"✓ Configuration saved to {env_file}")
+        return True
+
+    except Exception as e:
+        print(f"✗ Failed to save configuration: {e}")
+        return False
+
+
+def main():
+    # Parse command-line arguments
+    parser = argparse.ArgumentParser(
+        description='PaddleOCR-VL API Configuration Tool',
+        formatter_class=argparse.RawDescriptionHelpFormatter,
+        epilog="""
+Examples:
+  # Interactive mode
+  python configure.py
+
+  # CLI mode (non-interactive)
+  python configure.py --api-url "https://your-service.com/v1" --token "your_token"
+        """
+    )
+    parser.add_argument('--api-url', help='VL API URL (non-interactive mode)')
+    parser.add_argument('--token', help='VL access token (non-interactive mode)')
+    parser.add_argument('--quiet', action='store_true', help='Suppress output messages')
+
+    args = parser.parse_args()
+
+    # Find .env file location (project root, 2 levels up from script)
+    project_root = Path(__file__).parent.parent.parent
+
+    # ========================================
+    # CLI Mode (non-interactive)
+    # ========================================
+    if args.api_url and args.token:
+        try:
+            api_url = args.api_url.strip()
+            token = args.token.strip()
+
+            # Validate URL format
+            if not api_url.startswith(("http://", "https://")):
+                api_url = f"https://{api_url}"
+
+            # Validate token
+            if len(token) < 16:
+                print("Error: Token seems too short. Please check and try again.")
+                sys.exit(1)
+
+            # Save configuration
+            if save_config(api_url, token, project_root, quiet=args.quiet):
+                if not args.quiet:
+                    masked_token = token[:8] + "..." + token[-4:] if len(token) > 12 else "***"
+                    print("\n✓ Configuration complete!")
+                    print(f"  VL_API_URL: {api_url}")
+                    print(f"  VL_TOKEN: {masked_token}")
+                sys.exit(0)
+            else:
+                sys.exit(1)
+
+        except Exception as e:
+            print(f"Error: {e}")
+            sys.exit(1)
+
+    elif args.api_url or args.token:
+        print("Error: Both --api-url and --token are required for CLI mode")
+        print("Run without arguments for interactive mode")
+        sys.exit(1)
+
+    # ========================================
+    # Interactive Mode
+    # ========================================
+    print("=" * 60)
+    print("PaddleOCR-VL Configuration Wizard")
+    print("=" * 60)
+    print()
+
+    env_file = project_root / ".env"
+    print(f"Configuration will be saved to: {env_file}")
+    print()
+
+    # Read existing .env if it exists
+    existing_config = {}
+    if env_file.exists():
+        print("Found existing .env file, loading current values...")
+        with open(env_file, 'r', encoding='utf-8') as f:
+            for line in f:
+                line = line.strip()
+                if line and not line.startswith('#') and '=' in line:
+                    key, value = line.split('=', 1)
+                    existing_config[key.strip()] = value.strip()
+        print()
+
+    # Get current values
+    current_api_url = existing_config.get('VL_API_URL', '')
+    current_token = existing_config.get('VL_TOKEN', '')
+
+    print("Please provide your PaddleOCR-VL API credentials:")
+    print("(Press Enter to keep current value)")
+    print()
+
+    # Prompt for API URL
+    print("1. VL_API_URL - PaddleOCR-VL API endpoint")
+    print("   Example: https://your-service.com/v1")
+    if current_api_url:
+        print(f"   Current: {current_api_url}")
+
+    api_url_input = input("   Enter VL_API_URL: ").strip()
+    new_api_url = api_url_input if api_url_input else current_api_url
+
+    if not new_api_url:
+        print()
+        print("ERROR: VL_API_URL is required.")
+        print("Please run this wizard again and provide a valid API URL.")
+        sys.exit(1)
+
+    print()
+
+    # Prompt for Token
+    print("2. VL_TOKEN - Your access token")
+    if current_token:
+        masked_token = current_token[:8] + "..." + current_token[-4:] if len(current_token) > 12 else "***"
+        print(f"   Current: {masked_token}")
+
+    token_input = input("   Enter VL_TOKEN: ").strip()
+    new_token = token_input if token_input else current_token
+
+    if not new_token:
+        print()
+        print("ERROR: VL_TOKEN is required.")
+        print("Please run this wizard again and provide a valid token.")
+        sys.exit(1)
+
+    print()
+
+    # Save configuration
+    print("Saving configuration...")
+
+    if not save_config(new_api_url, new_token, project_root):
+        sys.exit(1)
+
+    print()
+
+    # Verify configuration
+    print("Verifying configuration...")
+    try:
+        sys.path.insert(0, str(Path(__file__).parent))
+        from _lib import Config
+
+        Config.load_env()
+
+        test_url = Config.get_vl_api_url()
+        test_token = Config.get_vl_token()
+
+        print("✓ VL_API_URL loaded successfully")
+        print("✓ VL_TOKEN loaded successfully")
+        print()
+
+    except Exception as e:
+        print(f"✗ Configuration verification failed: {e}")
+        print()
+        sys.exit(1)
+
+    # Next steps
+    print("=" * 60)
+    print("Configuration Complete!")
+    print("=" * 60)
+    print()
+    print("Next steps:")
+    print("  1. Test the configuration:")
+    print("     python scripts/paddleocr-vl/smoke_test.py")
+    print()
+    print("  2. Try parsing a document:")
+    print("     python scripts/paddleocr-vl/vl_caller.py --file-url \"URL\"")
+    print()
+
+
+if __name__ == '__main__':
+    main()

package/templates/paddleocr-vl/scripts/paddleocr-vl/optimize_file.py

@@ -0,0 +1,226 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""
+File Optimizer for PaddleOCR-VL
+
+Compresses and optimizes large files to meet size requirements.
+Supports images (PNG, JPG) and PDFs.
+
+Usage:
+    python scripts/paddleocr-vl/optimize_file.py input.pdf output.pdf --target-size 15
+    python scripts/paddleocr-vl/optimize_file.py input.png output.png --quality 85
+"""
+
+import argparse
+import sys
+from pathlib import Path
+
+
+def optimize_image(input_path: Path, output_path: Path, quality: int = 85, max_size_mb: float = 20):
+    """
+    Optimize image file by reducing quality and/or resolution
+
+    Args:
+        input_path: Input image path
+        output_path: Output image path
+        quality: JPEG quality (1-100, lower = smaller file)
+        max_size_mb: Target max size in MB
+    """
+    try:
+        from PIL import Image
+    except ImportError:
+        print("ERROR: Pillow not installed")
+        print("Install with: pip install Pillow")
+        sys.exit(1)
+
+    print(f"Optimizing image: {input_path}")
+
+    # Open image
+    img = Image.open(input_path)
+    original_size = input_path.stat().st_size / 1024 / 1024
+
+    print(f"Original size: {original_size:.2f}MB")
+    print(f"Original dimensions: {img.size[0]}x{img.size[1]}")
+
+    # Convert RGBA to RGB if needed (for JPEG)
+    if img.mode in ('RGBA', 'LA', 'P'):
+        # Create white background
+        background = Image.new('RGB', img.size, (255, 255, 255))
+        if img.mode == 'P':
+            img = img.convert('RGBA')
+        background.paste(img, mask=img.split()[-1] if img.mode in ('RGBA', 'LA') else None)
+        img = background
+
+    # Determine output format
+    output_format = output_path.suffix.lower()
+    if output_format in ['.jpg', '.jpeg']:
+        save_format = 'JPEG'
+    elif output_format == '.png':
+        save_format = 'PNG'
+    else:
+        save_format = 'JPEG'
+        output_path = output_path.with_suffix('.jpg')
+
+    # Try saving with specified quality
+    img.save(output_path, format=save_format, quality=quality, optimize=True)
+    new_size = output_path.stat().st_size / 1024 / 1024
+
+    # If still too large, reduce resolution
+    scale_factor = 0.9
+    while new_size > max_size_mb and scale_factor > 0.3:
+        new_width = int(img.size[0] * scale_factor)
+        new_height = int(img.size[1] * scale_factor)
+
+        print(f"Resizing to {new_width}x{new_height} (scale: {scale_factor:.2f})")
+
+        resized = img.resize((new_width, new_height), Image.Resampling.LANCZOS)
+        resized.save(output_path, format=save_format, quality=quality, optimize=True)
+        new_size = output_path.stat().st_size / 1024 / 1024
+
+        scale_factor -= 0.1
+
+    print(f"Optimized size: {new_size:.2f}MB")
+    print(f"Reduction: {((original_size - new_size) / original_size * 100):.1f}%")
+
+    if new_size > max_size_mb:
+        print(f"\nWARNING: File still larger than {max_size_mb}MB")
+        print("Consider:")
+        print("  - Lower quality (--quality 70)")
+        print("  - Use --file-url instead of local file")
+        print("  - Split multi-page documents")
+
+
+def optimize_pdf(input_path: Path, output_path: Path, max_size_mb: float = 20):
+    """
+    Optimize PDF by compressing images within it
+
+    Args:
+        input_path: Input PDF path
+        output_path: Output PDF path
+        max_size_mb: Target max size in MB
+    """
+    try:
+        import fitz  # PyMuPDF
+    except ImportError:
+        print("ERROR: PyMuPDF not installed")
+        print("Install with: pip install PyMuPDF")
+        sys.exit(1)
+
+    print(f"Optimizing PDF: {input_path}")
+
+    original_size = input_path.stat().st_size / 1024 / 1024
+    print(f"Original size: {original_size:.2f}MB")
+
+    # Open PDF
+    doc = fitz.open(input_path)
+    print(f"Pages: {len(doc)}")
+
+    # Create output PDF with compression
+    writer = fitz.open()
+
+    for page_num in range(len(doc)):
+        page = doc[page_num]
+
+        # Get page as pixmap with lower DPI if needed
+        dpi = 150  # Lower DPI for smaller file
+        mat = fitz.Matrix(dpi / 72, dpi / 72)
+        pix = page.get_pixmap(matrix=mat)
+
+        # Create new page
+        new_page = writer.new_page(width=page.rect.width, height=page.rect.height)
+
+        # Insert image with compression
+        new_page.insert_image(new_page.rect, pixmap=pix)
+
+        print(f"Processed page {page_num + 1}/{len(doc)}")
+
+    # Save with compression
+    writer.save(
+        output_path,
+        garbage=4,  # Maximum compression
+        deflate=True,
+        clean=True
+    )
+    writer.close()
+    doc.close()
+
+    new_size = output_path.stat().st_size / 1024 / 1024
+    print(f"Optimized size: {new_size:.2f}MB")
+    print(f"Reduction: {((original_size - new_size) / original_size * 100):.1f}%")
+
+    if new_size > max_size_mb:
+        print(f"\nWARNING: PDF still larger than {max_size_mb}MB")
+        print("Consider:")
+        print("  - Split into multiple files")
+        print("  - Process specific pages only")
+        print("  - Use --file-url instead")
+
+
+def main():
+    parser = argparse.ArgumentParser(
+        description='Optimize files for PaddleOCR-VL processing',
+        formatter_class=argparse.RawDescriptionHelpFormatter,
+        epilog="""
+Examples:
+  # Optimize image with default quality (85)
+  python scripts/paddleocr-vl/optimize_file.py input.png output.png
+
+  # Optimize with specific quality
+  python scripts/paddleocr-vl/optimize_file.py input.jpg output.jpg --quality 70
+
+  # Optimize PDF
+  python scripts/paddleocr-vl/optimize_file.py input.pdf output.pdf
+
+  # Target specific size
+  python scripts/paddleocr-vl/optimize_file.py input.pdf output.pdf --target-size 15
+
+Supported formats:
+  - Images: PNG, JPG, JPEG, BMP, TIFF
+  - Documents: PDF
+        """
+    )
+
+    parser.add_argument('input', help='Input file path')
+    parser.add_argument('output', help='Output file path')
+    parser.add_argument(
+        '--quality',
+        type=int,
+        default=85,
+        help='JPEG quality (1-100, default: 85)'
+    )
+    parser.add_argument(
+        '--target-size',
+        type=float,
+        default=20,
+        help='Target maximum size in MB (default: 20)'
+    )
+
+    args = parser.parse_args()
+
+    input_path = Path(args.input)
+    output_path = Path(args.output)
+
+    # Validate input
+    if not input_path.exists():
+        print(f"ERROR: Input file not found: {input_path}")
+        sys.exit(1)
+
+    # Determine file type
+    ext = input_path.suffix.lower()
+
+    if ext == '.pdf':
+        optimize_pdf(input_path, output_path, args.target_size)
+    elif ext in ['.png', '.jpg', '.jpeg', '.bmp', '.tiff', '.tif']:
+        optimize_image(input_path, output_path, args.quality, args.target_size)
+    else:
+        print(f"ERROR: Unsupported file format: {ext}")
+        print("Supported: PDF, PNG, JPG, JPEG, BMP, TIFF")
+        sys.exit(1)
+
+    print(f"\nOptimized file saved to: {output_path}")
+    print("\nYou can now process with:")
+    print(f'  python scripts/paddleocr-vl/vl_caller.py --file-path "{output_path}" --pretty')
+
+
+if __name__ == '__main__':
+    main()

package/templates/paddleocr-vl/scripts/paddleocr-vl/requirements-optimize.txt

@@ -0,0 +1,8 @@
+# File Optimization Dependencies
+# Install with: pip install -r scripts/paddleocr-vl/requirements-optimize.txt
+
+# Image processing
+Pillow>=10.0.0
+
+# PDF processing
+PyMuPDF>=1.23.0

package/templates/paddleocr-vl/scripts/paddleocr-vl/requirements.txt

@@ -0,0 +1,7 @@
+# PaddleOCR-VL Dependencies
+
+# HTTP client
+httpx>=0.24.0
+
+# Environment variables
+python-dotenv>=1.0.0