feat: add new examples from pocketflow-academy

cookbook/pocketflow-tool-pdf-vision/tools/pdf.py

@@ -0,0 +1,52 @@
+import fitz  # PyMuPDF
+from PIL import Image
+import io
+import base64
+from typing import List, Tuple
+
+def pdf_to_images(pdf_path: str, max_size: int = 2000) -> List[Tuple[Image.Image, int]]:
+    """Convert PDF pages to PIL Images with size limit
+    
+    Args:
+        pdf_path (str): Path to PDF file
+        max_size (int): Maximum dimension (width/height) for images
+        
+    Returns:
+        list: List of tuples (PIL Image, page number)
+    """
+    doc = fitz.open(pdf_path)
+    images = []
+    
+    try:
+        for page_num in range(len(doc)):
+            page = doc[page_num]
+            pix = page.get_pixmap()
+            
+            # Convert to PIL Image
+            img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
+            
+            # Resize if needed while maintaining aspect ratio
+            if max(img.size) > max_size:
+                ratio = max_size / max(img.size)
+                new_size = tuple(int(dim * ratio) for dim in img.size)
+                img = img.resize(new_size, Image.Resampling.LANCZOS)
+            
+            images.append((img, page_num + 1))
+            
+    finally:
+        doc.close()
+        
+    return images
+
+def image_to_base64(image: Image.Image) -> str:
+    """Convert PIL Image to base64 string
+    
+    Args:
+        image (PIL.Image): Image to convert
+        
+    Returns:
+        str: Base64 encoded image string
+    """
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    return base64.b64encode(buffer.getvalue()).decode('utf-8')

cookbook/pocketflow-tool-pdf-vision/tools/vision.py

@@ -0,0 +1,40 @@
+from PIL import Image
+from utils.call_llm import client
+from tools.pdf import image_to_base64
+
+def extract_text_from_image(image: Image.Image, prompt: str = None) -> str:
+    """Extract text from image using OpenAI Vision API
+    
+    Args:
+        image (PIL.Image): Image to process
+        prompt (str, optional): Custom prompt for extraction. Defaults to general OCR.
+        
+    Returns:
+        str: Extracted text from image
+    """
+    # Convert image to base64
+    img_base64 = image_to_base64(image)
+    
+    # Default prompt for general OCR
+    if prompt is None:
+        prompt = "Please extract all text from this image."
+    
+    # Call Vision API
+    response = client.chat.completions.create(
+        model="gpt-4o",
+        messages=[{
+            "role": "user",
+            "content": [
+                {"type": "text", "text": prompt},
+                {"type": "image_url", "image_url": {"url": f"data:image/png;base64,{img_base64}"}}
+            ]
+        }]
+    )
+    
+    return response.choices[0].message.content
+
+if __name__ == "__main__":
+    # Test vision processing
+    test_image = Image.open("example.png")
+    result = extract_text_from_image(test_image)
+    print("Extracted text:", result)