Update app.py

app.py

@@ -76,7 +76,7 @@ def analyze_construction_media(media):
         return [("No input", "Error: Please upload images or a video for analysis.")]
 
     try:
-        logger.info("Starting analysis")
+        logger.info(f"Starting analysis of {len(media)} files")
         results = []
 
         instruction = ("You are an AI assistant specialized in analyzing images for safety issues. "
@@ -85,77 +85,106 @@ def analyze_construction_media(media):
                        "and suggest steps to resolve them. If it's not a construction site, simply state that")
 
         for i, file in enumerate(media):
-            file_path = file.name  # Get the file path
-            file_type = file_path.split('.')[-1].lower()
-            
-            if file_type in ['jpg', 'jpeg', 'png', 'gif']:
-                # Handle image
-                image = Image.open(file_path)
-                resized_image = resize_image(image)
-                image_data_url = f"data:image/png;base64,{encode_image(resized_image)}"
-                messages = [
-                    {
-                        "role": "user",
-                        "content": [
-                            {
-                                "type": "text",
-                                "text": f"{instruction}\n\nAnalyze this image (File {i+1}/{len(media)}). First, determine if it's a construction site. If it is, explain the image in detail, focusing on safety aspects. If it's not, briefly describe what you see."
-                            },
+            try:
+                file_path = file.name  # Get the file path
+                logger.info(f"Processing file {i+1}/{len(media)}: {file_path}")
+                
+                if not os.path.exists(file_path):
+                    logger.error(f"File does not exist: {file_path}")
+                    results.append((f"File {i+1} analysis", f"Error: File does not exist: {file_path}"))
+                    continue
+
+                file_type = os.path.splitext(file_path)[1][1:].lower()
+                
+                if file_type in ['jpg', 'jpeg', 'png', 'gif']:
+                    # Handle image
+                    try:
+                        image = Image.open(file_path)
+                        logger.info(f"Successfully opened image: {file_path}")
+                        resized_image = resize_image(image)
+                        image_data_url = f"data:image/png;base64,{encode_image(resized_image)}"
+                        
+                        messages = [
                             {
-                                "type": "image_url",
-                                "image_url": {
-                                    "url": image_data_url
-                                }
+                                "role": "user",
+                                "content": [
+                                    {
+                                        "type": "text",
+                                        "text": f"{instruction}\n\nAnalyze this image (File {i+1}/{len(media)}). First, determine if it's a construction site. If it is, explain the image in detail, focusing on safety aspects. If it's not, briefly describe what you see."
+                                    },
+                                    {
+                                        "type": "image_url",
+                                        "image_url": {
+                                            "url": image_data_url
+                                        }
+                                    }
+                                ]
                             }
                         ]
-                    }
-                ]
-                completion = client.chat.completions.create(
-                    model="llama-3.2-90b-vision-preview",
-                    messages=messages,
-                    temperature=0.7,
-                    max_tokens=1000,
-                    top_p=1,
-                    stream=False,
-                    stop=None
-                )
-                result = completion.choices[0].message.content
-                results.append((f"Image {i+1} analysis", result))
-            elif file_type in ['mp4', 'avi', 'mov', 'wmv']:
-                # Handle video
-                frames = extract_frames_from_video(file_path)
-                for j, frame in enumerate(frames):
-                    image_data_url = f"data:image/png;base64,{encode_image(frame)}"
-                    messages = [
-                        {
-                            "role": "user",
-                            "content": [
-                                {
-                                    "type": "text",
-                                    "text": f"{instruction}\n\nAnalyze this frame from a video (File {i+1}/{len(media)}, Frame {j+1}/{len(frames)}). First, determine if it's a construction site. If it is, explain what you observe, focusing on safety aspects. If it's not, briefly describe what you see."
-                                },
+                        
+                        completion = client.chat.completions.create(
+                            model="llama-3.2-90b-vision-preview",
+                            messages=messages,
+                            temperature=0.7,
+                            max_tokens=1000,
+                            top_p=1,
+                            stream=False,
+                            stop=None
+                        )
+                        result = completion.choices[0].message.content
+                        results.append((f"Image {i+1} analysis", result))
+                        logger.info(f"Successfully analyzed image {i+1}")
+                    except Exception as img_error:
+                        logger.error(f"Error processing image {i+1}: {str(img_error)}")
+                        results.append((f"Image {i+1} analysis", f"Error processing image: {str(img_error)}"))
+                
+                elif file_type in ['mp4', 'avi', 'mov', 'wmv']:
+                    # Handle video
+                    try:
+                        frames = extract_frames_from_video(file_path)
+                        logger.info(f"Extracted {len(frames)} frames from video: {file_path}")
+                        for j, frame in enumerate(frames):
+                            image_data_url = f"data:image/png;base64,{encode_image(frame)}"
+                            messages = [
                                 {
-                                    "type": "image_url",
-                                    "image_url": {
-                                        "url": image_data_url
-                                    }
+                                    "role": "user",
+                                    "content": [
+                                        {
+                                            "type": "text",
+                                            "text": f"{instruction}\n\nAnalyze this frame from a video (File {i+1}/{len(media)}, Frame {j+1}/{len(frames)}). First, determine if it's a construction site. If it is, explain what you observe, focusing on safety aspects. If it's not, briefly describe what you see."
+                                        },
+                                        {
+                                            "type": "image_url",
+                                            "image_url": {
+                                                "url": image_data_url
+                                            }
+                                        }
+                                    ]
                                 }
                             ]
-                        }
-                    ]
-                    completion = client.chat.completions.create(
-                        model="llama-3.2-90b-vision-preview",
-                        messages=messages,
-                        temperature=0.7,
-                        max_tokens=1000,
-                        top_p=1,
-                        stream=False,
-                        stop=None
-                    )
-                    result = completion.choices[0].message.content
-                    results.append((f"Video {i+1}, Frame {j+1} analysis", result))
-            else:
-                results.append((f"File {i+1} analysis", f"Unsupported file type: {file_type}"))
+                            completion = client.chat.completions.create(
+                                model="llama-3.2-90b-vision-preview",
+                                messages=messages,
+                                temperature=0.7,
+                                max_tokens=1000,
+                                top_p=1,
+                                stream=False,
+                                stop=None
+                            )
+                            result = completion.choices[0].message.content
+                            results.append((f"Video {i+1}, Frame {j+1} analysis", result))
+                        logger.info(f"Successfully analyzed video {i+1}")
+                    except Exception as vid_error:
+                        logger.error(f"Error processing video {i+1}: {str(vid_error)}")
+                        results.append((f"Video {i+1} analysis", f"Error processing video: {str(vid_error)}"))
+                
+                else:
+                    logger.warning(f"Unsupported file type: {file_type}")
+                    results.append((f"File {i+1} analysis", f"Unsupported file type: {file_type}"))
+            
+            except Exception as file_error:
+                logger.error(f"Error processing file {i+1}: {str(file_error)}")
+                results.append((f"File {i+1} analysis", f"Error processing file: {str(file_error)}"))
 
         logger.info("Analysis completed successfully")
         return results