Update app.py

app.py

@@ -84,16 +84,18 @@ def analyze_mixed_input(input_files):
                        "If it does, identify any safety issues or hazards, categorize them, and provide a detailed description, "
                        "and suggest steps to resolve them. If it's not a construction site, simply state that")
 
-        for i, file_path in enumerate(input_files):
-            file_type = file_path.split('.')[-1].lower()
+        for i, file in enumerate(input_files):
+            file_type = file.name.split('.')[-1].lower()
             if file_type in ['jpg', 'jpeg', 'png', 'bmp']:
                 # Process image
-                image_data = encode_image(file_path)
+                image = Image.open(file.name)
+                resized_image = resize_image(image)
+                image_data_url = f"data:image/png;base64,{encode_image(resized_image)}"
                 content_type = "image"
             elif file_type in ['mp4', 'avi', 'mov', 'webm']:
                 # Process video
-                frames = extract_frames_from_video(file_path)
-                image_data = encode_image(frames[0])  # Use the first frame
+                frames = extract_frames_from_video(file.name)
+                image_data_url = f"data:image/png;base64,{encode_image(Image.fromarray(frames[0]))}"  # Use the first frame
                 content_type = "video"
             else:
                 results.append((f"File {i+1} analysis", f"Unsupported file type: {file_type}"))
@@ -110,7 +112,7 @@ def analyze_mixed_input(input_files):
                         {
                             "type": "image_url",
                             "image_url": {
-                                "url": f"data:image/jpeg;base64,{image_data}"
+                                "url": image_data_url
                             }
                         }
                     ]
@@ -132,7 +134,7 @@ def analyze_mixed_input(input_files):
             # If it's a video, analyze additional frames
             if content_type == "video" and len(frames) > 1:
                 for j, frame in enumerate(frames[1:], start=2):
-                    frame_data = encode_image(frame)
+                    frame_data_url = f"data:image/png;base64,{encode_image(Image.fromarray(frame))}"
                     messages = [
                         {
                             "role": "user",
@@ -144,7 +146,7 @@ def analyze_mixed_input(input_files):
                                 {
                                     "type": "image_url",
                                     "image_url": {
-                                        "url": f"data:image/jpeg;base64,{frame_data}"
+                                        "url": frame_data_url
                                     }
                                 }
                             ]