Update agent.py

agent.py

@@ -367,8 +367,57 @@ def reverse_text(text: str) -> str:
     """Fix reversed text patterns"""
     return text[::-1].replace("\\", "").strip() if text.startswith(('.', ',')) else text
 
+
 # 3. Unified Document Creation
 
+def create_documents(data_source: str, data: list) -> list:
+    """Handle both Gradio chat and JSON questions"""
+    docs = []
+    
+    for item in data:
+        content = ""
+        # Process different data sources
+        if data_source == "json":
+            raw_question = item.get("question", "")
+            content = raw_question  # Adjust as per your content processing logic
+        else:
+            print(f"Skipping invalid data source: {data_source}")
+            continue
+
+        # Ensure metadata type safety
+        metadata = {
+            "task_id": str(item.get("task_id", "")),
+            "level": str(item.get("Level", "")),
+            "file_name": str(item.get("file_name", ""))
+        }
+
+        # Check if content is non-empty
+        if content.strip():  # Only append non-empty content
+            docs.append(Document(page_content=content, metadata=metadata))
+        else:
+            print(f"Skipping invalid entry with empty content: {item}")
+
+    return docs
+
+# Path to your data.json
+file_path = "/home/wendy/Downloads/data.json"
+
+def load_data(file_path: str) -> list[dict]:
+    """Safe JSON data loading with error handling"""
+    if not os.path.exists(file_path):
+        raise FileNotFoundError(f"Data file not found: {file_path}")
+        
+    if not file_path.endswith('.json'):
+        raise ValueError("Invalid file format. Only JSON files supported")
+        
+    try:
+        with open(file_path, "r", encoding="utf-8") as f:
+            return json.load(f)
+    except json.JSONDecodeError:
+        raise ValueError("Invalid JSON format in data file")
+    except Exception as e:
+        raise RuntimeError(f"Error loading data: {str(e)}")
+
     
 
 # 4. Vector Store Integration
@@ -392,6 +441,20 @@ class MyVector_Store:
 # Process JSON data and create documents
 # -----------------------------
 
+file_path = "/home/wendy/Downloads/data.json"
+
+try:
+    with open(file_path, "r", encoding="utf-8") as f:
+        data = json.load(f)
+        print(data)
+except FileNotFoundError as e:
+    print(f"Error: {e}")
+except json.JSONDecodeError as e:
+    print(f"Error decoding JSON: {e}")
+
+docs = create_documents("json", data)
+texts = [doc.page_content for doc in docs]
+
 
 # -----------------------------
 # Initialize embedding model