Update app.py

app.py

@@ -225,8 +225,6 @@ gemini = os.getenv("GEMINI")
 genai.configure(api_key=gemini)
 model = genai.GenerativeModel("gemini-1.5-flash")
 
-
-
 def read_project_files(project_path):
     """Reads all files in the project directory and its subdirectories."""
     file_paths = []
@@ -298,6 +296,69 @@ def identify_required_functions(project_path, functionality_description):
     # Process and return the response
     return response.text
 
+def generate_qwen_documentation(function_list, project_path, functionality_description):
+    """
+    Iteratively generates documentation for functions using Qwen and ensures all are fully documented.
+    Args:
+        function_list (list): Initial list of functions related to the specified functionality.
+        project_path (str): Path to the project directory.
+        functionality_description (str): Description of the functionality.
+    Returns:
+        dict: A dictionary with Qwen-generated documentation for each function.
+    """
+    completed_documentation = {}
+    remaining_functions = function_list
+
+    while remaining_functions:
+        # Generate documentation for remaining functions using Qwen
+        qwen_outputs = query_qwen_for_documentation(remaining_functions, project_path)
+
+        # Clean and validate Qwen's output using Gemini
+        completed, truncated = clean_and_validate_qwen_output(qwen_outputs, remaining_functions)
+
+        # Update documentation and determine remaining functions
+        completed_documentation.update(completed)
+        remaining_functions = truncated  # Retry only truncated or missing functions
+
+    return completed_documentation
+
+def query_qwen_for_documentation(function_list, project_path):
+    """
+    Queries Hugging Face API to generate documentation for functions.
+    Args:
+        function_list (list): List of functions for documentation.
+        project_path (str): Path to the project directory.
+    Returns:
+        dict: Hugging Face API's output for each function.
+    """
+
+    passkey = os.getenv("QWEN")  # Hugging Face API key from environment variables
+    api_url = "https://api-inference.huggingface.co/models/Qwen/Qwen2.5-Coder-14B-Instruct"
+    headers = {"Authorization": f"Bearer {passkey}"}
+
+    outputs = {}
+    for file_name, function in function_list:
+        prompt = f"Generate detailed documentation for the following function in file `{file_name}`:\n{function}"
+        payload = {"inputs": prompt, "parameters": {"max_length": 1024, "truncate": True}}
+
+        # Send the request to the Hugging Face API
+        try:
+            response = requests.post(api_url, headers=headers, json=payload)
+            response.raise_for_status()
+            result = response.json()
+
+            # Extract generated text from the API response
+            if "generated_text" in result:
+                outputs[(file_name, function)] = {"prompt": prompt, "output": result["generated_text"]}
+            else:
+                st.error(f"No generated text for function: {function}. Response: {result}")
+
+        except requests.exceptions.RequestException as e:
+            st.error(f"Error querying Hugging Face API for function {function}: {e}")
+            outputs[(file_name, function)] = {"prompt": prompt, "output": None}
+
+    return outputs
+
 
 def extract_cleaned_gemini_output(gemini_output):
     """
@@ -330,14 +391,12 @@ def extract_cleaned_gemini_output(gemini_output):
     return "\n".join(line for line in cleaned_output if line)
 
 
-
-
-
-
 def split_into_chunks(content, chunk_size=1000):
     """Splits large content into smaller chunks."""
     return [content[i:i + chunk_size] for i in range(0, len(content), chunk_size)]
 
+
+
 def generate_detailed_documentation(file_contents, functionality_description):
     """
     Generates detailed documentation using Gemini directly.