Update app.py

app.py

@@ -273,6 +273,12 @@ def identify_required_functions(project_path, functionality_description):
     return response.text
 
 
+
+
+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)]
+
 # Hugging Face Inference API endpoint for the model
 API_URL = "https://api-inference.huggingface.co/models/Qwen/Qwen2.5-Coder-32B-Instruct"
 qwen = os.getenv("QWEN")
@@ -280,7 +286,7 @@ headers = {"Authorization": f"Bearer {qwen}"}
 
 def validate_and_generate_documentation(api_url, headers, gemini_output, file_contents, functionality_description):
     """Uses the Hugging Face Inference API to validate functions and generate documentation."""
-    # Generate the prompt for the Qwen model
+    # Generate the base prompt for the Qwen model
     prompt = f"""
     User-specified functionality: '{functionality_description}'
     Functions identified by Gemini:
@@ -289,9 +295,10 @@ def validate_and_generate_documentation(api_url, headers, gemini_output, file_co
     Project files:
     """
     for file_path, content in file_contents.items():
-        # Truncate content to avoid excessively large payloads
-        truncated_content = content[:1000] if len(content) > 1000 else content
-        prompt += f"File: {os.path.basename(file_path)}\n{truncated_content}\n\n"
+        # Split content into manageable chunks if necessary
+        chunks = split_into_chunks(content, chunk_size=3000)  # Adjust chunk size based on model's max input length
+        for idx, chunk in enumerate(chunks):
+            prompt += f"File: {os.path.basename(file_path)} (Part {idx + 1})\n{chunk}\n\n"
 
     prompt += """
     Task:
@@ -323,10 +330,17 @@ def validate_and_generate_documentation(api_url, headers, gemini_output, file_co
 
     # Handle the API response
     if response.status_code == 200:
-        return response.json().get("generated_text", "No output generated.")
+        api_response = response.json()
+        if isinstance(api_response, list):  # If response is a list, extract the first element
+            return api_response[0].get("generated_text", "No output generated.")
+        elif isinstance(api_response, dict):  # Handle dict response
+            return api_response.get("generated_text", "No output generated.")
+        else:
+            raise ValueError("Unexpected response format from Hugging Face API.")
     else:
         raise ValueError(f"Error: {response.status_code}, {response.text}")
 
+
 def generate_documentation_page():
     st.subheader(f"Generate Documentation for {st.session_state.current_project}")
     st.write("Enter the functionality or parts of the project for which you'd like to identify relevant functions.")