import streamlit as st
import sqlite3
import hashlib
import os
import google.generativeai as genai
import zipfile
from git import Repo
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
import requests
# Database setup
DB_FILE = "users.db"
def create_user_table():
conn = sqlite3.connect(DB_FILE)
cursor = conn.cursor()
cursor.execute("""
CREATE TABLE IF NOT EXISTS users (
username TEXT PRIMARY KEY,
password TEXT
)
""")
conn.commit()
conn.close()
def add_user(username, password):
conn = sqlite3.connect(DB_FILE)
cursor = conn.cursor()
hashed_password = hashlib.sha256(password.encode()).hexdigest()
try:
cursor.execute("INSERT INTO users (username, password) VALUES (?, ?)", (username, hashed_password))
conn.commit()
except sqlite3.IntegrityError:
st.error("Username already exists. Please choose a different username.")
conn.close()
def authenticate_user(username, password):
conn = sqlite3.connect(DB_FILE)
cursor = conn.cursor()
hashed_password = hashlib.sha256(password.encode()).hexdigest()
cursor.execute("SELECT * FROM users WHERE username = ? AND password = ?", (username, hashed_password))
user = cursor.fetchone()
conn.close()
return user
def initialize_session_state():
if "authenticated" not in st.session_state:
st.session_state.authenticated = False
if "username" not in st.session_state:
st.session_state.username = None
if "page" not in st.session_state:
st.session_state.page = "login"
if "current_project" not in st.session_state:
st.session_state.current_project = None
if "project_uploaded" not in st.session_state:
st.session_state.project_uploaded = False
def main():
st.title("SimplifAI")
# Initialize session state
initialize_session_state()
# Initialize database
create_user_table()
# Page routing logic
if st.session_state.page == "login":
login_page()
elif st.session_state.page == "workspace":
workspace_page()
elif st.session_state.page == "project_view":
project_view_page()
elif st.session_state.page == "generate_documentation":
generate_documentation_page()
elif st.session_state.page == "view_documentation":
view_documentation_page()
def login_page():
st.subheader("Please Log In or Register to Continue")
auth_mode = st.radio("Choose an Option", ["Log In", "Register"], horizontal=True)
if auth_mode == "Log In":
st.subheader("Log In")
username = st.text_input("Username", key="login_username")
password = st.text_input("Password", type="password", key="login_password")
# Handle single-click login
if st.button("Log In"):
if authenticate_user(username, password):
st.session_state.authenticated = True
st.session_state.username = username
st.session_state.page = "workspace"
else:
st.error("Invalid username or password. Please try again.")
elif auth_mode == "Register":
st.subheader("Register")
username = st.text_input("Create Username", key="register_username")
password = st.text_input("Create Password", type="password", key="register_password")
# Handle single-click registration
if st.button("Register"):
if username and password:
add_user(username, password)
st.success("Account created successfully! You can now log in.")
else:
st.error("Please fill in all fields.")
def workspace_page():
# Sidebar with logout button
st.sidebar.title(f"Hello, {st.session_state.username}!")
if st.sidebar.button("Log Out"):
st.session_state.authenticated = False
st.session_state.username = None
st.session_state.page = "login"
# User's folder for projects
user_folder = os.path.join("user_projects", st.session_state.username)
os.makedirs(user_folder, exist_ok=True)
# Refresh project list dynamically
projects = [d for d in os.listdir(user_folder) if os.path.isdir(os.path.join(user_folder, d))]
# Display "Projects" dropdown
selected_project = st.sidebar.selectbox("Projects", ["Select a project"] + projects)
if selected_project != "Select a project":
st.session_state.current_project = selected_project
st.session_state.page = "project_view"
st.rerun()
# Display success message if a project was uploaded
if st.session_state.project_uploaded:
st.success(f"Project '{st.session_state.current_project}' uploaded successfully!")
st.session_state.project_uploaded = False # Reset after showing the message
# Main content area
st.subheader("Workspace")
st.write("You can create a new project by uploading files or folders, or by cloning a GitHub repository.")
# Existing upload functionality...
# User action selection
action = st.radio("Choose an action", ["Upload Files or Folders", "Clone GitHub Repository"], horizontal=True)
project_name = st.text_input("Enter a project name")
if action == "Upload Files or Folders":
st.subheader("Upload Files or Folders")
uploaded_files = st.file_uploader(
"Upload one or more files or a .zip archive for folders", accept_multiple_files=True
)
if uploaded_files and project_name:
if st.button("Upload Project"):
project_folder = os.path.join(user_folder, project_name)
os.makedirs(project_folder, exist_ok=True)
for uploaded_file in uploaded_files:
# Save uploaded .zip files or regular files
file_path = os.path.join(project_folder, uploaded_file.name)
with open(file_path, "wb") as f:
f.write(uploaded_file.getbuffer())
# If a .zip file is uploaded, extract its contents
if uploaded_file.name.endswith(".zip"):
try:
with zipfile.ZipFile(file_path, "r") as zip_ref:
zip_ref.extractall(project_folder)
os.remove(file_path) # Remove the .zip file after extraction
st.success(f"Folder from {uploaded_file.name} extracted successfully!")
except zipfile.BadZipFile:
st.error(f"File {uploaded_file.name} is not a valid .zip file.")
else:
st.success(f"File {uploaded_file.name} saved successfully!")
# Update session state and trigger a rerun
st.session_state.current_project = project_name
st.session_state.project_uploaded = True
st.rerun()
elif action == "Clone GitHub Repository":
st.subheader("Clone GitHub Repository")
repo_url = st.text_input("Enter the GitHub repository URL")
if repo_url and project_name:
if st.button("Upload Project"):
project_folder = os.path.join(user_folder, project_name)
os.makedirs(project_folder, exist_ok=True)
try:
Repo.clone_from(repo_url, project_folder)
# Update session state and trigger a rerun
st.session_state.current_project = project_name
st.session_state.project_uploaded = True
st.rerun()
except Exception as e:
st.error(f"Failed to clone repository: {e}")
#------------------------------------------------------------------------------------------------------------------------------------------------------------------------
# Configure Gemini API
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 = []
for root, _, files in os.walk(project_path):
for file in files:
# Skip .git files or folders from GitHub clones
if ".git" not in root:
file_paths.append(os.path.join(root, file))
return file_paths
def read_files(file_paths):
"""Reads content from a list of file paths."""
file_contents = {}
for file_path in file_paths:
if os.path.exists(file_path):
try:
# Attempt to read the file as UTF-8
with open(file_path, 'r', encoding='utf-8') as file:
file_contents[file_path] = file.read()
except UnicodeDecodeError:
print(f"Skipping binary or non-UTF-8 file: {file_path}")
else:
print(f"File not found: {file_path}")
return file_contents
def generate_prompt(file_contents, functionality_description):
"""Generates a prompt for Gemini to analyze the files."""
prompt = "Analyze the following code files to identify all functions required to implement the functionality: "
prompt += f"'{functionality_description}'.\n\n"
for file_path, content in file_contents.items():
prompt += f"File: {os.path.basename(file_path)}\n{content}\n\n"
prompt += "For each relevant function, provide:\n"
prompt += "1. Which file the function is found in.\n"
prompt += "2. The function name.\n"
prompt += "3. Dependencies on other functions or modules.\n"
prompt += """
Return your output in the following format providing no commentary:
Project Summary:
Functionality:
Functions:
:
-:
-Function Dependencies:
"""
return prompt
def identify_required_functions(project_path, functionality_description):
"""Identifies required functions for a specified functionality."""
# Gather all file paths in the project directory
file_paths = read_project_files(project_path)
# Read file contents
file_contents = read_files(file_paths)
# Generate a refined prompt for Gemini
prompt = generate_prompt(file_contents, functionality_description)
# Call the Gemini model
response = model.generate_content(prompt)
# Process and return the response
return response.text
def extract_cleaned_gemini_output(gemini_output):
"""
Removes the 'Functions' section and any irrelevant content from the Gemini output.
Args:
gemini_output (str): The raw output returned by Gemini.
Returns:
str: Cleaned output without the 'Functions' section.
"""
lines = gemini_output.splitlines()
cleaned_output = []
skip_section = False
for line in lines:
line = line.strip()
# Detect the start of the "Functions:" section and skip it
if line.startswith("Functions:"):
skip_section = True
# Detect the end of the "Functions:" section and resume processing
if skip_section and (line.startswith("Tasks:") or line.startswith("Project Summary:")):
skip_section = False
# Append lines outside the skipped section
if not skip_section:
cleaned_output.append(line)
# Return the cleaned output with blank lines filtered out
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)]
# 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")
headers = {"Authorization": f"Bearer {qwen}"}
def extract_cleaned_gemini_output(gemini_output):
"""
Extracts and formats the cleaned output from Gemini to send to Qwen.
Args:
gemini_output (str): The output returned by Gemini.
Returns:
str: Cleaned and formatted output for Qwen.
"""
lines = gemini_output.splitlines()
cleaned_output = []
functions_section = False
for line in lines:
line = line.strip()
if line.startswith("Project Summary:") or line.startswith("Functionality:"):
cleaned_output.append(line)
elif line.startswith("Functions:"):
cleaned_output.append(line)
functions_section = True
elif functions_section and line:
cleaned_output.append(line)
elif line.startswith("File:") or "Qwen," in line:
break
return "\n".join(cleaned_output)
def clean_output(output):
"""
Cleans the final output to include only documentation sections.
"""
lines = output.splitlines()
filtered_lines = []
in_valid_section = False
for line in lines:
line = line.strip()
# Include only valid sections
if line.startswith("Project Summary:") or line.startswith("Functionality Summary:") or line.startswith("Functionality Flow:"):
in_valid_section = True
filtered_lines.append(line)
elif line.startswith("Function Documentation:"):
in_valid_section = True
filtered_lines.append(line)
# Stop processing at any undesired section
elif line.startswith("User-specified functionality:") or line.startswith("Tasks identified by Gemini:"):
in_valid_section = False
elif in_valid_section and line:
filtered_lines.append(line)
return "\n".join(filtered_lines)
def validate_and_generate_documentation(api_url, headers, gemini_output, functionality_description):
"""
Uses the Hugging Face Inference API to generate clean and relevant documentation using Qwen.
"""
# Clean Gemini output
cleaned_gemini_output = extract_cleaned_gemini_output(gemini_output)
# Generate the refined prompt for Qwen
prompt = f"""
User-specified functionality: '{functionality_description}'
Functions identified by Gemini:
{cleaned_gemini_output}
Tasks:
1. Generate a project summary:
'
Project Summary:
'
2. Refine the user-defined functionality:
'
Functionality Summary:
'
3. Describe the functionality flow:
'
Functionality Flow:
'
4. Generate detailed documentation for each function:
'
Function Documentation:
For each relevant function:
- Summary:
- Inputs:
- Outputs:
- Dependencies:
- Data structures:
- Algorithmic Details:
- Error Handling:
- Assumptions:
- Example Usage:
'
5. Return only the required information for the above tasks, and exclude everything else.
"""
# Prepare payload and call API
payload = {"inputs": prompt, "parameters": {"max_new_tokens": 1024}}
response = requests.post(api_url, headers=headers, json=payload)
# Handle API response
if response.status_code == 200:
api_response = response.json()
output = api_response.get("generated_text", "") if isinstance(api_response, dict) else api_response[0].get("generated_text", "")
return clean_output(output)
else:
raise ValueError(f"Error during API call: {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 generate documentation.")
# Prompt user for functionality description
functionality = st.text_area(
"Describe the functionality",
placeholder="e.g., Explain the function of the file `main.py`",
)
# Button to start analyzing functionality
if st.button("Analyze"):
if functionality.strip():
st.write("Analyzing project files... Please wait.")
# Get the path of the current project
user_folder = os.path.join("user_projects", st.session_state.username)
project_folder = os.path.join(user_folder, st.session_state.current_project)
if os.path.exists(project_folder):
try:
# Call Gemini to identify required functions
gemini_result = identify_required_functions(project_folder, functionality)
# Generate documentation using Qwen
final_documentation = validate_and_generate_documentation(
API_URL, headers, gemini_result, functionality
)
# Display the final documentation
st.success("Documentation generated successfully!")
st.text_area("Generated Documentation", final_documentation, height=600)
except Exception as e:
st.error(f"An error occurred: {e}")
else:
st.error("Project folder not found. Ensure the GitHub repository was cloned successfully.")
else:
st.error("Please enter the functionality to analyze.")
#------------------------------------------------------------------------------------------------------------------------------------------------------------------------
def view_documentation_page():
st.subheader(f"View Documentation for {st.session_state.current_project}")
st.write("This page will display the generated documentation for the selected project.")
if st.button("Back to Project"):
st.session_state.page = "project_view"
st.rerun()
def project_view_page():
# Sidebar with logout and return buttons
st.sidebar.title(f"Project: {st.session_state.current_project}")
if st.sidebar.button("Back to Workspace"):
st.session_state.page = "workspace"
st.rerun()
if st.sidebar.button("Log Out"):
st.session_state.authenticated = False
st.session_state.username = None
st.session_state.page = "login"
st.rerun()
# Main content for project page
st.subheader(f"Project: {st.session_state.current_project}")
st.write("Manage your project and explore its files.")
# Buttons for documentation functionality
if st.button("Generate Documentation"):
st.session_state.page = "generate_documentation"
st.rerun()
if st.button("View Documentation"):
st.session_state.page = "view_documentation"
st.rerun()
# Toggle file structure display (if required)
if "show_file_structure" not in st.session_state:
st.session_state.show_file_structure = False
if st.button("Show File Structure"):
st.session_state.show_file_structure = not st.session_state.show_file_structure
if st.session_state.show_file_structure:
user_folder = os.path.join("user_projects", st.session_state.username)
project_folder = os.path.join(user_folder, st.session_state.current_project)
st.write("File structure:")
for root, dirs, files in os.walk(project_folder):
level = root.replace(project_folder, "").count(os.sep)
indent = " " * 4 * level
if level == 0:
st.write(f"📂 {os.path.basename(root)}")
else:
with st.expander(f"{indent}📂 {os.path.basename(root)}"):
sub_indent = " " * 4 * (level + 1)
for file in files:
st.write(f"{sub_indent}📄 {file}")
if __name__ == "__main__":
main()