Repo: https://github.com/AndreaKinder/HuChat-Chat-Bot-Assistant-Desktop/tree/master

Introduction

In today’s AI and natural language processing (NLP)-driven world, HuChat emerges as a user-friendly desktop application that facilitates seamless interaction with chatbots. Leveraging the power of the Hugging Face API, an open-source Python library for simplifying interactions with language models and chatbots, HuChat aims to provide an engaging and personalized assistant experience.

Technology Stack

Programming Language: Python

APIs and Dependencies:

• os
• json
• hugchat
• Tkinter
• Async
• Threading

User Interface Design:

UI Colors:

Color Usage	Color Reference
Primary Color	d8d8d8
Secondary Color (buttons, …)	fe9801
Text Color	000000

Alpha Version Functionality Overview

Primary Use Case

Desktop chatbot assistant.

Code

Code reference: https://github.com/AndreaKinder/HuChat-Chat-Bot-Assistant-Desktop/blob/master/src/main.py

Imports

• os: Provides tools for interacting with the operating system (creating folders, working with files).
• json: Handles JSON (JavaScript Object Notation) data, used to store login details.
• customtkinter as ctk: A library to make modern-looking interfaces with tkinter.
• hugchat: A library designed for interacting with a Hugging Face chatbot (which is likely a conversational AI model).
• hugchat.login: Provides functions to log into the Hugging Face chatbot service.
• tkinter as tk: The foundation of Python’s GUI toolkit.
• ttk: A tkinter extension for themed widgets (buttons, labels, etc.) that look nicer.
• asyncio: Enables asynchronous programming in Python, which helps manage tasks running concurrently.
• threading: Allows code to run in separate threads, preventing the main GUI from freezing.

Global Variables

• folder_name: The name of the folder where data will be saved (‘storage’).
• file_path: The complete path to the login information file (‘storage/log.json’).
• capture_id_chat: A list to keep track of chat conversation IDs.
• global_chatbot: Stores the main chatbot object that can be accessed from anywhere in the code.

Functions

• capture_log(us, passwd):
• Purpose: Stores the user’s username and password as a JSON file (‘log.json’).
• Creates the ‘storage’ folder if it’s missing.
• read_log():
• Purpose: Fetches the user’s username and password from ‘log.json’.
• check_directory(directory):
• Purpose: Checks to see if a folder exists.
• Example: if check_directory(‘storage’): print(‘The storage folder exists’)
• check_file_true():
• Purpose: A simple function to print a message indicating a directory and file were found.
• check_file_false():
• Purpose: Prints a message when a folder or file cannot be found. Also triggers the creation of the login window.
• check_file_log():
• Purpose: Looks for the login file (‘log.json’). If found, it allows automatic login. Otherwise, shows the login window.
• import_log():
• Purpose: Loads the user’s login information from ‘log.json’, if it exists.
• create_cookies():
• Purpose: Generates login cookies to use with the Hugging Face chatbot.
• How it works: It uses the import_log function to get login details and the Login object from the hugchat.login module.
• capture_id_chat_id(id_chat):
• Purpose: Stores chat conversation IDs to keep a history of conversations.
• create_chatbot():
• Purpose: Initializes the hugchat.ChatBot object. Also, it handles creating new chat instances when needed.
• chat_init(send): (Asynchronous)
• Purpose: Starts a chat and sends the user’s first message.
• How it works: It may create a new chat if none exist. Leverages the Hugging Face backend for the chatbot’s responses.
• chat_continue(send): (Asynchronous)
• Purpose: Continues a chat conversation that’s already in progress.
• chat_end():
• Purpose: Deletes all chats linked to the user’s login (this assumes the Hugging Face backend has the ability to do this).
• run_asyncio_coroutine(coroutine):
• Purpose: A helper to run an asynchronous coroutine (defined with async and await) within a thread. This is necessary for updating the GUI.
• send_question_async(question):
• Purpose: Asynchronously sends a question to the chatbot and updates the chat interface with the reply.
• How it works: Calls either chat_init or chat_continue, depending on whether there’s a chat in progress.
• clear_chat():
• Purpose: Deletes the chat history from the interface.
• send_question(event=None):
• Purpose: Gets the user’s text from the input box and starts the asynchronous sending process.

Classes

• InputUs(ctk.CTkFrame): A custom frame for entering the username.
• InputPasswd(ctk.CTkFrame): A custom frame for entering the password.
• ButtonLog(ctk.CTkButton): A custom button for logging in.
• App(ctk.CTk): The main window of the app; uses custom tkinter widgets for styling.