Skip to content

Gemini Chatbot Code

Published 4 min read191 views

Gemini Chatbot Code

Gemini Chatbot Code Documentation

This documentation provides an overview of a React-based chatbot application using the Gemini Language Model API. The chatbot interface allows users to type questions and receive generated responses.

Quick Preview

<iframe
src=“https://manishtamang.tech/"\ style={{
width: ‘100%’,
height: ‘auto’,
aspectRatio: ‘16 / 9’,
}}
title=“Live Preview”
/>

Components and Functions

Imports

import { useState, useEffect } from "react";
import axios from "axios";
import ReactMarkdown from "react-markdown";
import { Input } from "@/components/ui/input";
import { Button } from "@/components/ui/button";
  • useState and useEffect are React hooks for state management and side effects.
  • axios is used for making HTTP requests.
  • ReactMarkdown renders markdown content.
  • Input and Button are custom UI components.

State Variables

const [question, setQuestion] = useState("");
const [messages, setMessages] = useState([]);
const [generatingAnswer, setGeneratingAnswer] = useState(false);
  • question: Stores the current input question from the user.
  • messages: Keeps track of the chat messages between the user and the bot.
  • generatingAnswer: Indicates if the bot is currently generating an answer.

useEffect Hooks

Load Messages from Local Storage

useEffect(() => {
  const savedMessages =
    JSON.parse(localStorage.getItem("chatMessages")) || [];
  setMessages(savedMessages);
}, []);
  • Loads previously saved messages from localStorage when the component mounts.

Save Messages to Local Storage

useEffect(() => {
  localStorage.setItem("chatMessages", JSON.stringify(messages));
}, [messages]);
  • Saves messages to localStorage whenever the messages state changes.

How to get Gemini API key?

At first go to here

It will look like the below image:\

Click on the Get an API Key as shown in the picture.

Then it will redirect you to another page as shown in below image:\

Click on Create API key and copy the given api Key then Paste it in .env file (Remember to rename .env.example file as .env)

NEXT_PUBLIC_API_GENERATIVE_LANGUAGE_CLIENT=XXXXXXXXXXXXXXXXXXXXXXXX

generateAnswer Function

async function generateAnswer(e) {
  setGeneratingAnswer(true);
  e.preventDefault();
  const newMessage = { sender: "user", text: question };
  setMessages((prev) => [...prev, newMessage]);
  setQuestion("");
  
  try {
    const response = await axios({
      url: `https://generativelanguage.googleapis.com/v1beta/models/gemini-pro:generateContent?key=${process.env.NEXT_PUBLIC_API_GENERATIVE_LANGUAGE_CLIENT}`,
      method: "post",
      data: {
        contents: [{ parts: [{ text: newMessage.text }] }],
      },
    });

    const botMessage = {
      sender: "bot",
      text: response.data.candidates[0].content.parts[0].text,
    };
    setMessages((prev) => [...prev, botMessage]);
  } catch (error) {
    console.log(error);
    const errorMessage = {
      sender: "bot",
      text: "Sorry - Something went wrong. Please try again!",
    };
    setMessages((prev) => [...prev, errorMessage]);
  }

  setGeneratingAnswer(false);
}
  • Handles the submission of a user’s question.
  • Prevents the default form submission behavior.
  • Adds the user’s question to the messages state.
  • Sends the question to the Gemini API to get a response.
  • Adds the bot’s response to the messages state.
  • Handles errors by logging them and displaying an error message.
  • Disables the generating answer state when done.

JSX Structure

Main Container

<div className="flex items-center justify-center h-screen bg-gray-100 dark:bg-gray-900">
  • Centers the chatbot vertically and horizontally on the screen with background color.

Chat Window

<div className="flex flex-col h-[90vh] max-w-md w-full bg-white dark:bg-gray-950 rounded-2xl shadow-lg overflow-hidden">
  • Creates a vertical layout with defined height and width, and applies styles for the chat window.

Header

<header className="bg-gray-100 dark:bg-gray-900 px-4 py-3 flex items-center gap-3">
 <svg
            xmlns="http://www.w3.org/2000/svg"
            viewBox="0 0 256 256"
            fill="currentColor"
            className="w-6 h-6"
          >
            <path d="M230.92 212c-15.23-26.33-38.7-45.21-66.09-54.16a72 72 0 1 0-73.66 0c-27.39 8.94-50.86 27.82-66.09 54.16a8 8 0 1 0 13.85 8c18.84-32.56 52.14-52 89.07-52s70.23 19.44 89.07 52a8 8 0 1 0 13.85-8ZM72 96a56 56 0 1 1 56 56 56.06 56.06 0 0 1-56-56Z"></path>
          </svg>
  <div className="text-lg font-medium">Chatbot</div>
</header>
  • Displays the chatbot header with a title and icon.

*Message List

<div className="flex-1 overflow-y-auto p-4 space-y-4">
  {messages.map((msg, index) => (
    <div key={index} className={`flex ${msg.sender === "user" ? "justify-end" : "justify-start"}`}>
      <div className={`${msg.sender === "user" ? "bg-blue-500 text-white" : "bg-gray-100 dark:bg-gray-800"} px-4 py-3 rounded-2xl max-w-xs`}>
        <ReactMarkdown>{msg.text}</ReactMarkdown>
      </div>
    </div>
  ))}
</div>
  • Displays the list of messages with conditional styling based on the sender (user or bot).
  • Uses ReactMarkdown to render message text.

Input Form

<form onSubmit={generateAnswer} className="flex items-center gap-2 p-4 bg-gray-100 dark:bg-gray-900">
  <Input
    type="text"
    placeholder="Type your message..."
    className="flex-1 bg-transparent"
    value={question}
    onChange={(e) => setQuestion(e.target.value)}
    disabled={generatingAnswer}
  />
  <Button
    type="submit"
    variant="ghost"
    size="icon"
    className="text-blue-500"
    disabled={generatingAnswer}
  >
    Send
  </Button>
</form>
  • Handles user input and form submission.
  • Disables input and button while waiting for the bot’s response.
  • Uses Input and Button components with styles and interactivity.

Summary

This React component creates a chatbot interface that saves chat history to local storage and interacts with the Gemini Language Model API to generate responses to user questions. The component is structured with a header, message list, and input form, and includes error handling and state management for a responsive user experience.

Repo Link : Github