npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2024 – Pkg Stats / Ryan Hefner

@david-vectara/react-chatbot

v0.1.6

Published

A Vectara-powered Chatbot component

Downloads

126

Readme

React-Chatbot

React-Chatbot is a UI widget for adding Vectara-powered chatbot to your React apps with a few lines of code.

[!TIP]

Looking for something else? Try another open-source project:

  • React-Search: Add Vectara semantic search to your React apps with a few lines of code.
  • Create-UI: The fastest way to generate a working React codebase for a range of generative and semantic search UIs.
  • Vectara Answer: Demo app for Summarized Semantic Search with advanced configuration options.
  • Vectara Ingest: Sample templates and crawlers for pulling data from many popular data sources.

Demo

Try out the demo!

UI

React-Chatbot adds a button to the lower right corner of your application:

When the button is clicked, a chat window opens, where your users can start a conversation with your data:

The Vectara platform responds to user messages and presents a list of references that support its answer:

When the chat API returns an error, the UI allows the user to retry sending the message:

React-Chatbot is also ready for use on mobile sites, taking the fullscreen when the chat window is opened:

Use it in your application

Use the chatbot directly

Install React-Chatbot:

npm install --save @vectara/react-chatbot

Then use it in your application like this:

import { ReactChatbot } from "@vectara/react-chatbot";

/* snip */

<ReactChatbot
  customerId="CUSTOMER_ID"
  corpusId={["CORPUS_ID_1", "CORPUS_ID_2", "CORPUS_ID_N"]}
  apiKey="API_KEY"
  title="My Chatbot"
  placeholder="Chat with your AI assistant"
  inputSize="large"
  emptyStateDisplay={<MyEmptyStateDisplayComponent />}
  isInitiallyOpen={false}
  zIndex={ /* (optional) number representing the z-index the component should have */ }
/>;

Configuration Options

customerId (required)

Every Vectara account is associated with a customer ID. You can find your customer ID by logging into the Vectara Console and opening your account dropdown in the top-right corner.

corpusIds (required)

After you create a corpus, you can find its ID by navigating to the corpus and looking in the top-left corner, next to the corpus name.

apiKey (required)

API keys enable applications to access data inside of corpora. Learn how to create a QueryService API key.

title (optional)

Configure the title in the header of the chatbot window.

placeholder (optional)

Configure the placeholder text in the chatbot's input.

emptyStateDisplay (optional)

Configure JSX content to render in the messages window when there are no messages to display.

isInitiallyOpen (optional)

Set the chat window to be opened on initial render.

inputSize (optional)

Control the size of the input - either "large" (18px) or "medium" (14px)

zIndex (optional)

Customize the z-index of the chatbot widget

Use your own views with the useChat hook

Install React-Chatbot:

npm install --save @vectara/react-chatbot

Then use the useChat hook in your application like this:

import { useChat } from "@vectara/react-chatbot/lib";

/* snip */

const { sendMessage, startNewConversation, messageHistory, isLoading, hasError } = useChat(
  "CUSTOMER_ID",
  ["CORPUS_ID_1", "CORPUS_ID_2", "CORPUS_ID_N"],
  "API_KEY"
);

The values returned by the hook can be passed on to your custom components as props or used in any way you wish.

Hook Values

sendMessage: async ({ query: string; isRetry?: boolean }) => void;

This is used to send a message to the chat API. Send true for the optional isRetry flag to if this is retrying a previously failed message. This allows the internal logic to correctly link the next successful answer to the failed query.

startNewConversation: () => void;

This is used to reset the conversational context of the chat. The message history will be cleared and the chatbot will "forget" everything that's been discussed.

messageHistory: ChatTurn[]

This is an array of objects representing question and answer pairs from the entire conversation. Each pair is a ChatTurn object. More information on types can be found here.

isLoading: boolean

A boolean value indicating whether or not a chat API request is still pending

hasError: boolean

A boolean value indicating whether or not the most recent chat API request encountered an error

Usage with SSR Frameworks

Using React-Chatbot with SSR frameworks may require additional infrastructure. Here are some common gotchas:

Next.js

Since React-Chatbot requires a few browser-only features to function, we need to defer the rendering of the component to the client. In order to do this, you will need to:

  1. Use the "use client" directive in the file that imports React-Chatbot.
  2. Use a one-time useEffect call in React-Chatbot's consumer. The useEffect callback should import and set the rendered widget as a state on the consumer component. We do the import here as some declarations in imported file also require resources only available in the browser.
  3. Include the rendered widget state value in the rendered consumer component.

Example:

"use client";

export const App = (props: Props): ReactNode => {
  const [chatWidget, setChatWidget] = useState<ReactNode>(null);

  /* the rest of your code */

  useEffect(() => {
    const importAndCreateWidget = async () => {
      const { ReactChatbot } = await import("@vectara/react-chatbot");

      setChatWidget(
        <ReactChatbot
          customerId="CUSTOMER_ID"
          corpusIds={["CORPUS_ID_1", "CORPUS_ID_2", "CORPUS_ID_N"]}
          apiKey="API_KEY"
        />
      );
    }

    importAndCreateWidget();
  }, []);

  return (
    <>
      { /* other content */ }
    </>
  )
};

Set up your data

React-Chatbot uses the data in your Vectara corpus to provide factual responses. To set this up:

  1. Create a free Vectara account.
  2. Create a corpus and add data to it.
  3. Create a QueryService API key.

Pro-tip: After you create an API key, navigate to your corpus and click on the "Access control" tab. Find your API key on the bottom and select the "Copy all" option to copy your customer ID, corpus ID, and API key. This gives you all the data you need to configure your <ReactChat /> instance.

Copy all option

Maintenance

This codebase comes with a development environment to facilitate enhancements and bug fixes. It allows maintainers to quickly iterate on the code and verify changes instantly.

Running the development environment

From the root directory, run:

npm run installAllDependencies

This completes the following steps:

  • installs dependencies for React-Chatbot
  • builds React-Chatbot
  • installs the freshly-built React-Chatbot package in the docs site directory

Once this completes, run:

npm run docs

This spins up an application running at http://localhost:8080/. Your latest changes will be reflected here.

Making changes to the component

Once the development environment is running, any changes made to .ts and .tsx files in the /src directory will trigger a rebuild of the component and a reload of the webpage.

Additionally, any changes to the development app source code at /docs/index.tsx will also trigger a rebuild + reload.

License

Vectara React-Chatbot is an open-sourced software licensed under the Apache 2.0 license.