Integrating Search UI with NextJS Router

Next JS is a popular React framework that provides a number of features out of the box, including server-side rendering, routing, and more.

To take advantage of being able to save the search criteria in the URL, you need to override the existing routing options and use Next Router.

Below is an example of how to do achieve this.

To highlight, we're using React memo to prevent the component from re-rendering the Search UI chrome when the URL changes.

import { memo } from 'react'

export default function App() {
  // useNextRouting is a custom hook that will integrate with Next Router with Search UI config
  // config is search-ui configuration.
  // baseUrl is the path to the search page
  const combinedConfig = useNextRouting(config, "<baseUrl>");
  return <MemoSearch config={combinedConfig} />;
}

const MemoSearch = memo(({ config }) => {
  return (
    <SearchProvider config={config}>
      <>
        <ResultsPerPage />
        <Results config={config} />
        <Paging />
      </>
    </SearchProvider>
    )
})

Integration with Elasticsearch Connector

We do not advise exposing your Elasticsearch instance to the public. Fortunately with NextJS, we do not need to do this. NextJS makes it very simple to build an API route so we can move the connector to the server side.

To start, we create a connector that will send the requestState and queryConfig over to the client side.

// located in <root>/services/APIConnector.js

class APIConnector {
  constructor() {}

  onResultClick() {
    // optional. Called when a result has been clicked
  }
  onAutocompleteResultClick() {
    // optional. Called when an autocomplete result has been clicked
  }

  async onSearch(requestState, queryConfig) {
    const response = await fetch("/api/search", {
      method: "POST",
      headers: {
        "Content-Type": "application/json"
      },
      body: JSON.stringify({
        requestState,
        queryConfig
      })
    });
    return response.json();
  }

  async onAutocomplete(requestState, queryConfig) {
    const response = await fetch("api/autocomplete", {
      method: "POST",
      headers: {
        "Content-Type": "application/json"
      },
      body: JSON.stringify({
        requestState,
        queryConfig
      })
    });
    return response.json();
  }
}

export default APIConnector;

then add an api folder within <root>/pages folder. Add two files, autocomplete.js and search.js

// located in <root>/pages/api/search.js
import ElasticsearchAPIConnector from "@elastic/search-ui-elasticsearch-connector";

const connector = new ElasticsearchAPIConnector({
  host: "<elasticsearch host>",
  index: "<elasticsearch index>",
  apiKey: "<api key>" // optional
});

export default async function handler(req, res) {
  const { requestState, queryConfig } = req.body;
  const response = await connector.onSearch(requestState, queryConfig);
  res.json(response);
}
// located in <root>/pages/api/autocomplete.js
import ElasticsearchAPIConnector from "@elastic/search-ui-elasticsearch-connector";

const connector = new ElasticsearchAPIConnector({
  host: "<elasticsearch host>",
  index: "<elasticsearch index>",
  apiKey: "<api key>" // optional
});

export default async function handler(req, res) {
  const { requestState, queryConfig } = req.body;
  const response = await connector.onAutocomplete(requestState, queryConfig);
  res.json(response);
}

then lastly swap the elasticsearch connector to use the APIConnector. With this change, when a user searches, the search will be sent to the server and the API routes will fetch the results from Elasticsearch and will be sent back to the client.

import Connector from "../services/APIConnector";

const apiConnector = new Connector();

const searchConfig = {
  apiConnector: apiConnector
  // ...truncated search configuration
};

and then restart your nextjs app. To confirm that this is working, you should see network requests within chrome developer tools performing requests with requestState and queryConfig to the /api/search & /api/autocomplete API routes and results being returned back to the client.

Below is an example of this running within codesandbox.