UI libraries / Autocomplete / Core concepts

Displaying items with Templates

Once you’ve set up your data sources, you need to define how they display in your autocomplete experience. It encompasses the structure for each item and the way they look.

Autocomplete provides a Templates API to let you fully customize the render of each item.

Rendering each item

The rendering system of Autocomplete uses an agnostic virtual DOM implementation. This ensures great performance even with many renders, safety against cross-site scripting (XSS) attacks, inline event handling, and more.

You can return anything from each template as long as they’re valid virtual DOM elements (VNodes).

For example, templates can return a string:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
import { autocomplete } from '@algolia/autocomplete-js';

autocomplete({
  // ...
  getSources() {
    return [
      {
        // ...
        templates: {
          item({ item }) {
            return item.name;
          },
        },
      },
    ];
  },
});

Or an HTML string using the provided html tagged template:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
import { autocomplete } from '@algolia/autocomplete-js';

autocomplete({
  // ...
  getSources() {
    return [
      {
        // ...
        templates: {
          item({ item, html }) {
            return html`<div>${item.name}</div>`;
          },
        },
      },
    ];
  },
});

Or a Preact component, using JSX:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
/** @jsx h */
import { h } from 'preact';
import { autocomplete } from '@algolia/autocomplete-js';

autocomplete({
  // ...
  getSources() {
    return [
      {
        // ...
        templates: {
          item({ item }) {
            return <div>{item.name}</div>;
          },
        },
      },
    ];
  },
});

Autocomplete uses Preact 10 to render templates by default. It isn’t compatible with earlier versions.

Returning HTML

Native HTML elements aren’t valid VNodes, which means you can’t return a template string that contains HTML, or an HTML element. But if you’re not using a virtual DOM implementation in your app, you can still return HTML with the provided html tagged template.

Every Autocomplete template provides an html function that you can use as a tagged template. Using html lets you safely provide templates as an HTML string. It works directly in the browser, no need for a transpiler or a build step.

The html function is only available starting from v1.6.0.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
import { autocomplete } from '@algolia/autocomplete-js';

autocomplete({
  // ...
  getSources() {
    return [
      {
        // ...
        templates: {
          item({ item, html }) {
            return html`<div>
              <img src="${item.image}" alt="${item.name}" />
              <div>${item.name}</div>
            </div>`;
          },
        },
      },
    ];
  },
});

Internet Explorer 11 doesn’t support tagged template literals. If you need to support Internet Explorer 11, check out the suggested solutions.

Components and layouts

You can use provided components in your templates using either their function form, or by interpolating them.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
import { autocomplete } from '@algolia/autocomplete-js';

autocomplete({
  // ...
  getSources() {
    return [
      {
        // ...
        templates: {
          item({ item, components, html }) {
            return html`<div>
              <img src="${item.image}" alt="${item.name}" />
              <div>
                ${components.Highlight({ hit: item, attribute: 'name' })}
              </div>
            </div>`;
          },
        },
      },
    ];
  },
});

The html function is also exposed in render and renderNoResults to customize the panel layout.

Loops and conditional rendering

You can use plain JavaScript to build dynamic templates.

For example, you can use Array.map to loop over an array and display a list.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
import { autocomplete } from '@algolia/autocomplete-js';

autocomplete({
  // ...
  getSources() {
    return [
      {
        // ...
        templates: {
          item({ item, html }) {
            return html`<div>
              <img src="${item.image}" alt="${item.name}" />
              <div>${item.name}</div>
              <ul>
                ${item.categories.map(
                  (category) =>
                    html`<li key="${category.id}">${category.label}</li>`
                )}
              </ul>
            </div>`;
          },
        },
      },
    ];
  },
});

Passing a unique key attribute is helpful when mapping over items. It helps the virtual DOM keep track of each element when they change, and update the UI efficiently.

To conditionally render a part of your UI, you can use a short-circuit operator or a ternary.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
import { autocomplete } from '@algolia/autocomplete-js';

autocomplete({
  // ...
  getSources() {
    return [
      {
        // ...
        templates: {
          item({ item, html }) {
            return html`<div>
              <img src="${item.image}" alt="${item.name}" />
              <div>${item.name}</div>
              <div>
                ${item.rating !== null ? `Rating: ${item.rating}` : 'Unrated'}
              </div>
            </div>`;
          },
        },
      },
    ];
  },
});

HTML in variables

Only the HTML provided as a string literal is evaluated. HTML in variables (for example, stored in Algolia records) isn’t supported and is rendered as-is.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
import { autocomplete } from '@algolia/autocomplete-js';

autocomplete({
  // ...
  getSources() {
    return [
      {
        // ...
        templates: {
          item({ item, html }) {
            const name = '<p>Jimmie Barninger</p>';

            // This returns a `<div>` with text "<p>Jimmie Barninger</p>"
            return html`<div>${name}</div>`;
          },
        },
      },
    ];
  },
});

Internet Explorer 11 support

Tagged template notation isn’t supported in Internet Explorer 11, meaning you can’t use html as a tagged template in this browser. Depending on your project setup, you can work around this problem to still use html while providing compatible code to your Internet Explorer 11 users.

With Babel

If you have a build step and you’re already using Babel to compile your code for legacy browsers, you can transform all html expressions into regular function calls.

The recommended setup is to use @babel/preset-env, which provides this transformation along with other common ones based on a list of browsers to support.

1
2
3
{
  "presets": [["@babel/preset-env"]]
}

You can also get this transform as a standalone Babel plugin.

With a shim

If you don’t have a build step in your project, you can write a shim. Tagged templates are regular functions with a specific signature, so you can wrap their calls with a friendlier API to avoid using tagged template notation.

This function takes templates either as static strings, or as an array of interspersed chunks, splits them, and passes them to the html function.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
function htmlShim(template, html) {
  if (typeof template === 'string') {
    return html([template]);
  }

  const parts = template.reduce(
    (acc, part, index) => {
      const isEven = index % 2 === 0;

      acc[Math.abs(Number(!isEven))].push(part);

      return acc;
    },
    [[], []]
  );

  return html(parts[0], ...parts[1]);
}

The documented shim assumes every even array entry is a template string, and every odd entry is a dynamic value. You can adapt it if you need a different behavior.

Further optimizations

The provided html function works in the browser without any build step, with a negligible impact on memory and bundle size (< 600 bytes).

For optimal performance, you can compile html away using the babel-plugin-htm Babel plugin.

To use this plugin, you need to adapt your code so that the compiler directive to replace html calls with is always accessible. For example, instead of destructuring in the signature, you need to name the parameter and destructure it in the function body—or not destructure it at all. The parameter (here, params) must have the same name in every template.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
autocomplete({
  // ...
  getSources() {
    return [
      {
        // ...
        templates: {
-         item({ item, html }) {
+         item(params) {
+           const { item, html } = params;

            return html`<div>${item.name}</div>`;
          },
        },
      },
    ];
  },
- render({ children, render, html }, root) {
+ render(params, root) {
+   const { children, render, html } = params;

    render(html`<div>${children}</div>`, root);
  },
});

Then, you can set up the Babel plugin.

1
2
3
4
5
6
7
8
9
10
{
  "plugins": [
    [
      "htm",
      {
        "pragma": "params.createElement"
      }
    ]
  ]
}

If you’re destructuring objects, ensure you also transpile it using @babel/preset-env.

Returning VNodes directly

Using JSX

The JSX syntax compiles down to VNodes. If you’re using JSX in your project, you can directly return JSX templates.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
/** @jsx h */
import { h } from 'preact';
import { autocomplete } from '@algolia/autocomplete-js';

autocomplete({
  // ...
  getSources() {
    return [
      {
        // ...
        templates: {
          item({ item }) {
            return <div>{item.name}</div>;
          },
        },
      },
    ];
  },
});

By default, Autocomplete uses Preact 10 internally to render templates. If you’re using another virtual DOM implementation, you can pass a custom renderer.

Using createElement

Each template function provides access to createElement and Fragment to create VNodes.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
import { autocomplete } from '@algolia/autocomplete-js';

autocomplete({
  // ...
  getSources() {
    return [
      {
        // ...
        templates: {
          item({ item, createElement, Fragment }) {
            return createElement(Fragment, {}, item.name);
          },
        },
      },
    ];
  },
});

By default, createElement and Fragment default to Preact’s preact.createElement (or h) and preact.Fragment. If you’re using another virtual DOM implementation, you can replace them.

In addition to rendering items, you can customize what to display before and after the list of items using the header and footer templates.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
autocomplete({
  // ...
  getSources({ query }) {
    return [
      {
        // ...
        templates: {
          header() {
            return 'Suggestions';
          },
          item({ item }) {
            return `Result: ${item.name}`;
          },
          footer() {
            return 'Footer';
          },
        },
      },
    ];
  },
});

Using components

Autocomplete exposes components to all templates to share them everywhere in the instance.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
autocomplete({
  // ...
  getSources({ query }) {
    return [
      {
        getItems() {
          return [
            /* ... */
          ];
        },
        templates: {
          item({ item, components }) {
            return components.Highlight({ hit: item, attribute: 'name' });
          },
        },
      },
    ];
  },
});

Four components are registered by default:

  • Highlight to highlight matches in Algolia results.
  • Snippet to snippet matches in Algolia results.
  • ReverseHighlight to reverse highlight matches in Algolia results.
  • ReverseSnippet to reverse highlight and snippet matches in Algolia results.

Highlighting and snippeting

Templates expose a set of built-in components to handle highlighting and snippeting. You can use these components to highlight or snippet Algolia results returned from the getAlgoliaResults function.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
autocomplete({
  // ...
  getSources({ query }) {
    return [
      {
        // ...
        templates: {
          item({ item, components, html }) {
            return html`<div>
              <img class="thumbnail" src="${item.image}" />
              <a href="${item.url}">
                ${components.Highlight({
                  hit: item,
                  attribute: 'name',
                  tagName: 'em',
                })}
              </a>
            </div>`;
          },
        },
      },
    ];
  },
});

Rendering a no results state

When there are no results, you might want to display a message to inform users or let them know what to do next. You can do this with the noResults template.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
autocomplete({
  // ...
  getSources({ query }) {
    return [
      {
        // ...
        templates: {
          // ...
          noResults() {
            return 'No results.';
          },
        },
      },
    ];
  },
});

Styling items

Since you’re fully controlling the rendered HTML, you can style it the way you want or use any class-based CSS library.

For example, if you’re using Bootstrap:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
import { autocomplete } from '@algolia/autocomplete-js';

import 'https://cdn.jsdelivr.net/npm/bootstrap@latest/dist/css/bootstrap.min.css';

autocomplete({
  // ...
  getSources() {
    return [
      {
        // ...
        templates: {
          item({ item, html }) {
            return html`<div class="list-group-item-action">${item.name}</div>`;
          },
        },
      },
    ];
  },
});

Or Tailwind CSS:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
import { autocomplete } from '@algolia/autocomplete-js';

import 'https://unpkg.com/tailwindcss@latest/dist/tailwind.min.css';

autocomplete({
  // ...
  getSources() {
    return [
      {
        // ...
        templates: {
          item({ item, html }) {
            return html`<div
              class="py-2 px-4 rounded-sm border border-gray-200"
            >
              ${item.name}
            </div>`;
          },
        },
      },
    ];
  },
});

Reference

Parameter Description
templates
type: AutocompleteTemplates

A set of templates to customize how items are displayed. You can also provide templates for header and footer elements around the list of items.

You must define templates within your sources.

See template for what to return.

template ➔ template

Parameter Description
header
type: (params: { state: AutocompleteState<TItem>, source: AutocompleteSource<TItem>, items: TItem[], createElement: Pragma, Fragment: PragmaFrag, components: AutocompleteComponents }) => VNode | string

A function that returns the template for the header (before the list of items).

item
type: (params: { item: TItem, state: AutocompleteState<TItem>, createElement: Pragma, Fragment: PragmaFrag, components: AutocompleteComponents }) => VNode | string

A function that returns the template for each item of the source.

type: (params: { state: AutocompleteState<TItem>, source: AutocompleteSource<TItem>, items: TItem[], createElement: Pragma, Fragment: PragmaFrag, components: AutocompleteComponents }) => VNode | string

A function that returns the template for the footer (after the list of items).

noResults
type: (params: { state: AutocompleteState<TItem>, source: AutocompleteSource<TItem>, createElement: Pragma, Fragment: PragmaFrag, components: AutocompleteComponents }) => VNode | string

A function that returns the template for when there are no items.

Did you find this page helpful?