Concepts / Building Search UI / Routing URLs
Nov. 19, 2019

Routing URLs

Overview

Vue InstantSearch provides the necessary API entries to let you synchronize the state of your search UI (e.g., refined widgets, current search query) with any kind of storage. This is possible via the routing option. This guide focuses on storing the UI state in the browser URL.

Synchronizing your UI with the browser URL is considered good practice. It allows your users to share one of your results page by copying its URL. It also improves the user experience by enabling the use of the back and next browser buttons to keep track of previous searches.

Basic URLs

In this guide, we’re using the router from instantsearch.js. You need to add instantsearch.js to your project dependencies along with vue-instantsearch.

Vue InstantSearch provides a basic way to activate the browser URL synchronization with the routing option. You can find a live example on this sandbox.

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
27
28
29
<template>
  <ais-instant-search
    :search-client="searchClient"
    index-name="movies"
    :routing="routing"
  >
    <!-- add the other components here -->
  </ais-instant-search>
</template>

<script>
import { history as historyRouter } from 'instantsearch.js/es/lib/routers';
import { simple as simpleMapping } from 'instantsearch.js/es/lib/stateMappings';

export default {
  data() {
    return {
      searchClient: algoliasearch(
        'latency',
        '6be0576ff61c053d5f9a3225e2a90f76'
      ),
      routing: {
        router: historyRouter(),
        stateMapping: simpleMapping(),
      }
    };
  },
};
</script>

Assume the following search UI state:

  • Query: “galaxy”
  • Menu:
    • categories: “Cell Phones”
  • Refinement List:
    • brand: “Apple”, “Samsung”
  • Page: 2

The resulting URL in your browser URL bar will look like this:

1
https://website.com/?menu[categories]=Cell Phones&refinementList[brand][0]=Apple&refinementList[brand][1]=Samsung&page=2&query=galaxy

This URL is accurate, and can be translated back to a search UI state. However, this isn’t the most human-readable, or optimized for search engines. We’ll see in the next section how to make it more SEO-friendly.

SEO-friendly URLs

In this guide we’ll use the router from instantsearch.js so you may need to add instantsearch.js to your project dependencies in addition to vue-instantsearch.

URLs are more than query parameters. Another important part is the path. Manipulating the URL path is a common e-commerce pattern that allows you to better reference your page results. In this section, you’ll learn how to create this kind of URLs:

1
https://website.com/search/Cell+Phones/?query=galaxy&page=2&brands=Apple&brands=Samsung

This URL is composed of the path which now includes /search, then /Cell+Phones for the category. The query parameters are also simplified by only using the attribute name as key. This requires you to have only one widget per attribute.

Implementation example

Here’s an example where we’re storing the brand in the path name, and the query and page as query parameters. You can find a live example on this sandbox.

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
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
import { history as historyRouter } from 'instantsearch.js/es/lib/routers';

// Returns a slug from the category name.
// Spaces are replaced by "+" to make
// the URL easier to read and other
// characters are encoded.
function getCategorySlug(name) {
  return name
    .split(' ')
    .map(encodeURIComponent)
    .join('+');
}

// Returns a name from the category slug.
// The "+" are replaced by spaces and other
// characters are decoded.
function getCategoryName(slug) {
  return slug
    .split('+')
    .map(decodeURIComponent)
    .join(' ');
}

const routing = {
  router: historyRouter({
    windowTitle({ category, query }) {
      const queryTitle = query ? `Results for "${query}"` : 'Search';

      if (category) {
        return `${category}${queryTitle}`;
      }

      return queryTitle;
    },

    createURL({ qsModule, routeState, location }) {
      const urlParts = location.href.match(/^(.*?)\/search/);
      const baseUrl = `${urlParts ? urlParts[1] : ''}/`;

      const categoryPath = routeState.category
        ? `${getCategorySlug(routeState.category)}/`
        : '';
      const queryParameters = {};

      if (routeState.query) {
        queryParameters.query = encodeURIComponent(routeState.query);
      }
      if (routeState.page !== 1) {
        queryParameters.page = routeState.page;
      }
      if (routeState.brands) {
        queryParameters.brands = routeState.brands.map(encodeURIComponent);
      }

      const queryString = qsModule.stringify(queryParameters, {
        addQueryPrefix: true,
        arrayFormat: 'repeat',
      });

      return `${baseUrl}search/${categoryPath}${queryString}`;
    },

    parseURL({ qsModule, location }) {
      const pathnameMatches = location.pathname.match(/search\/(.*?)\/?$/);
      const category = getCategoryName(
        (pathnameMatches && pathnameMatches[1]) || ''
      );
      const { query = '', page, brands = [] } = qsModule.parse(
        location.search.slice(1)
      );
      // `qs` does not return an array when there's a single value.
      const allBrands = Array.isArray(brands)
        ? brands
        : [brands].filter(Boolean);

      return {
        query: decodeURIComponent(query),
        page,
        brands: allBrands.map(decodeURIComponent),
        category,
      };
    },
  }),

  stateMapping: {
    stateToRoute(uiState) {
      return {
        query: uiState.query,
        page: uiState.page,
        brands: uiState.refinementList && uiState.refinementList.brand,
        category: uiState.menu && uiState.menu.categories,
      };
    },

    routeToState(routeState) {
      return {
        query: routeState.query,
        page: routeState.page,
        menu: {
          categories: routeState.category,
        },
        refinementList: {
          brand: routeState.brands,
        },
      };
    },
  },
};

We are now using the history router to explicitly set options on the default router mechanism used in the first example. You can notice that we use both the router and stateMapping options to map uiState to routeState, and vice versa.

Using the routing option as an object, we can configure:

  • windowTitle: a method to map the routeState object returned from stateToRoute to the window title.
  • createURL: a method called every time we need to create a URL. This should be done when:
    • you want to synchronize the routeState to the browser URL,
    • you want to render a tags in the menu widget,
    • you call createURL in one of your connectors’ rendering methods.
  • parseURL: a method called every time the user loads or reloads the page, or clicks on the back or next buttons of the browser.

Making URLs more discoverable

In real-life applications, you might want to make specific categories easier to access by associating them with readable and memorable URLs.

Given our dataset, we can make some categories more discoverable:

  • “Cameras & Camcorders” → /Cameras
  • “Car Electronics & GPS” → /Cars

In this example, anytime users visit https://website.com/search/Cameras, it pre-selects the “Cameras & Camcorders” filter.

You can achieve this with a dictionary.

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
27
28
29
30
31
32
33
34
35
36
// Step 1. Add the dictionaries to convert the names and the slugs
const encodedCategories = {
  Cameras: 'Cameras & Camcorders',
  Cars: 'Car Electronics & GPS',
  Phones: 'Cell Phones',
  TV: 'TV & Home Theater'
};

const decodedCategories = Object.keys(encodedCategories).reduce((acc, key) => {
  const newKey = encodedCategories[key];
  const newValue = key;

  return {
    ...acc,
    [newKey]: newValue
  };
}, {});

// Step 2. Update the getters to use the encoded/decoded values
function getCategorySlug(name) {
  const encodedName = decodedCategories[name] || name;

  return encodedName
    .split(' ')
    .map(encodeURIComponent)
    .join('+');
}

function getCategoryName(slug) {
  const decodedSlug = encodedCategories[slug] || slug;

  return decodedSlug
    .split('+')
    .map(decodeURIComponent)
    .join(' ');
}

Note that you can build these dictionaries from your Algolia records.

With such a solution, you have full control over what categories are discoverable via the URL.

About SEO

For your search results to be part of search engines results, you have to be selective. Trying to index too many search results pages could be considered as spam.

To do that, you can create a robots.txt and host it at https://website.com/robots.txt.

Here’s an example based on the URL scheme we created.

1
2
3
4
5
User-agent: *
Allow: /search/Audio/
Allow: /search/Phones/
Disallow: /search/
Allow: *

Combining with Vue Router

The previous examples were using the InstantSearch router. This is fine in almost all of the use cases, but if you plan on using Vue Router too, and also plan on reading the URL with Vue Router in the search page to show something outside of the InstantSearch life cycle, you can choose to synchronize with Vue Router. Note that this is not necessary if you are using Vue Router and are not planning to read from the URL.

The API we use of InstantSearch is the router, but instead of using historyRouter, a new one is written from scratch. The router key expects an object with the following keys as value:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
const routing = {
  router: {
    read() {
      /* read from the URL and return a routeState */
    },
    write(routeState) {
      /* write to the URL */
    },
    createURL(routeState) {
      /* return a URL as a string */
    },
    onUpdate(callback) {
      /* call this callback whenever the URL changed externally */
    },
    dispose() {
      /* remove any listeners */
    },
  }
};

We will fill in all these functions Vue Router. For simplicity, in this example we will not fill in a stateMapping, and we will also synchronize towards the query string completely. The default configuration of Vue Router does not allow for deeply nested URLs, so we have to implement that first in main.js:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
import qs from 'qs';

const router = new Router({
  routes: [
    // ...
  ],
  // set custom query resolver
  parseQuery(query) {
    return qs.parse(query);
  },
  stringifyQuery(query) {
    const result = qs.stringify(query);

    return result ? `?${result}` : '';
  },
});

Then we can fill in the router key on the routing object in the data 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
25
26
27
28
29
30
31
32
33
34
35
36
37
const vueRouter = this.$router; /* get this from Vue Router */

const routing = {
  router: {
    read() {
      return vueRouter.currentRoute.query;
    },
    write(routeState) {
      vueRouter.push({
        query: routeState,
      });
    },
    createURL(routeState) {
      return vueRouter.resolve({
        query: routeState,
      }).href;
    },
    onUpdate(cb) {
      this._onPopState = ({state}) => {
        const routeState = state;
        // at initial load, the state is read from the URL without
        // update. Therefore the state object is not there. In this
        // case we fallback and read the URL.
        if (!routeState) {
          cb(this.read());
        } else {
          cb(routeState);
        }
      };
      window.addEventListener('popstate', this._onPopState);
    },
    dispose() {
      window.removeEventListener('popstate', this._onPopState);
      this.write();
    },
  },
};

The live version of this example is also available here.

An ideal implementation would also listen to updates from Vue Router and propagate them into InstantSearch, which isn’t done here for brevity.

Combining with Nuxt.js

The first steps to enable routing in a Nuxt application are:

  1. Enabling routing in createInstantSearch.
  2. Parsing the current URL and map the parameters to resultsState using findResultsState in asyncData.

On the front end, Nuxt merges anything that is returned from asyncData with data. This calls the hydrate method, but only in the client (via the beforeMount method). As a result, search parameters added for back end search are shown if you regularly navigate to a page.

Finally, add the rootMixin to provide the instance to the component.

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
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
<template>
  <ais-instant-search-ssr>
    <!-- add the other components here -->
  </ais-instant-search-ssr>
</template>

<script>
import {
  AisInstantSearchSsr,
  createInstantSearch
} from 'vue-instantsearch';
import algoliasearch from 'algoliasearch/lite';
import qs from 'qs';

const searchClient = algoliasearch(
  'latency',
  '6be0576ff61c053d5f9a3225e2a90f76'
);

const { instantsearch, rootMixin } = createInstantSearch({
  searchClient,
  indexName: 'instant_search',
  // Enable routing
  routing: true
});

export default {
  asyncData({ route }) {
    // Parse current url
    const params = qs.parse(route.fullPath.replace(/^.*\?/, ''));
    return (
      instantsearch
        // Map URL parameters to `resultsState`
        .findResultsState({
          query: params.query || '',
          disjunctiveFacets: ['brand'],
          disjunctiveFacetsRefinements: params.refinementList
        })
        .then(() => ({
          algoliaState: instantsearch.getState()
        }))
    );
  },
  beforeMount() {
    // Nuxt merges `asyncData` and `data` on the client
    instantsearch.hydrate(this.algoliaState);
  },
  mixins: [rootMixin],
  components: {
    AisInstantSearchSsr,
    // Add your other components here
  },
};
</script>

Did you find this page helpful?