A first-party Elastic Site Search jQuery plugin for search autocomplete.
Requirements:
- Site Search account. Sign up at swiftype.com.
- Site Search engine with some data in it.
Include the following in the header of your webpage:
- the latest version of jQuery
- the Swiftype jQuery plugin
- (optional) the Swiftype Autocomplete stylesheet
All together it should look like this:
<script type="text/javascript" src="https://ajax.googleapis.com/ajax/libs/jquery/1.7.1/jquery.min.js"></script>
<script type="text/javascript" src="jquery.swiftype.autocomplete.js"></script>
<link type="text/css" rel="stylesheet" href="autocomplete.css" media="all" />
Note: This client has been developed for the Elastic Site Search API endpoints only. You may refer to the Elastic Site Search API Documentation for additional context.
Simply apply the swiftype method to an existing search input field on your webpage. For example, add it to a search input field with id st-search-input
as follows:
$('#st-search-input').swiftype({
engineKey: 'jaDGyzkR6iYHkfNsPpNK'
});
Be sure to change the engineKey
attribute shown above to match the one assigned to your Swiftype search engine. If you are using the web interface, the search engine key is listed on the first page of your dashboard.
This simple installation will by default match the string a user is typing into your search input to any of the items indexed in your search engine. By default, the string is matched against any fields you have indexed with type string
.
This plugin is written to be flexible based on your specific use-case. For example you might want to retrieve more data for each element in the dropdown, customize the way data is display to the user, or restrict the autocomplete query to certain elements of your search engine.
Let's go through an example that does all of this. For this example, let's assume you followed the QuickStart tutorial for our Ruby Gem, and now you have data for a Bookstore indexed in your example search engine.
To specify the number of results you would like returned from the API, set the resultLimit
attribute as follows:
$('#st-search-input').swiftype({
engineKey: 'jaDGyzkR6iYHkfNsPpNK',
resultLimit: 20
});
To specify the fields you would like returned from the API, set the fetchFields
attribute to a hash containing an array listing the fields you want returned for each document type. For example, if you have indexed title
, genre
, and published_on
fields for each document, you can have them returned as follows:
$('#st-search-input').swiftype({
fetchFields: { 'books': ['title', 'genre', 'published_on'] },
engineKey: 'jaDGyzkR6iYHkfNsPpNK'
});
These additional fields will be returned with each item in the autocomplete, and they can be accessed in the rendering function as shown in the next section.
Now that you have more data for each autocomplete item, you'll want to customize the item rendering function to make use of them.
The default rendering function is shown below:
var defaultRenderFunction = function(document_type, item) {
return '<p class="title">' + Swiftype.htmlEscape(item['title']) + '</p>';
};
The additional fields are available as keys in the item dictionary, so you could customize this to make use of the genre
field as follows:
var customRenderFunction = function(document_type, item) {
var out = '<a href="' + Swiftype.htmlEscape(item['url']) + '" class="st-search-result-link">' + item.highlight['title'] + '</a>';
return out.concat('<p class="genre">' + item.highlight['genre'] + '</p>');
};
Now simply set the renderFunction
attribute in the options dictionary to your customRenderFunction
to tell our plugin to use your function to render results:
$('#st-search-input').swiftype({
renderFunction: customRenderFunction,
fetchFields: { 'books': ['url'] }, // Fetch the URL field as a raw field.
highlightFields: { 'books': { 'title': { 'size': 60, 'fallback': true }, 'genre': { 'size': 60, 'fallback': true }, 'published_on': { 'size': 15, 'fallback':true } } },
engineKey: 'jaDGyzkR6iYHkfNsPpNK'
});
By default, the Swiftype autocomplete library will match the string the user is typing to any string
field indexed for your documents. So if you would like to ensure that it only matches entries in the title
field, for example, you can specify the searchFields
option:
$('#st-search-input').swiftype({
renderFunction: customRenderFunction,
fetchFields: { 'books': ['title', 'genre', 'published_on'] },
searchFields: { 'books': ['title'] },
engineKey: 'jaDGyzkR6iYHkfNsPpNK'
});
Similarly to the fetchFields
option, searchFields
accepts a hash containing an array of fields for each document_type on which you would like the user's query to match.
Now let's say you only want your autocomplete to display books that are of the fiction genre
and are in_stock. In order to restrict search results, you can pass additional query conditions to the search API by specifying them as a dictionary in the filters
field. Multiple clauses in the filters field are combined with AND logic:
$('#st-search-input').swiftype({
renderFunction: customRenderFunction,
fetchFields: { 'books': ['title', 'genre', 'published_on'] },
filters: { 'books': { 'genre': 'fiction', 'in_stock': true } },
searchFields: { 'books': ['title'] },
engineKey: 'jaDGyzkR6iYHkfNsPpNK'
});
In order to change what happens when clicking on an item, you need to provide an onComplete
handler function.
By default, clicking on an item in the dropdown will execute the following onComplete
handler, which routes a user to the url
property of the selected item:
var defaultOnComplete = function(item, prefix) {
window.location = item['url'];
};
To change this, simply provide a new handler function in the onComplete
option when initializing your autocomplete.
Here is an example that updates the input value with the selected item's title:
var input; // Save a reference to the autocomplete dropdown
input = $('#st-search-input').swiftype({
onComplete: function(selectedItem) {
input.val(selectedItem['title']); // Update the autocomplete dropdown's value
},
fetchFields: { 'books': ['title'] },
engineKey: 'jaDGyzkR6iYHkfNsPpNK'
});
Yes! If you are looking for core search functionality, checkout out the Swiftype Search Plugin.
If something is not working as expected, please open an issue.
Your best bet is to read the documentation.
You can checkout the Elastic Site Search community discuss forums.
We welcome contributors to the project. Before you begin, a couple notes...
- Before opening a pull request, please create an issue to discuss the scope of your proposal.
- Please write simple code and concise documentation, when appropriate.
Thank you to all the contributors!