A first-party Elastic Site Search jQuery plugin for website search.
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 Search jQuery plugin
- (optional) the Swiftype Search stylesheet
All together it should look like this:
<script type="text/javascript" src="https://ajax.googleapis.com/ajax/libs/jquery/1.11.1/jquery.min.js"></script>
<script type="text/javascript" src="jquery.swiftype.search.js"></script>
<link type="text/css" rel="stylesheet" href="search.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.
Start by having at least these three elements on the page: a form, an input field within the form, and a container for results.
<form>
<input type="text" placeholder="Search" id="st-search-input" />
</form>
<div id="st-results-container"></div>
Apply the swiftype method to an existing search input field within a form on your webpage and provide a container for results. For example, add it to a search input field with id st-search-input
as follows:
$('#st-search-input').swiftypeSearch({
resultContainingElement: '#st-results-container',
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 plugin is written to be flexible based on your specific use-case. For example you might want to retrieve more data for each result, customize the way data is display to the user, or restrict the search 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 per page, use the perPage
attribute.
$('#st-search-input').swiftypeSearch({
engineKey: 'jaDGyzkR6iYHkfNsPpNK',
perPage: 20
});
The maximium value that will be honored by the API is 100.
Any fields that are queried during a search will return the top match (if any) in the highlight property of the results. All snippets in this form have HTML entities from the original text encoded. Actual highlighting is specified using (unencoded) <em>
tags.
You can customize which fields are returned in the highlight property by using the highlightFields
option:
$('#st-search-input').swiftypeSearch({
renderFunction: customRenderFunction,
fetchFields: {'books': ['title','genre','published_on']},
engineKey: 'jaDGyzkR6iYHkfNsPpNK'
highlightFields: {'books': {'body': {'size': 300, 'fallback': true }}}
});
The highlightFields
option accepts a hash containing the fields you want to have highlighted for each object of each document type. For each field, specify size
as the maximum number of characters to include in the snippet. Set fallback
to true to force inclusion of a non-highlighted snippet if a highlight is not available for that field.
See the custom.html file for an additional example of highlightFields
.
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').swiftypeSearch({
fetchFields: {'books': ['title','genre','published_on']},
engineKey: 'jaDGyzkR6iYHkfNsPpNK'
});
These additional fields will be returned with each item, and they can be accessed in the rendering function as shown in the next section.
Now that you have more data for each result 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 '<div class="st-result"><h3 class="title"><a href="' + item['url'] + '" class="st-search-result-link">' + item['title'] + '</a></h3></div>';
};
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="' + item['url'] + '" class="st-search-result-link">' + item['title'] + '</a>';
return out.concat('<p class="genre">' + item['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').swiftypeSearch({
renderFunction: customRenderFunction,
fetchFields: {'books': ['title','genre','published_on']},
engineKey: 'jaDGyzkR6iYHkfNsPpNK'
});
By default, the Swiftype search library will match the submitted query to any string
or text
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').swiftypeSearch({
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 results 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').swiftypeSearch({
renderFunction: customRenderFunction,
fetchFields: {'books': ['title','genre','published_on']},
filters: {'books': {'genre': 'fiction', 'in_stock': true}},
searchFields: {'books': ['title']},
engineKey: 'jaDGyzkR6iYHkfNsPpNK'
});
In come cases, you may need to manually trigger a new search. The following method can be used to do so. An example of such a case would be if a user chooses to apply a filter and a new query needs top be executed.
Swiftype.reloadResults();
Yes! If you are looking for autocomplete functionality, checkout out the Swiftype Autocomplete 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!