Search REST API: Search

Overview

Instructions to use REST API Search service.

Step-by-step guide

Service Name

search

Service URL

URL_SEARCH/INSTANCE_ID/search?INPUT_PARAMETERS

Input Parameters

Parameter	Description	Data Type	Mandatory
q	The query for the request	String	Yes
lang	Set the language of the query	String	Yes
start	It is used to paginate results from a query	Integer	Yes
rows	It is used to paginate results from a query	Integer	Yes
topTrends.rows	Specify a number of rows for the topTrends section. If this parameter is not specified, the rows param will be used	Integer	No
filter	Set query filters	String	No
boost	boosting query in lucene query format	String	No
facet	Set query facets	String	No
sort	Sort results	String	No
jsonCallback	Name of the callback that will be executed after the request	String	No
suggestions.suggestion.docs	Number of results to include for each suggestion	String	No

Sorting

Send us a request with the fields you want to use for SORTING and we will configure it.

You can find below some code samples of the api syntax for sorting.

title A-Z

sort=name_sort asc

title Z-A

sort=name_sort desc

price min to max

sort=price asc

price max to min

sort=price desc

Boosting

Boosting allows you to modify the default ranking pushing to the top the documents matching the specified queries.

Boosting Nike and Rebook products

boost=brand:Nike^2.1 OR brand:Rebook^1.3

boost parameter can be specified multiple times, so multiple queries could be specified as part of a single query using OR (as specified above) or specifying those queries in individual boost parameters.

boost=brand:Nike^2.1&brand:Rebook^1.3

Faceting

Send us a request with the list of fields that you want to use for FACETING and we will add it to the config.

Example of the api syntax for faceting, with the field brand_facet:

facet={!ex=brand_facet}brand_facet

To add more facets to the query, you should simply repeat the facet parameter.

Top Trends Faceting

It requires EB version 3.8+. Send us a request if you like to upgrade your EB instance.

Once the feature is activated, you don't need to specify any new parameter on the query. See below (section "Output - Results + topTrends") to check how the format of the response is.

Examples

NOTE: The following code is just an example, do not copy and paste this for your use

Output

There are different outputs depending of the search results.

Results + topTrends (empathize) (including categories for the most popular query)

By default, the service will return two json nodes, one for the search results (content) and another one for the topTrends (more popular queries and categories for the first one).

{
    content: {},
    topTrends: [ ]
}

{
    content: {
        numFound: 14,
        docs: [
                {
                    name: "Samsung Galaxy Note 10.1"
                },
                {
                    name: "Samsung SNH-1010N Smartcam"
                },
        ],
        facets: [
                {
                    facet: "brand",
                    values: [
                        {
                            value: "samsung",
                            count: 13,
                            filter: "brand:samsung"
                        },
                        {
                            value: "SBS",
                            count: 1,
                            filter: "brand:SBS"
                        }
                    ]
                }
        ]
    },
    topTrends: [ 
        {
            title: "<b>sams</b>ng",
            title_raw: "samsung",
            facets: [
                {
                    facet: "rootCategories_facet"
                    values: [
                        {
                             value: "Mobile Phones",
                             count: 5,
                             filter: "{!tag=rootFilter}rootCategories_facet:Mobile Phones"
                        },
                        {
                             value: "Phone Accessories",
                             count: 3,
                             filter: "{!tag=rootFilter}rootCategories_facet:Phone Accessories"
                        }
                    ]
                }
            ]
        },
        {
            title: "<b>sams</b>ung galaxy",
            title_raw: "samsung galaxy"
        },
        {
            title: "tablet <b>sams</b>ung",
            title_raw: "tablet samsung"
        }
    ]
}

According to the previous example:

numFound → number of the documents returned by the service
docs → Json array with the documents returned by the service. Each one will contain all the fields configured to be returned by the search service.
facets → filters, will contain one entry for each facet contained in the request using the facets param or directly configured in the search service by the EB Team.
topTrends → Popular trends, including category facet for the first one.

Spellcheck + topTrends (empathize)

The search engine can detect grammatical errors and automatically correct it, for example searching by "sammsung" the system will suggest "samsung" in the spellchecked node.

{
    content: {
        numFound: 2,
        docs: [
                {
                    name: "Samsung Galaxy Note 10.1"
                },
                {
                    name: "Samsung SNH-1010N Smartcam"
                },
        ],
        facets: [
                {
                    facet: "brand",
                    values: [
                        {
                            value: "samsung",
                            count: 13,
                            filter: "brand:samsung"
                        },
                        {
                            value: "fitbit",
                            count: 1,
                            filter: "brand:fitbit"
                        }
                    ]
                }
        ],
        spellchecked: "samsung"
    },
    topTrends: [ 
        {
            title: "<b>samsung</b>",
            title_raw: "samsung"
        }
    ]
}

Suggestions + topTrends (empathize)

For example searching by "samsung panasonic" the search engine detects than there're no products with both terms but there are products with each term separately.

{
    content: {
        suggestions: [
                    {
                        suggestion: "panasonic",
                        numFound: 101,
                        docs: [
                            {
                                name: "Panasonic KX-TG 8621"
                            },
                            {
                                name: "Panasonic KX-TGP 550"
                            }
                        ]
                    },
                    {
                        suggestion: "samsung",
                        numFound: 14,
                        docs: [
                           {
                                name: "Samsung Galaxy Note 10.1"
                           },
                           {
                                name: "Samsung SNH-1010N Smartcam"
                           }
                        ]
                    }
        ]
   },
   topTrends: [ ]
}

The suggestions node will contain one or more suggestions for each term or combination of terms contained in the query typed by the user.

suggestion → term suggested by the search engine
numFound → number of results that the search engine will return searching using this suggestion
docs → By default, the search engine will return a sample of the results using this suggestion

Empathy Knowledge Base