Querying with the Compound Operator

Overview

In this guide, you will learn how to use multiple search operators in a compound query by using the aggregation pipeline.

Time required: 5 minutes

What You'll Need

A MongoDB account. See Sign Up for a MongoDB Account.
An Atlas cluster. See Create a Cluster.
Sample datasets loaded into your cluster.
A search index with default settings and dynamic indexing. To create a default dynamic index, see Build a Dynamic Index.

Procedure

In Atlas, go to the Clusters page for your project.

If it is not already displayed, select the organization that contains your desired project from the Organizations menu in the navigation bar.
If it is not already displayed, select your desired project from the Projects menu in the navigation bar.
If the Clusters page is not already displayed, click Database in the sidebar.

Click on your cluster name.

click to enlarge

Click the Collections tab.

click to enlarge

Click the sample_supplies database, then click the sales collection.

Run a compound query on your index in the Aggregations tab.

Click the Aggregation tab.

Click + Add Stage to begin creating your aggregation pipeline.

In the stage window, click on the dropdown menu labeled Select and select the $search stage.

Replace the placeholder code with the following query.

{
   "compound": {
      "filter": [{
         "text": {
         "query": "Online",
            "path": "purchaseMethod"
         }
      }],
      "should": [{
         "text": {
         "query": "notepad",
            "path": "items"
         }
      }]
   }
}

This query uses the compound operator with:

The filter clause to return only documents with a purchaseMethod of Online.
The should clause to return documents with an items field that contains notepad to have a higher score than those without.

Review the results of the compound query.

In the right-hand pane, your results should include a list of documents that resemble the following sample document:

items: Array
   0: Object
   1: Object
   2: Object
      name: "notepad"
   ...
storeLocation: "Denver"
customer: Object
couponUsed: false
purchaseMethod: "Online"
_id: ObjectID('5bd761dcae323e45a93cd06e')
saleDate: 2013-11-22T18:49:45.212+00:00

Note

The fields don't always appear in the same order for every returned document.

You can expand Array and Object fields to view their contents by clicking on them.

Summary

If you successfully completed the procedure in this guide, you have created an Atlas Search aggregation stage with a compound operator, and have seen how different clauses in the operator interact to focus your search results across multiple axes. In the next guide, you will learn how to group search results using facets.

What's Next

Querying with Facets

10 mins

Retrieve documents in MongoDB with counts of specific values.

Start Guide

Chapter 3

Atlas Search

Start with Guides