How to Search Non-Alphabetical Data as Strings

Deployment Type

Interface

This tutorial demonstrates how to run string-specific queries against non-string fields by converting the fields to strings and storing them in a materialized view. The materialized view lets you use string-specific operators to query the converted fields and keep the original data intact in the source collection.

This tutorial takes you through the following steps:

Create a materialized view on a sample collection
Create MongoDB Search indexes on the materialized view
Run MongoDB Search queries against the converted fields in the materialized view

Before you begin, ensure that your cluster meets the requirements described in the Prerequisites.

Create a Materialized View

This section demonstrates how to create a materialized view named airbnb-mat-view on the sample_airbnb.listingsAndReviews collection. This view stores various numeric and date fields from the source collection as string fields.

Create MongoDB Search Indexes on the Materialized View

To run the queries in the Run Queries on the Converted Fields stage of this tutorial, you need to create MongoDB Search indexes on the converted string fields in your airbnb_mat_view materialized view.

The following JSON definitions define MongoDB Search indexes on the airbnb_mat_view materialized view. You can use dynamic or static mappings to specify which fields you want to index in your collection, depending on the type of queries you want to run. To learn more about field mappings or MongoDB Search index syntax, see Define Field Mappings or Index Reference, respectively.

The following JSON MongoDB Search index definition uses dynamic mappings to index the fields in the materialized view. You can use this index to run queries using the queryString operator.

You can't run queries using the autocomplete operator against dynamically indexed fields.

{
  "mappings": {
    "dynamic": true
  }
}

The following JSON MongoDB Search index definition uses static mappings to index the fields in the materialized view as the autocomplete type. You can use this index to run queries using the autocomplete operator.

You can't run queries using the queryString operator against fields indexed as type autocomplete.

{
  "mappings": {
    "dynamic": false,
    "fields": {
      "accommodatesNumber": [
        {
          "dynamic": true,
          "type": "document"
        },
        {
          "minGrams": 1,
          "type": "autocomplete"
        }
      ],
      "lastScrapedDate": [
        {
          "dynamic": true,
          "type": "document"
        },
        {
          "type": "autocomplete"
        }
      ],
      "maximumNumberOfNights": [
        {
          "dynamic": true,
          "type": "document"
        },
        {
          "minGrams": 1,
          "type": "autocomplete"
        }
      ]
    }
  }
}

To learn how to create the indexes defined above using your preferred interface, see Supported Clients.

Run Queries on the Converted Fields

You can run queries against the numeric and date fields that were converted to strings. This tutorial uses queryString and autocomplete operators to search for properties. The query uses the following pipeline stages:

$search stage to search the collection
$limit stage to limit the output to 5 results
$project stage to exclude _id

In this section, you connect to your cluster and run the sample queries using the operator against the fields in the airbnb_mat_view collection.

Note

You can't run near or range queries against the date and number fields that were converted to strings in your materialized view.

Back

Design Search for Your Data Model

Embedded Documents