Keyword Analyzer

The keyword analyzer accepts a string or array of strings as a parameter and indexes them as a single term (token). Only exact matches on the field are returned. It leaves all text in its original letter case.

Tip

For exact matching, instead of using the keyword analyzer, you can index the field as the Atlas Search token type and use the equals operator to search the field.

If you select Refine Your Index, the Atlas UI displays a section titled View text analysis of your selected index configuration within the Index Configurations section. If you expand this section, the Atlas UI displays the index and search tokens that the keyword analyzer generates for each sample string. You can see the tokens that the keyword analyzer creates for a built-in sample document and query string when you create or edit an index in the Atlas UI Visual Editor.

Important

Atlas Search won't index string fields where analyzer tokens exceed 32766 bytes in size. If using the keyword analyzer, string fields which exceed 32766 bytes will not be indexed.

Example

The following example index definition specifies an index on the title field in the sample_mflix.movies collection using the keyword analyzer. To follow along with this example, load the sample data on your cluster and either use mongosh or navigate to the Create a Search Index page in the Atlas UI following the steps in the Create an Atlas Search Index tutorial.

Then, using the movies collection as your data source, follow the example procedure to create an index from mongosh or the Atlas UI Visual Editor or JSON editor.

➤ Use the Select your language drop-down menu to set the language of the example on this page.

Click Refine Your Index to configure your index.
In the Index Configurations section, toggle Dynamic Mapping to off.
In the Field Mappings section, click Add Field to open the Add Field Mapping window.
Click Customized Configuration.
Select title from the Field Name dropdown.
Click the Data Type dropdown and select String if it isn't already selected.

Expand String Properties and make the following changes:

Index Analyzer	Select `lucene.keyword` from the dropdown.
Search Analyzer	Select `lucene.keyword` from the dropdown.
Index Options	Use the default `offsets`.
Store	Use the default `true`.
Ignore Above	Keep the default setting.
Norms	Use the default `include`.

Click Add.
Click Save Changes.
Click Create Search Index.

Replace the default index definition with the following index definition.

{
  "mappings": {
    "fields": {
      "title": {
        "type": "string",
        "analyzer": "lucene.keyword"
      }
    }
  }
}

Click Next.
Click Create Search Index.

1 db.movies.createSearchIndex(
2   "default",
3     {
4       "mappings": {
5         "fields": {
6           "title": {
7             "type": "string",
8             "analyzer": "lucene.keyword"
9           }
10         }
11       }
12     }
13 )

The following query searches for the phrase Class Action in the title field.

Click the Query button for your index.
Click Edit Query to edit the query.
Click on the query bar and select the database and collection.

Replace the default query with the following and click Find:

[
  {
    "$search": {
      "text": {
        "query": "Class Action",
        "path": "title"
      }
    }
  }
]

SCORE: 4.346973419189453   _id:  "573a1399f29313caabcec6b7"
   awards: Object
   cast: Array (4)
   countries: Array (1)
   directors: Array (1)
   fullplot: "Jeb Ward is an attorney who specializes in whistle blower, David vs. G…"
   genres: Array (2)
   imdb: Object
   languages: Array (1)
   lastupdated: "2015-09-06 00:42:51.167000000"
   metacritic: 58
   num_mflix_comments: 2
   plot: "Jeb Ward is an attorney who specializes in whistle blower, David vs. G…"
   poster: "https://m.media-amazon.com/images/M/MV5BNWY5Mjk4ZmItMTAzYS00NWE3LWEzYz…"
   rated: "R"
   released: 1991-03-15T00:00:00.000+00:00
   runtime: 110
   title: "Class Action"
   tomatoes: Object
   type: "movie"
   writers: Array (3)
   year: 1991

db.movies.aggregate([
  {
    "$search": {
      "text": {
         "query": "Class Action",
         "path": "title"
      }
    }
  },
  {
    "$project": {
      "_id": 0,
      "title": 1
    }
  }
])

[
  {
    title: 'Class Action'
  }
]

Atlas Search returned the document because it matched the query term Class Action to the single token Class Action that it creates for the text in the field using the lucene.keyword analyzer. By contrast, Atlas Search doesn't return any results for the following query:

db.cases.aggregate([
  {
    "$search": {
      "text": {
        "query": "action",
        "path": "title"
      }
    }
  }
])

Many documents in the collection contain the string action, but the keyword analyzer only matches documents in which the search term matches the entire contents of the field exactly. For the preceding query, the keyword analyzer wouldn't return any results. However, if you indexed the field using the Standard Analyzer or Simple Analyzer, Atlas Search would return multiple documents in the results, including the document with the title field value Class Action, because it would create tokens similar to the following, which it would then match to the query term:

Analyzer	Output Tokens	Matches `action`	Matches `Class Action`
Keyword Analyzer Tokens	`Class Action`	X	√
Standard Analyzer Tokens	`class`, `action`	√	√
Simple Analyzer Tokens	`class`, `action`	√	√

Back

Whitespace

Language

1	db.movies.createSearchIndex(
2	"default",
3	{
4	"mappings": {
5	"fields": {
6	"title": {
7	"type": "string",
8	"analyzer": "lucene.keyword"
9	}
10	}
11	}
12	}
13	)