Navigation

$text

MongoDB Atlas Search

Atlas Search makes it easy to build fast, relevance-based search capabilities on top of your MongoDB data. Try it today on MongoDB Atlas, our fully managed database as a service.

Definition

$text

$text performs a text search on the content of the fields indexed with a text index. A $text expression has the following syntax:

Changed in version 3.2.

{
  $text:
    {
      $search: <string>,
      $language: <string>,
      $caseSensitive: <boolean>,
      $diacriticSensitive: <boolean>
    }
}

The $text operator accepts a text query document with the following fields:

Field Type Description
$search string A string of terms that MongoDB parses and uses to query the text index. MongoDB performs a logical OR search of the terms unless specified as a phrase. See Behavior for more information on the field.
$language string

Optional. The language that determines the list of stop words for the search and the rules for the stemmer and tokenizer. If not specified, the search uses the default language of the index. For supported languages, see Text Search Languages.

If you specify a language value of "none", then the text search uses simple tokenization with no list of stop words and no stemming.

$caseSensitive boolean

Optional. A boolean flag to enable or disable case sensitive search. Defaults to false; i.e. the search defers to the case insensitivity of the text index.

For more information, see Case Insensitivity.

New in version 3.2.

$diacriticSensitive boolean

Optional. A boolean flag to enable or disable diacritic sensitive search against version 3 text indexes. Defaults to false; i.e. the search defers to the diacritic insensitivity of the text index.

Text searches against earlier versions of the text index are inherently diacritic sensitive and cannot be diacritic insensitive. As such, the $diacriticSensitive option has no effect with earlier versions of the text index.

For more information, see Diacritic Insensitivity.

New in version 3.2.

The $text operator, by default, does not return results sorted in terms of the results’ scores. For more information on sorting by the text search scores, see the Text Score documentation.

Behavior

Restrictions

  • A query can specify, at most, one $text expression.
  • The $text query can not appear in $nor expressions.
  • The $text query can not appear in $elemMatch query expressions or $elemMatch projection expressions.
  • To use a $text query in an $or expression, all clauses in the $or array must be indexed.
  • You cannot use hint() if the query includes a $text query expression.
  • You cannot specify $natural sort order if the query includes a $text expression.
  • You cannot combine the $text expression, which requires a special text index, with a query operator that requires a different type of special index. For example you cannot combine $text expression with the $near operator.
  • Views do not support text search.

If using the $text operator in aggregation, the following restrictions also apply.

  • The $match stage that includes a $text must be the first stage in the pipeline.
  • A text operator can only occur once in the stage.
  • The text operator expression cannot appear in $or or $not expressions.
  • The text search, by default, does not return the matching documents in order of matching scores. To sort by descending score, use the $meta aggregation expression in the $sort stage.

$search Field

In the $search field, specify a string of words that the text operator parses and uses to query the text index.

The text operator treats most punctuation in the string as delimiters, except a hyphen-minus (-) that negates term or an escaped double quotes \" that specifies a phrase.

Phrases

To match on a phrase, as opposed to individual terms, enclose the phrase in escaped double quotes (\"), as in:

"\"ssl certificate\""

If the $search string includes a phrase and individual terms, text search will only match the documents that include the phrase.

For example, passed a $search string:

"\"ssl certificate\" authority key"

The $text operator searches for the phrase "ssl certificate".

Negations

Prefixing a word with a hyphen-minus (-) negates a word:

  • The negated word excludes documents that contain the negated word from the result set.
  • When passed a search string that only contains negated words, text search will not match any documents.
  • A hyphenated word, such as pre-market, is not a negation. If used in a hyphenated word, $text operator treats the hyphen-minus (-) as a delimiter. To negate the word market in this instance, include a space between pre and -market, i.e., pre -market.

The $text operator adds all negations to the query with the logical AND operator.

Match Operation

Stop Words

The $text operator ignores language-specific stop words, such as the and and in English.

Stemmed Words

For case insensitive and diacritic insensitive text searches, the $text operator matches on the complete stemmed word. So if a document field contains the word blueberry, a search on the term blue will not match. However, blueberry or blueberries will match.

Case Sensitive Search and Stemmed Words

For case sensitive search (i.e. $caseSensitive: true), if the suffix stem contains uppercase letters, the $text operator matches on the exact word.

Diacritic Sensitive Search and Stemmed Words

For diacritic sensitive search (i.e. $diacriticSensitive: true), if the suffix stem contains the diacritic mark or marks, the $text operator matches on the exact word.

Case Insensitivity

Changed in version 3.2.

The $text operator defaults to the case insensitivity of the text index:

  • The version 3 text index is case insensitive for Latin characters with or without diacritics and characters from non-Latin alphabets, such as the Cyrillic alphabet. See text index for details.
  • Earlier versions of the text index are case insensitive for Latin characters without diacritic marks; i.e. for [A-z].

$caseSensitive Option

To support case sensitive search where the text index is case insensitive, specify $caseSensitive: true.

Case Sensitive Search Process

When performing a case sensitive search ($caseSensitive: true) where the text index is case insensitive, the $text operator:

  • First searches the text index for case insensitive and diacritic matches.
  • Then, to return just the documents that match the case of the search terms, the $text query operation includes an additional stage to filter out the documents that do not match the specified case.

For case sensitive search (i.e. $caseSensitive: true), if the suffix stem contains uppercase letters, the $text operator matches on the exact word.

Specifying $caseSensitive: true may impact performance.

See also

Stemmed Words

Diacritic Insensitivity

Changed in version 3.2.

The $text operator defaults to the diacritic insensitivity of the text index:

  • The version 3 text index is diacritic insensitive. That is, the index does not distinguish between characters that contain diacritical marks and their non-marked counterpart, such as é, ê, and e.
  • Earlier versions of the text index are diacritic sensitive.

$diacriticSensitive Option

To support diacritic sensitive text search against the version 3 text index, specify $diacriticSensitive: true.

Text searches against earlier versions of the text index are inherently diacritic sensitive and cannot be diacritic insensitive. As such, the $diacriticSensitive option for the $text operator has no effect with earlier versions of the text index.

Diacritic Sensitive Search Process

To perform a diacritic sensitive text search ($diacriticSensitive: true) against a version 3 text index, the $text operator:

  • First searches the text index, which is diacritic insensitive.
  • Then, to return just the documents that match the diacritic marked characters of the search terms, the $text query operation includes an additional stage to filter out the documents that do not match.

Specifying $diacriticSensitive: true may impact performance.

To perform a diacritic sensitive search against an earlier version of the text index, the $text operator searches the text index which is diacritic sensitive.

For diacritic sensitive search, if the suffix stem contains the diacritic mark or marks, the $text operator matches on the exact word.

See also

Stemmed Words

Text Score

The $text operator assigns a score to each document that contains the search term in the indexed fields. The score represents the relevance of a document to a given text search query. The score can be part of a sort() method specification as well as part of the projection expression. The { $meta: "textScore" } expression provides information on the processing of the $text operation. See $meta projection operator for details on accessing the score for projection or sort.

Examples

The following examples assume a collection articles that has a version 3 text index on the field subject:

db.articles.createIndex( { subject: "text" } )

Populate the collection with the following documents:

db.articles.insert(
   [
     { _id: 1, subject: "coffee", author: "xyz", views: 50 },
     { _id: 2, subject: "Coffee Shopping", author: "efg", views: 5 },
     { _id: 3, subject: "Baking a cake", author: "abc", views: 90  },
     { _id: 4, subject: "baking", author: "xyz", views: 100 },
     { _id: 5, subject: "Café Con Leche", author: "abc", views: 200 },
     { _id: 6, subject: "Сырники", author: "jkl", views: 80 },
     { _id: 7, subject: "coffee and cream", author: "efg", views: 10 },
     { _id: 8, subject: "Cafe con Leche", author: "xyz", views: 10 }
   ]
)

Search for a Single Word

The following query specifies a $search string of coffee:

db.articles.find( { $text: { $search: "coffee" } } )

This query returns the documents that contain the term coffee in the indexed subject field, or more precisely, the stemmed version of the word:

{ "_id" : 2, "subject" : "Coffee Shopping", "author" : "efg", "views" : 5 }
{ "_id" : 7, "subject" : "coffee and cream", "author" : "efg", "views" : 10 }
{ "_id" : 1, "subject" : "coffee", "author" : "xyz", "views" : 50 }

Match Any of the Search Terms

If the search string is a space-delimited string, $text operator performs a logical OR search on each term and returns documents that contains any of the terms.

The following query specifies a $search string of three terms delimited by space, "bake coffee cake":

db.articles.find( { $text: { $search: "bake coffee cake" } } )

This query returns documents that contain either bake or coffee or cake in the indexed subject field, or more precisely, the stemmed version of these words:

{ "_id" : 2, "subject" : "Coffee Shopping", "author" : "efg", "views" : 5 }
{ "_id" : 7, "subject" : "coffee and cream", "author" : "efg", "views" : 10 }
{ "_id" : 1, "subject" : "coffee", "author" : "xyz", "views" : 50 }
{ "_id" : 3, "subject" : "Baking a cake", "author" : "abc", "views" : 90 }
{ "_id" : 4, "subject" : "baking", "author" : "xyz", "views" : 100 }

Search for a Phrase

To match the exact phrase as a single term, escape the quotes.

The following query searches for the phrase coffee shop:

db.articles.find( { $text: { $search: "\"coffee shop\"" } } )

This query returns documents that contain the phrase coffee shop:

{ "_id" : 2, "subject" : "Coffee Shopping", "author" : "efg", "views" : 5 }

See also

Phrases

Exclude Documents That Contain a Term

A negated term is a term that is prefixed by a minus sign -. If you negate a term, the $text operator will exclude the documents that contain those terms from the results.

The following example searches for documents that contain the words coffee but do not contain the term shop, or more precisely the stemmed version of the words:

db.articles.find( { $text: { $search: "coffee -shop" } } )

The query returns the following documents:

{ "_id" : 7, "subject" : "coffee and cream", "author" : "efg", "views" : 10 }
{ "_id" : 1, "subject" : "coffee", "author" : "xyz", "views" : 50 }

Search a Different Language

Use the optional $language field in the $text expression to specify a language that determines the list of stop words and the rules for the stemmer and tokenizer for the search string.

If you specify a language value of "none", then the text search uses simple tokenization with no list of stop words and no stemming.

The following query specifies es, i.e. Spanish, as the language that determines the tokenization, stemming, and stop words:

db.articles.find(
   { $text: { $search: "leche", $language: "es" } }
)

The query returns the following documents:

{ "_id" : 5, "subject" : "Café Con Leche", "author" : "abc", "views" : 200 }
{ "_id" : 8, "subject" : "Cafe con Leche", "author" : "xyz", "views" : 10 }

The $text expression can also accept the language by name, spanish. See Text Search Languages for the supported languages.

Text Search Score Examples

Return the Text Search Score

The following query performs a text search for the term cake and uses the $meta operator in the projection document to append the relevance score to each matching document:

db.articles.find(
   { $text: { $search: "cake" } },
   { score: { $meta: "textScore" } }
)

The returned document includes an additional field score that contains the document’s relevance score:

{ "_id" : 3, "subject" : "Baking a cake", "author" : "abc", "views" : 90, "score" : 0.75 }

See also

$meta

Sort by Text Search Score

  • Starting in MongoDB 4.4, you can specify the { $meta: "textScore" } expression in the sort() without also specifying the expression in the projection. For example,

    db.articles.find(
       { $text: { $search: "cake" } }
    ).sort( { score: { $meta: "textScore" } } )
    

    As a result, you can sort the resulting documents by their search relevance without projecting the textScore.

    In earlier versions, to include { $meta: "textScore" } expression in the sort(), you must also include the same expression in the projection.
  • Starting in MongoDB 4.4, if you include the { $meta: "textScore" } expression in both the projection and sort(), the projection and sort documents can have different field names for the expression.

    For example, in the following operation, the projection uses a field named score for the expression and the sort() uses the field named ignoredName.
    db.articles.find(
       { $text: { $search: "cake" } } ,
       { score: { $meta: "textScore" } }
    ).sort( { ignoredName: { $meta: "textScore" } } )
    

    In previous versions of MongoDB, if { $meta: "textScore" } is included in both the projection and sort, you must specify the same field name for the expression.

  • In MongoDB 4.2 and earlier, to sort by the text score, include the same $meta expression in both the projection document and the sort expression. The following query searches for the term coffee and sorts the results by the descending score:

    db.articles.find(
       { $text: { $search: "coffee" } },
       { score: { $meta: "textScore" } }
    ).sort( { score: { $meta: "textScore" } } )
    

    The query returns the matching documents sorted by descending score.

See also

$meta

Return Top 2 Matching Documents

Use the limit() method in conjunction with a sort() to return the top n matching documents.

The following query searches for the term coffee and sorts the results by the descending score, limiting the results to the top two matching documents:

db.articles.find(
   { $text: { $search: "coffee" } },
   { score: { $meta: "textScore" } }
).sort( { score: { $meta: "textScore" } } ).limit(2)

See also

$meta

Text Search with Additional Query and Sort Expressions

The following query searches for documents where the author equals "xyz" and the indexed field subject contains the terms coffee or bake. The operation also specifies a sort order of ascending date, then descending text search score:

db.articles.find(
   { author: "xyz", $text: { $search: "coffee bake" } },
   { score: { $meta: "textScore" } }
).sort( { date: 1, score: { $meta: "textScore" } } )
←   $regex $where  →