MongoDB Full-text Search

In this chapter, we will introduce MongoDB's full-text search, which is different from SQL's "like" operator. MongoDB's full-text search is more efficient for text searching compared to the implementation of "like" in SQL.

Prepare test data

Insert a few records into the stores collection

db.stores.insert(
   [
     { _id: 1, name: "Java Hut", description: "Coffee and cakes" },
     { _id: 2, name: "Burger Buns", description: "Gourmet hamburgers" },
     { _id: 3, name: "Coffee Shop", description: "Just coffee" },
     { _id: 4, name: "Clothes Clothes Clothes", description: "Discount clothing" },
     { _id: 5, name: "Java Shopping", description: "Indonesian goods" }
   ]
)

Create text index

To use the ability of full-text search, you need to create a text index. Example:

db.stores.createIndex( { name: "text", description: "text" } )

Explanation:

• Create a text type index for the name and description fields.

Note: A collection only allows one text index, but a text index can include multiple fields.

$text operator

Use the $text operator for text search.

Syntax:

{ $text: { $search: "search keyword" } }

Example:

db.stores.find( { $text: { $search: "java coffee shop" } } )

Explanation:

• Search for documents containing the keywords "coffee", "shop", and "java" in the name and description fields.

Output:

{ "_id" : 3, "name" : "Coffee Shop", "description" : "Just coffee" }
{ "_id" : 1, "name" : "Java Hut", "description" : "Coffee and cakes" }
{ "_id" : 5, "name" : "Java Shopping", "description" : "Indonesian goods" }

Phrase search

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

Output

{ "_id" : 3, "name" : "Coffee Shop", "description" : "Just coffee" }

Relevance sorting

By default, the results of a $text query are unsorted. The $text query calculates a score (textScore) for the relevance of each document. We can use this score to sort and display documents with higher relevance at the top.

db.stores.find(
   { $text: { $search: "java coffee shop" } },
   { score: { $meta: "textScore" } }  // Declare to return the textScore relevance score
).sort( { score: { $meta: "textScore" } } ) // Specify sorting using textScore

Note: MongoDB full-text search does not provide very good support for Chinese phrases.