How to Use Text Search in MongoDB: Syntax and Examples

To use text search in MongoDB, first create a text index on the fields you want to search. Then, use the $text operator in your query to find documents matching the search terms.

Syntax

Text search in MongoDB requires two main steps: creating a text index on the fields you want to search, and querying with the $text operator.

• Creating text index: Use db.collection.createIndex({ field: "text" }) to enable text search on a field.
• Querying text: Use { $text: { $search: "search terms" } } to find documents matching the terms.

mongodb

db.collection.createIndex({ fieldName: "text" })
db.collection.find({ $text: { $search: "your search terms" } })

Example

This example shows how to create a text index on the description field and search for documents containing the word "coffee".

mongodb

use testdb

// Create collection and insert sample documents
db.products.insertMany([
  { name: "Coffee Maker", description: "Makes great coffee" },
  { name: "Tea Kettle", description: "Boils water for tea" },
  { name: "Espresso Machine", description: "Brews espresso coffee" }
])

// Create a text index on the description field
db.products.createIndex({ description: "text" })

// Search for documents containing 'coffee'
db.products.find({ $text: { $search: "coffee" } })

Output

[
  { "_id": ObjectId("..."), "name": "Coffee Maker", "description": "Makes great coffee" },
  { "_id": ObjectId("..."), "name": "Espresso Machine", "description": "Brews espresso coffee" }
]

Common Pitfalls

Common mistakes when using text search in MongoDB include:

• Not creating a text index before querying with $text. Queries will fail or return no results.
• Trying to create multiple text indexes on the same collection. MongoDB allows only one text index per collection.
• Expecting case-sensitive or partial word matches. Text search is case-insensitive but matches whole words or stems.
• Not using quotes for phrase search, which can lead to unexpected results.

mongodb

/* Wrong: Querying without text index */
db.products.find({ $text: { $search: "coffee" } })

/* Right: Create text index first */
db.products.createIndex({ description: "text" })
db.products.find({ $text: { $search: "coffee" } })

Quick Reference

Action	Command Example
Create text index	db.collection.createIndex({ field: "text" })
Search text	db.collection.find({ $text: { $search: "term" } })
Search phrase	db.collection.find({ $text: { $search: '"exact phrase"' } })
Exclude words	db.collection.find({ $text: { $search: "coffee -tea" } })
Sort by relevance	db.collection.find({ $text: { $search: "term" } }, { score: { $meta: "textScore" } }).sort({ score: { $meta: "textScore" } })

Key Takeaways

Always create a text index on the fields before using $text search.

$text queries match whole words and are case-insensitive by default.

Only one text index is allowed per collection in MongoDB.

Use quotes in $search to find exact phrases.

Sort results by relevance using the textScore metadata.