Text indexes for search in MongoDB - Time & Space Complexity

Time Complexity: Text indexes for search

O(log n + k)

Understanding Time Complexity

When using text indexes in MongoDB, it's important to understand how the search time changes as the amount of data grows.

We want to know how the search speed changes when we look for words in bigger collections.

Scenario Under Consideration

Analyze the time complexity of the following code snippet.

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

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

This code creates a text index on a field and then searches for documents containing a specific word.

Identify Repeating Operations

Identify the loops, recursion, array traversals that repeat.

• Primary operation: The database looks up the search word in the text index structure.
• How many times: The lookup happens once per search query, but the index internally may check multiple entries depending on the word frequency.

How Execution Grows With Input

As the number of documents grows, the index helps keep search fast by avoiding scanning every document.

Input Size (n)	Approx. Operations
10	Few index lookups, very fast
100	More index entries checked, still quick
1000	More index entries, but search remains efficient

Pattern observation: The search time grows slowly compared to the number of documents because the index narrows down the search quickly.

Final Time Complexity

Time Complexity: O(log n + k)

This means the search time grows slowly as the collection grows, thanks to the index structure, plus a factor depending on the number of matching documents.

Common Mistake

[X] Wrong: "Searching with a text index scans every document one by one."

[OK] Correct: The text index lets MongoDB jump directly to relevant documents without checking all of them, making search much faster.

Interview Connect

Understanding how text indexes speed up search shows you know how databases handle big data efficiently, a useful skill for many projects.

Self-Check

"What if we searched without a text index? How would the time complexity change?"

Practice

1. What is the main purpose of creating a text index in MongoDB?

easy

A. To encrypt text data for security

B. To enable fast searching of text within string fields

C. To store large text files inside the database

D. To sort documents by text fields alphabetically

Solution

1. Step 1: Understand what a text index does

   A text index allows MongoDB to quickly search text inside string fields of documents.

2. Step 2: Compare options with the purpose

   Sorting, storing files, or encrypting are not functions of text indexes.

3. Final Answer:

   To enable fast searching of text within string fields -> Option B

4. Quick Check:

   Text index = fast text search [OK]

Hint: Text index = fast search inside text fields [OK]

Common Mistakes:

• Confusing text index with sorting index
• Thinking text index stores files
• Assuming text index encrypts data

2. Which of the following is the correct syntax to create a text index on the description field in a MongoDB collection named products?

easy

A. db.products.createTextIndex({ description: 1 })

B. db.products.createIndex({ description: 'string' })

C. db.products.createIndex({ description: 'text' })

D. db.products.createIndex({ description: 1 })

Solution

1. Step 1: Recall the syntax for creating a text index

   The correct method is createIndex with the field set to 'text' type.

2. Step 2: Check each option

   db.products.createIndex({ description: 'text' }) uses createIndex with { description: 'text' }, which is correct. Others use wrong methods or types.

3. Final Answer:

   db.products.createIndex({ description: 'text' }) -> Option C

4. Quick Check:

   Use createIndex with 'text' type [OK]

Hint: Use createIndex({field: 'text'}) to make text index [OK]

Common Mistakes:

• Using 'string' instead of 'text' for index type
• Using createTextIndex (not a valid method)
• Creating normal ascending index (1) instead of text

3. Given a collection articles with a text index on fields title and content, what will the following query return?

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

medium

A. Documents containing both words 'mongodb' and 'tutorial' in title or content

B. Documents containing exactly the phrase 'mongodb tutorial' only

C. Documents containing either 'mongodb' or 'tutorial' in any field

D. Documents sorted alphabetically by title

Solution

1. Step 1: Understand $text and $search behavior

   The $search operator looks for documents containing all search words (AND behavior) in indexed text fields.

2. Step 2: Analyze the query

   The query searches for documents with both 'mongodb' and 'tutorial' in either title or content fields.

3. Final Answer:

   Documents containing both words 'mongodb' and 'tutorial' in title or content -> Option A

4. Quick Check:

   $search finds all words, not exact phrase [OK]

Hint: $search finds all words, not exact phrase [OK]

Common Mistakes:

• Thinking $search matches exact phrase only
• Assuming $search matches any single word (OR)
• Expecting sorting by title automatically

4. You tried to create two text indexes on a collection: one on title and another on description. What is the problem with this approach?

medium

A. MongoDB allows only one text index per collection

B. Text indexes must be created on numeric fields only

C. You must create text indexes using createTextIndex method

D. Text indexes cannot be created on multiple fields

Solution

1. Step 1: Recall MongoDB text index limitation

   MongoDB allows only one text index per collection, but it can cover multiple fields.

2. Step 2: Evaluate the options

   MongoDB allows only one text index per collection correctly states the limitation. Others are incorrect about field types or methods.

3. Final Answer:

   MongoDB allows only one text index per collection -> Option A

4. Quick Check:

   One text index per collection rule [OK]

Hint: Only one text index allowed per collection [OK]

Common Mistakes:

• Trying to create multiple text indexes on same collection
• Using wrong method name for creating text index
• Thinking text indexes only work on numbers

5. You want to create a text index on title and summary fields but give title twice the search weight of summary. Which is the correct way to create this weighted text index?

hard

A. db.books.createIndex({ title: 'text', summary: 'text' }, { weights: { summary: 2, title: 1 } })

B. db.books.createIndex({ title: 'text', summary: 'text' }, { weight: { title: 2, summary: 1 } })

C. db.books.createIndex({ title: 'text', summary: 'text' }, { weight: { summary: 2, title: 1 } })

D. db.books.createIndex({ title: 'text', summary: 'text' }, { weights: { title: 2, summary: 1 } })

Solution

1. Step 1: Recall syntax for weighted text indexes

   Weights option is an object named weights specifying field weights.

2. Step 2: Check options for correct spelling and values

   db.books.createIndex({ title: 'text', summary: 'text' }, { weights: { title: 2, summary: 1 } }) uses weights correctly and assigns title weight 2, summary 1 as required.

3. Final Answer:

   db.books.createIndex({ title: 'text', summary: 'text' }, { weights: { title: 2, summary: 1 } }) -> Option D

4. Quick Check:

   Use 'weights' option with correct field weights [OK]

Hint: Use 'weights' option with field weights object [OK]

Common Mistakes:

• Using 'weight' instead of 'weights' option
• Swapping weights between fields
• Not specifying weights object correctly