Convert Text to snake_case

Converting Text to snake_case

snake_case replaces all spaces and delimiters with underscores (_) and lowercases every letter. It is called "snake case" because the underscores resemble a snake lying on the ground between the words.

Basic Conversion

Input:  User First Name
Output: user_first_name

Input:  backgroundColor
Output: background_color

Input:  GET-API-RESPONSE
Output: get_api_response

Where snake_case Is Used

snake_case is the dominant convention in several ecosystems:

# Python — variables, functions, methods, modules (PEP 8)
user_name = "Alice"
def get_user_profile(user_id):
    pass
import json_parser

# Ruby — variables, methods, file names
first_name = "Alice"
def calculate_total_price(items)
end

Database Column Naming

snake_case is the de facto standard for database schemas:

CREATE TABLE user_profiles (
    user_id       SERIAL PRIMARY KEY,
    first_name    VARCHAR(100),
    last_name     VARCHAR(100),
    email_address VARCHAR(255),
    created_at    TIMESTAMP DEFAULT NOW(),
    updated_at    TIMESTAMP
);

PostgreSQL, MySQL, and SQLite all conventionally use snake_case for table and column names.

API Design

REST APIs often use snake_case for JSON keys, especially in Ruby on Rails and Python ecosystems:

{
  "user_id": 42,
  "first_name": "Alice",
  "created_at": "2025-01-15T10:30:00Z"
}

Conversion Algorithm

1. Insert an underscore before each uppercase letter that follows a lowercase letter (for camelCase input).
2. Replace all hyphens, dots, and spaces with underscores.
3. Collapse consecutive underscores into one.
4. Lowercase the entire string.
5. Trim leading and trailing underscores.

Handling Acronyms

"XMLParser"     → "xml_parser"
"getHTTPSUrl"   → "get_https_url"
"userID"        → "user_id"

Acronyms are split at the boundary where an uppercase run meets a lowercase character.

Edge Cases

• Single word: "hello" → "hello".
• Already snake_case: passes through unchanged.
• Consecutive uppercase: "HTMLParser" → "html_parser".
• Numbers: "item2Count" → "item_2_count" or "item2_count" depending on the algorithm.