camelCase vs snake_case vs PascalCase vs kebab-case — Which to Use?

Updated October 2026 · 4 min read

Every programming language, framework, and platform has its own convention for writing multi-word identifiers. Getting it wrong does not usually break code, but it marks you as unfamiliar with the ecosystem — and in some systems (CSS properties, URL routing) the wrong case causes real bugs.

The Four Main Naming Conventions

Which Case for Which Language?

JavaScript / TypeScript

• camelCase: variables, functions, method names
• PascalCase: classes, React components, TypeScript interfaces and types
• SCREAMING_SNAKE_CASE: module-level constants — MAX_RETRY_COUNT, API_BASE_URL

Python (PEP 8)

• snake_case: variables, functions, module names — the default for nearly everything
• PascalCase: class names only
• UPPER_SNAKE_CASE: constants

CSS

• kebab-case: class names and custom properties — .user-profile, --primary-color
• Hyphens are part of CSS syntax; underscores technically require escaping in some selector contexts

URLs and Slugs

• kebab-case always: spaces become %20 or + in URLs. Google treats hyphens as word separators and underscores as word joiners, so kebab-case is better for SEO

Databases

• PostgreSQL / MySQL: snake_case for table and column names
• Some ORMs use PascalCase for model names and map to snake_case tables automatically

The Golden Rule: Consistency Over Convention

The specific convention matters less than applying it consistently across a codebase. A project that mixes getUserEmail and get_user_email in the same file is harder to read than one that picks either and sticks with it everywhere. Use a linter (ESLint for JS, Pylint for Python, Stylelint for CSS) to enforce conventions automatically rather than relying on manual review.