snake_case Converter

snake_case: Python, databases, and file names

snake_case writes multi-word identifiers in all lowercase with an underscore between each word, as in total_order_value. It is the house style of Python, Ruby, Rust, C, and virtually every SQL schema - an ecosystem large enough that snake_case is probably the most-typed naming convention in software. Readability studies going back to the 1980s consistently find underscore-separated identifiers slightly faster to read correctly than camelCase, at the cost of a few extra keystrokes.

Where snake_case is the documented standard:

• Python - PEP 8 prescribes it for variables, functions, methods, and modules: user_profile, get_user_by_id, calculate_total_price
• SQL - table and column names: first_name, created_at, is_active
• Rust - the compiler emits a non_snake_case warning if you name a function otherwise
• Ruby - methods and locals: find_by_email, per the community style guide
• Terraform and Ansible - resource names and variables: instance_type, vpc_security_group_ids
• C standard library - str_len-style names predate camelCase's popularity entirely

Why databases push you toward snake_case

There is a technical reason SQL schemas are almost universally snake_case, not just a stylistic one. PostgreSQL folds unquoted identifiers to lowercase, so a column created as createdAt is silently stored as createdat and must be written "createdAt" - quotes and all - forever after. Oracle folds to uppercase. Underscores sidestep the whole problem because case folding cannot damage a name that has no case distinctions to lose.

The classic friction point is the boundary between a snake_case backend and a camelCase JSON API. Pick one side to convert on and be consistent: serializer-level mapping (Django REST Framework's djangorestframework-camel-case, Jackson's SNAKE_CASE naming strategy, Ecto's field sources) beats ad-hoc renaming scattered through controllers. Also watch the acronym trap - parse_HTTP_response should be parse_http_response, and a camelCase source like parseHTTPResponse is the input most converters mangle.

This converter splits on spaces, hyphens, dots, and existing underscores, and also detects camelCase and PascalCase boundaries, so helloWorld, Hello-World, and hello world all land on hello_world. Need the constant form instead? SCREAMING_SNAKE_CASE (MAX_RETRIES) has its own converter. Nothing you paste here is uploaded - the transformation is plain JavaScript running inside your own browser tab.