FAQ
Which naming styles are supported?
Seven in total, covering camelCase, PascalCase, snake_case, kebab-case and CONSTANT_CASE among other common conventions.
How are word boundaries detected?
By combining underscores, hyphens, spaces, camel-case capitals and digit boundaries, so mixed forms like userID2Name split correctly.
Can Chinese variable names be converted?
No, only Latin words are handled. Translate to English first.
About this tool
Converts a name across seven styles: camelCase, PascalCase, snake_case, kebab-case, CONSTANT_CASE, dot.case and plain spaces. The copy button on each row grabs that variant.
Each style has a job, which is why picking one matters: camelCase for JS / Java variables and methods; PascalCase for classes, types and React components, and in Go the leading case decides whether something is exported; snake_case for Python, Rust and SQL column names; kebab-case for CSS classes, URLs and file names; CONSTANT_CASE for constants and environment variables; dot.case for config keys and package namespaces.
Splitting rules: words break on _ - . space / \ and on case boundaries (userName becomes user + Name), while a run of capitals stays together (HTTPServer becomes HTTP + Server). Digits do not split off: sha256Sum gives sha256_sum, not sha_256_sum, and the constant form of v2 is V2 - chopping digits out is a common converter bug that produces names nobody can use.
Two boundaries to know: it works on ASCII only, so CJK and other scripts are ignored, and when the input contains no Latin letters or digits the page says so instead of showing seven empty strings. And a fully lowercase word cannot be re-split: username is one word, so camelCase and snake_case both return username - not an error. To keep the boundaries, type user_name, user-name or userName.