Snake Case Converter

What is snake_case?

Snake case writes a multi-word name in lowercase with underscores between the words, so HelloWorld example becomes hello_world_example. The underscores stand in for the spaces that identifiers are not allowed to contain.

Paste any text and the converter works out where the words are before joining them. It splits on case changes, on whitespace, and on symbol boundaries, so it handles input that is already camelCase or kebab-case rather than only plain prose.

Why the word splitting matters

A naive converter lowercases everything and swaps spaces for underscores, which mangles anything that was already an identifier. Feed it getUserID and you get getuserid, one word, meaning lost.

Detecting the boundaries first is what makes the conversion useful on real input: existing identifiers, column headers pasted out of a spreadsheet, keys from a JSON payload, filenames. That is most of what people actually convert, and it is the part a find-and-replace cannot do.

Are there alternatives?

The other conventions solve the same problem differently. camelCase and PascalCase mark word boundaries with capitals instead of underscores. kebab-case uses hyphens, which suits URLs and filenames but breaks in most languages as an identifier because it parses as subtraction. SCREAMING_SNAKE_CASE is the same shape as snake case and conventionally means a constant.

Which one is correct is decided by the language and the codebase, not by preference. Python and Ruby lean on snake case for variables and functions, and SQL column names commonly use it too.

Fun facts and more info

Acronyms are where every case converter shows its seams. There is no universally right answer for HTTPServer, so expect runs of capitals to be normalised in a way you may want to check rather than trust.

Snake case is a convention rather than a specification, which is why implementations disagree at the edges. Punctuation, leading digits, and repeated separators all get handled slightly differently depending on who wrote the converter.

Round trips are not guaranteed either. Convert an identifier to snake case and back to camelCase and you may not land where you started, because the acronym normalisation in the first step has already thrown information away.