Identifier naming case detector

An identifier such as getUserName is camelCase; UserProfile is PascalCase; http_status_code is snake_case; content-type is kebab-case; and MAX_RETRY_COUNT is SCREAMING_SNAKE_CASE. Empty, missing, or illegal identifiers fail with invalid_input instead of a silent empty split. Successful API calls are billed at $0.002; validation failures are not charged.

What identifier case detection means and how each style is recognized

PascalCase, also called UpperCamelCase, starts with a capital and is the usual shape of type names, React components, and C# classes: HttpServer is http plus server. snake_case writes every letter in lowercase and joins words with a single underscore, the default for Python functions, Ruby methods, and many SQL columns. kebab-case does the same job with hyphens and appears in CSS classes, URL slugs, and HTTP header names such as content-type. SCREAMING_SNAKE_CASE is snake_case in full capitals and is reserved for constants and environment keys such as MAX_RETRY_COUNT. This detector also reports three honest extras that renaming tools meet every day: lower for a single lowercase token such as user, upper for a single capital token such as OK, and mixed when underscores, hyphens, and interior capitals are combined in one symbol, for example get_userName. Underscore-separated lowercase parts are snake_case; the same parts in capitals are SCREAMING_SNAKE_CASE. Leading underscores used for privacy or Python dunders are stripped before the style is named so _foo_bar stays snake_case and __init__ stays snake_case, while the returned identifier still echoes the original spelling.

Input contract, word splitting, and error behaviour

Length is capped at four thousand ninety-six characters: a longer paste fails with an explicit limit message rather than a silent cut. The identifier must start with a letter, underscore, or dollar sign and then use only ASCII letters, digits, underscores, or hyphens. Leading or trailing hyphens and consecutive hyphens are rejected because they are not legal names in the languages this tool is built for. At least one letter is required: a token made only of digits or symbols has no naming case to detect. There is no locale-specific title casing and no partial success that drops tokens without saying so. POST a JSON body with one required field, text, holding the identifier to inspect. Aliases identifier, name, value, and input are accepted so spreadsheets, query strings, and typed forms can reuse the same contract without renaming a column. The value must be a non-empty string after trim; numbers, booleans, objects, and arrays are rejected so a mistaken JSON body cannot become the word true or an object dump. On success the body returns five stable fields. identifier echoes the trimmed input so you can audit what was classified. case_style is one of camelCase, PascalCase, snake_case, kebab-case, SCREAMING_SNAKE_CASE, lower, upper, or mixed. words is the ordered array of lowercase tokens after splits at underscores, hyphens, a lowercase-or-digit to uppercase edge, an acronym followed by a Titlecase word, and letter-to-digit or digit-to-letter edges, so XMLHttpRequest becomes xml, http, and request, and getUser2FA becomes get, user, 2, and fa. word_count is the length of that array. readable joins the words with single spaces for labels and logs. Missing text, an empty string, and an illegal character set fail with invalid_input naming the field; those failures are not billed.

Where renaming tools, linters, and migrations use a case detector

A rename refactor is unsafe if the tool assumes every symbol is already camelCase. Documentation generators turn the readable phrase into headings and accessibility labels while still knowing the original convention for cross links. Failed validation does not consume balance. Keep requests to one identifier, pin the words array in tests, and you get a reproducible classification for rename pipelines, lint rules, and classroom checks without standing up a parser generator of your own. Front-end codebases mix React components in PascalCase, hooks in camelCase, CSS modules in kebab-case, and environment constants in SCREAMING_SNAKE_CASE; a detector that returns both the style and the words lets a codemod rebuild getUserName as get_user_name or GetUserName without a human pointing at each file. Backend migrations face the same split between Python snake_case functions, HTTP kebab-case headers, and Java PascalCase types when an OpenAPI document, an ORM model, and a client SDK must stay aligned. Style linters and review bots can flag a function that arrived as Http_Server instead of guessing whether the author meant snake_case or PascalCase. Teachers and coding-interview platforms can ask a student to name the style of ten identifiers and mark the answer against this oracle instead of a spreadsheet macro that treats UserID and user_id as the same string.