Text Transform

What is a text transform tool?#

A text transform tool rewrites a string according to a fixed rule — change the case, switch the naming convention, flip half-width and full-width forms, or reverse the characters. Most of these jobs have a one-line answer in your head until you actually try to do them: “convert this header to kebab-case” is obvious for Content Type but ambiguous for HTTPServer, getUserById_v2, or X-Rate-Limit-Remaining. Which parts are words? Where do the boundaries fall?

This tool answers that with a single, predictable word-splitter. It normalises separators (underscore, dash, whitespace, punctuation), then inserts boundaries at camelCase and PascalCase joints, at acronym-to-word joints (HTTPServer → HTTP Server), and at letter-to-digit joints (v2 → v 2). The resulting tokens feed twelve operations: five naming conventions (camelCase, PascalCase, snake_case, kebab-case, CONSTANT_CASE), four letter cases (UPPER, lower, Title, Sentence), East-Asian width conversion (half ↔ full), and reverse. Reverse walks by Unicode code point, so emoji and other surrogate-pair characters survive intact instead of turning into broken halves.

How to use it#

1. Pick the operation from the dropdown in the toolbar. The twelve options are grouped logically: naming conventions first, then letter cases, then width conversion, then reverse.
2. Type or paste into the input pane on the left, or hit Sample to load an example that exercises the splitter.
3. Read the result in the output pane on the right — it updates the moment input or the selected operation changes.
4. Click Copy in the output header to grab the transformed string, or Clear to start over.

The operations are total: empty input yields empty output, and no operation can throw. Switching the dropdown re-runs the same input through the new rule instantly, which makes this a fast way to compare, say, snake_case against kebab-case for the same identifier.

Key features#

• One splitter, twelve outputs. The same word boundaries feed every naming convention, so getUserById_v2 produces consistent getUserByIdV2 / GetUserByIdV2 / get_user_by_id_v2 / get-user-by-id-v2 / GET_USER_BY_ID_V2.
• Acronym-aware splitting. HTTPServer splits into HTTP + Server, not H T T P Server — the recognised pattern (a run of capitals followed by a capital-plus-lowercase word) keeps acronyms readable.
• Letter-to-digit boundaries. v2 becomes two tokens, so version2 slugifies cleanly instead of gluing a digit onto a word.
• East-Asian width conversion. Half-width ASCII Hello 123 becomes full-width Ｈｅｌｌｏ　１２３ (space included), and the inverse restores ASCII — useful when filling East-Asian forms or aligning CJK text.
• Code-point reverse. Reversing a😀b yields b😀a, not a broken emoji — the splitter walks code points rather than UTF-16 code units.
• Local and instant. No network, no upload; every transform is pure string math in the page.

Worked example#

Paste a human-readable label into the input:

user first name

Switch the operation through the five naming conventions and watch the output pane:

• camelCase → userFirstName
• PascalCase → UserFirstName
• snake_case → user_first_name
• kebab-case → user-first-name
• CONSTANT_CASE → USER_FIRST_NAME

Now try a messier identifier that mixes conventions:

HTTPServer v2

Because the splitter recognises the HTTP + Server boundary and the v + 2 boundary, the outputs stay readable rather than collapsing into httpservver-v2:

• kebab-case → http-server-v-2
• snake_case → http_server_v_2
• CONSTANT_CASE → HTTP_SERVER_V_2

That is the difference between a transform that actually understands identifier structure and one that just lowercases and joins on a separator.

FAQ#

Why does HTTPServer not become h-t-t-p-server?#

Because the splitter treats a run of capitals followed by a capital-plus-lowercase word as an acronym plus a word. HTTPServer is read as HTTP + Server, which is almost always what you meant. A naive splitter that broke on every capital would produce unreadable slugs like h-t-t-p-server.

What is the difference between Title Case and Sentence Case?#

Title Case capitalises the first letter of every whitespace-delimited word (the quick fox → The Quick Fox), which is right for headlines and book titles. Sentence Case capitalises only the first alphabetic character of the whole string (the quick fox → The quick fox), which is right for prose and sentences. Pick based on what the text actually is.

When would I use half-width ↔ full-width conversion?#

East-Asian typography and some government/business forms expect full-width Latin letters and digits (ＡＢＣ１２３) for visual alignment with CJK characters. Conversely, text copied from a Japanese source often arrives in full-width form and needs to be restored to ASCII before a regex or code parser will accept it. The two operations are exact inverses of each other.

Does reverse handle emoji correctly?#

Yes. The operation iterates by Unicode code point (via Array.from), not by UTF-16 code unit, so a character outside the basic multilingual plane — an emoji like 😀, a rare CJK ideograph, or a mathematical symbol — is reversed as one unit. Reversing abc😀def gives fed😀cba, with the emoji intact.

Is this useful for renaming a batch of files or variables?#

For a single identifier, yes — paste it in, pick the convention, copy out. For batch work you would normally script it, but this page is a quick way to confirm the exact output a convention produces before you commit it to a regex or codemod, especially for the awkward mixed-case inputs that tempt bugs.