Paste your text in the box and every case appears at once, then hit Copy on the one you want. Nothing is sent anywhere, so it is safe for text you would not paste into a random website.
The cases split into two families. UPPERCASE, lowercase, Sentence case and Title Case keep your spaces and punctuation exactly where they were, because they are about letters. camelCase, PascalCase, snake_case, kebab-case, CONSTANT_CASE, dot.case and path/case throw the spaces away and rebuild the text from words, because they are about naming things in code. That is why "hello, world" turns into "Hello, World" in one family and "hello_world" in the other.
What title case rules actually say
Proper title case does not capitalise every word. It leaves small words in lowercase unless they open or close the title, which is why "The Lord of the Rings" has a lowercase "of" and "the" in the middle. The three choices here do genuinely different things:
- AP style capitalises every word of four letters or more. So articles and short words like a, an, the, and, but, or, for, nor, in, of, on, to and up drop to lowercase, but "From" and "Over" stay capitalised because they are long enough.
- Chicago style lowercases every preposition however long, so "from", "over", "between" and "through" all go down. It also lowercases "to" in front of a verb and "as" everywhere.
- Capitalise every word is the one most converters give you without saying so. Useful for headings in a product table, wrong for a book title.
The active rule set is printed next to the Title Case row, so you always know which one produced the result. Both real styles capitalise the first and last word no matter what, and both treat each half of a compound word as its own word.
Why other converters break on real text
Because they assume English letters and nothing else. Three things go wrong, and all three are fixed here.
The first is Turkish. In Turkish the capital of "i" is "İ" and the lowercase of "I" is "ı", so a naive uppercase turns "istanbul" into "ISTANBUL" when the city is "İSTANBUL". Set Language rules to Turkish and you get the right letters. The same applies to Azerbaijani.
The second is German. The uppercase of "ß" is "SS", so "straße" becomes "STRASSE". A converter that only shifts letters by a fixed amount leaves the ß sitting there.
The third is Greek. Lowercase sigma has two shapes: "σ" in the middle of a word and "ς" at the end. "ΟΔΟΣ" has to come back as "οδός" shape, with the final sigma, and a per character loop gets that wrong every time. This page checks the position of the letter before it picks the shape, even in aLtErNaTiNg and inverse case where letters are handled one at a time.
Emoji, accents and counting characters
An emoji is usually several characters glued together. The family 👨👩👧👦 is four people joined by invisible characters, and a flag is two regional letters. Cut between them and you get rubbish, which is why so many tools mangle emoji when they reverse text or convert to snake_case. Everything here walks the text by grapheme, meaning whole visible characters, so emoji and accented letters come out intact, and the character count matches what your eyes see rather than how many bytes it took.
Remove accents works on Latin letters only. "Crème Brûlée" becomes "Creme Brulee" and ß becomes ss, while Greek, Arabic, Hindi and Thai are left alone, because stripping marks there would destroy the words instead of simplifying them.
Which case should you use
Follow whatever the thing you are typing into already uses. JavaScript variables and JSON keys use camelCase. React components and classes use PascalCase. Python variables, database columns and most APIs use snake_case. URLs, CSS classes and file names use kebab-case, because search engines read a hyphen as a word break and an underscore as glue. Environment variables use CONSTANT_CASE. Analytics events and translation keys tend to use dot.case.
What the cleanup buttons do
Each one rewrites the box in place and stacks an undo, so you can chain them. Trim whitespace cuts the ends off every line. Collapse spaces turns any run of spaces or tabs into one, leaving line breaks alone. Remove line breaks joins everything into one paragraph, which is the fix for text copied out of a PDF. Line break per sentence does the opposite. Remove duplicate lines keeps the first copy of each line. The sorts ignore blank lines and compare letters the way your chosen language does, so "ä" lands next to "a" rather than at the bottom, and numbers inside the text sort 2 before 10. Reverse the text flips visible characters, not bytes. Strip HTML tags drops the tags, throws away script and style blocks with their contents, turns block tags into line breaks and decodes entities like & back into real characters. Remove extra punctuation squashes "!!!" to "!" and pulls stray spaces off the front of commas and full stops.
Does anything leave my browser
No. There is no server to send it to. The page keeps your text in this browser's local storage under the key oddlyplay.caseconv.v1 so it survives a reload, and Clear wipes both the box and that copy. Private browsing or blocked storage changes nothing except that the text will not come back after a reload.