Case converter: uppercase, lowercase and ten more formats

Title Case is not the same in every language

This is the thing nearly every converter gets wrong. English titles capitalise almost every word — The Lord of the Rings — and that habit has leaked into online converters, which apply it to any language whatsoever. Applied to Spanish it produces titles that are simply incorrect.

The Spanish rule is straightforward: a title capitalises the first word and proper nouns, and nothing else. It is Cien años de soledad, never Cien Años De Soledad. So this page defaults to the rule of the language version you are on, with a switch at the top for when you are formatting a title in the other one — which happens more often than you would think.

The English rule has its own detail, too. It is not enough to lowercase the articles. The convention people expect — the news style one — lowercases articles, coordinating conjunctions and prepositions of fewer than four letters, but capitalises those of four or more. That is why it is Gone With the Wind rather than Gone with the Wind. The first and last words always take a capital whatever they are: What to Look For. Acronyms are left alone, so the NASA report becomes The NASA Report and not The Nasa Report.

For the Spanish rule, proper nouns cannot be guessed without understanding the text, so what happens instead is that the clue you left is respected: a word that already had a capital and is not the first keeps it. The one hopeless case is a text arriving entirely in capitals, where there is no clue at all — and when that happens, you are told.

The programming formats, and why they strip accents

The six formats in the second row exist because every ecosystem names things its own way:

• camelCase — variables and functions in JavaScript, Java and C#.
• PascalCase — classes and components; camelCase with the first letter up too.
• snake_case — Python, Ruby and almost every database column name.
• SCREAMING_SNAKE_CASE — constants and environment variables.
• kebab-case — URLs, CSS class names and filenames.
• dot.case — configuration keys and namespaces.

All six transliterate accents and tell you what changed. It is not sloppiness: an identifier with an accent looks like it works in JavaScript or Python, but breaks the moment that name travels — database columns, JSON keys passing through older systems, filenames moving between Windows and Linux, anything that ends up in a URL. So número de teléfono comes out as numeroDeTelefono, and you are told. The ordinary text formats leave accents intact.

The word splitting is careful too: it recognises words that were already joined. XMLHttpRequest is read as three words and gives xml_http_request, not xmlhttprequest. And since the code formats apply line by line, you can paste a whole list of labels and get the whole list of identifiers back at once.

The cleanup panel is what most people actually came for

Plenty of people arrive at a case converter with a problem that has nothing to do with case: they pasted something from Word or a PDF and it came out broken. That is why the cleanup panel lives on this page rather than hidden in another tool.

The commonest case is the PDF one: copying pastes every visual line as its own line, so a six-line paragraph arrives in six pieces. Turning stray line breaks into paragraphs joins the ones inside a paragraph and keeps a break only where there was a blank line. That is different from removing all line breaks, which welds the document into one block — which is why the two cannot both be ticked.

The rest goes after the invisible: double spaces, trailing spaces nobody sees but that break an import, leftover blank lines, curly quotes and em dashes that sneak in from Word. Before applying anything you see how many characters are about to go, and you can always undo.

Slugs, one per line

A slug is the readable part of an address: in utiliobox.com/en/case-converter, the slug is case-converter. It comes from the title by stripping accents, lowercasing, and putting hyphens where there were no letters or digits. It matters because a URL with accents or spaces escapes itself and ends up full of codes like %C3%B1: it breaks when pasted into an email and helps nobody.

Here the slug is generated as you type, with one line per slug: paste twenty headlines and get twenty addresses. And if what you want is to count words or estimate reading time for that same text, that is what the word counter does, also without uploading anything.

And when the cleanup you need does not fit in a checkbox — keeping only the lines that match a pattern, or reordering fields — that is where the regex tester comes in, replace tab included.