Lowercase Converter

Letters that don't lowercase the way you'd expect

toLowerCase() and its relatives use the default Unicode rules, which are the same in every country. Most of the time that's what you want. A few letters behave differently:

• Greek has two small sigmas. A capital Σ at the end of a word becomes the final form ς, and σ everywhere else: ΟΔΟΣ becomes οδος.
• Turkish and Azerbaijani have a dotted and a dotless i. "ISTANBUL".toLowerCase() gives istanbul, but Turkish readers expect ıstanbul, which toLocaleLowerCase("tr") returns.
• "İstanbul".toLowerCase() gives an i followed by a combining dot (U+0307), so the 8-character word comes back with 9. A length check after lowercasing can fail because of it.
• The reverse bug hits code that lowercases identifiers with the user's locale. On a Turkish system, "TITLE".toLocaleLowerCase("tr") is tıtle, and a keyword check fails. Java's toLowerCase() with no argument uses the system locale, which is why toLowerCase(Locale.ROOT) is the safe choice for keys and commands.
• The capital ẞ lowercases to ß, and Cyrillic and Armenian convert like Latin letters (ПРИВЕТ becomes привет). Scripts with no capitals, like Arabic, Chinese and Hindi, come out unchanged.

Lowercase for logins, emails and URLs

Lowercasing both sides is the usual way to compare text without caring about case. Email domains are not case-sensitive. RFC 5321 allows the part before the @ to be, but almost no mail provider treats it that way, so most sign-up forms store a lowercase copy and compare against that.

For comparisons, Python's casefold() goes further than lower(): "Straße".lower() doesn't match "STRASSE".lower(), but the casefolded values are equal. In SQL, WHERE LOWER(email) = LOWER(?) can't use a plain index on email. PostgreSQL can index the expression (CREATE INDEX ON users (LOWER(email));) or use the citext type, and MySQL's default collations already ignore case.

Paths are where mixed case causes real bugs. Domain names ignore case, but the path after them usually doesn't, so /Blog/Post and /blog/post can be two pages, or one page and a 404. File systems differ as well. macOS and Windows ignore case by default and Linux doesn't, so an import of ./header that finds Header.tsx on a laptop breaks in a Linux CI build. Keeping URLs and file names lowercase from the start avoids both problems.

Programming cases built on small letters

The other buttons split your text into words at spaces, punctuation and changes from small to capital letters, and keep acronyms together, so XMLHttpRequest becomes xml_http_request. These are the formats based on small letters:

• snake_case: Python and Ruby variables and functions, and database tables and columns (total_price_usd).
• kebab-case: URL slugs, CSS classes and properties (background-color), file names and command-line flags (--dry-run).
• camelCase: variables and functions in JavaScript, TypeScript and Java, and keys in many JSON APIs (totalPriceUsd).
• Sentence case: ordinary prose. Everything is lowercased first, then the first letter of each sentence gets a capital.
• The capital-based formats, CONSTANT_CASE and PascalCase, are explained on the Uppercase Converter page.

Lowercase text in code

Every language has a lowercase function. Prefer the Unicode-aware one, because a few older functions only touch the letters A to Z.

"Hello WORLD".toLowerCase();            // 'hello world'
"ISTANBUL".toLocaleLowerCase("tr");     // 'ıstanbul' (Turkish rules)

"Hello WORLD".lower()      # 'hello world'
"Straße".casefold()        # 'strasse', for comparisons

mb_strtolower("ÉCOLE");    // "école"
strtolower("ÉCOLE");       // "École": only A to Z change

"ÉCOLE".downcase           # "école"

"Hello WORLD".toLowerCase(Locale.ROOT)

"Hello WORLD".ToLowerInvariant()

strings.ToLower("Hello WORLD")

SELECT LOWER(email) FROM users;
-- SQLite's built-in LOWER() only changes A to Z

echo "Hello WORLD" | tr '[:upper:]' '[:lower:]'

name="Hello WORLD"
echo "${name,,}"           # Bash 4 or newer