Slugify String

Examples

Each row was produced by the tool’s own function with the default options (hyphen, Lowercase on, Trim on):

What happens in order: full-width forms such as Ｗ, ％ and the ideographic space are converted to ASCII (NFKC); the text is decomposed (NFD) and combining marks are dropped; the symbols and letters in the tables below are replaced; any other non-ASCII character becomes a space; every remaining character that is not a letter, digit, space, hyphen, underscore or dot becomes a space; runs of spaces become one separator; and runs of -, _ and . collapse into one separator. That last step is why user_profile.v2 loses its underscore and dot: existing separators are normalised to the one you picked. With Lowercase off and the underscore separator, the first example gives 10_Tips_and_Tricks_for_Node_js_2026_Edition, a common form for file names and Python identifiers.

Because a removed character counts as a space, words around CJK text stay apart: Vue3入门Vite教程 gives vue3-vite, and Hello　World with a full-width space gives hello-world.

Character Mappings

Symbols become a separate word, so Rock & Roll gives rock-and-roll and Tom&Jerry gives tom-and-jerry. Accented letters lose their marks (Unicode NFD decomposition, then combining marks are removed): é → e, ñ → n, Łódź → lodz, Tiếng Việt → tieng-viet. Letters that do not decompose use the second table, with values from sindresorhus/transliterate. Uppercase letters keep their case (Ø → O, Œ → OE) until Lowercase is applied. The currency signs in the table become the word in their Unicode character name; others, such as $ and ¢, are removed (YEN SIGN → yen, WON SIGN → won; see the Unicode currency symbols chart).

Limitations

• Marks are removed without language-specific rules: German ä ö ü become a o u (not ae oe ue), and Turkish İ becomes I.
• Non-Latin scripts are removed, so input written only in Chinese, Japanese or Korean gives an empty slug with Trim on. With Trim off it gives a single separator: 東京 → -.
• An apostrophe (', ’, ‘, ʼ, ʻ, ´, ′) after a letter or digit and before a letter is deleted: John's Guide → johns-guide, 1990's Music → 1990s-music. Next to a space or at the start or end of the text it acts as a separator: rock 'n' roll → rock-n-roll.
• With Trim off, separators at the start and end stay: Hello World! → hello-world-.
• Words are never shortened and stop words (a, the, of) are kept. There is no length limit, so a long sentence gives a long slug; trim it by hand if your CMS caps slug length.
• # becomes the word hash, so C# gives c-hash rather than c-sharp. Edit such names before converting.
• The tool converts one string at a time; each line break counts as a space, so pasting a list produces one long slug.

Differences From sindresorhus/slugify

sindresorhus/slugify, a widely used npm package, shares the & → and rule and the letter table, but the README examples show where results part ways. The same inputs here give:

The package also splits camelCase and, through transliterate, turns Fußgängerübergänge into fussgaengeruebergaenge; this tool gives fussgangerubergange. An apostrophe between letters is deleted here, straight or curly (Conway's Law and Conway’s Law both give conways-law), as WordPress does; the package deletes it only before a final s or t, so Rock'n'Roll becomes rock-n-roll there and rocknroll here. If your site already has published URLs, match the generator that made them, and use this page for one-off slugs or to check what a title would turn into. To encode a string for a query parameter instead of a path, use the URL encoder.

When to Use Slugs

• Blog post URLs: /blog/how-to-use-slugs
• Product pages: /products/red-running-shoes
• File names: my-important-document.pdf
• Database keys: Unique, human-readable identifiers without spaces.

Consistent slug generation ensures your URLs remain stable and SEO-friendly even when the original title changes formatting or punctuation.