JavaScript string utilities

The string module of modern-fns, a dependency-free JavaScript and TypeScript utility library.

import { slugify, truncate, mask, camelCase } from 'modern-fns';
import slugify from 'modern-fns/string/slugify';

Every function accepts a string and tolerates null/undefined/non-strings by returning '' (or [] for the extract* pair) rather than throwing. Character handling is unicode-aware: the converters, truncate, initials and mask count by code point, so emoji and non-Latin scripts are never cut in half.

---

capitalize(value) / capitalizeWords(value)

capitalize('hello world'); // 'Hello world'
capitalizeWords('hello vue world'); // 'Hello Vue World'
capitalizeWords('  spaced   out  '); // '  Spaced   Out  '

Signatures — (value: string) => string.

Edge cases — capitalizeWords preserves the original spacing and punctuation instead of re-joining on single spaces. Neither lower-cases the rest of the word.

---

camelCase / pascalCase / kebabCase / snakeCase / constantCase

One unicode-aware word splitter behind all five, so they agree with each other and round-trip.

function camelCase(value: string): string; // and the other four

Input	camel	pascal	kebab	snake	constant
user_first_name	userFirstName	UserFirstName	user-first-name	user_first_name	USER_FIRST_NAME
XMLHttpRequest	xmlHttpRequest	XmlHttpRequest	xml-http-request	xml_http_request	XML_HTTP_REQUEST
hello world 42	helloWorld42	HelloWorld42	hello-world-42	hello_world_42	HELLO_WORLD_42

Edge cases — acronym runs split correctly (XMLHttp → XML, Http); digits attach to the preceding word (user2 is one word); punctuation-only input returns ''; the functions are idempotent.

TypeScript — mapKeys(apiResponse, camelCase) is the usual pairing.

---

slugify(value, options?)

function slugify(
  value: string,
  options?: { separator?: string; lower?: boolean; maxLength?: number },
): string;

slugify('Hello Vue World!'); // 'hello-vue-world'
slugify('Café & Bar — 2026'); // 'cafe-and-bar-2026'
slugify('Straße'); // 'strasse'
slugify('Hello World', { separator: '_' }); // 'hello_world'
slugify('one two three four', { maxLength: 11 }); // 'one-two'

Edge cases — diacritics are folded to ASCII, & becomes and, ß/æ/œ/ø/ł/þ are transliterated, and punctuation and emoji are dropped. maxLength trims on a separator boundary where possible. Scripts with no ASCII equivalent (Chinese, Arabic, Devanagari) reduce to '' — always keep a fallback such as the record id.

---

truncate(value, length, options?) / truncateWords(value, wordCount, options?)

function truncate(
  value: string,
  length: number,
  options?: { omission?: string; separator?: string | RegExp },
): string;
function truncateWords(value: string, wordCount: number, options?: { omission?: string }): string;

truncate('The quick brown fox', 10); // 'The quick…'  (exactly 10 chars)
truncate('The quick brown fox', 10, { separator: ' ' }); // 'The…'
truncate('abcdefgh', 6, { omission: '...' }); // 'abc...'
truncateWords('The quick brown fox jumps', 3); // 'The quick brown…'

Edge cases — length includes the omission, so the result never exceeds the width you budgeted. Strings already short enough are returned unchanged. A length of 0 or less returns ''. Emoji are never split. truncateWords trims the input and collapses whitespace when counting.

---

initials(value, options?)

function initials(value: string, options?: { max?: number; uppercase?: boolean }): string;

initials('Vishnu M'); // 'VM'
initials('john ronald reuel tolkien'); // 'JT'
initials('mary-jane watson'); // 'MW'
initials('Vishnu M', { max: 1 }); // 'V'

Edge cases — with the default max: 2 it takes the first and last parts, skipping middle names; max: 1 takes the first. Splits on whitespace, ., _ and -; leading punctuation is stripped; punctuation-only or blank input returns ''. Unicode-safe (Ángel Ñuñez → ÁÑ).

---

mask(value, options?)

function mask(
  value: string,
  options?: {
    visible?: number; // default 4
    from?: 'start' | 'end'; // default 'end'
    maskChar?: string; // default '*'
    maskLength?: number; // fixed mask width
  },
): string;

mask('9876543210', { visible: 4 }); // '******3210'
mask('4111111111111111', { visible: 4, maskLength: 4 }); // '****1111'
mask('secret', { visible: 2, from: 'start' }); // 'se****'

Edge cases — when the value is shorter than visible, everything is masked rather than leaking the value. Empty or non-string input returns ''. This is a display helper, not redaction: mask on the server before logging.

---

stripHtml / stripWhitespace / removeSpaces / normalizeWhitespace

stripHtml('<p>Hello <b>world</b></p>'); // 'Hello world'
stripHtml('<script>alert(1)</script>Safe'); // 'Safe'
stripWhitespace(' 4111 1111 1111 1111 '); // '4111111111111111'
removeSpaces('a b\tc\nd'); // 'ab\tc\nd'
normalizeWhitespace('  Hello \n\t world  '); // 'Hello world'

Edge cases — stripHtml removes <script>/<style> with their contents and drops HTML comments, but it is not a sanitiser: never feed its output back into innerHTML. stripWhitespace removes every unicode space; removeSpaces removes only ASCII spaces, keeping line structure; normalizeWhitespace collapses runs to one space and trims — the fix for text pasted from Word, Slack or a PDF.

---

escapeHtml(value) / unescapeHtml(value)

escapeHtml('<script>alert("x")</script>'); // '&lt;script&gt;alert(&quot;x&quot;)&lt;/script&gt;'
unescapeHtml('caf&#233;'); // 'café'

Edge cases — escapeHtml covers &, <, >, " and '. unescapeHtml also decodes numeric entities (&, &) and leaves unknown entities untouched.   decodes to a real non-breaking space (U+00A0) — follow with normalizeWhitespace if you want it collapsed. The pair round-trips.

---

isEmail(value) / isUrl(value, options?)

function isEmail(value: unknown): boolean;
function isUrl(
  value: unknown,
  options?: { protocols?: string[]; allowRelative?: boolean; requireTld?: boolean },
): boolean;

isEmail('vishnu+tag@example.co.in'); // true
isEmail('user@localhost'); // false — no TLD
isUrl('https://example.com/path?q=1'); // true
isUrl('javascript:alert(1)'); // false
isUrl('example.com', { allowRelative: true }); // true

Edge cases — isEmail is pragmatic, not RFC 5322: it accepts plus-addressing and unicode locals, rejects spaces, consecutive dots, over-long locals (>64) and over-long addresses (>254), and trims first. isUrl uses the platform URL parser, defaults to http/https only (so javascript: and data: are rejected), and validates the hostname. Validation is not verification.

---

extractNumbers(value) / extractEmails(value)

extractNumbers('Order #123 total ₹1,299.50, -5 items'); // [123, 1299.5, -5]
extractEmails('Mail a@x.com or B@X.COM, also a@x.com'); // ['a@x.com', 'b@x.com']

Edge cases — extractNumbers understands thousands separators, decimals and negatives. extractEmails strips trailing punctuation, lower-cases, de-duplicates and validates each match with isEmail. Both return [] for non-strings and for no matches.