Detection

import { isUrdu, urduRatio, hasUrduSpecificLetters } from "urdu-text-utils";

isUrdu("آپ کیسے ہیں؟"); // true
isUrdu("hello world"); // false

It is a ratio, not a contains-check

isUrdu("The word پاکستان appears in this English sentence");
// false — only a small share of the letters are Urdu

isUrdu("The word پاکستان appears", { threshold: 0.1 });
// true

The default threshold is 0.5: more than half of the letters must be Arabic-script. Digits, punctuation, spaces and emoji are ignored entirely, so "12345 !!!" is not Urdu and not English either — it returns false.

isUrdu(text, {
  threshold: 0.5, // share of letters that must be Arabic-script
  minLetters: 1,  // require at least this many Arabic-script letters
});

Raw ratio

urduRatio("پاکستان Pakistan"); // 0.47
urduRatio("پاکستان");          // 1
urduRatio("12345");            // 0 — no letters at all

Useful for ranking mixed content, or for a "this post looks mostly English" warning in a CMS.

Urdu versus Arabic

isUrdu measures script, not language. Arabic, Persian, Pashto, Sindhi and Urdu share the Arabic script, so no ratio can separate them.

What can separate them is the alphabet itself — Urdu has letters Arabic does not:

hasUrduSpecificLetters("لڑکی");      // true  — ڑ
hasUrduSpecificLetters("پاکستان");   // true  — پ, ک
hasUrduSpecificLetters("كتاب مدرسة"); // false — pure Arabic letters

The signals are ٹ ڈ ڑ ں ے ۓ ہ ھ ک گ چ پ ژ ی. Note this is shared with Persian for some letters (پ چ گ ژ), so a true means "not Arabic" rather than "definitely Urdu". Combining it with a short stopword check (ہے, اور, کی) gets you the rest of the way if you need it.

Try it

normalizeUrdu(text) stripDiacritics

Input

كيا حال ہے

Output

کیا حال ہے