> ## Documentation Index
> Fetch the complete documentation index at: https://persian-tools.usestrict.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Arabic Detection

> Check whether a string is in the Arabic script or contains Arabic characters.

Detect Arabic script in input. Useful for routing multilingual content to the right NLP pipeline or separating Arabic from Persian (Farsi) input.

## Functions

### `isArabic(str, trimPattern?)`

Returns `true` when **all** characters of `str` (after trimming) are in the Arabic Unicode block **and** match contextual Arabic letterforms.

```ts theme={null}
import { isArabic } from "@persian-tools/persian-tools";

isArabic("السلام عليكم"); // true
isArabic("Hello مرحبا"); // false
isArabic(""); // false
```

#### Parameters

| Name          | Type     | Default          | Description                                |
| ------------- | -------- | ---------------- | ------------------------------------------ |
| `str`         | `string` | —                | text to test                               |
| `trimPattern` | `RegExp` | `/["'-+()\s.]/g` | characters stripped before the script test |

### `hasArabic(str)`

Returns `true` when `str` contains **at least one** Arabic character.

```ts theme={null}
hasArabic("Hello مرحبا"); // true
hasArabic("Hello"); // false
```

## Persian vs. Arabic

Persian uses the Arabic script **plus** `پ چ ژ گ` **plus** different code points for ya/kaf. Two practical consequences:

1. A pure Persian string typed on a Persian keyboard **also passes `isArabic`** — the characters all fall in U+0600–U+06FF.
2. A pure Arabic string **fails `isPersian`** (default mode) only because of trim/punctuation rules; in complex mode the verdict can flip.

If you need "Arabic but **not** Persian", combine the checks:

```ts theme={null}
import { isArabic, hasPersian } from "@persian-tools/persian-tools";

const isPureArabic = (s: string) => isArabic(s) && !hasPersian(s);

isPureArabic("السلام"); // true
isPureArabic("علی"); // false — Persian-specific ya present
```

## Edge cases

| Input              | `isArabic(...)` | Why                                             |
| ------------------ | --------------- | ----------------------------------------------- |
| `""`               | `false`         | Nothing to match                                |
| `"السلام"`         | `true`          | All chars in U+0600–U+06FF                      |
| `"عليكم 123"`      | `false`         | Latin digits are not in the Arabic block        |
| `"عليكم ١٢٣"`      | `true`          | Arabic-Indic digits **are** in the Arabic block |
| `"سلام"` (Persian) | `true`          | Persian letters live in the same block          |

## Pitfalls

* `isArabic` is **not** the negation of `isPersian`. Use the combined check shown above when you need a clean partition.
* Arabic-Indic digits (`٠-٩`) pass; Persian digits (`۰-۹`) pass too. For digit-level distinction, use `arDigitsRegex` and `faDigitsRegex` from the [Digits](../numbers/digits) module.

## Source

`src/modules/isArabic/index.ts` · Tests: `test/isArabic.spec.ts`