Unicode Property Escapes — \p{Letter}, \p{Script=Hiragana}

Unicode Property Escapes

Unicode property escapes (ES2018) let you match characters by Unicode metadata rather than enumerating code points. They require the u flag.

Basic Syntax

\p{PropertyName}      // matches characters with this property
\P{PropertyName}      // matches characters WITHOUT this property
\p{Property=Value}    // for properties that take values

Common Properties

Escape	Matches
\p{Letter} or \p{L}	Any letter (any script)
\p{Number} or \p{N}	Any digit or numeral
\p{White_Space}	Any whitespace
\p{Punctuation}	Any punctuation
\p{Emoji}	Any emoji code point
\p{Extended_Pictographic}	Emoji-like (broader than \p{Emoji})
\p{Lowercase}	Lowercase letters
\p{Uppercase}	Uppercase letters

Script Property

\p{Script=Hiragana}    // matches ひらがな
\p{Script=Katakana}    // matches カタカナ
\p{Script=Han}         // matches 漢字 (Chinese characters / Kanji)
\p{Script=Latin}       // matches abc, é
\p{Script=Cyrillic}    // matches Привет

Tested Examples

Pattern	Input	Match
\p{L}+	"hello world 123"	hello, world
\p{Script=Hiragana}+	"今日は"	は
\p{Script=Han}+	"今日は"	今日
\p{Emoji}+	"hi 👋✨"	👋✨
\P{ASCII}+	"hello 世界"	世界

Validating Multilingual Names

A name field that should accept letters in any script:

^[\p{L}\p{M}\s'-]+$

\p{M} covers combining marks (accents and diacritics).

Splitting Mixed-Script Text

"hello世界こんにちは".match(/\p{L}+/gu)
// ["hello", "世界こんにちは"]

To separate scripts:

"hello世界".match(/\p{Script=Latin}+|\p{Script=Han}+/gu)
// ["hello", "世界"]

Browser Support

Supported in all modern engines. Node 10+, Chrome 64+, Safari 11.1+, Firefox 78+. For older targets, transpile with Babel or fall back to explicit character ranges.

Why It Matters

Without property escapes, internationalized validation requires hand-maintained character lists, which inevitably drift behind Unicode updates. Property escapes track the Unicode standard automatically.