regexUnicodeCodepointEscapes

Reports surrogate pair escapes in regular expressions that can be replaced with Unicode codepoint escapes.

✅ This rule is included in the ts stylistic presets.

Reports surrogate pair escapes in regular expressions that can be replaced with Unicode codepoint escapes. When the u or v flag is present, surrogate pairs like \uD83D\uDE00 can be written as \u{1F600}. Unicode codepoint escapes are clearer and more maintainable.

Examples

Surrogate Pair in Regex Literal

❌ Incorrect

const emoji = /\uD83D\uDE00/u;

✅ Correct

const emoji = /\u{1F600}/u;

Surrogate Pair in Character Class

❌ Incorrect

const emoji = /[\uD83D\uDE00]/u;

✅ Correct

const emoji = /[\u{1F600}]/u;

Surrogate Pair with v Flag

❌ Incorrect

const emoji = /\uD83D\uDE00/v;

✅ Correct

const emoji = /\u{1F600}/v;

Surrogate Pair in RegExp Constructor

❌ Incorrect

const emoji = new RegExp("\\uD83D\\uDE00", "u");

✅ Correct

const emoji = new RegExp("\\u{1F600}", "u");

Options

This rule is not configurable.

When Not To Use It

If you need to support environments that do not support Unicode codepoint escapes, or if you prefer surrogate pairs for consistency with existing code, you might need to disable this rule. Without the u or v flag, surrogate pairs are required for astral characters.

Further Reading

• MDN: Unicode character class escape

Equivalents in Other Linters

• ESLint: regexp/prefer-unicode-codepoint-escapes