Unicode Escape Decode Online

What a Unicode escape is

A Unicode escape writes a character as a source-code sequence instead of the glyph. The character 你 is U+4F60, so a JavaScript or JSON string can spell it \u4f60. That is not a byte encoding. UTF-8 is how the character is stored as bytes; \uXXXX is how a source file names a code unit.

ECMA-262 defines two JavaScript forms:

• \uXXXX is four hex digits and one UTF-16 code unit. A BMP character such as a common Han character needs only one sequence.
• \u{...} is a code-point escape. The hex inside the braces is the scalar value, so an emoji does not have to be split. See MDN lexical grammar.

JSON only requires the four-digit form (RFC 8259 §7). Python source adds the eight-digit \UXXXXXXXX form (Python lexical analysis). \xHH names a single value in U+0000–U+00FF.

Nearby tools

• Decimal code points: Unicode code point.
• UTF-8 bytes: UTF-8 encoder.
• Percent-encoding: URL encoder.

This page only converts text to escape sequences and back. It runs in the browser and does not upload the text.

Encode text

Type Chinese, ASCII, or emoji on the left. The right side follows the selected format. The default is JavaScript / JSON: Han characters become \uXXXX, and emoji become a surrogate pair.

For 你好:

• JavaScript / JSON: \u4f60\u597d
• ES6 code point: \u{4f60}\u{597d}
• Python: \u4f60\u597d as well; characters outside the BMP use eight-digit \U

Printable ASCII stays literal, so the letters in hello, 山月 remain readable. Turn on “Escape printable ASCII too” when the whole string should be escapes. A backslash is still written as \\ when ASCII is left literal, so decoding does not treat the following characters as an escape.

“Use \xHH when escaping U+00FF and below” only changes characters that are already being escaped. é can become \xe9.

Decode escapes

Paste escapes on the right. The left side shows the text. Forms can be mixed:

• \u4f60 and \u4F60 are both 你
• \u{1F600}, \U0001F600, and \uD83D\uDE00 are all 😀
• \n, \t, and \\ follow ordinary string rules

An incomplete sequence such as \u12, or a code point above U+10FFFF, shows an error and leaves the last successful text in place.

A JSON string copied with its quotes keeps those quotes. A doubled backslash (\\u4f60) is one extra escape layer: the first decode yields \u4f60, and a second decode yields 你.

API

These snippets match the three formats on the page. Use codePointAt and fromCodePoint so emoji stay intact. See MDN String.fromCodePoint.

JavaScript

JSON.stringify does not escape Han characters by default. RFC 8259 allows raw Unicode in strings. To force \u escapes, walk code points and emit UTF-16 units:

function toJsonUnicodeEscape(text) {
  return Array.from(text, (ch) => {
    const cp = ch.codePointAt(0)
    if (cp <= 0xffff) return '\\u' + cp.toString(16).padStart(4, '0')
    const rest = cp - 0x10000
    const hi = 0xd800 + (rest >> 10)
    const lo = 0xdc00 + (rest & 0x3ff)
    return (
      '\\u' +
      hi.toString(16).padStart(4, '0') +
      '\\u' +
      lo.toString(16).padStart(4, '0')
    )
  }).join('')
}

toJsonUnicodeEscape('你') // \u4f60
toJsonUnicodeEscape('😀') // \ud83d\ude00
String.fromCodePoint(0x1f600) // 😀

An ES6 code-point escape is written \u{1F600} in source. The grammar is in ECMA-262.

Python

'你好'.encode('unicode_escape')
# b'\u4f60\u597d'

'😀'.encode('unicode_escape')
# b'\U0001f600'

b'\\u4f60\\u597d'.decode('unicode_escape')
# 你好

unicode_escape leaves ASCII as-is, often uses \xHH through U+00FF, and uses \U for supplementary characters. The page's Python format plus the \xHH option is the closest match. The full list is in the Python string literal docs.

FAQ

How is this different from the UTF-8 tool?

\u4f60 names the code point U+4F60. The UTF-8 bytes of 你 are E4 BD A0, not 4F 60. Use the UTF-8 encoder for bytes, the code point tool for decimal scalars, and the URL encoder for %E4%BD%A0.

In JSON, \u escapes a UTF-16 code unit. That rule is in RFC 8259, and it is not a UTF-8 encoding.

Why does an emoji become two \u sequences?

JavaScript and JSON \uXXXX hold one 16-bit code unit. 😀 is U+1F600, outside the BMP, so the JSON form is the surrogate pair \uD83D\uDE00. The same fact makes '😀'.length === 2 in JavaScript (MDN strings).

The ES6 format keeps one escape, \u{1F600}. The Python format uses \U0001F600 (Python escapes).

Is the text uploaded?

No. Encoding and decoding run in the browser. The text is not sent to a server.

When do \xHH and \UXXXXXXXX show up?

\xHH is a value from 0 to 255, often used for Latin-1 such as \xe9 for é. \UXXXXXXXX is the eight-digit code-point escape in Python and some C# source, used for characters outside the BMP. Decoding accepts both. Encoding emits them when you pick the matching option.