9314JIUsi
Developer tools Β· Unicode

Unicode Converter

Text, \uXXXX escapes, HTML entities and U+ code points - both directions, surrogate pairs done right.

TYPE DIAL UTF-16 surrogate-aware Β· character by character

The escape form you meet in JS / Java / Python string literals; emoji come out as surrogate pairs (like \\ud83d\\ude00).

Plain text
Escaped / \uXXXX

Your text is converted inside this device's browser only - nothing is uploaded, nothing is collected, and it works offline. The style preference lives in this browser's localStorage; switching devices or clearing site data loses it. "Clear my data" in the footer removes it.

First characters mapped (character β†’ hexadecimal code point):

Three steps

  1. Pasteplain text or escape sequences both work - the tool picks the direction: \\u, &# or U+ prefixes trigger decoding.
  2. Pick a styletap the notation you need; emoji render as surrogate pairs in escape mode, merged as U+1F600 in code point mode.
  3. Verifythe character grid below maps each glyph; hit Copy to take the result, or the round button to convert back and prove the round trip.

Facts and limits

InputAny UTF-8 text; or escape sequences in \uXXXX / &#NNNN; / &#xHHHH; / U+XXXX form (mixed freely); up to 1MB per run
OutputThree styles: \uXXXX escapes (surrogate pairs) / &#...; decimal entities (merged code points) / U+XXXX code points (4-6 digits)
Key parametersDefault style is \uXXXX; ?style=entity or ?style=codepoint links straight to it; direction auto-detected (majority of escapes means decode)
ProcessingBrowser-local JavaScript, no upload, no account; style preference in localStorage (nothing written on first load)
Limits1MB input per run (larger input is rejected with a message); the character grid shows the first 12 characters only
Known edgesNo named HTML entities (  and friends need an HTML escape tool); grid rendering depends on system fonts - a missing glyph shows as a box but the converted value is still correct

FAQ

How do I turn \u4f60\u597d back into readable text?

Paste it straight into the left pane. The tool detects that the input is mostly \u escapes and flips to decode direction automatically - line breaks are fine, each line is handled on its own. After converting, hit the round swap button to convert back and eyeball the round trip, character by character.

Why does one emoji become two escapes like \ud83d\ude00?

πŸ˜€ sits at code point U+1F600, above the 0xFFFF ceiling a single UTF-16 value can hold, so it is stored as a surrogate pair: \ud83d is the high surrogate, \ude00 the low one, and only together do they make that one emoji. This is exactly where naive converters break - they emit half a pair and the result pastes as a tofu box elsewhere. All three styles here handle it correctly: escape mode outputs the pair as-is, code point mode merges it into U+1F600.

What is the difference between 😀 and \ud83d\ude00?

Same character, three notations for three consumers. 😀 is a decimal HTML entity for HTML parsers; \ud83d\ude00 is the UTF-16 escape form used in JS/Java/Python string literals; U+1F600 is the raw code point notation humans and spec documents use. Pick whichever style you need - the output follows the selected button, and the swap button converts back and forth between them.

Does it work for rare scripts and historic characters?

Yes, as long as your system has a font with the glyph. Characters in planes above the BMP (code points above 0xFFFF - rare CJK extensions, historic scripts, symbols) go through the same surrogate-pair logic as emoji. The only way you would "see nothing" is a missing font rendering a box - that is a display issue, not a conversion error: copy the result elsewhere or switch fonts and the character is there.

Is my text uploaded anywhere?

No. All conversion happens inside this device's browser - no upload endpoint, works offline. Only the output style preference lives in localStorage; it does not follow you to another device, and "Clear my data" in the footer removes it.

Related tools